Loading...
The URL can be used to link to this page
Your browser does not support the video tag.
Home
My WebLink
About
Assessment of Energy Technologies 1983
Report to Alaska Power Authority The Department of Commerce and Economic Development LIBRARY COPY State of Alaska —— Assessment of Energy Technologies Report to The Department of Commerce and Economic Development June 1983 Assessment of Energy Technologies AX Arthur D. Little, Inc. Reference 88376 DISCLAIMER This study was prepared as an account of work ‘sponsored by the State of Alaska. Neither the State of Alaska; nor the State of Alaska's Department of Commerce and Economic Development; nor any of their employees; nor any of their contractors, subcontractors; or their employees; makes any warranty, expressed or implied, or assumes any legal liability or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. | 1, - TABLE OF CONTENTS is | . | Page ©. 1.0 Introduction. .. 2... ee ee ee ee ew 1 if ~ 1.1 Background . . ... 2. 2. 2 © © + © © © © © © we ee ew 1 & 1.2 Purpose and Scope. ... +. + © « «© © «© © © «© © o « i Pe 1.3 Energy Technology Sectors. . . . «+ + + «© «© «© © « 2 1 1.4 Selection of Energy Technologies .... . +. +, ++ - 3 1.5 Format of Technology Discussions . ..... + +s -s 3 1.6 Costs. 2. 2. 2 2 «6 6 © ee we ee ee ee we ew ee 7 2.0 Overview of Central Energy Systems. . .... ++ +e ees 13 7 2.1 Introduction and Summary ..... «2 « «© «© «© «© «© « 13 2.2 Fuel Processing and Conversion . .... 4+ + « « « © 15 c 2.3 Thermal Energy . . . . 2 © «© © « © © © © © © @ wo 20 | 2.4 Electric Power... . a 30 2.5 Central Electric Power Grid Systems. . ... se «-s 39 | 2.6 Large-Scale Energy Transport ......-.-+-e-e«e-s 41 3.0 Oil and Gas Fired Boilers ..... Le o Ce ees 43 - 3.1 Introduction and Summary .........24 e+ +e. 438 i 3.2 Description of Technology. .......4.-+ +e «=e 45 3.3 Economic Implications. . ......-.. «+ « « ee @ 55 \ 3.4 Impact . . . 2. 2 © © © © © © © © © © © ee we 60 a 4.0 Conventional Solid Fuel Fired Boilers ........+. © 63 a 4.1 Introduction and Summary .......:. 2. «. . 63 i: 4.2 Description of the Technology. . ......:.... 67 4.3 Economic Implications. i 81 4.4 Impact ... .%. «© «© © «© © © © © © © we ew ew ew ew 88 5.0 Fluidized Bed Combustion. .....- +--+ eee ees 98 l. 5.1 Introduction and Summary ....... «© «+ «+ «+ « «© « 93 . 5.2 Description of the Technology. ........-+e-e 96 ? 5.3 .Economic Implications. . . ... «6 « . 2 «© © » » » © 109 4 5.4 Impact... . 2. 2 2 2 6 © © © © © © we ow ew eh ew wl he) «6118 i . ons 6.0 Combustion Turbine Combined Cycles... ~~... « «+. 6 « « 1238 .! 6.1 Introduction and Summary ........ . - + 123 ‘ 6.2° Description of Technology. . ...... +... 6. « 125 \ 6.3 Economic Implications. soe ee ew ew we we ww ew 185 { 6.4 Impact . . . 2. 2 2 2 6 © ee ew oe we ew we ew ww we hw he) «140 iii 7.0 8.0 9.0 10.0 Transportation Fuels from Natural Gas Introduction and Summary . Description of Technology. ... Economic Implications. . ° 2 e Impact . . 2. 2. © «© ©» © © © © Large-Scale Solid Fuel Gasification . Coal Introduction and Summary. . Description of the Technology Economic Implications. ... Impact ........-s«e- Slurry Pipelines ........ Introduction and Summary . Description of Technology. Economic Implications. ... Impact . . . 2. 6. 2. 2 © we ew ew Overview of Dispersed Energy Systems. 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 Introduction. .......e2-. Diesel Engines. .... rr Rankine Bottoming Cycle Engines Small-Scale Gasification. . . Fuel Cells. . .....-e-. Stirling Engines. ..... Small-Scale Hydro ..... Wind Power Systems... . Photovoltaic Power Units. 2 0 © © © 11.0 Diesel Engines. ........4+s.6-s 12.0 11.1 11.2 11.3 11.4 Introduction and Summary. Description of Technology... Economic Implications 2. Impact. . . . . © © © © © © «© Photovoltaics ses ees seeeas 12.1 12.2 12.3 12.4 Introduction and Summary. . Technology Description. .. Economic Implications .... Impact. . . . 2. 2. e+ eee iv eo o o 8 © © © © © 8 ee we oo @ o 8 © 2 © © o ee © © © © ee eo ew © 6© © 8 © 0 © © o 6 0 o # © © © © © 8 ee o@ 143 143 147 161 163 169 169 171 185 193 197 197 202 211 215 219 219 223 227 229 229 233 235 237 239 243 243 245 261 267 271 271 273 284 289 a 13.0 7 14.0 1 4, : 15.0 is © i: 16.0 {; UL fe 17.0 M ihe y ht 18.0 ab . . ' a 19.0 ‘17.3 Economic Implications . . Rankine Bottoming Cycle Engines . 13.1 Introduction and Summary. . 13.2 Description of Technology . 13.3 Economic Implications . : 13.4 Impact. . . 2... 6 2 «© © @ Stirling Engines. .....+.-+-. 14.1 Introduction and Summary. . 14.2 Description of Technology . 14.3 Economic Implications .. . 13.4 Impact. .......2.-s Small-Scale Hydro Power . Lee 15.1 Introduction and Summary. 15.2 Description of Technology 15.3 Economic Implications . . 15.4 Impact. . . 2... ee 2 Wind. . 2. 2. 2 es 6 we ew ww ew ww 16.1 Introduction and Summary. 16.2 Description .of Technology 16.3 Economic Implications . . 16.4 Impact. .......6-. Small-Seale Gasification... . Introduction and Summary. Description of Technology 17.1 17.2 17.4 Impact. . 2. . 6 © 2 «© « « Fuel Cells. . . ... 1. 2. 2 © « © 18.1 Introduction and Summary... 18.2 Description of Technology . 18.3 Economic Implications... 18.4 Impact. . ..... 2. 2 «© « oe 8 co 8 oe ee eo ee © © ee ee ee oe ee ee ee © ee @ © Overview of Residential and Commercial End-Use Systems. 19.1 Introduction. . . ee 19.2 Comfort and Space Heating . 19.3 Water Heating ........ 19.4 Refrigeration and Cooking . 19.5 Energy Storage. . . see 19.6 Controls and Energy Manageme 19.7 Lighting and Appliances . . nt . . . . . ° ° ee 8 © @ eo 8 ee we ow ee ee ee ee, 295 295 297 307 309 313 313 315 340 344 349 345 350 363 367 371 371 375 391 393 403 403 406 418 424 427 427 429 448 455 459 459 460 471 473 474 474 475 20.0 21.0 22.0 23.0 24.0 25.0 26.0 Heat Pumps. ........-e. 20.1 Introduction and Summary. 20.2 Description of Technology 20.3 Economic-Implications . . 20.4 Impact. . ..... +. © Condensing Furnaces and Boilers 21.2 Introduction and Summary. ' 21.2 Description of Technology 21.3 21.4 Economic Implications . . Impact. 2. 6. 2. 6 ee eee District Heating. ....... 22.1 Introduction and Summary. 22.2 Description of Technology 22.3 22.4 Economic Implications . . Impact. . . .. 2. 2. 2 ee Energy Storage. ....... 23.1 Introduction and Summary. 23.2 Description of Technology 23.3 Economic Implications . . 23.4 Impact. ........ Solar Space Heating - Passive . 24.1 Introduction and Summary. 24.2 Description of Technology 24.3 Economic Implications . . 24.4 Impact. . ..... ee Solid Fuel Fired Stoves... . 25.1 Introduction and Summary. 25.2 Description of Technology 25.3 Economic Implications . . 25.4 Impact. ...... 2... Solar Domestic Hot Water. ... 26.1 Introduction and Summary. 26.2 Description of Technology 26.3 Economic Implications . . 26.4 Impact.-......-. oe ee eo eo 8 477 477 481 501 505 509 509 511 519 524 527 527 528 538 544 549 549 553 563 568 571 571 573 585 589 591 591 593 607 611 615 615 617 638 645 27.0 28.0 29.0 30.0 31.0 Transportation Ed-Use Systems - Overview. 27.1 27.2 27.3 27.4 27.5 27.6 Aircraft Engine Modification. . 28.1 Introduction and Summary’. 28.2 Description of Technology 28.3 Economic Implications . . 28.4 Impact. . . . 2. 2.0 e « « Railroad Electrification. ... 29.1 Introduction and Summary. 29.2 Description of Technology 29.3 Economic Implications . . 29.4 Impact. ......6e. Alternate Fueled Highway Vehicles 30.1 Introduction and Summary. 30.2 Description of Technology 30.3 Economic Implications . . 30.4 Impact. ......22. Slow Steaming of Marine Engines 31.1 Introduction and Summary. 31.2 Description of Technology 31.3 Economic Implications . . 31.4 Impact. ...... 2... © e@ © «© eo e «@ oe © © 0 ee © 0 © @ 2 © © © o 8 © oe ee 2 2 © @ coe ee 2 0 ee ee oe oe ee Introduction - Transportation in Alaska . Summary of Energy Technologies for Highway Sector Summary of Energy Technologies for Aviation Sector Summary of Energy Technologies for Rail Sector. Summary of Energy Technologies for Marine Sector. Summary of Potential Improvements in Alaska's Tran portation Infrastructure as an Energy Technology . e 0 @ ee ee 2 © © oe ee eo 8 ee eo 8 © s- 649 649 651 661 664 -669 670 677 677 681 689 692 697 _ 697 700 713 720 725 725 729 753 759 765 765 768 775 782 1.0 INTRODUCTION 1.1 Background Unlike the aggregate of the United States, Alaska produces far more energy than it consumes. It exports about 1.6 million barrels per day of North Slope crude oil to the other states and about 25 billion cubic feet per year of liquefied natural gas to Japan. In contrast, Alaska's own consumption of energy amounts to the equivalent of about 10% of these two export streams alone. The state also has other significant, but largely untapped, resources of peat, biomass, coal, uranium, geothermal, hydro, tidal, wind, and solar energy. — Also unlike. the Lower 48, Alaska has a sparse, geographically dispersed population, vast total land area, harsh climate, and frequently rugged terrain. These factors lead to a variety of problems which together make Alaska's energy situation unique. For example, there is very little interconnection of electricity networks except in the more urban areas. of Anchorage, Fairbanks, and the Southeast. This, coupled with the relatively low summer electricity. demand, results in very low capacity utilization’ of generation equipment. Further, the distribution of fossil fuels in Alaska for both power systems and thermal needs is a costly process outside the population centers. Fuel costs in the Bush can often be. as high as $10 per million Btu; and the cost of power generated by small, inefficient diesel power units, in excess of $0.50 per kWh. The distribution system is often a torturous combination of barge and air, which in itself is a very energy-intensive operation. These observations highlight the point. that energy use and costs in Alaska often bear little resemblance to those in the Lower 48. As a result, judgments made on the applicability and economics of energy systems based on experience in the Lower 48 are not necessarily valid in Alaska. 1.2 Purpose and Scope The purpose of this assessment of energy technologies is to provide a background document which may be useful to Alaskans in: making energy related decisions--at the government level, at the business level, and at the consumer level. It is recognized that, by itself, the document cannot answer all questions or provide site-specific analysis for any given location.in Alaska. Rather, the intent is to provide the reader with the perspective from which he may draw appropriate initial judgments as to the applicability of energy technologies to various situations. From this preliminary review, the reader may then decide to access the references cited or obtain’ other data to make the best possible informed decision. This report comprises thirty-one chapters, an introduction, four technology overviews, and _ twenty-six individual technology discussions. The introduction presents the overall basis which guided the preparation of the other chapters. The four overviews, categorized into four energy technology sectors, provide background on the alternative technologies which are usually considered, the relative advantages of each leading technology, the bases on which decisions are made, and technological and economic trends which may alter the sector in the future. The individual technology discussions follow a consistent format to describe how the technology works, practical constraints, environmental issues, commercial status, costs, the potential impact on Alaska, and future trends. 1.3 Energy Technology Sectors The overall picture has been divided into four sectors: e central energy systems; e dispersed energy systems; e residential/commercial end-use systems; and e transportation end-use systems. The division between central and dispersed systems acknowledges the existence of two very diverse energy environments in Alaska. Central energy systems, which exist primarily in the regions of high population, take advantage of economies of scale to produce electricity and other energy forms. Dispersed systems, on the other hand, serve individuals and small groups of individuals in rural villages where the economies of scale are not possible. In principle, the resources which serve the needs of central and dispersed energy systems are selected from the same alternatives including the full range of renewables and nonrenewables. In practice, the choices are affected by local resource availability, existing infrastructure, weather, economics, and many other factors. The available technology for residential and commercial systems is largely the same in both rural and urban areas. Again, choices may be importantly affected by needs of the user and the relative economics which can vary dramatically from one locale to another, particularly in Alaska. . . : Highway and rail transportation in Alaska are largely limited to the populated areas of the state. However, air and water transport are the major linkages among all the residents. This sector represents nearly half of the energy use in the state. Yet the choices of technology are severely limited by the fact that most equipment for this sector is produced out of state and designed to meet the requirements of mass markets whose needs are quite different from many of those in Alaska. 1.4 Selection of Energy Technologies The energy technologies assessed in this report were selected based on several criteria: : e@ cost e technical status e reliability e simplicity of operation e impact on Alaska's energy picture e environmental benevolence e socioeconomtic benefits -e@ performance (efficiency) In addition, there was consideration as to whether a proper mix of technologies represented both large-scale and small-scale systems. Also, certain .technologies areas, which have already been _ well covered and widely publicized in Alaska (e.g., large-scale hydro), were omitted from consideration. The technologies selected. are listed in Table 1.1. Some of the "technologies" selected are more aggregations of related technologies than they are individual technologies per se. For example, alternate fueled vehicles include treatment of compressed natural gas, liquefied natural gas, ethanol, and methanol. Other of the "technologies" selected are techniques or practices applied to achieve an energy related objective. for example, passive solar heating involves proper design and construction of a new building or adaptation of an existing building to conserve fuel by capturing and retaining solar radiation. , 1.5 Format of Technology Discussions Table 1.2 shows the abbreviated outline which is common to all twenty-six technology chapters. Therefore, in overall structure, each chapter. has a reasonably parallel treatment of the issues. However, within each subsection, this was not practical in all instances, since the issues differ substantially among technologies -- strict parallelism is, of course, possible, but it would lead to considerable redundant discussion and dilution of the major thrust of each chapter. : TABLE 1.1 ENERGY TECHNOLOGIES ADDRESSED IN THIS REPORT Central Energy Systems Oil and Gas-Fired Boilers Conventional Solid Fuel Fired Boilers Fluidized Bed Combustion Combustion Turbine Combined Cycles Transportation Fuels from Natural Gas Large-Scale Solid Fuel Gasification Coal Slurry Pipelines Dispersed Energy Systems Diesel Engines Photovoltaics Rankine Bottoming Cycle Engines Stirling Small-Scale Hydro Wind Small-Scale Gasification Fuel Cells Residential and Commercial End-Use Systems Transportation End-Use Systems Heat Pumps ~ . Condensing Furnaces and Boilers District Heating Energy Storage Solar Space Heating-Passive Solid Fuel Fired Stoves Solar Domestic Hot Water Aircraft Engine Modification Railroad Electrification Alternate Fuel Highway Vehicles Slow Steaming of Marine Vessels ft ae) eae 1.0 2.0 3.0 4.0 TABLE 1.2 OUTLINE OF TECHNOLOGY DISCUSSION Introduction and Summary 1,1 Technical Overview 1.2 Alaskan Perspective 1.3 Significance of the Technology Description of the Technology Operating Principles Technical Characteristics Environmental Issues Commercial Status NNNN see Fone Economic Implications 3.1 Costs 3.2 Socioeconomic Factors Impact 4.1 Effect on Overall Energy Supply and Use 4.2 Future Trends Anchorage (Residential, Commercial Transportation End Use) Railbelt Outside Anchorage (Central Energy Systems) Rural with Water Access (Dispersed Energy Systems) Small-Scale Hydro (Rural) TABLE 1.3 CAPITAL cost MULTIPLIERS (Alaska Cost + Lower 48 Cost) Equipment Direct and Materials Site Labor 1.15 1.85 1.03 1.82 1.20 2.74 1.24 3.31 Indirect Site Labor 1.17 1.38 1.66 1.67 Engineering and Home Office Costs 1.35 1.35 1.47 1.51 Senet Po) 1.6 Costs One of the motivations for preparing this report is to present costs on a consistent basis. This means using both a common calculation basis and a common presentation format. Many technologies have not been studied in the Alaska context, and direct cost estimates are not available. For this reason, a set of cost multipliers was prepared to translate various categories of cost which may be known for a typical Lower 48 location to the relevant regions of Alaska. These multipliers, shown in Table 1.3, are implicit in all Alaskan costs shown in the technology chapters, even in the majority of cases where they are not mentioned at all. Where possible, of course, Alaska-specific costs have been used; where not possible, the factors assisted in making the available costs more applicable to Alaska. The multipliers in Table 1.3 are specific not only to the regions shown but also to the categories of equipment which are most applicable to those regions. As shown in the table, this report develop develops. representative costs for end-use systems in Anchorage, control energy systems in the Railbelt outside Anchorage, and dispersed energy systems in the Bush. Thus, the cost factors are not universal, but limited to certain technology categories. For example, it may appear unusual that equipment in Anchorage would reflect a multiplier of 1.15 while in the Railbelt outside Anchorage the multiplier would be 1.03. The former reflects the fact that the "Anchorage-based" technologies. would be based on __ standard warehoused equipment: and would include the cost of that warehousing and handling. However, the central systems outside Anchorage would be special-ordered from the Lower 48 and thus would only reflect an incremental transportation cost. ‘ The costs of owning and operating energy conversion equipment fall generically into two categories: capital investment (first cost) and operating costs (recurring costs). To annualize the first cost, we have adopted the "capital charge factor" approach. By multiplying the first cost by a capital charge factor, one obtains an equivalent annual investment component which may be added directly to the recurring costs. The total annual cost obtained in this way may be divided by the total annual output of the facility to calculate the cost per unit of output, sometimes referred to as the life cycle cost. For this.study, we have utilized a capital charge factor of 9.5% throughout.* It includes interest on debt, depreciation, federal taxes, and return on equity investment. Since the analyses are performed in constant 1982 dollars (i.e., no inflation), this rate may *This value was selected to be consistent with the midpoint value used in the 1983 Long-Term Energy Plan. TABLE 1.4 AVERAGE DELIVERED FUEL PRICES TO END USER No, 2 Distillate, $/gallon Gasoline, $/gallon Natural Gas, $/million Btu Coal, Utility, $/million Btu Coal, Residential, $/million Btu Electricity, ¢/kWh Source: 1983 Long-Term Energy Plan. Anchorage 1.31 1.37 1.70 5.00 3.9 Railbelt Outside Anchorage 1.38 1.44 1.70 1.88 5.00 9.2 Bush with © Water Access 1.80 cn e { t seem rather low when contrasted with the recent high money rates which contain a high premium for inflation. Although a universal rate of 9.5% was applied to all technologies, actual rates will vary depending on certain factors such as expected life of the technology and perceived riskiness of the investment. Therefore, the reader may wish to adjust this factor accordingly. Table 1.4 summarizes the average 1982 values of fuel delivered to _various regions in Alaska. These values have been used in the estimation of technology operating costs. Tables 1.5 and 1.6 show the formats used to display capital and operating cost estimates for each technology. Since each technology does not necessarily have each cost component associated with it, tables in certain of the chapters are somewhat abbreviated, although each follows the overall format. Certain groups of technologies are likely to be applied more frequently--although not exclusively--in certain regions of Alaska. For this reason, each technology sector focuses on one region as follows: e central energy systems - Railbelt outside Anchorage e dispersed energy systems - Bush with water access e residential/commercial end-use - Anchorage e transportation end-use - Anchorage Although the last two sectors are not limited to Anchorage in practice, most Alaskans outside the municipal area are probably reasonably familiar with translating such costs to their own situation. The bottom line of Table 1.4 is the unit (or life-cycle) cost of the energy form of interest. These costs are expressed in different units, depending on the energy form itself. For example, fuels are usually valued in dollars per conventional unit of weight or volume of the fuel, such as $/ton (coal), $/ gallon (oil), or $/cubic foot (natural gas). Thermal energy is valued in $/million Btu of either steam or space heat, and electric power in ¢/kWh. Because of such conventions, it is virtually impossible to adhere to a common, yet meaningful set of units to express the results across all technologies. The units selected in each chapter are those felt to be the most meaningful to the reader. TABLE 1.5 CAPITAL COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD CAPITAL COST Equipment and Materials Direct Labor Indirect Labor Home Office Costs Contingency TOTAL CAPITAL INVESTMENT 1982 constant dollars 10 fe. an mn. + 1 TECHNOLOGY BASIS Location Year CAPACITY Input Output OPERATING FACTOR Operating Costs VARIABLE COSTS Fuel Electricity Other TOTAL VARIABLE COSTS FIXED COSTS ' Labor Operating Maintenance TABLE 1.6 OPERATING COST SUMMARY Annual Consumption Unit Cost Maintenance Materials Taxes and Insurance Capital Charges TOTAL FIXED COSTS TOTAL ANNUAL OPERATING COSTS TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST 11 Annual Cost units /unit Despite the effort expended to adapt the cost estimates to appropriate regions of Alaska and to place them on a reasonably consistent basis, the reader is cautioned not to use generalized estimates for any purpose other than preliminary assessment. Any investment decision should first carry the effort farther to account for site-specific factors such as site conditions, space availability, infrastructure, labor availability, climate, local regulations, etc. Actual costs may vary considerably from those presented in this report. 12 2.0 OVERVIEW OF CENTRAL ENERGY SYSTEMS 2.1 Introduction and Summary Central energy systems typically are those that convert or transport energy at a scale sufficient to serve many residences and/or one or more sizable industrial enterprises. Although the technologies used can theoretically be scaled down to serve dispersed needs, central systems generally use technologies that exhibit the best economies of scale. Other benefits of central technologies are that: needs for excess or redundant capacity can. be minimized; risk and reliability can be effectively managed; and the optimal mix of technologies can be used. . Central systems can be categorized in terms of the form of energy that is produced (or transported), which may be one or a combination of: , e fuels e thermal energy (heat) e electricity . The generalized flow of energy required to produce these energy forms from renewable and nonrenewable sources is shown in Figure 2.1. In those energy conversion chains involving fuels, fuel production generally occurs as the first step. Fuels may. be sold for use in the transportation sector or exported to other markets. Or they may be converted further using other central systems to produce heat for commercial or industrial needs or to generate electric power ' through a thermomechanical cycle (heat engine) to drive the shaft of an electricity generator. The figure shows other options that impact the overall flows in other ways. These include: e nuclear reaction which .can substitute for fuel combustion needed to produce heat; , e electrochemical devices such as fuel cells which can create an - electric current flow directly from a fuel; e internal’ combustion engines, including gas turbines and -diesels, which are heat engines where combustion of. fuel is an integral part of the cycle; , e turbomechanical devices which transfer the energy of a natural flow (hydro, tidal, wind) directly to shaft power; e thermoelectrics which induce current flow directly from a heat source; and : 13 oT Non- Renewables Renewables ————d Engine Shaft Power Conversion External eS] Thermo: Fossil, Peat, ’— Biomass Fuels! Upgrading Con bustion Mechanical ; Turbo- Hydro, Wind, Fuels Heat Electricity Waste Heat Electrochemical Nuclear : Fuel J Reaction Waste Heat |) ——— Int. Comb. Mechanical Tidal Energy Geothermal, Ocean Thermat Solar Photo- Energy Voltaic a —_—— —— ee —— ; . Thermal . Electric Fuel Processing Energy Power and Conversion FIGURE 2.1 FLOW OF ENERGY, CENTRAL SYSTEMS @ photovoltaics which induce current flow directly from a light source. The remainder of this overview chapter gives a general perspective on the role of various energy technologies in, these central system applications. The emphasis is on technologies currently in use or readily available for use in Alaska; and on options currently in development which may be available soon enough to have an appreciable impact by the end of this century. The discussion is divided first into the three categories listed above: 2.2 Fuel Processing and Conversion 2.3 Thermal Energy 2.4 Electric Power The final sections deal with technologies which are employed to link energy systems: 2.5 Central Electric Power Grid Systems 2.6 Large-Scale Energy Transport Technologies which are more amenable to non-central (dispersed) than to central applications--although they can in principle be used for either--are discussed later in Chapter 10. 2.2 Fuel Processing and Conversion The function. of this energy technology category is to produce marketable fuels from naturally occurring resources. A prime example is petroleum refining, where crude oil ‘is upgraded to produce a variety of refined products. Natural gas processing is another example--corrosive, toxic, and/or diluent components are removed from the gas before it is placed in the pipeline. An alternative to these conventional fuel processing options would be to produce the required products: synthetically through a series of chemical transformations of other cheaper or more plentiful fuels. For example, natural gas can be broken down and_ subsequently rearranged to produce’ fuel-grade methanol or even liquid hydrocarbon fuels such as gasoline and heating oils. In addition, since coal, wood, or peat can also be gasified to produce an intermediate gas similar in composition to that obtained by breaking down natural gas, the same range of liquid products can be synthesized... Other approaches, known as pyrolysis and direct liquefaction, are being . developed to produce liquids from solid fuels, but generally the products of these processes would require further upgrading before being marketed for end use. 15 9T TABLE 2.1 PRODUCTION AND CONSUMPTION OF DISTILLATE PETROLEUM PRODUCTS IN ALASKA (barrels per day) . . Total Total Refinery , Chevron Mapco Tesoro Production Consumption Propane --- --—- 300 300 300 Gasoline 1,100 -— 11,000 12,100 12,000 Avaiation Gasoline - --- --- --- -——- 1,100 Jet Fuels 3,800 7,200 12,800 23,800 28,000 Diesel, Heating Oil 2,700 4,400 3,900 11,000 30,000 7,600 11,600 28,000 47,200 71,400 Nominal Capacity* 22,000 47,000 48,500 113,500 * Balance between nominal capacity and production is residual fuel, which is largely exported, and unutilized capacity. Source: Estimates based on data from refiners (1982) and the State of Alaska (1980-1981). 2.2.1 Petroleum Refining Petroleum is, of course, the major energy source currently produced in Alaska. Although most of the crude oil is exported to the Lower 48, three refineries in Alaska produce a major: fraction. of the state's needs for refined products, as summarized in Table 2. 1... The North Pole Refinery, run by Mapco, recovers middle distillates from the Trans-Alaska Pipeline at Fairbanks and returns the heavy fraction to the pipeline. The Kenai refineries of Chevron and Tesoro produce roughly enough gasoline to meet the needs of the entire state and demands for commercial and military jet fuels, the major consumption of which is concentrated in the Anchorage area. In addition, a small refinery is operated at Prudhoe Bay to obtain motor fuels and other products needed to support the oil field and pipeline operations there. All refineries involve distillation steps, in which various fractions of the crude oil are separated (boiled off) from successively heavier fractions. Since many refineries serve markets that demand more light products (such as gasoline) than are naturally yielded from the erude oil, some of the heavier materials may be processed catalytically or thermally (e.g., catalytic cracking, hydrocracking) to adjust the overall refinery yields. Further processing steps such as_ those required to meet gasoline octane requirements (e.g., catalytic reforming, alkylation) are often incorporated. In Alaska, only Tesoro is sufficiently integrated. (has enough processing steps) to meet a major portion of the state's needs for gasoline and jet fuels. Refineries generally possess some flexibility as to the range and relative quantity of their refined products. This flexibility is constrained, however, by the equipment installed and the composition of the crude oil feedstocks being refined. The two Kenai refineries were originally designed to process crude oil from Cook Inlet. As this light, sweet crude is depleted, it is being replaced as feedstock by the heavier, sour (high sulfur) crude from the North Slope. Since the natural result of: heavier crude runs will be a reduction in the light products such as gasoline, new investment will be required in these refineries to retain the current yield of light products from the heavier feedstock. Light products are generally. more valuable than heavy products, and there is thus.some incentive to make such investments consistent with the needs of the Alaskan market (and to some extent, the West Coast market). 2.2.2 Natural Gas Processing Natural gas, as. it emerges from the ground, contains several hydrocarbon compounds. in addition to diluents and impurities. Higher hydrocarbons such as ethane, propane, and butane are more valuable than methane because they are used as chemical feedstocks. As a result, they are ordinarily recovered from the natural gas prior to its sale. This is accomplished by condensing and separating out 17 these of these gas liquids from gaseous methane at elevated pressure * and reduced temperature. If carbon dioxide or sulfur compounds are present in unacceptably high concentrations, they can be removed by any of several commercially proven absorption techniques. Dehydration of the gas must also be accomplished early in the process to avoid the corrosive action of any condensed liguid water on gas processing equipment and downstream pipeline. Nearly all the natural gas produced in Alaska today comes from Cook Inlet and the North Slope. Most of the North Slope production is associated with the production of crude oil there, and the gas is largely reinjected into the formation to be stored for future use. That which is not reinjected is used to drive the four most northerly pumping stations on the Alyeska pipeline and generally to support the Prudhoe Bay operations. The Cook Inlet gas is relatively dry (free of gas liquids). .As a consequence, the need to process it is minimal. The net gas production at the Cook Inlet has three major market outlets; limited mainly by the infrastructure needed to transport the gas itself and by the equipment required to convert the gas to a more readily transportable form: e end-use markets and electric power generation in the Anchorage area, where Alaska's only major gas distribution system exists; e the Phillips liquefied natural gas (LNG) plant at Kenai, which exports its output to Japan; and . e the Union Oil ammonia and urea complex, also on the Kenai Peninsula, which exports its products mainly to the Lower 48 and Southeast Asia. The other noteworthy production of natural gas in Alaska is at Barrow, which is dedicated to local consumption only. 2.2.3 Transportation Fuels from Natural Gas Natural gas has long been recognized as a potential source of transportation. fuels. Currently, ENSTAR Natural Gas Company is running a fleet of about 60 light-duty vehicles in the Anchorage area with a dual fuel capability including compressed natural gas. Similar demonstrations, including some using LNG, have been conducted in the Lower 48. Natural gas can also be chemically converted to methanol, a mixture of hydrocarbon liquids, or even high octane gasoline. The most common method for producing chemical-grade methanol is steam reforming, where a hydrocarbon source, usually natural gas, is 18 reacted with steam at a.temperature of about 1600°F in the presence of a catalyst. The result is a mixture of predominantly hydrogen and carbon monoxide gases. These compounds are then recombined in the presence of another catalyst to form methanol which can be. used either neat or in mixtures to produce high octane motor: fuel. Alternatively, the methanol can be reacted with isobutylene to produce methyl tertiary butyl ether (MTBE), a high octane gasoline extender which, together with ethanol, methanol, tertiary butyl alcohol, and similar compounds, is expected to make up about 2-3% of the entire U.S. gasoline pool by 1985. Another approach is the Mobil M-Gasoline process, which uses a proprietary zeolite catalyst to convert. the entire methanol stream to high octane gasoline. This process is slated for its first commercial test in a 14,700-barrel per day facility planned for 1985 startup in New Zealand, which has ample natural gas reserves but negligible oil: reserves. In principle, the Fischer-Tropsch ‘process represents another. means of reassembling the hydrogen and carbon monoxide intermediates to produce a fairly wide range of distillate hydrocarbons. However, since the process is not highly selective, i.e., able to produce only © the most desired product(s), subsequent steps are required to separate the product into appropriate fractions and to refine them. The Fischer-Tropsch process was used in Germany during World: War II to produce liquid fuels from coal. A plant. based on this technology has been operating in South Africa since the mid-1950's and was recently expanded to produce about half of that nation's liquid fuel needs. 2.2.4 Liquid and Gaseous Fuels from Coal and Other Solid Fuels Coal, peat, and wood’ are the predominant solid fuel resources in Alaska. Since they are relatively expensive to transport, especially given the sparse infrastructure in Alaska, a central facility to convert them should ideally be situated close to the resource base. Fortunately, adequate tonnages of each are located in’ reasonable proximity to the major population. areas of the Extended Railbelt and the Southeast. - In addition to direct combustion to produce thermal energy (discussed in 2.3 below), solid fuels can be processed by techniques such as pyrolysis, gasification, or direct liquefaction to produce gases, liquids, or solids. that may be better suited to available markets than the raw resource itself. Pyrolysis, the oldest of these techniques, is also known as destructive distillation. In this process the fuel is subjected. to elevated temperatures (1000-1400°F) which alter the chemical structure to yield a mixture of gaseous, liquid, and solid products. Depending on the nature of the feedstock, the temperature of treatment and time of exposure can be selected to adjust the ratio of these products and, to some extent, their composition. Commercialization of solid fuel pyrolysis has been limited because it has not traditionally resulted in products sufficiently similar to conventional fuels to be generally accepted as substitutes. 19 In urbanized areas where coal was available, solid fuel gasification was long practiced as a means of using pipeline distribution to deliver gaseous fuels to various end uses. Town gas, as it was called, was later displaced from such applications as cheaper natural gas became available. But coal gasification has remained in practice for purposes such as. ammonia and methanol production and for synthetic liquid fuel production. It is also applied to produce synthetic liquid fuel in South Africa, as a means of minimizing its vulnerability to disruptions in world crude oil supply. In recent years, many other nations, including the United States, have contemplated similar measures. Many configurations have been used in solid fuel gasification, but generally all accomplish gasification by heating the fuel to higher temperatures than pyrolysis (1500-3000°F) and reacting it with steam to form hydrogen and carbon monoxide. In most cases, the heating is effected by using either oxygen or air to combust a portion of the fuel within the gasifier. The actual composition of the resultant gas is importantly determined by the temperature, pressure, and configuration of the gasifier. Production of transportation fuels from such a gas requires some gas purification and composition adjustment, followed by catalytic synthesis, in. a process conceptually identical to that described in Section 2.2.3 for natural gas. The Beluga Project has been conceived to use such an approach to manufacture methanol for export to California. In addition, processes are under development to produce liquids directly from coal without forming a gaseous intermediate. Such techniques, which have advanced importantly as a result of Department of Energy funding, are improvements on the old Bergius process which was carried out under extremely high pressures (10,000 psig) in the presence of hydrogen. The products formed are comparable, although not identical, to crude oil. and its primarily heavy derivatives. Coals which give the best yields and lowest capital costs are generally the higher quality bituminous coals, which are not plentiful in Alaska. This fact, coupled with the as yet undemonstrated nature of the technology and the very high projected capital cost, makes direct liquefaction an unlikely candidate for commercialization in Alaska for the foreseeable future. 2.3 Thermal Energy Thermal energy is usually provided by direct combustion of a fuel. Heat may be transferred either directly to the stream which will use it (such .as heating air for space heating), or to a convenient intermediary stream (frequently steam). The combustion device can be the familiar oil or gas burner or, in the case of solid or unconventional fuels, it can involve: 20 e stoker firing, where a grate is used in any of a variety of configurations, depending on the properties of the fuel, the scale of operation, etc.; e pulverized fuel. firing (usually limited to coal and coal-like fuels) where a stream of finely powdered fuel particles is burned in suspension in a flame much the same as ‘in oil and gas combustion; and. e fluidized bed combustion, where fuel and air are combusted as the air passes through a layer (or bed) of solid particles; these particles are kept in constant floating motion by the rising gas stream:and act collectively as a bubbling turbulent fluid. 2.3.1 Natural Gas Combustion Of all fossil fuels, natural gas is the cleanest and easiest to burn. Most problem impurities and diluents such as sulfur and carbon dioxide are removed before the gas is transferred into-the natural gas transport and distribution system. The major advantages associated with the use of this fuel are that: (1) the combustion products can be released directly to the atmosphere without treatment and still meet relevant pollutant guidelines; and (2) that no costs for gas storage facilities or inventory need be borne by the fuel user, since gas is obtained directly from the distribution line on an as-needed basis. Disadvantages of natural gas are that: (1) it is not immediately available to those who are not adjacent to existing distribution lines of sufficient capacity; and (2) for some large users, supply contracts are frequently set up on an interruptible basis. This means that in times of high demand, such as cold winter days, the gas company may require the interruptible user to suspend his gas use until peak demand has subsided. To insure against such eventualities, the gas user may install a backup fuel system, such as fuel oil, that can be stored in sufficient quantity to cover anticipated needs during the curtailment. 2.3.2 Oil Combustion In large part, oil combustion for thermal energy is quite similar to natural gas combustion, especially if the oils are light distillate materials such. as No. 1 and No. 2 fuel oil. Because of their cleanliness and relative ease of combustion, these light distillates are ordinarily used for residential and small commercial heating purposes. However, since their costs are higher, distillate oils are seldom used for large-scale central systems, where heavier residual fuels are more common. 21 oo TABLE 2.2 SUMMARY OF ALASKA AND U.S. EMISSION LIMITS FOR PARTICULATE AND SULFUR DIOXIDE Heat Input Million Btu/hr Wood Waste All Municipal Waste (incinerator) All Coal, Oil, Gas < 250 Coal, Oil, Gas (industrial) > 250 Oil and Gas (utility) > 250 Coal (utility) > 250 QQ) been calculated for inclusion in this table for comparative purposes. (3) Source: References (1) and (2). Particulate Emission Limit, 1b/million Btu Input State 0.257 0.137 0.085 0.085 0.085 0.085 (1) Federal 0.12) 0.137 (2) 0.10 0.03 0.03 (3) © Sulfur Dioxide Emission Limit, 1b/million Btu Input state!) 0.5 (2) 0.5 0.5 0.5 0.5 Actual state regulations are stated as flue gas concentration, grains/cubic foot. Federal 1.2) Gy (2) 1.2 0.2 or 90% reduction, whichever is higher Minimum reduction of 70-90%; rises with fuel sulfur content Numbers shown have Since the calculation involves assumptions about excess air required for consumption, the state limits are only approximate. yo standard established below 250 million Btu/hr input. Calculated based on standard stated as 0.08 grains/standard cubic foot. Residual oils contain both sulfur and mineral matter in concentrations relating to both the initial composition of the crude oil and the extent of processing. Since these impurities affect both the life of the combustion. equipment and the quantity of pollutant emissions, they must be considered in designing any device which is to burn residual oil. Thus; although: equipment capable of burning residual oil can usually also burn distillate’ fuels (or gas) with. a minimum of modification or adjustment, the reverse is seldom true. 2.3.3 Solid Fuel Combustion Solid fuels such as coal, wood and peat require different combustion processes than those for oil and natural gas. Historically, grate methods of firing solid fuels were first developed for manual feeding. But as requirements for higher outputs emerged, so.too did mechanical stokers, which enabled large. quantities of fuel to be fed evenly onto the grate while at the same time effecting ash removal. Many configurations of grates and feed mechanisms have been used over the years, depending on such factors as fuel properties, capacity, range of desired operation, emission limitations, efficiency requirements and the like. Pulverized coal, or powdered coal, has burning properties that are qualitatively similar to those of a gas. Because oil prices rose rapidly in the. United States during the 1890's, this technique gained wide use at that time to fire cement kilns. After World War I, pulverized coal firing spread throughout the electric utility industry because of its economies at larger sizes, higher combustion efficiency, and ease of control compared to stoker firing. Since noncombustible solids (ash) present in ‘the fuel become entrained in the stream of combustion products, particulate removal devices have been required to limit emissions to reasonable levels. More recently, since: the late 1960's, many coal combustors-~both stokers and pulverized coal units--have also added flue gas desulfurization equipment, also known as scrubbers, in response to pollution control regulations. Depending on the specific application (electric power plant or industrial boiler), the need for scrubbers is either eliminated or reduced when low sulfur solid fuels such as wood or peat are burned alone or in combination with coal, according to currently applicable regulations (Table 2.2). The advent of sulfur emission. standards has given considerable impetus to. the research and development of fluidized bed combustion systems.. This is because the fluidized. bed, where the combustion takes place, is itself an ideal location to add an appropriate reagent such as limestone to effect the removal of sulfur before it ever leaves the combustor. This eliminates the need for any subsequent’ equipment: to scrub the gases before they: pass to the atmosphere. Another advantage. of fluidized beds is their ability to burn a wide variety of solid and liquid fuels including many otherwise 23 difficult-to-burn materials. Commercialization of the fluidized. bed concept to date has been limited to small industrial units, where SO removal requirements are minimal or not required at all. Scale-up to larger utility sized units will depend on adequate demonstration of the sulfur removal capability of the concept at reasonable cost, combustion efficiency, and reagent consumption levels and on the development of fuel feeding techniques suitable for very large systems. — : 2.3.4 Waste Heat Recovery Any energy transformation process cannot practically be. totally efficient. The limitation is partly thermodynamic in nature, since the heat is available at a lower temperature level than that at which it can be utilized. The limitation is partly an economic one, where the investment to recover additional heat cannot be justified by the potential value of the energy saved. Some energy is invariably rejected or lost to the environment. It is usually economically infeasible to transport low temperature heat over long distances. Sometimes processes can be strategically located so that their waste heat is recovered to satisfy an existing process or space heating need (heat sink). Alternatively, the heat sink may be located to exploit an available waste heat source. Either way, the synergistic benefit of improving the overall energy utilization of the two processes may be significant relative to operating each plant independently. Frequently,- economic circumstances may change and alter. the incentive to invest in waste heat recovery. For example, energy values may rise sharply as happened twice in the 1970's; or in manufacturing operations, production rates may exceed earlier expectations. In either example, a given investment would result in a higher return as measured by the total value of energy recovery. 2.3.5 Geothermal Energy The. result of naturally occurring radioactive decay within the earth, geothermal energy reaches near enough to the earth's surface in places ‘to make it exploitable by man. Alaska has numerous such locations, some capable of producing temperatures above 150°F. Geothermal energy has some of the same economic limitations discussed for industrial waste heat. As a low temperature energy form, it is relatively expensive to transport. The energy utilization must therefore take place very near the source. Further, the temperature requirement must be compatible with the temperature available from this resource. Because of the remoteness of presently known Alaskan geothermal resources from inhabited areas and because of the narrow range of practical applications for this energy form, geothermal energy will have a limited impact in Alaska for the foreseeable future, particularly in terms of large-scale, central conversion technologies. 24 2.3.6 Solar Industrial ° Process Heat Sunlight incident on a flat plate collector can be utilized to heat a fluid which passess through that collector. However, most process heating requires higher temperatures than a flat plate collector can achieve. Designs that concentrate sunlight at appropriate collection points can produce higher temperatures commensurate. with the usual needs of industrial processes, but at lower overall efficiencies as the collection temperature rises. Examples of concentrating collectors are: @ compound parabolic concentrators (Figure 2.2), which can produce temperatures of 100-35 °F; although they do not need to track the arigle of solar incidence for optimal performance, they may be tilted monthly or seasonally to maximize the total energy collected; e@ parabolic trough concentrators (Figure 2.3), which can produce temperatures of 200-65 °F;. these devices require single-axis tracking (rotation about the major axis of the collector) to keep the incoming energy focussed on the receiver; and e@ parabolic dish collectors (Figure 2.4), which can produce temperatures of 700-1500+°F; dish collectors need to track in two dimensions to assure focussing at the collection point. Parabolic’ troughs and compound parabolic concentrators § are commercially available. Parabolic dishes, although ‘they can be purchased, still need further development and demonstration before they are applied on a widespread basis. In any application of solar energy processes, thermal requirements during periods of low or no insolation must be supplied either from previously stored energy or from a backup system such as fossil fuel combustion. Both these options add to the cost of using solar energy. This situation is exacerbated in Alaska relative to most industrialized parts of the world because of the disparity between the number of daylight hours in the winter and those in the summer, and because the relatively low sun angle for most of the year causes a larger fraction of total sunlight to be absorbed in the atmosphere rather than by 'a collector. 2.3.7 Nuclear The potential of nuclear fission as a peaceful energy source was understood almost from the discovery of the process itself. Nuclear reactors producing useful energy developed largely in parallel with, but slightly behind, nuclear weapons. The first practical large-scale 25 92 Parabolic Reflector Source: General Electric Evacuated Tube Receiver Heat Transfer Heat Transfer Fluid Iniet ’ Fluid Outlet FIGURE 2.2 COMPOUND PARABOLIC CONCENTRATOR Le Source: Acurex FIGURE 2.3 Parabolic Reflectors - Receiver Sade PARABOLIC TROUGH CONCENTRATOR Insulated Receiver Concentric Reflector Declination Drive System (Jack-Screws}, Rotation Drive System Elevation & Stow Drive System (Jack-Screws) FIGURE 2.4 PARABOLIC DISH CONCENTRATOR 28 application of nuclear fission reactors aS a power ‘source was in submarines. Very soon the reactors moved to land as a source of electric utility power generation. In the first part of the nuclear fuel cycle, uranium ore is mined and processed. gyre key step in processing involves increasing the fraction of U atom in the product from the naturally occurring 0.7% to 2.5-3.5%, to facilitate fission. This process, called uranium enrichment, may consume a few percent of the power which the fissioning uranium will ultimately produce. This enriched uranium is then fabricated into fuel elements for shipment to the power plant. Nuclear fission is the splitting of a heavy nucleus into two lighter ones. This reaction is easily induced in certain heavy nuclei by; bombarding them with low speed neutrons. Such, nuclei, notably the isotope of uranium containing 143 neutrons, U°**, yield more free neutrons with the. fission, along with a large amount of energy. A fission event produces two to three free neutrons which. collectively cause another fission, and so on. A nuclear power. reactor, a carefully designed arrangement of nuclear fuel and other materials, conducts this process at temperatures, pressures, and power densities which allow extraction of the fission energy as useful heat. Directly or indirectly, they raise steam which is used to drive turbine generators. Four U.S. manufacturers currently market nuclear power reactors. General Electric markets. a system which raises steam directly in the reactor core--a “boiling, water reactor". or BWR. Westinghouse, Combustion Engineering, and Babcock and Wilcox all market systems which cool the reactor core with high pressure water--a PWR--and use this hot water to raise steam in external steam generators. Another power reactor concept is high temperature gas-cooled reactor, or HTGR. This reactor has a core built primarily of graphite, which could operate at very high temperatures. It is cooled by high pressure helium, which is used to raise and superheat steam in external steam generators. . Developed by General Atomics, the HTGR has not achieved market acceptance and is not currently being actively pursued. : A fuel element will remain in the reactor core for a few years before its fuel value is spent. These highly radioactive spent. fuel elements must then be removed and ultimately shipped to disposal. Whether or not to reprocess these fuel elements to extract fissile material. and how to dispose of the highly radioactive wastes are controversial political issues. Lack of. resolution of these issues and regulatory problems affecting nuclear power plant construction and operation cast doubt upon the viability of installation of new nuclear generating capacity in the United States. 29 2.4 Electric Power The overall objective of an "integrated central electric utility operation is. to provide electric power to the customers in its service area at minimum cost and maximum reliability. To do this, the utility has to make a number of strategic choices: e how many plants to build and/or shut down? e using what fuels? e where? e what size? e what technologies? — These choices, constrained by the company's financial resources and ability to manage several operating plants, are complicated by a -number of limiting factors discussed in the following paragraphs. Economic outlook. This affects the anticipated growth in power demand and, hence, the utility's potential need for new generating capacity. The economic outlook, including projected construction costs, will have an important bearing on the timing of any new projects. Fuel price outlook. Although fuel prices are related to the overall economic picture, recent history has shown some to be affected by international politics. To the extent that such’ dislocations can be anticipated, the utility may be able to take mitigating actions, such as switching to more stably priced, locally available resources which in Alaska include coal, peat, and a wide variety of renewables. Location of the resource base. A resource base close to its point of use provides a measure of security of supply and helps keep the transportation costs to a minimum. Infrastructure. Another factor to be weighed is the existence of infrastructure to: (1) move fuels to existing or proposed generating stations; (2) distribute power to the customer; and/or (3) move construction equipment and labor. The existing infrastructure has an important bearing on where new units should be located, how quickly they can be brought on line, and which resources they should utilize. Environmental impact. Any central energy conversion project will have some environmental impact, whether it be in terms of land use, emissions and effluents, noise, visual impact, etc. These factors must be weighed against the expected benefits of the project. 30 Demand profile. This consideration, perhaps as much as any other, accounts for the variety of. technologies and. fuels that are used within a given service area. For example, in the Cook Inlet region, power generating capacity consists of gas turbines, steam turbines, hydro, and diesels, although the latter are principally on a standby basis. The importance of the demand profile derives from the fact that the annual average operating output of a total power system is limited by the actual demand for power. Since there is no appreciable use of electricity storage technology either in Alaskan or other utility systems, the generation capacity in place must be sufficient to cover the full peak load demand and provide a reasonable margin of safety to assure a reliable supply. However, on an annual average basis, this capacity can be utilized no more extensively than the output required by the customers. Thus, the utility is faced with the question of whether to invest in technology with a high first cost.to save on its future energy bills (through the use of cheaper resources and/or less of them); or to invest in less costly technology knowing that future energy costs will be higher. The solution chosen usually combines both approaches: the former with the intent of operating it to cover that .portion of the load which is steady and persistent throughout most of the year (base load); and the latter to cover the peak demands. Figure 2.5 shows the approximate trend between the first cost and energy cost (comprising both the price and quantity of the energy form needed) of conventional central station technologies. In Alaska the ratio of peak to base load capacity is much higher than in the Lower 48, owing to the fact that there is no summer cooling load in Alaska. This tends to lower the capacity utilization in Alaska as compared with other locations, which partially’ explains why ’ generating capacity in the state is of lower capital cost intensity and higher fuel cost intensity. A related factor is that with the relatively low cost of natural gas in the Anchorage area, the technologies based on this fuel tend to be viewed more favorably here than Figure 2.5 implies to be true more generally. Another factor which has contributed to this capacity mix is that most generating units are relatively small, consistent with the population base. Because of economies of scale, the capital intensive options are most economical at larger sizes and, hence, have not been extensively deployed in Alaska. Salient features of the major central electric power generation technologies are discussed below. Other technologies are covered in the Glossary, along with a thermodynamic overview of central, dispersed, transportation, and residential/commercial end-use sectors. 31 HIGHER ENERGY COST —______» 4 9 : = 6 3 <6 a «~ & FF w c o og ® 2 4 5 65 3 ££ & 8 @ > 35S uwoeo6¢ = 2 8 8 6 & HIGHER INVESTMENT *Hydro is very site-specific and may not necessarily incur the highest investment. ** A notable exception to gas turbines involving the highest energy cost is in the Anchorage area where natural gas is available at a relatively low cost. _ Source: Arthur D. Little, Inc. FIGURE 2.5 GENERALIZED RANKING OF INVESTMENT COSTS AND ENERGY COSTS FOR CONVENTIONAL CENTRAL POWER GENERATING TECHNOLOGIES 32 ome 2.4.1 Steam-Electric Power Generation Steam-electric power plants use a Rankine cycle (Figure 2.6) with water (steam) as the working fluid to convert thermal energy to electric energy. The source of heat can be fossil fuels, geothermal energy, solar energy, nuclear, or waste heat from any of these. The factors impacting the choice of heat source for steam-electric or other use have already been discussed in 2.3.0 Thermal Energy. In general, a Rankine cycle can be made to run more efficiently by raising the pressure at which steam is generated and by incorporating: other improvements such as regeneration and reheat. The benefits from these measures are somewhat counterbalanced by the cost of implementing them. For example, high pressures and reheat. temperatures require more exotic materials of construction to ‘ withstand the extreme conditions. _ Since the incentives become greater as unit sizes increase and because of economies of scale, larger systems tend to be designed for higher temperatures and. pressures than smaller systems. However, this trend levels out rapidly for even larger systems (moderate to large size base load units) because of the limitation of materials available today. Steam cycles run relatively efficiently through much of the operating range. They are amenable to both base load and peak load operation, determined in large. part by the type of fuel used. 2.4.2 Gas Turbines A gas turbine is a Brayton engine, where the combustion gases --mainly nitrogen, carbon dioxide, water and oxygen gases--are the working fluid. As shown in Figure 2.7, gas turbines are very similar in design to jet airplane engines and can burn similar fuels as well as gaseous fuels which are more readily available for stationary use. Today, they are limited to burning relatively clean, light fuels because contaminant byproducts of combustion such as_ sulfur compounds and dust particles would rapidly deteriorate the turbine blades. Gas turbines by themselves are much less efficient than most other heat engines, and their performance (efficiency) degenerates at reduced loads. Because of its need for premium quality fuels, the gas turbine is best suited to applications where the capacity it provides is only needed for a relatively small fraction of the time. Of course, in the Anchorage area where natural gas is still fairly inexpensive, gas turbine operation at higher capacity factors can be justified much more readily than in most other parts of the world. 2.4.3 Combined Cycles This term refers to a combination of any two power cycles that are designed and operated in tandem in order to recover more heat than either cycle when operated separately. In commercial practice today, 33 Heat FIGURE 2.6 High Pressure Steam Electricity Boiler Feedwater Pump SIMPLE RANKINE CYCLE POWER PLANT WITH STEAM AS THE WORKING FLUID 34 sé Air Inlet Source: General Electric Company ew Peed — aes woe wm. Fuel Nozzle Stage Gas Turbine 17-Stage Air Compressor FIGURE 2.7 SIMPLE CYCLE GAS TURBINE Combustion -Gas Outlet 9€ Energy Utilization (Btu Products-.Btu Fuel) 1.0 Source: Reference 3. T T T TT tT Tt Conventional Steam Cycie *-Gas Turbine with Heat Recovery Combined Cycle with Heat Recovery Separate Power and Stéam Plants Four Stroke Diesel with Heat Recovery Low-Speed Two Stroke Diesel with Heat Recovery 2 3 4 5 6 7 Ratio of Thermal Energy to Electric Energy FIGURE 2.8 ENERGY UTILIZATION CHARACTERISTICS OF COGENERATION SYSTEMS ee a combined cycle most frequently means a gas turbine followed by a steam (Rankine) cycle. In effect, the latter captures waste heat which would otherwise be lost by the former and converts it to more electricity. This requires a greater capital investment than for the gas turbine alone. If high capacity utilization is: expected, the combined cycle can be justified on the basis on the fuel saving. However, since all the fuel is combusted in the gas turbine, this combined cycle concept is limited to those fuels which it can accept. In the Cook Inlet region, approximately 300 MW -of base load capacity is comprised of gas turbine combined cycle units using relatively inexpensive natural gas. Other commonly considered combined cycle concepts are fuel cells combined with steam-Rankine cycles and diesels combined with organic-fluid-Rankine cycles. : 2.4.4 Cogeneration Cogeneration is the coproduction of electricity and thermal energy, usually as steam. Its attraction rests in its: ability to produce both energy forms: simultaneously with less total energy than required to produce each separately. Thus, if thermal energy is valued at. the cost of producing it alone, incremental electricity can be provided at about twice the efficiency of a power plant where cogeneration is not used. . A variety of thermal-to-electric ratios may be selected for a cogeneration system. Diesel units with waste heat recovery offer the lowest ratio, while steam topping units: represent the highest (Figure 2.8). Once a cogeneration design is implemented, the system must be operated at or near the optimum ratio or the thermodynamic advantage is quickly lost. This can be a problem in applications without -grid connection, since the ratio of electric and thermal demand can fluctuate and the grid is not available to buffer system output and demand. Recognizing this constraint as well as the potential national benefit of saving primary energy through cogeneration, Congress included a provision in the Public Utilities Regulatory Policies Act of 1979 (PURPA) requiring that electric utility companies purchase cogenerated power from their customers. at a fair price to. be established by each state on the basis of Department of Energy guidelines. In the face of tardy state price guidelines, frequently unattractive rates, an uncertain economic and energy outlook, and even constitutional questions regarding the Act, cogeneration projects have not evolved very rapidly. Some of these conditions are likely to continue to forestall the development of more cogeneration, even in highly industrialized states. In Alaska, where the industrial base in rather thin, cogeneration is not likely to have a major impact for many years. : 37 2.4.5 Hydroelectric Hydroelectric power is the energy of a river or stream flowing from a higher to a lower elevation is converted to shaft power via a flow turbine which, in turn, is converted to electricity in a generator. The greater the vertical distance (head) that the water drops en route through the turbine, the more force it exerts on the turbine blades and, hence, the more energy that can be extracted per unit of flow. A hydro project can be implemented to recover "run-of-the-river" power with minimum civil work and water impoundment, recognizing that flow varies seasonally and yearly. Discussed more fully under dispersed energy systems, small-scale hydro is usually implemented in this manner as a supplement to other power sources. In contrast, large-scale hydro incorporates large-scale water storage to allow the operator to control the timing of generation. An additional benefit of building a dam is to levelize the seasonal and annual flow variations, enabling power extraction at more nearly the rate required by the grid; and to increase the head and, therefore, the quantity of electricity. that the turbines can produce. The actual flow of water through the turbogenerators is largely controllable by the operator and, in principle, can be set to follow power demand within a wide range. Because of this load-following capability, the relative economics of base load and peak power within the grid system is a major factor affecting the optimum project design. Another hydro option, which does not necessarily involve energy extraction from a natural stream, is pumped storage hydro. Here, water is pumped up from one reservoir (usually man-made) to an adjacent higher one using base load electricity generated in off-peak ‘periods. The energy stored this way is then available for use in generating power during peak periods by passing it back down through the pump which in this case acts as a turbine. Thus, pumped storage hydro is a load leveling technique to store relatively inexpensive base load power for use during more costly peak periods. Pumped storage hydro is not ordinarily an economically viable approach for utilities with access to hydro power at reasonable cost. The reason is that a single hydroelectric project is already optimized in terms of the split between base load and peak power (as described above). As a separate project, pumped hydro cannot often - successfully’ compete because of the extra step involved in pumping compared to merely storing water for peak usage in a hydroelectric project. Therefore, one would conclude that in Alaska, given its vast hydroelectric potential, pumped storage hydro is likely to find little application. 38 2.5 Central Electric Power Grid Systems 2.5.1 Central Electric Power Grid Systems The portfolio of conventional central power generation technologies has been discussed, along with the issues which impact their selection. Implicit in some of these issues is the knowledge that, - since central system generators are interconnected, they offer several advantages: e the peak demand of a central system is lower than the sum of the peaks of individual areas; : e the requirement for excess capacity is less than it would be to maintain equal reliability of supply in several disconnected sub-areas; @ pooled demand: strengthens the negotiating position for purchase of energy resources; e pooled demand permits economies of scale in system design and installation; . e the existing infrastructure can be most effectively utilized; e generating stations. can usually be strategically dispersed within the service area; and ® power generation can be dispatched most economically on an almost instantaneous basis. Simply put, interconnection of generating stations and/or electric ‘companies involves running power lines of sufficient capacity to permit the full range of expected current flow, in much the same manner as the connection of an individual residence but on a more elaborate scale. Compared to most urbanized areas of the Lower 48, Alaskan grid connections are rather modest. Several conventional options are used to transmit small to moderate amounts of power over short to long distances. Table 2.3 below lists common voltages - and nominal capacities of single-circuit systems. Power lines or cables operating at voltages from 4 to 69 kilovolts (kV) are usually .classified as distribution lines or cables; transmission systems operate at voltages of 138 kV and higher. 39 TABLE 2.3 TYPICAL TRANSMISSION LINE VOLTAGES AND CAPACITIES Usual Maximum Voltage Capacity Underground Option Distance Type (kilowatts) (megawatts) Available (miles) AC 4 1 Yes . 2 AC 13 4 Yes : 5 AC 23 8 Yes 10 AC 69. 40 Yes . 40 AC 138 80 Yes : 300 DC +200 400* No No Limit For distances greater than 400 miles, DC systems become both more efficient and economical and are preferred as a technical option. A DC system using two separate conductors, positive and negative, also has the advantage of maintaining transmission capability (at slightly reduced capacity) if one of the conductors is downed. In this case, the earth serves as a return conductor. In contrast, if one of the three conductors of an AC circuit is downed, the system becomes essentially useless. Underground systems using conventional oil/paper cables are from 10 to 20 times more expensive than overhead lines of equivalent capacity, on a per-mile basis. The length of such cables rarely exceeds 20 miles because of capacitive limitations. These limitations can be eliminated if inductive compensation is employed, but this adds significantly to the cost of a cable system. , Careful evaluation of applicability of long distance DC transmission systems for use in Alaska may be of additional importance in that it may determine the viability of interconnecting the larger but currently unconnected electric grids. An interconnection such as the planned Anchorage-Fairbanks intertie could augment individual grid reliability and reduce combined system fuel costs. It could also facilitate use of larger fossil-based or hydroelectric generating units. 2.5.2 Rural Grid Systems Although rural grid systems offer the same advantages in principal as their urban analogs, their costs are much greater, owing to distance between adjacent villages, difficult construction on permafrost and *Nominal values; higher or lower capacities can be selected depending” on the level of eurrent desired. 40 tundra, and scarcity of local skilled construction workers. These factors have hindered the development of a grid system in the bush. As a positive step in the direction of interconnecting bush communities, the state has sponsored an 8 1/2 mile demonstration of the single wire concept, where the earth is used as a return conductor. This 14.4 kV single-phase AC line connects Napakiak and Bethel at a capacity of 105 kVA (about 100 kW). Another similar line connects Shungnak and Kobuk 2.6 Large-Scale Energy Transport Limited infrastructure exists in Alaska for moving: oil and gas by pipeline; oil and coal by rail; ; oil by tanker and barge; and liquefied natural gas by cryogenic tanker. These have been developed over the years primarily to transport Alaskan. resources to world markets. Given Alaska's limited population base, it is difficult to envision that any _ similar transportation projects could be justified by in-state needs alone. The impetus of world trade would be needed to enlarge the capacity to take advantage of economies of scale and to attract the financing needed. For example, the Alaska Natural Gas Transportation System (ANGTS) is contemplated to move North Slope gas into the Trans- Canada Pipeline and into the north central United States. And a project has been proposed to convert Beluga coal into methanol for transport to the West Coast. A coal transport project could possibly materialize in the foreseeable future. Alaskan coals, though relatively low in energy content, are exceptionally low in sulfur. This feature could be very attractive in some parts of the world. Slurry pipeline represents one option that could compete with rail transport. When coal is ground finely in a suitable liquid such as water, it can be made to flow readily through a pipeline. Slurry pipelines are already commercial throughout the world and carry a variety of minerals, including copper concentrates, taconite fines, limestone, sulfur, and coal. Coal slurry transport from rich deposits in the Rocky Mountains has been proposed to move coal to markets in the central, southwest, and western portions of the Lower 48. But these proposals have been embroiled in controversy regarding (1) water rights in the arid regions where the coal is located and (2) rights of way, especially where the pipelines would cross competitive railroads. Although an Alaskan coal slurry pipeline would not likely encounter these difficulties, there would be other environmental impacts and, of course, numerous’ construction difficulties associated with fragile terrain, climate, accessibility to construction sites, and the need to transport and house construction workers--issues with which Alaskans have become very familiar during the construction of the Alyeska Pipeline. 41 REFERENCES State of Alaska Register 74, Title 18 Environmental Conservation, Chapter 50, Air Quality Control, July 1980. Federal Register, Volume 44, No. 113, June Il, 1979, p. 33580. Arthur D. Little, Inc., Westinghouse Electric Corp., Gibbs & Hill, Inc., Industrial Cogeneration Optimization Program, U.S. Department of Energy, January 1980. : 42 3.0 OIL AND GAS FIRED BOILERS 3.1 Introduction. and Summary 3.1.1 Overview Oil and gas fired boilers burn natural gas and fuel oils and use the heat. of combustion to raise steam. They have numerous advantages over their solid fuel fired counterparts as a source of steam. These include: e lower capital cost; e ‘minimum operator attention; e minimum maintenance; and e higher reliability. Collectively, these advantages make oil and gas boilers preferable to raise steam for most commercial and many industrial applications. These applications tend to be those characterized by relatively small steam demands and relatively low capacity factors. In the Anchorage area, natural gas fired boilers will the added advantage of having low fuel costs. Adding this advantage to their lower capital cost, higher reliability, and lesser operation and maintenance requirements, gas fired boilers are the lowest cost source of steam for all applications in the Anchorage area. . 3.1.2 Alaskan Perspective Oil and gas fired boilers are used today for commercial and industrial steam raising in Alaska. They provide reliable and, in many cases, very low cost steam in support of large process plants as well as for many smaller scale uses. Oil fired boilers also provide steam in Alaska's paper industry. Gas fired boilers enjoy a particular economic advantage in the Anchorage area. With the low cost natural gas available there, they are and are expected to continue. to be the cheapest source of steam in the immediate vicinity of Anchorage. 3.1.3 Significance of the Technology Oil and gas fired boilers are technologically mature. They provide a highly reliable supply of steam with minimal operator attention and minimal expenditures from maintenance. They can be expected to provide the least expensive steam to serve many relatively small and/or low capacity factor steam demands throughout the state and for essentially any commercial or industrial application in the Anchorage area. 43 Scanner and Observation Ports Air Register Water Cooled Throat Gas Spuds Main Oil “3 Stabilizer Burner f Source: Reference 1. FIGURE 3.1 NATURAL GAS BURNER 44 7 Evaluation of a 100,000 lb per hour gas fired boiler operating with a capacity factor of 65% in the Anchorage area yields an estimated average annual steam cost of $3.14 per million Btu of steam raised. This represents a cost of steam which no other system will be able to match in the Anchorage. 3.2 Description of Technology 3.2.1 Operating Principles Gas and oil fired boilers burn natural gas, fuel oil, and/or similar fluid fuels to raise steam. These boilers may be described as carrying out two’ processes: fuel combustion and heat transfer from the combustion exhaust to boiler tubes. Combustion of natural gas differs significantly from combustion of fuel oil; therefore, the two classes of burners are treated separately. The: processes of the heat exchange from the combustion exhaust to the boiler tubes are very similar for boilers fired with gas and. those fired with oil; accordingly, they are described for a gas fired boiler with significant differences between gas and oil fired systems noted separately. Fuel Combustion In boilers utilizing natural gas, the fuel is discharged into the open furnace zone of the boiler at high velocity through gas jets (commonly called spuds). These jets are surrounded by orifices of the combustion air supply system, known as air registers or wind boxes, as shown in Figure 3.1. ‘The rate of combustion is controlled by manipulation of the gas jet/wind box. configuration and velocities of gas and air entering the boiler. These collectively control the rate of mixing of gas and air, hence the rate of combustion. , The furnace is designed to allow complete fuel burnout before the flame contacts any heat transfer surface. This. is done because premature flame-tube contact -will quench the -flame resulting in incomplete fuel combustion. In older designs, it was customary to maximize the rate of mixing so as to limit flame length and allow complete gas combustion in the most compact furnace possible. More recent designs incorporate capability to limit gas/air mixing, producing a longer, cooler flame cooled over its length by radiation. Combustion in this cooler flame minimizes the formation of oxides of nitrogen whose emissions are regulated by the New Source Performance Standards of 1971 and the stricter Standards of 1979. Combustion of fuel oil is in many ways very similar to combustion of gas. Fuel is introduced into the furnace at high velocities through a nozzle which is surrounded by wind boxes. Fuel and air velocity and oil nozzle and wind box configurations are manipulated to control flame mixing and thereby limit formation of oxides of nitrogen. 45 4O0°- Sac° Pree aa aaaeaee 46 Source: Reference 2. FIRE TUBE BOILER FIGURE 3.2 However, introducing fuel oil into a furnace is substantially more complex than for natural gas. For high combustion efficiency, the oil must be sprayed into the furnace as an atomized fog. Three classes of atomizers are currently used: mechanical atomizers, air atomizers, and steam atomizers. By far. the most popular is the steam atomizer. (1) Here, fuel oil and steam at high pressure (approximately 300 psig) are pumped into the furnace. Upon leaving the fuel nozzle, the steam expands, breaking up the stream of oil into an atomized fog. Fuel oils heavier than No. 2 must be preheated in order to flow acceptably through fuel handling systems and-to respond properly to atomization. . Preheat temperatures range from 135°F for a No. 4 fuel oil to 200-220°F for a No. 6 fuel oil.(1) As with natural. gas, oil fired furnaces are sized. to allow complete fuel burnout before the flame contacts any heat transfer surface. In very large boilers where flame length is the dominant consideration in furnace design, an inherently longer oil flame requires that an oil fired furnace be larger than a gas fired furnace of the same capacity. For smaller systems with capacity ratings which are likely to: be of most interest in Alaska (that is, smaller than 500,000 pounds of steam per hour), other considerations constrain furnace design, resulting in use of essentially identically sized furnaces for burning oil and gas. Boilers burning natural gas or fuel oil display. very high combustion efficiencies. In the case of natural gas fired boilers, loss of heating value due to incomplete combustion is considered to be totally negligible. ‘ Heat Transfer In order to raise steam it is necessary to extract the usable heat from the combustion gases. This process begins with radiation of energy to the water-cooled furnace walls. Immediately following flame burnout, more efficient heat transfer by. convection begins. Two basic classes of boiler design offer alternate approaches to heat exchange: firetube boilers and watertube boilers. In a firetube boiler, combustion typically occurs in a furnace which is surrounded by a chest of water. Combustion exhaust passes out of this furnace through tubes which extend upward through the top of the furnace through the water chest, as shown in Figure 3.2. Such a design is called a firetube boiler because the fire exhaust is passed through the heat transfer tubes. It provides for very efficient and compact heat transfer, but since it contains the boiling water in a chest, the firetube boiler concept is limited in size and achievable steam pressure. With no boilers fabricated which produce more than approximately 40,000 Ib of steam per hour, the fire tube boiler design is of limited use for large-scale industrial steam raising.(3) It has— essentially no application for electric utility steam raising. 47 Source: Reference 1. FIGURE 3.3 CUTAWAY DRAWING OF A PACKAGED NATURAL GAS FIRED BOILER 48 The watertube boiler design concept allows raising and superheating of very high pressure steam and is amenable to scale-up to enormous size. Furnace exhaust is forced to flow across row after row of boiler tubes. As it does so, the furnace exhaust is cooled by heating water or steam in the tubes. Figure 3.3 displays a cutaway drawing of such a boiler. . Boilers which fire heavy ash-bearing. oils must have convective heat transfer zones which are designed slightly differently from those designed for burning natural gas or light fuel oils. In order to minimize tube erosion and plugging by fuel ash, the boiler tubes in the convective section must be spaced farther apart: and gas velocities must be limited. This increases boiler size substantially; to a lesser _ extent, it makes the boiler more expensive. In industrial and electric utility boiler applications, exhaust may be cooled to its final discharge temperature by heat transfer to boiler tubes. In large-scale installations, the final cooling is more likely to take place in an air preheater to preheat incoming combustion air. In either case, final discharge temperature will be limited by the following considerations: , e if a sulfur-bearing fuel (as is the case with most fuel oils) is burned, condensation of sulfuric acid in the coolest area exposed to boiler exhaust must be avoided: This typically requires that the temperature be maintained ‘at levels as high as 350°F, depending on sulfur content. e if an essentially sulfur free fuel, such as natural gas, is burned, discharge temperature will be limited by considerations of plume buoyancy and economic trade-offs between higher capital cost for more heat exchange capability and higher fuel costs resulting from lower boiler efficiency. Based on these considerations, large installations typically have discharge temperatures between 250 and 300°F. 3.2.2 Technical Characteristics Despite the differences noted between oil burners and gas burners, oil fired boilers and gas fired boilers have the same rating and tend to be quite similar. The similarities are so pronounced that most sales of oil or gas fired water tube boilers in the United States today are of boilers which are capable of burning either fuel. (3) These boilers differ markedly from solid fuel fired boilers. Their technical characteristics can best be understood by description in contrast with those of solid fuel fired boilers, as is done in Table 3.1. 49 Table 3.1 | COMPARISON OF BOILERS BUILT FOR FLUID AND Fuel Type Boiler Characteristfi Fluid Fuels Natural Gas, Fuel Oil, Refinery Gas, etc. Ratings Available (1000 1b Steam/hr) Shop Fabricated Up to 550 Field Erected 100 - 10000 Capital Cost Low Economies of Scale Moderate Boiler Efficiency 85% Operation and Maintenance Difficulties Minimal Reliability Very High 50 SOLID FUELS Solid Coal, Peat, Wood, etc. 5 - 50 10 - 10000 High Substantial 84% Substantial Moderate to High Fluid (oil and gas) and solid fuel fired boilers are both available in essentially any capacity - range’ of interest for industrial or electric utility use. A major advantage of oil and gas fired boilers over solid fuel fired boilers can be seen by comparing the capacity range of factory assembled packaged designs. Here the limit on boiler capacity is set by the need to be able to ship the boiler to its destination in no more than four pieces. (4) Oil and gas fired boilers are available in shop fabricated designs with rating of ten or more times the maximum size of a shop. fabricated coal fired boiler. This results directly from. fundamental differences between oil and gas and solid fuel fired boiler technology. The oil/ gas fired boiler requires a much smaller furnace, and water tubes in the convective zone of an oil/gas fired boiler can be packed much more closely together than their. counterparts in the solid fuel fired boiler. This leads to the capability to pack far more steam generating capability into a smaller volume. Oil. and gas fired boilers having. rated capacities greater than approximately 550,000 pounds of steam per hour must be field erected, even smaller for some manufacturers. Field erected oil and gas fired boilers are also offered in smaller sizes to. suit particular application needs. Boilers are marketed today which produce steam at temperatures up to 1010°F and pressures up to 3500 psig. Maximum steam pressure and temperature ratings increase with increasing boiler capacity, with the highest temperature and pressure steam raising capabilities available only in boilers generating millions of pounds of steam per hour. Shop fabricated or field erected, oil and gas fired boilers have much lower capital costs than solid fuel fired boilers able to generate the same output. This advantage is most pronounced: in small sizes, particularly where shop fabrication is an option. Field erected oil/ gas fired boilers are also less expensive than their solid fuel fired counterpart; however, in scale-up this advantage is relatively less important. A-boiler fired with oil or natural gas will be slightly more efficient than a boiler fired with Beluga coal. The 85% boiler efficiency calculated for a gas fired boiler operating in the climate of Anchorage is quite insensitive to variations in ambient temperature or to expectable variations in natural gas composition (boiler efficiency varies 0.017% with ambient temperature variations of 1°F from the mean value of 35°F found in Anchorage). Operation and maintenance difficulties ‘posed by gas fired boilers are minimal. Those: posed: by oil fired boilers are only slightly greater. The nature of fuel feed and of the flame itself minimize. problems of flame stability which plague operation of pulverized coal fired boilers. 51 Fuel handling and cooling of boiler exhaust result in practically no wear related system damage; accordingly, maintenance requirements are minimized. Thus, operation and maintenance of gas fired boilers requires a minimum of full-time labor. The -reliability of oil and gas fired boilers is very high. The principal causes of failure of solid fuel fired boiler--boiler tube failure. due to fireside damage, and failure of fuel handling and preparation systems--have no counterparts in oil and gas fired boilers. : 3.2.3 Environmental Issues All boilers pose one or more of the following pollution problems: e air pollutant emissions; e water pollutant discharges; e "solid waste production; and e noise emissions. Air Pollutant Emissions Three classes of regulated air pollutants are of concern: sulfur dioxide, oxides of nitrogen (NO_), and particulate matter. Levels of emissions are determined by fuel composition, boiler design, and application of post-combustion controls. The quantities of sulfur dioxide emitted from a boiler firing oil or gas are determined directly by the sulfur content of the fuel fired. Since natural gas is practically a sulfur free fuel, only negligible amounts of sulfur dioxide are emitted’ when it is burned. Fuel oils are available with sulfur contents ranging from less than 0.1% to more than 3% by weight. Oil sulfur content can readily adjusted in the refinery. Current commercial practice calls for a user to buy fuel oil with such sulfur content as will result in compliance with sulfur dioxide. emission regulations without post-combustion controls (scrubbers). Use of proper boiler design technique readily controls formation and -emission of NO_. -Practically all of the NO_ emitted from an oil or gas fired .boiler* results from very high température chemical reactions between atmospheric nitrogen and atmospheric oxygen in or near the flame. By limiting the rate of fuel-air mixing in the vicinity of the flame, combustion can be slowed and can be caused to take place throughout the length of a longer flame. Over the length of this flame, the combustion gases are continually cooled by radiation; as a result of this continuous cooling,‘ peak flame temperature can be limited, thus limiting formation of NO,. 52 o. Limiting. the rate of fuel-air mixing also creates areas within the flame in which oxides of nitrogen already formed. are reduced to nitrogen gas by chemical reactions such as: NO + CO + 1/2N, + CO, By judicious application of these techniques, emissions of oxides of nitrogen can readily be limited to levels below 0.2 lb of NO_ as NO, per million Btu of fuel fired if natural gas is burned or 0.3 1B of NOY as NO, per million Btu of fuel fired if fuel oil is burned. , Particulate emissions from oil and gas fired boilers are inconsequential unless residual fuel oil is burned. In. such cases, use of an electrostatic precipitator can readily reduce emissions to less than 0.03 Ib per million Btu of fuel fired. Water Pollutant Discharges Operation of any boiler will produce wastewater streams, some of which contain pollutants. Oil and gas-fired boilers will produce boiler blowdown which often contains unacceptably large quantities of dissolved metals. Periodically, they will produce boiler cleaning wastes from the chemical cleaning of the inside of the boiler tubes and steam drum. These water pollution problems are common to all boilers. They are not entirely inconsequential; however, they are minor compared to those posed by coal utilization. Oil and gas fired boilers produce no water pollutant discharges comparable to coal pile runoff or coal ash handling waters. : Solid Waste Production Solid waste production of oil and gas fired boilers ranges from minimal in the case of residual oi) firing to none at all in the case of distillate oil and natural gas. Oil and gas firing produce no counterpart of the large quantities of solid wastes generated in firing of coal, wood, etc. Noise Noise generated by oil and gas fired boilers will be dominated by that produced by the fans used to drive gas. through the boiler and pumps used to circulate water through the boiler. These modest sources of noise should: pose no problem outside of the buildings in which these specific machines are housed. Noise level at the facility property line which results from boiler operation can readily be controlled and should be inconsequential. 53 Table 3.2 COMMERCIAL OFFERINGS OF OIL AND GAS FIRED BOILERS BY USA FIRMS Firm Packaged Boilers Field Erected Boilers Aqua-Chem, Inc. Cleaver Brooks Division Babcock & Wilcox Combustion Engineering, Inc. Foster Wheeler Energy Corp. International Boiler Works Co. E. Keeler Co. Nebraska Boiler Co. Riley Stoker Corporation Seattle Boiler Works, Inc. Henry Vogt Machine. Co. Zurn Industries, Inc. Source: Reference 4, updated by x x X X x x x xX X X X X X X x Xx x Xx Xx Arthur D. Little, Inc. 54 r Resource Utilization Oil and gas fired boilers burn their fuel quite efficiently, with losses of unburned combustibles being considered to be completely negligible for most purposes. Deviation of boiler efficiency from 100% (actual obtainable figure approximately 85%) results from inevitable loss of latent and sensible heat in the exhaust gases. 3.2.4 Commercial Status Oil and gas fired boilers are technologically mature. Literally tens of thousands of watertube oil and gas fired boilers are in operation today in the United States alone. Leading American vendors of such boilers are listed in Table 3.2. Constraints to use of oil and gas fired boilers in Alaska are quite similar to those encountered in the Lower 48. These include: e access to natural gas; in many localities, the use of natural gas for some purposes is prohibited, also, the boiler site must be located convenient to a gas pipeline with sufficient gas capacity where available to support such a boiler; and e oil delivery problems; residual fuel oil is shipped mostly by special tanker and barge, which keep it hot enough to be off-loaded by pumping; it is seldom considered a practical fuel at sites which cannot take delivery by water. The principal basis for prohibition of natural gas use as a boiler fuel _— is the Power Plant and Industrial Fuel Use Act of 1978. This act has been used with very limited success to prevent use of natural gas. Its language provides for comparative economic evaluation of gas versus other fuels and other bases for exempting a facility from its requirements. It is not expected to be a factor in determining fuel use in boilers in Alaska. Physical access to natural gas has limited its use as a boiler fuel primarily to the Cook Inlet area. This situation could be changed by the building of a gas pipeline from Prudloe Bay to Fairbanks to the southern coast, bringing gas service to communities along the way. Use of residual fuel oil in boilers will similarly be limited to large installations located on navigable water. 3.3 Economic Implications 3.3.1 Costs Table 3.3 presents the general design specifications for a 100,000 Ib per hour natural gas fired boiler. It raises steam at 1,200 psig, and superheats it to 925°F. These are typical steam conditions for a 55 Table 3.3 CAPITAL COST SUMMARY* TECHNOLOGY: Natural Gas Fired Packaged Boiler 1200 psig/925°F Steam BASIS: Location: Railbelt outside Anchorage Year: 1982 constant dollars CAPACITY: Input: (Natural Gas) 145 million Btu/hr Output: (Steam) 100,000 1b/hr ESTIMATED USEFUL LIFE: 20 years CONSTRUCTION PERIOD: 6 months CAPITAL COST: Equipment and Materials: $ 707,000 Direct Labor: $ 135,000 Indirect Costs: $ 76,000 Home Office Costs: $ 120,000 Contingency: $ 155,000 TOTAL CAPITAL INVESTMENT: $ 1,193,000 *Based on data from References 5 and 6. 56 {3 1.25 1.00 é S75 a > 1 8 \ > % 50 o - z ( °o re 25 r i. 0.00 7 0 50 100 150 200 ‘ 250 300 re Heating Duty, Million’Btu/hr .. . Source: Arthur D. Little, Inc., estimated, based on References 5 and 6. ‘ FIGURE 3.4 CAPACITY COSTS AS A FUNCTION OF STEAM HEATING DUTY: laa SHOP FABRICATED NATURAL GAS FIRED BOILERS i Ls 57 Table 3.4 OPERATING COST SUMMARY TECHNOLOGY : Natural Gas Fired Boiler BASIS: Location: Railbelt outside Anchorage Year: 1982 CAPACITY: 145 million Btu/hr Natural Gas : Input: Output: 100,000 1b/hr steam @ 1200 psig, 925°F OPERATING FACTOR: 65% OPERATING COSTS UNIT COST VARIABLE COSTS: Energy: Fuel: $1.70/Million Btu Electricity: $ 92/MWh Other (E.G., Catalysts and Chemicals) TOTAL VARIABLE COSTS FIXED COSTS: Labor: Maintenance Materials: Taxes and Insurance: Capital Charges: TOTAL FIXED COSTS: TOTAL ANNUAL OPERATING COSTS: TOTAL ANNUAL OUTPUT: TOTAL LIFE CYCLE: 58 877,000 Million Btu ANNUAL ANNUAL CONSUMPTION COST $ 1,410,000 2,480 MWh 230,000 1,640,000 20,000 110,000 560,000 2,200,000 $ 2,210,000 703,000 Million Btu Steam $ 3.14/Million Btu - Steam awe boiler this size which is used for steam topping cogeneration. In raising steam at these conditions, it delivers effective heating duty of 123.4 million Btu per hour. The boiler is shop fabricated. Table 3.3 also presents the estimated capital cost of the complete boiler system. It is broken down into the cost of equipment and materials delivered to the site, site construction labor, site indirect costs, home office costs, and contingency. The costs are compiled based on the assumption of installation at a site in the Railbelt outside of Anchorage at which other process facilities exist. These facilities might be an industrial plant which would use the steam or simply the balance of an electric utility power plant (turbine generator, switch yard, etc.). Total system cost and estimated time to construct are based on construction at a site where other development has taken or is taking place. No allowance is made for site accessing work (building railroads, docks, etc.), and it is assumed that other concurrent activities at the site do not hinder construction. Time and funds are allowed for ground clearing, grading, and foundation work for the boiler and its auxiliaries. Figure 3.4 displays unit boiler capacity cost as a function of boiler heating duty. Steam heating duty is chosen as the basis, since this parameter tends to dictate boiler system cost, little affected by variations in steam flow rates and conditions. As can be seen, gas fired boilers display only modest improvements in unit capacity costs with increasing scale. Table 3.4 projects the total annual cost of owning and operating this 100,000 Ib per hour natural gas fired boiler. The evaluation is made based on a capacity factor of 65%. This might be expected to be consistent with boiler use for power generation in the Railbelt. This boiler will consume 145 million Btu per hour of natural gas and 0.45 kW of electricity. Total annual consumption will be 877 million Btu of natural gas and 2,480 MWh of electricity.(9) Total cost of operation and maintenance, materials, and labor is projected to be $430,000 per year. This is dominated by operating labor requirements. These operating costs, together with capital charges, property taxes, and insurance constitute total lifecycle steam costs of $3.14 per million Btu. 3.3.2 Socioeconomic Factors Installation, operation, and maintenance of such a boiler will require very little technical, clerical, and managerial support. Altogether, operation of the 100,000 lb per hour gas fired boiler considered in the preceding section could be expected to create approximately six full-time jobs at the plant. These will be essentially. blue collar positions, typically filled by hiring high school graduates for training in the plant. 59 Periodic nonroutine boiler maintenance procedures must be performed on the natural gas fired boiler. However, these procedures are sufficiently minor and infrequent that they will not contribute significantly to employment in the state. Installation of this shop fabricated boiler at a site in the Railbelt outside of Anchorage will require a minimum of direct and indirect site labor. Altogether, it is expected to require only about 3% as much direct labor at the site as field erection of the coal fired boiler of the same capacity. Indirect Benefits Operation of a natural gas fired boiler will require purchase of natural gas extracted in Alaska. Payment for such gas will have limited benefits in terms of job creation in Alaska, due to the limited labor requirements of oil and gas extraction. 3.4 Impact 3.4.1 Effect on Overall Energy Supply and Use Oil and gas fired boilers play a vital economic role by serving essentially all commercial and many industrial steam needs.. They have four great advantages over solid fuel fired boilers: e they serve steam needs with maximum reliability and minimum operator attention; @ they provide steam raising capacity at minimum capital cost; e they raise steam while producing minimal environmental impact; and e simplicity of operation, maintenance, fuel procurement, permitting, etc. yields a steam supply which places minimal burdens on the owning and operating organization. In- the Lower 48, and in other locations where oil and gas are available--but at higher prices than solid fuel such as coal--oil and gas fired boilers tend to be used to serve steam demands which: e@ are characterized by relatively low annual capacity factors; 6 must have reliable steam supply as support to some economic enterprise; and e are created by economic enterprise in which the cost of steam is a very small part of total cost. 60 Examples of such applications would be for space heating of commercial buildings and for the space heating and limited process heating demands of many mechanical manufacturing operations. In large installations with high annual steam system capacity factors, the most economical steam supply tends to: be that which is based on . the cheapest available fuel: coal in most of the Lower 48. In the | Anchorage area today, natural gas fired boilers hold this advantage as well. As such, it will be the most economical source of steam for all classes of steam users. Despite this advantage, however, gas fired steam electric power generation is not expected to have a significant future in Alaska. This is because natural gas can be used much more efficiently and economically for base and intermediate load power generation in a combined cycle power plant and for peak load power generation in a simple cycle gas turbine as is being done today in Anchorage and elsewhere (see Chapter 6). In the Lower 48, substantial efforts have been made by government to encourage substitution of solid fuel fired. steam generating capacity for oil and gas fired boilers. Such efforts are based: primarily on the Power Plant and Industrial Fuel Use Act of 1978. .The act contains provision for steam cost evaluation based on site specific availability and price of competing fuels which has the effect of rendering it inapplicable on economic grounds in the vicinity of Anchorage where gas is available. 3.4.2 Future Trends Oil and gas fired boilers can be expected to play an expanding role in industrial and commercial steam raising in Alaska. Much of industrial and commercial enterprise requires steam in quantities and demand patterns which the oil or gas fired boiler is the most appropriate system to serve. As the Alaskan economy grows, more’ oil and gas-fired boilers will be needed to supply commercial and industrial steam at specific sites. For electricity production, oil and gas fired boilers must compete with more competitive technologies and, therefore, are not expected to have a significant future in Alaska or elsewhere (unless residual fuel oil is available at very low prices). For electricity production, oil and gasfired boilers are technologically mature, and cannot be expected to exhibit significant improvements in efficiency or capital costs in the foreseeable future. 61 REFERENCES Babcock and Wilcox Company, Steam: Its Generation and Use, 39th Edition, New York, 1978. Johnston Boiler Company, private communication. American Boiler Manufacturers Association, private communication. R. Schwieger, "Industrial Boilers: What's Happening Today," Power, February 1977. : The Richardson Rapid System: Process Plant Construction Estimating Standards, Vol. 4, Richardson Engineering Services, Inc., Solana Beach, California, 1980. United Technologies Corporation, Cogeneration Technology Alternative Study, Vol. 4, Prepared for NASA Lewis Research Center for the U.S. Department of Energy, NASA CR-159762, 1980. 62 4.0 CONVENTIONAL SOLID FUEL FIRED BOILERS 4.1 Introduction and Summary 4.1.1 Technical Overview This chapter treats the two principal solid fuel fired boiler technologies in use today: fixed bed combustion (commonly referred to as stoker fired combustion) and entrained bed combustion (commonly referred to as pulverized fuel combustion). The most prominent emerging solid fuel fired boiler technology, fluidized bed combustion, is treated separately in another chapter. Stoker fired boilers burn solid fuel particles at rest on a bed of unburned fuel or ash at the bottom of the furnace. This bed is basically fixed with respect to the flow of the combustion air; typically the combustion air is forced upward through the bed. The bed itself is supported by some form of grate. Stokers are classified by the mode of fuel feed to the boiler. Many older boilers rely on pushing the fuel in from the side of the bed, commonly called "“underbed feeding." The most important modern design concepts, however, feed the fuel onto the top of the bed, commonly called "overbed feeding." Utilizing such overbed feed, almost any coarse solid fuel can be burned. Additionally, through proper allowances in furnace design, some modest fraction of the fuel can be burned in suspension above the bed, if all fuel particles are not so large as to settle onto the bed after feeding. This capability to utilize a wide variety of coarse solid fuels makes stoker fired boilers well suited to utilization of wood wastes and refuse derived fuels, as well as coarse particles of coal and peat. The grate serves to remove ash from the bottom of the furnace. Stoker fired boilers offer simplicity in fuel preparation and feeding and other aspects of boiler design which make them particularly attractive at smaller scales. Problems of fuel feed, ash handling, and heat release limit the practical maximum size of stoker fired boilers to approximately 400,000 pounds of steam per hour (many engineers consider the maximum practical capacity to be much smaller than this). (1) The most important class of solid fuel fired boiler in the world today is the pulverized fuel fired boiler. Such boilers can fire coal, peat, or other solid materials which are easily pulverized. The technology is most amenable to and most highly developed for use of coal, by far its most important application. Accordingly, the discussion of pulverized fuel fired boilers in this chapter focusses on coal. Pulverized firing of other fuels will be discussed as appropriate to note important differences from pulverized coal fired boilers. 63 This pulverized fuel (size consistency finer than face powder, with 70% of the feed being particles smaller than 110 microns) burns readily in suspension in its own combustion air. This mode of combustion yields high fuel efficiency, both by producing highly efficient fuel burnout and by allowing utilization of all coal or peat brought into the plant regardless of particle size at delivery. Pulverized coal fired boilers can be scaled up to enormous size. Some have been built to produce up to 10 million pounds of steam per hour. (1) : Despite these differences between the two boiler technologies, both offer the same basic benefits. Both raise steam by burning a solid fuel. These fuels, be they coal, peat; refuse derived fuel, or wood wastes, will be available to the user at prices based on the cost of their production and transport. In many cases, these fuels can be expected to be available to the user at cost-based prices lower than world oil prices. In utilizing. solid. fuels, either of these boiler systems would be creating a demand for such fuels in Alaska. Serving such a demand results in new jobs for Alaskans, more jobs than would result from serving that same demand from Alaskan oil or natural gas. The stoker fired boiler does offer some important advantages over the pulverized coal fired boiler. First, over the capacity range in which it is available, it has a lower capital cost than pulverized coal fired boilers of the same rating. Second, it can readily utilize a variety of fuels, especially cellulosic fuels such as wood waste and refuse derived fuels which cannot be pulverized easily. In utilizing these fuels, a stoker fired boiler can serve to extract energy while reducing the magnitude of a solid waste disposal problem. The pulverized coal fired boiler offers two principal advantages over the stoker fired boiler. First, it offers a substantially higher boiler efficiency when utilizating fuels which are readily pulverized, such as coal and peat. Second, it can: be scaled up to very large size. 4.1.2 Alaskan Perspective Use of solid fuel fired boilers has been limited in Alaska. In addition to the pulverized coal fired boilers in the Fairbanks area and the wood fired boilers at paper plants, there are small stoker fired boilers at- Fort Richardson and elsewhere. . Solid fuel fired boilers may be used more extensively in Alaska in the near future than they are today. This could result from their deployment for one or more of the following purposes: 64 r ™ e electricity utility power generation in the Railbelt; e steam raising for large process industry plants; or e extraction of energy from municipal solid waste in Anchorage and/or Fairbanks. Electric power generation based on solid fuel could have an important impact in the Railbelt. Today, power generation is conducted independently in several non-interconnected electric utility systems. The principal systems, those.serving Fairbanks and those serving Anchorage, will be interconnected by the end of 1984.(2) Then for purposes of generation planning, they. can be considered to be one system accounting for most of the power generation in the Railbelt. Within this system, most of the power is generated today by firing natural gas in. gas turbines and combined cycle generating units. Power has been generated from pulverized coal in the Fairbanks area since the late 1960's.(3) Additional coal fired generating capacity could displace use of natural gas for power generation, reserving it uses where its versatility is more important. Uncertainties in generation planning cloud the prospects for coal fired power plants. Current plans call for only small amounts of new capacity to be deployed between now and the early 1990's. In 1993 it is expected that some power generation capacity in the Susitna hydroelectric project will be commercialized, eliminating the need for additional thermal electric capacity for the balance of this century. Deployment of coal fired power generation within interconnected Railbelt system might offer an alternative to Susitna. : Demands for industrial steam in Alaska are now _ concentrated principally at a few installations: petroleum refineries, an LNG plant, an ammonia plant, and paper mills. Of these, all but the paper mills operate on the premise that hydrocarbon fuels (principally natural gas) are readily available at low prices. It is difficult to imagine natural gas price scenarios which might allow continued operation of an LNG plant or ammonia plant but call for installation of solid fuel fired boilers to provide the required steam for the plant; therefore, it is not expected that solid fuel fired boilers will play an important role in the future of these two installations. Solid fuel fired boilers already provide approximately half of the steam requirements in the state's paper plants, based on wood and wastes. Additional ones may be deployed to serve the balance of this demand and other steam demands of plants which may be sited in Alaska in the future. Opportunities certainly exist for utilization of fuels derived from the municipal wastes of the cities of Anchorage and Fairbanks. Fairbanks currently bales and stacks its waste, a system which many believe is not satisfactory. in the long term. Anchorage landfills its waste, but is expecting to exhaust its landfill capacity in the late 1980's. 65 No basic technical obstacles exist to more extensive use of solid fuel fired boilers in ‘Alaska. Rather, the key question in evaluation of economic viability of such boilers in the Anchorage area--and in the future in the electrically interconnected portion of the Railbelt--is the future price of natural gas. Current natural gas prices per million Btu are lower in the Anchorage area than the projected delivered cost of Beluga coal. If they do not increase in the future at rates substantially faster than other fuels, little economic incentive will exist for solid fuel fired boiler deployment in this part of the state. If coal is an economically viable fuel for Railbelt power generation in the future, the extent of its use will be determined by whether or not the Susitna hydroelectric project is built. If it is’ built, it is possible that no new thermal electric generating capacity of any sort would be needed in the interconnected Railbelt prior to the year 2000. If Susitna is not built, demand for new base load thermal electric capacity, which could be served effectively by coal, could exceed 500 MW by the year 2000. 4.1.3 Significance of Technology The basic technology of both stoker fired and pulverized coal fired boilers is well established. Their use in the Alaskan climate arid: the burning of Alaskan fuels in them pose no new fundamental problems. Both systems are characterized by high capital cost. Unit capital cost displays. substantial economies of scale as system size increases. Consequently, both boilers are most economically attractive when used to serve relatively. high steam demands which are fairly constant throughout the year. Such steam demands are typical of electric utility power generation: and the process industries: chemicals, petroleum refining, paper, etc. Evaluation of a pulverized coal fired system of moderate size (100,000 lb of steam per hour) but high capacity factor (65%) yields an estimated average annual steam cost of $7.04 per million Btu of steam. The importance of solid fuel fired boilers during the balance of this century depends on developments in several related areas: e@ for power generation, the importance of solid fuel fired boilers depends upon the. future price and availability of natural gas for power generation and on whether or not the Susitna hydroelectric power project is built; e for future process industry steam raising, the importance of solid fuel fired boilers depends- upon the availability of natural gas at the sites of new and existing industrial installations and upon how much new process industry will move to the state; and 66 e for extraction of energy from municipal solid waste, the importance of solid fuel fired boilers depends primarily upon solid waste disposal policy; such policy may serve to make utilization economical or may serve to make simple landfilling of all waste more attractive. 4.2 Description of the Technology 4.2.1 Operating Principles Design of boilers to burn solid fuels is determined by many considerations. By far the most important of these is the selection of the solid fuel combustion mode. Three distinct modes can be identified, each characterized by the pattern of interaction of the fuel and the combustion air: e fixed bed combustion / @ entrained bed combustion e fluidized bed combustion Fixed bed combustion is the oldest, simplest, and most familiar mode — of solid fuel combustion. .In this case, large stationary fuel particles remain in a fixed bed. _ Combustion air is’ blown’ over or possibly upward through this bed of particles. Such combustion air draft may be provided by natural convection, as in a fireplace grate, or by forced convection as in cases of stoker firing of solid material on a moving grate in a boiler. : By contrast, entrained bed combustion is characterized by combustion of very small fuel particles in suspension in their own combustion air. This relatively newer mode of solid fuel combustion allows for very high heat release rates. It is practical only for fuels which are readily reduced to very small particle size, such as coal or peat. The third and most complex mode, fluidized bed combustion, is discussed in detail in Chapter 5. This chapter focuses:on fixed bed combustion and entrained bed combustion and the boilers which are designed to use these combustion modes. Each class of boiler is discussed separately, followed by consideration of two problems common to both: control of sulfur dioxide, and principles of air preheater application in the Arctic environment. Stoker Firing Fixed bed combustion in modern boilers depends on use of a mechanical feed system which continuously supplies fuel to the bed and remove ash from the bed. The most common modern systems for this duty are called stokers. In a boiler based on stoker feeding of fuel, commonly referred to as stoker firing, the solids are deposited and burned on a bed supported by a grate at the bottom of the furnace. : 67 . rif] Overfire Coa! Hopper e Air Overthrow ~ Rotor r Air Seal Stoker F Air Seal Chain - aan Ash Hopper Source: Reference 1. FIGURE 4.1 FEED SYSTEM AND FURNACE ZONE OF A SPREADER STOKER FIRED BOILER 68 3 q a E esa = & i esata r Stokers ‘are classified according to how material is introduced into t bed. Many older systems are.based on underbed feeding, in whi the fuel bed is continuously maintained by pushing fresh fuel into from one side at levels below those at which combustion occurs. Th continual feed pushes spent ash out through an opening at the oth side of the bed. These systems are in limited application tod< because: : e the fuel feeding mode can be used only for narrow beds which limit the distance ‘over which the fuel must be pushed; e the fuel feeding system is incapable of supplying fine materi to the bed and keeping it there until it is completely burned; e caking coals will tend to agglomerate in such a bed, plugging it and shutting it down. Overbed feed stokers, by contrast, continuously feed fuel onto the top of the bed from above. The bed is. supported by a moving grate which provides for the removal of spent material from the bed, by vibration, by rotary dump action,. or by its slow continuous movement across the bottom of the bed. ‘ The most important class of overbed feed stoker is the spreader stoker. Older overbed feed stoker designs were limited in terms of ‘fuel selection by inability to burn very fine particles which would become entrained in the combustion air and escape from the furnace zone prior to complete combustion. The modern spreader stoker has been developed to alleviate these limitations by continuously feeding material onto the top of the bed to allow for effective suspension firing of some fuel above the bed. In effect, it combines fixed bed and entrained bed combustion. Figure 4.1 displays the conceptual operation of a modern spreader stoker fired boiler. The fuel is continuously fed into the furnace by the spreader stokers.. The coarser particles settle onto the grate where they devolatilize. Their carbon content remains on the grate and burns slowly there. The volatile matter from the coarser particles and any finer particles introduced into furnace burn in suspension above the bed. At any given time; most of the grate is covered by a layer of ash which insulates it from the hotter burning fuel on the top of the bed. Roughly 80% of the combustion air is fed from below the travelling grate. This sustains combustion in the bed as well as some combustion above the bed. Above the bed, combustion is completed by introduction of the remaining air as overfire air. This allows burning of substantial quantities of the particles above the bed while burning coarser material on the bed. The travelling grate moves at a speed of 10-20 feet per hour, slowly removing burned out ash. 69 Stoker Fired Boiler Sized Coal 02 Boiler Feedwater Source: Arthur D. Little, Inc. Unburned Carryover Recycle to Disposal FIGURE 4.2 Preheated Combustion Air Particulate Control Air Preheater Water Steam STOKER FIRED BOILER PLANT SCHEMATIC Dry Ash to Disposal Passage of most of the combustion air upward through the bed also serves to keep the grate and the bed relatively cool. With only part of the combustion typically taking place on the top surface of the bed, the temperature of this surface is controlled at levels which will tend to minimize the melting of ash and formation of the fused clinkers. Some ash in suspension in the flame will melt, and must be allowed to cool by radiation to the furnace walls so that it resolidifies before it makes contact with convective. heat transfer surface. This requirement sets the height of the furnace. Figure 4.2 is a conceptual diagram of a complete stoker fired boiler system. Air is forced through an. air preheater by a forced draft fan. The preheater air is then forced into the furnace,. where it supports combustion as described above. When the combustion exhaust has reached the top of the furnace (by which time combustion should have been completed and molten ash resolidified), it is further eooled in convective heat transfer surface, and is then sent to a coarse particulate removal system, typically ceramic-lined cyclones. Here, coarse particulate matter which contains substantial unburned carbon is collected to be recycled to the furnace. The gas from the cyclones is routed to the air preheater to heat incoming combustion air. . Finally, the gas is cleaned in a particulate control system (an electrostatic precipitator or'a fabric filter) and a flue gas desulfurization. system (FGD), if required. Then, it is discharged to the environment through a stack. On the steam side, preheated boiler feedwater is introduced first into the economizer in the convective heat transfer section of the boiler. Next, it is passed upward through the vertical tubes which comprise the walls of the furnace itself, where it boils. If the desired steam product is to be saturated steam, more water may be boiled in the convective heat transfer sections of. the boiler. _ If superheated steam is required, the superheating will be accomplished in the convective sections. . Pulverized Coal Firing Figure 4.3 displays a modern pulverized coal fired boiler.. In ‘such a boiler, the fuel is fired in suspension in its own combustion air. Essentially any solid fuel can be burned in suspension, if only it can be reduced to a sufficiently small particle size. Fuel such as coal and to a lesser extent peat can be reduced easily to the consistency of a powder, and are well suited to entrained bed combustion. Materials which are difficult to pulverize, such as municipal wastes and wood, would be much more readily burned in a stoker fired boiler. . . 71 Reheater q Economizer —_———t = TS | | / 1 | | — Air Preheater Nor owe oT aso Source: Reference 4. FIGURE 4.3 PULVERIZED COAL FIRED BOILER 72 Superheaters _—_— Coa! Burners _——— Coal Storage Coal Pulverizers Fuel is introduced into this boiler pneumatically. Typically, it is blown into the boiler through a nozzle using a’ small fraction of total combustion air. Additional combustion air is blown into the boiler through air registers (called "wind boxes") around the burners. Fuel may be introduced into the boiler from one or two walls of. the furnace, or from the corners of the furnace, depending upon boiler design. The furnace is simply an open volume in which combustion takes place.(5) Its walls are constructed of vertical water-filled tubes called waterwalls. It must be sized for essentially complete burnout of the fuel and to allow the spent flame to cool by radiation to the furnace walls to permit resolidification of ‘the ash prior to allowing this ash to come into contact with convective heat transfer surface at the top of the furnace. , : The gas is cooled further in this convective heat transfer surface, then routed to an air preheater where it preheats the incoming combustion air. From there, the exhaust is passed through a particulate control system (electrostatic precipitator or fabric filter). If required by regulation, it may be passed through an FGD system as well. From there it may be discharged to the atmosphere through a stack. On the steam side, boiler feedwater is introduced first into the economizer in the convective heat transfer section of the boiler were it is heated to very near its boiling point. From the economizer it is passed upward through the waterwalls of the furnace where boiling takes place. If the desired product is saturated steam, additional steam raising may take place in convective heat transfer sections of the boiler. If superheated steam is required, steam raised in the waterwalls will be routed to superheaters in the convective portion of the boiler. Flue Gas Desulfurization As noted above, air pollution control. regulations may require application of the system to remove sulfur dioxide from the flue gas. These requirements vary with boiler size, boiler use, and local air quality.(6) Where such systems are required, the most common involve scrubbing the flue gas with an aqueous slurry of lime or limestone. The systems are commonly used to remove approximately 90% of the sulfur dioxide from the flue gas, as required for large utility boilers burning sulfur coal. They are typically applied to treat flue gas which has passed through a particulate control system. The cool, water-saturated gas may require reheating prior to discharge to the stack. 73 Table 4.1 CHARACTERISTICS OF COMBUSTION MODES Characteristic Fuels Fuel Preparation Coal Utilization Fuel Feeding Combustion Efficiency (on Coal) Boiler Efficiency (on Coal) Flue Gas Desulfuriza- tion Particulate Control Capital Cost Capacities Available (1000's 1b steam/hr) * May burn other fuels if a bottom grate is included for coarse material burnup. Fixed Bed Combustion (Spreader Stoker) Coal, Wood, Peat, Refuse-Derived Fuel Size Reduction and Classification Size Limited Fines Utilization Mechanical Stoking 97% 80% As Required by Regulations Yes Approximately 90% of Entrained Bed 10-400 74 Entrained Bed Combustion (Pulverized Coal) Coal, Peat* Pulverization Size Uses Run-of-the- Mine Coal Pneumatic 99.5% 84% As Required by Regulations Yes High Up to 10,000 e iz Utility boilers burning low sulfur coals are required to remove a smaller percentage of flue gas sulfur. When burning very low sulfur coals such as the Beluga coal, 70% SO, removal is required. An alternative which appears to be quite attractive for removal of approximately 70% of the sulfur dioxide from these flue gases is so-called dry scrubbing. This system is typically applied to flue gas after discharge from the air preheater but prior to passage through a particulate control system. Here, an aqueous slurry of lime is sprayed into the flue gas. In contrast to wet lime or limestone slurry scrubbing, however, this low moisture slurry completely dries in the flue gas. Sulfur dioxide is absorbed into the slurry and reacted with the lime during and after the drying of the slurry. This gas is then passed through the particulate control system (typically a fabric filter), which removes the dried sulfated lime at the same time that it removes the fly ash. This gas may then be discharged immediately to the environment through a stack. Unlike so-called wet scrubbing systems, no reheat of the gas is typically required. Combustion Air Preheat When burning. a sulfur-bearing fuel in any boiler, it is necessary to prevent sulfuric acid from condensing on downstream equipment as it is contacted by the boiler exhaust gas. Acid condensation will corrode and severely damage any affected surface. The problem is exacerbated in colder climates. In Alaska, steam (or possibly ethylene glycol heated by steam) should be used throughout much of the year to preheat the incoming combustion air before it enters the counterflow air preheater. This will ensure that the incoming combustion air does not cool the combustion exhaust to below 270°F as it leaves the cold end of the counterflow air preheater. This will serve to prevent sulfuric acid condensation -and the resulting corrosion. . 4.2.2 Technical. Characteristics Table 4.1 lists key characteristics of solid fuel fired boilers and displays how boilers based on fixed bed combustion (represented here by the spreader stoker fired boiler) and boilers based on entrained bed combustion (represented here by the pulverized coal fired boilers) compare. Substantial differences exist between the two types of boilers, which .are discussed below. Despite these numerous differences, however, the two boilers have fundamentally the same characteristics and tend to find economical application in very similar situations. — The stoker fired boiler appears to offer substantially greater fuel flexibility than do boilers based on. entrained bed combustion. Coal, municipal solid waste, wood, peat, and a variety of solid industrial wastes can readily be utilized in such boilers. ‘The only requirement is that the material be available with. sufficiently coarse size consist to 75 allow the bulk of it to remain in place on the bed until it has been largely consumed. The entrained bed combustion system, by contrast, depends upon a very short in-boiler residence time for essentially complete burnout of fuel material. Accordingly, it will operate well only with materials that are amenable to pulverization to fine particle sizes. Resilient cellulosic materials such as wood or municipal wastes have been burned in such boilers. But to accomplish this with acceptable carbon burnout, it is necessary to install facilities which allow for burning of some fuel material on a grate at the bottom of the boiler. These specialized boilers are, in effect hybrid fixed bed/entrained bed boilers. Fuel fed to a stoker fired boiler is prepared by first reducing -the maximum particle size for all material fed to a level which can be handled acceptably by the feed system. In the course of such size reduction, the production of very fine material must be limited. This is because the boiler cannot effectively burn a very large fraction of its feed in suspension above the bed, and because such fine material will not remain in the bed itself and burn there as coarser material would. While this is seldom a problem with cellulosic fuels, it does pose substantial problems in coal utilization. Crushing the coal to the desired size (typically so that the largest particle is smaller than 1-1/4") usually produces too much fine material for use even in a spreader fired boiler. Either an alternate use must be found for the excess fine particles or they must be discarded. By contrast, the pulverized coal fired boiler requires only that essentially all’ particles be reduced to very small size. As such, it is readily capable of using all of a stream of run-of-the-mine coal. When firing coal, the efficiency of a spreader stoker fired boiler tends to be somewhat lower than that of pulverized coal fired boiler. As noted in the table, the efficiency of the fixed bed boiler is approximately 80%, while that-of an entrained bed boiler could be expected to be approximately 84% when burning Beluga coal in an environment with an average ambient temperature of 35°F.(7) The difference between the two boiler efficiencies results from the greater degree of carbon burnout achieved in the entrained bed combustor and the lesser excess air requirement of this system. Both boiler efficiencies are relatively insensitive to ambient temperature, varying less than +1.0% with ambient temperature swings of +40°F. Both boiler efficiencies will decline substantially if a high moisture. cellulosic fuel such as wood is burned. Neither spreader stoker nor pulverized coal fired boilers have an inherent capability to control sulfur emissions within the combustion process itself. Thus, depending upon the fuel burned and the regulations applicable to the boiler, such boilers may require use of flue gas desulfurization to control emissions of sulfur dioxide. This point is discussed further in Section 4.2.3. 76 rm Installations of solid fuel fired boilers in the United States, require controls to limit emissions of particulate ‘matter to regulated levels. Such controls will typically be provided either. by a fabric filter or an electrostatic precipitator. : An advantage of the spreader stoker fired boiler is that it. offers lower capital costs than the pulverized coal fired boiler. In comparison of two systems designed to raise 100,000 pounds of steam per hour, without flue gas desulfurization, the total. capital cost of the spreader stoker fired boiler. system is estimated to be approximately 90% of that of the pulverized coal fired boiler system. Stoker fired boilers are available to raise between approximately 10,000 and 400,000 pounds of steam per hour. ‘This covers the size range of all but the very largest industrial boilers; these boilers are too small to be of interest for new coal fired electric utility power generation in most of the Lower 48. The size range is limited by the grate area which the stoker system can feed: and for which a grate can be built to manage solids flow. The pulverized coal fired boiler, . by contrast, is available in sizes ranging up to 10 million pounds of steam per hour. They may be used to serve medium to large industrial steam demands and electric utility generating needs. Notwithstanding the differences between the two classes of solid fuel fired boilers, comparison of them with oil or gas fired boilers indicates quite clearly that the two solid fuel fired boilers are much more alike than they are different. As Table 4.2 points out, the solid fuel fired boilers share a common advantage with respect to fuel supply: fuel prices tend to be based on the cost of production and delivery to the user. However, the prices of fluid fuels. in most locations will tend to be subject to many political unknowns tied to world oil prices and/or public policy. The former prices can readily be predicted and are expected to increase roughly at the rate of inflation. The latter prices are subject to numerous political unknowns. Additionally, prices of .fluid fuels will be boosted by depletion effects over the next 20-30 years (a reasonable operating life for a boiler to be installed today). This advantage of solid fuel fired boilers must be weighed against their disadvantages. Their capital costs are much higher than those of their oil and gas fired counterparts. While benefiting substantially. from economies of scale at very large size, solid fuel fired boilers are at a particularly great disadvantage in terms of capital cost at small scale... Solid’ fuel fired boilers are plagued with numerous fireside reliability problems which have no counterpart in boilers which burn fluid fuels. These and related problems tend to reduce boiler reliability and to require substantially greater expenditures for operation and maintenance of a solid fuel fired boiler than of an oil or gas fired boiler. . 77 Table 4.2 COMPARISON OF BOILERS BUILT FOR SOLID AND FLUID FUELS Fuel Type Boiler Characteristic Fuels Basis of Fuel Prices Capital Cost Economies of Scale Reliability Operation and Maintenance Difficulty Solid Coal, Peat, Wood, MSW, etc. Cost of Production High Substantial Moderate to High Substantial 78 Fluid (Liquid or Gas) Oil, Natural Gas, Refinery Gas, etc. World Oil Prices, Public Policy Low Moderate Very High Minimal rm m These characteristics collectively define a niche in which solid fuel fired boilers are likely to be economically competitive. This niche consists of serving steam demands which are relatively large and relatively constant throughout the year. Such steam demands tend to occur in electric utility power generation and in providing: process heat in industrial installations which operate seven days per week, 24 hours per day, all year long. In these situations, economies of scale can be exploited most fully. Also, the investment in capital and in | building the technical and managerial capability required to utilize such boilers .will be more fully utilized in plants which run at nearly constant load all year long. 4.2.3 Environmental Issues The environmental impact of solid fuel combustion in stoker or pulverized coal fired boilers can be substantial. The impact of either will be significantly greater than the impacts of burning oil or gas in ‘a boiler. While the: environmental effects of stoker and pulverized coal fired boilers tend to be ‘quite similar, important distinctions between the two do exist and will be noted below. Four classes of environmental impact are of significance: air pollution water pollution solid waste disposal noise The differing capabilities of the two classes of boilers to utilize various fuels efficiently is also of environmental significance. It was treated above in Section 4.2.2. Three classes of air pollutants emitted by solid fuel fired boilers are of concern: sulfur oxides, nitrogen oxides, and particulate matter. Emission beds are determined by fuel composition, boiler design, and application of postcombustion controls. Combustion of Beluga coal in either’ a stoker fired or a pulverized fired boiler will result in emission of a substantial fraction. of the sulfur contained in that coal. . Even if 100% of the sulfur were emitted: as sulfur dioxide, this rate of emissions (0.44 Ib per million Btu) would conform to regulations applicable for all new or existing boilers except new ‘electric utility boilers, without use of flue gas desulfurization. Regulations applicable to new electric utility boilers require that flue gas desulfurization be used in order to achieve 70% control of emissions of sulfur dioxide. Combustion of wood: wastes or municipal wastes in a stoker fired boiler will result in emission of essentially all of the sulfur contained in these fuels as sulfur dioxide. These are inherently low sulfur 79 fuels, however, and their combustion without sulfur dioxide control is expected to be permitted at all facilities which might practically be used in Alaska. By contrast, combustion of natural gas results in practically no emission of sulfur. dioxide. Combustion of distillate oil results in emission of essentially all of the sulfur in the oil as sulfur dioxide, typically about .0.05-0.2 lb per million Btu. Oil sulfur content will be adjusted as needed to meet regulations without use of post-combustion controls. Both stoker and pulverized coal fired boilers are expected to be able to meet emission standards for oxides of nitrogen, with emissions in the range: of 0.4-0.5 lb of NO_, as NO,:per million Btu. Emissions should be somewhat lower if lover nitro én, cooler burning fuels such as municipal waste, wood, or peat are fired. Research is now underway to limit emissions of nitrogen. oxide from pulverized firing to levels significantly below those quoted above, perhaps as low as 0.2 lb per million Btu. Even at this rate, however, emissions will be somewhat higher than those typically achieved in oil and gas fired boilers: 0.2-0.3 Ib per million Btu. Any new stoker or pulverized fuel fired boiler will require particulate control to limit these emissions to regulated levels. With application of electrostatic precipitators or baghouse filters, fly ash generated can readily be controlled to levels of 0.03 lb per million Btu of fuel fired or less. Three classes of solid wastes are produced by pulverized and stoker fired boilers: fly ash, bottom ash, and sludge from flue gas desulfurization (if sulfur dioxide control is required). Fly ash consists of the particulate matter carried unburned out of the furnace _section of the boiler by the exhaust gases. While it may contain some unburned combustible matter, especially in the case of cellulosic fuels, it will consist primarily of noncombustible mineral matter. It is collected dry in the electrostatic precipitator or in the fabric filter. It may then be handled and disposed of either wet or dry. In pulverized combustion of coal, approximately 80% of the ash content of the coal will be carried over as fly ash. In stoker fired coal combustion, generally approximately 25% of the coal ash content will be carried over as fly ash. The fly ash/bottom ash split obtained when burning other fuels is highly variable. Bottom ash is the material removed from the bottom of the furnace section of the boiler. In pulverized firing of coal, typically 20% of the ash content of the coal leaves the boiler as bottom ash. In stoker firing, generally 75% of the coal ash content leaves the boiler as bottom ash. Bottom ash is usually collected in a water quench tank. This wet collection may lead to water pollution control. problems, as discussed below. 80 ) Flue gas desulfurization systems based on wet lime or limestone slurry scrubbing produce a _ wet sludge. While processes exist for converting this sludge into usable gypsum, the most economical systems usually produce this material as a waste. The sludge poses problems with dewatering and with structural stability once placed in a disposal site. Much work has been done to alleviate these problems; however, it cannot be stated with certainty how effectively many of these procedures would work in the Alaskan climate. With so-called dry, lime-based scrubbing, by contrast, the flue gas desulfurization solid waste is produced dry and collected with the fly ash. A stoker or pulverized fuel fired boiler will use large quantities of water. The principal potential for water pollution from this system consists of fuel storage (i.e., coal pile) runoff and discharge of water which has been used in collecting and handling ash and/or flue gas desulfurization wastes. Such discharges can be minimized by maximizing use of dry collection of fly ash and by use of dry flue gas desulfurization if sulfur control is required. These two classes of solid fuel fired boilers generate roughly the same level of noise and with roughly the pattern. They will generate substantially more noise than will an oil or gas fired boiler. The noise will be dominated at the property line by that generated in solid fuel delivery and handling. 4.2.4 Commercial Status Table 4.3 lists American corporations which currently offer stoker or pulverized boilers for sale, together with information on their products. These technologies are highly developed and_ well understood. They are applied throughout the world. No fundamental technical constraints inhibit their application in Alaska. 4.3 Economic Implications 4.3.1 Costs Table 4.4 presents the general design specifications for a modern pulverized coal fired boiler*. Raising 100,000 Ib of steam per hour, it is representative of moderate to large scale industrial boilers. Such a boiler might find application in process industry in Alaska; alternatively it might be used for a small. power plant. The boiler considered raises steam at 1,200 psig, superheating this steam to 925°F. In raising steam at -these conditions, it delivers effective steam heating duty of 123.4 million Btu per hour. * A pulverized coal fired boiler was chosen to allow most consistent comparison with the other coal fired option (fluidized bed combustion) treated in this study. 81 Table 4.3 COMMERCIAL OFFERINGS OF SOLID FUEL FIRED BOILERS BY USA FIRMS Firm Coal Solid Waste Fuels Babcock and Wilcox Co. x xX Combustion Engineering, Inc. x xX Foster Wheeler Energy Corp. x x International Boiler Works x x E. Keeler and Co. , xX x Nebraska Boiler Co. xX Riley Stoker Corp. xX x Seattle Boiler Works, Inc. x Trane Co. : x Henry Vogt Machine Co. x Zurn Industries, Inc. x x Source: Reference 8. 82 Table 4.4 CAPITAL COST SUMMARY TECHNOLOGY: : Pulverized Coal Fired Boiler : BASIS: 1200 psig 3 925°F steam ~ . Location: Railbelt, outside Anchorage | a- Year: 1982 constant dollars i CAPACITY: . Input (Beluga Coal): 10.7 T/hr i Output (Steam): 100,000 1b/hr ESTIMATED USEFUL LIFE: 20 years CONSTRUCTION PERIOD: 18 months { CAPITAL COST: a Equipment and Materials: $4.49 million Direct Labor: $4.24 million Indirect Costs: $2.50 million i Home Office Costs: $1.26 million { Contingency: $1.87 million TOTAL CAPITAL INVESTMENT: $14.36 million 83 Capacity Cost, ¢/Btu/hr 20 = oa 10 0 100 200 300 400 500 Heating Duty. Million Btu/hr Source: Arthur D. Little, Inc., estimated based on Reference 1. FIGURE 4.4 CAPACITY COSTS AS A FUNCTION OF STEAM HEATING DUTY: FIELD ERECTED PULVERIZED COAL FIRED BOILERS 84 600 Pulverized coal firing requires that the boiler furnace zone consists of a large open volume. Such boiler designs are not amenable to shop fabrication or shipment to the site of application. Accordingly, such boilers require extensive field erection work. Table 4.4 also presents the estimated capital costs of this boiler. It is broken down into costs of equipment and materials delivered to the site, site construction labor, site indirect costs, home office costs, and contingency. The cost estimate is compiled based on installation at a site in the Railbelt outside of Anchorage at which other process facilities exist. These facilities might be an industrial plant which will use the steam or simply the balance of an electric utility power plant (turbine generator, switch yard, etc.). The cost estimates were developed based on vendor quotes for the boilers themselves.(9) Total cost of the boiler erected at. the site as well as the cost of auxiliary systems such as coal handling, ash handling, boiler feedwater treatment, etc. were developed based on data published by the United Technologies Corporation in their 1980 Cogeneration Technology Alternatives Study.(10) Data from these sources was used to estimate the actual cost of construction at a hypothetical site in the Railbelt outside of Anchorage. Figure 4.4 displays the variation in capital costs per unit steam raising capacity with boiler size. Calculated for pulverized coal fired boilers without flue gas desulfurization, the curve indicates that substantial economies of scale benefit larger coal fired boilers. Most important economy of scale affects in sizes under approximately 100 million Btu per hour result from the scale-insensitive total cost of solid handling facilities which constitutes a substantial fraction of total plant costs in the small size range. Economies of scale exhibited by larger systems result primarily from savings in scale-up of the steam generator itself. Table 4.4 also provides an estimate of the length of time required for actual field erection of the 100,000 lb per hour boiler. As with the cost figures described above, this estimate is based on the construction of the site where other development has taken or is taking place. No time is allowed for site accessing work (building railroads, docks, etc.), and it is assumed concurrent activities of the site do not hinder construction. Time is allowed for ground clearing, grading, and foundation work for the boiler and its auxiliaries. Table 4.5 projects total annual costs of owning and operating this 100,000 Ib per hour pulverized coal fired boiler. A unit such as this could be expected to be. operated at a very high capacity factor, constrained only by boiler availability and variations in demand for the steam. As discussed in Section 4.2, the boiler would most likely be located where steam demand variability is minimal. It is projected that the boiler could obtain an annual capacity factor of approximately 65%. . 85 OPERATING COST SUMMARY TECHNOLOGY: BASIS: Location: Year: CAPACITY: Input: Output: OPERATING FACTOR: OPERATING COSTS VARIABLE COSTS: Fuel: Electricity: Solid Waste Disposal: TOTAL VARIABLE COSTS FIXED COSTS: Table 4.5 Pulverized Coal Fired Boiler Railbelt outside Anchorage 1982 10.7 ton/hr Beluga Coal 100,000 1b/hr steam @ 1200 0.65 UNIT COST $25.80/ton $92/Mwh $20/dry ton Operation and Maintenance Labor and Materials: Taxes and Insurance: Capital Charges: TOTAL FIXED COSTS: TOTAL ANNUAL OPERATING COSTS: TOTAL ANNUAL OUTPUT: TOTAL LIFE CYCLE COST: 86 ANNUAL CONSUMPTION 61,000 tons 4,800 MWh 14,000 tons psig, 925°F | ANNUAL cost 1,570,000 440,000 280,000 2,290,000 1,011,000 290,000 1,360,000 2,660,000 $4,950,000 703,000 Million Btu Stear $7.04/Million Btu Steam Firing Beluga coal, this boiler would use 10.7 tons of coal per hour and 0.84 MW of electricity, totalling 61,000 tons of coal a year and 4,800 MWh of electricity per year. A total of 14,000 tons of ash would be produced annually. Total cost of operation and maintenance, labor and materials is projected to equal 7% of total capital costs. These operating costs, together with capital charges, property taxes, and insurance constitute total annual cycle steam costs of $7.04 per million Btu. If this steam were used for central station power generation, the power would cost approximately 8.6¢/kWh. Of this, 6.7¢/kWh would be accounted to the cost of the steam. It should be noted that the boiler design analyzed here includes no allowance for flue gas desulfurization. Regulations on emissions for sulfur dioxide will allow burning of low sulfur coal such as that available in the Beluga deposit without it. Inclusion of a dry flue gas desulfurization system would increase steam costs by approximately 308. 4.3.2 Socioeconomic Factors All together, installation of the 100,000 lb/hr boiler considered in the preceding section would create approximately 15 full-time jobs at the plant. This compares to six full-time jobs which would be created to operated and maintain a gas fired boiler with the same capacity. Technical support is provided by engineers hired directly from college with a Bachelor's degree in mechanical engineering. Practice in the Lower 48 typically results in filling the managerial positions which would be created by promotion of these engineers. Clerical positions typically are filled by college graduates with degrees in accounting. Blue collar positions typically are filled by hiring high school graduates for training in the plant. A solid fuel fired boiler periodically requires extensive maintenance. Typically, this is conducted by skilled craftsman (steam fitters, boilermakers, etc.) on a contracted basis. These highly skilled. personnel typically develop their skills within a union organization. When not engaged in boiler maintenance, they may work elsewhere in boiler construction and related activities. Construction of solid fuel fired boilers likewise produces substantially more temporary construction jobs than does construction of oil or gas fired boilers of the same size. This results because: e auxiliary systems required for a solid fuel fired boiler (fuel handling and storage, air pollution control, waste disposal, etc.) require substantial field erection labor; they have no analog in oil and gas fired boiler; and 87 e erection of the fuel preparation and steam generation system itself requires substantially more labor in the solid fuel fired case because in all situations the boilers involved are larger, and because in all but the very smallest sizes the solid fuel fired boilers must be field erected; in contrast, oil and gas fired boilers are available in factory-fabricated package designs in capacities up to approximately 400,000 lb of steam per hour. In the case of the 100,000 Ib of steam per hour boiler considered in preceding section, it is estimated that boiler erection would require more than ten times as much direct and indirect field labor and supervision as would erection of a gas fired package boiler of the same size. Indirect Benefits Operation and maintenance of a solid fuel fired boiler requires purchase of solid fuels mined, harvested, or recovered in Alaska. The delivered price of such fuels as coal, peat, and waste wood typically are cost based. As such, payments for these fuels largely flows through as payments to personnel involved in mining, harvesting, and transporting them. The relatively labor intensive nature of coal mining, wood harvesting, and peat harvest could be expected to create more jobs than oil and gas extraction, per unit of fuel energy. 4.4 Impact 4.4.2 Effect on Overall Energy Supply and Use Stoker fired boilers and pulverized coal fired boilers can be expected to have most application in relatively large scale, high load factor installations. In the Lower 48, typical applications would include power generation and steam raising in the process industries. In Alaska, the primary application would probably be for power generation of the Railbelt, and/or in very large industrial installations which might be built in the future. Power generation outside the Railbelt is generally conducted in response to demands which are too small and variable to support an economically sized baseloaded power generation system which might fire a solid fuel. Commercial steam users tend to be too small and have capacity factors which are too low to make solid fuel based systems competitive with oil and gas fired boilers. bs In such applications which favor solid fuel fired boilers, stoker fired boilers are likely to be the preferred choice when: 88 c e the steam demand to be served does not exceed approximately 100 million Btu per hour. In this size range, the difference between the capital cost of the stoker fired and pulverized fired systems is most significant; or : @ an opportunity exists to utilize inexpensive wood or refuse derived fuel. Stoker fired boilers are much more amenable to firing material such as refuse derived fuel and wood wastes than are pulverized fired boilers. Such applications might include production of steam for heat or power generation in Anchorage, Fairbanks, or in the forest products industry. Filling smaller steam demands, they may contribute importantly to economic viability of some industrial enterprises in Alaska, particularly in the forest products industry. By providing a means~ of extracting energy from solid wastes while simultaneously reducing their volume, they may contribute to economical alleviation of solid waste disposal problems. In these roles, however, it is unlikely that they will contribute as much as 1% to state energy supply. Pulverized coal. fired boilers could be expected to find application to fill relatively larger steam demands in the Railbelt, especially for power generation. They will be used where waste fuels are unavailable or offer insufficient supply to contribute meaningfully to fuel requirements. For. this purpose they might be built to fire coal or peat. In this role, they could contribute substantially to Alaska's overall energy supply, reserving natural gas and oil for higher value use and/or for shipment out of the state. 4.4.2 Future Trends Consideration of future prospects for use of solid fuel fired boilers in Alaska requires treatment of three application classes: : e. Moderate to large industrial boilers; e Coal fired utility power generation; and e Opportunistic waste utilization. Significance of solid fuel fired boilers in the first class depends upon future location of new process industrial plants in Alaska. Future requirements for thermal power generation in the Railbelt depend critically upon whether the Susitna hydroelectric project is built. Stoker fired boilers could be used to extract useful energy from municipal wastes in Anchorage and Fairbanks; economic viability will be driven by solid waste disposal policy, a complicated set of technical, economic, and political issues not related to energy. 89 Prospects for general industrial and utility use of solid fuels in the immediate vicinity of Anchorage are clouded by uncertainties in future natural gas prices. At today's prices, natural gas is a far more economical source of steam energy in this area than any commercial solid fuel could possibly be. Such gas prices act less directly to deter burning of fuels extracted from municipal wastes, as well, by lowering the sale value of the steam which is generated. 90 7 10. REFERENCES Babcock and Wilcox Company, Steam: Its Generation and Use, 39th Edition, New York, 1978. Aleska Power Authority, "“Anchorage-Fairbanks Transmission Intertic," Appropriate Energection, Vol 5, No. 1, Division of Energy and Power Development, Department of Commerce and Economic Development, July-August-September 1982. Hank Nikkels, et al, Municipality of Anchorage, private communication. Riley Stoker Corporation, private communication. Joseph G. Singer, Ed., Combustion: Fossil Power Systems, Combustion Engineering, Inc., Windsor, Connecticut, 1981. Title 40 Code of Federal Regulations Part 60, New Stationary Sources Performance Standards, Electric Utility Steam Generating Units. FR 44: 113, pp. 33580-33624. June 11, 1979. Davy McKee, Inc., Coal Methanol Feasibility Study: The Beluga Methanol Project, prepared for the Cook Inlet Region Inc., and Placer Amex, under DOE Grant DE-FG01-80RA-50299, 1981. R. Schwieger, "Industrial Boilers: What's Happening Today," Power, February 1977. : Confidential communication. United Technologies Corporation, Cogeneration Technolo Alternatives Study, Vol. 4, prepared for NASA Lewis Research Center for the U.S. Department of Energy, NASA CR-159762, 1980. ‘ 91 5.0 FLUIDIZED BED COMBUSTION 5.1 (Introduction and Summary 5.1.1 Technical Overview Fluidized bed combustion is the burning of a fuel in a churning bed of hot solid particles, set in motion by blowing air upward through it. The bed of solids under these circumstances behaves in many ways like a turbulent liquid. It is said to be fluidized. Fluidized bed combustion has been employed for incineration of some low grade liquid and gaseous fuels, but its applications of most commercial importance are for combustion of solid fuels. Likewise, heat from fluidized bed combustion has been put to two principal uses: generation of hot gas and steam raising. However, the latter is more important technically and economically. Accordingly, the focus here is on solid fuel fired fluidized bed boilers. Combusting solids in the fluidized bed as opposed to other possible methods offers several advantages. First, heat transfer within the bed itself is excellent. For example, steam tubes within the bed can extract more heat per. foot of tube than is possible in boilers of traditional design. This leads to smaller and less expensive boilers. By placing sufficient heat transfer surface within the bed, the temperature of combustion can be controlled. If this is done at about 1,550°F, formation and subsequent emission of nitrogen oxides are maintained at much lower levels than would be the case in conventional uncooled combustion. Further, if sulfur emissions emanating from the solid. fuel must be limited, this can be accomplished by burning the fuel in a bed composed. primarily of limestone, which will absorb most of the sulfur dioxide as it is formed. Fluidized bed combustion offers other advantages over its conventional alternatives in addition to those described above. Compared to stoker firing, fluidized bed combustion can achieve much higher boiler efficiencies. Compared to pulverized fuel firing, fluidized bed - combustion is much better adapted to burning alternate fuels such as wood waste, peat, and refuse derived fuel. Also, since bed temperatures are controlled at levels below the fusion temperature of most fuel ashes, . fuel-specific ash properties are _ relatively unimportant ‘to fluidized bed combustion. This facilitates fuel switching and also makes the boiler more forgiving of excursions in ash properties of fuel from a single source. 93 5.1.2 Alaskan Perspective Data available indicate that Alaskan coals are very well suited to fluidized bed combustion.(1) With their low sulfur contents and relatively -high ash contents, it is expected that sulfur control required by federal regulations for utility boilers can be accomplished within the bed solely by sulfation of calcium compounds available in the ash itself. Additionally, fluidized bed combustion enjoys an advantage over pulverized coal combustion or stoker fired coal combustion in simple compatibility with use of coals which contain large percentages of ash. : Fluidized bed combustion has ancilliary benefits of extracting energy from some waste streams, notably municipal wastes in Anchorage and possibly Fairbanks and wood wastes in forest product operations in the Southeast. But supplies of these fuels are limited; their use will mitigate a solid waste disposal problem but have little impact on overall Alaskan energy use. Prospects for substantial application of fluidized bed combustion have been limited by its technical immaturity. Commercial units have been operated which have steam generating capacities up to 200,000 ib steam per hour (a 300,000 lb per hour demonstration unit is now shut down), but no such units have been operated for more than a few years. Thus, design and operation of these units involves’ more uncertainty than do stoker or pulverized coal fired boilers. Additional uncertainties are posed by system scale-up to larger sizes. Current commercial offerings of fluidized bed boilers range up through 1.1 million Ib of steam per hour (capable of supporting approximately 150 MW of power generation).(2) This scale would be of interest for utility applications in Alaska if well demonstrated, but it represents boiler scale-up by roughly a factor of four from any commercially operated system. To date, no orders have been placed for boilers of such size. 5.1.3 Significance of Fluidized Bed Combustion Technology The technical feasibility of construction and operation of fluidized bed boilers is not in doubt. They are now being demonstrated commercially up through capacities of approximately .200,000 lb of steam per‘hour, and recent sales indicate that market acceptance is broadening. This complete size range covers most industrial applications that would probably be of interest in Alaska for the balance of this century. They are offered by boiler vendors with conventional commercial guarantees up through sizes of 1.1 million lb/hr but near-term demonstration of such boilers is in doubt. Boilers of this size might be expected to be applicable for utility power generation in Alaska, if developed. No Alaska-specific technical problems of fluidized bed combustion are anticipated. A fluidized bed boiler designed for use in Alaska will certainly differ from one designed for use in, say, Texas. But such 94 differences are minor and principally concern issues such as prevention of air preheater corrosion and of system freeze-up, issues common to all solid fuel fired boilers, not just those based on fluidized bed combustion. Economically, fluidized bed boilers are characterized by high capital costs with good economies of scale, reliable and low cost fuel supply, and substantial maintenance requirements and operational difficulties. In all of these respects, they are characteristic of all solid fuel fired boilers. They contrast sharply with oil and gas fired boilers, which are characterized by low capital cost, high fuel costs, and ease of r operation and maintenance. Each of these two boiler groups has a og. niche among boiler applications: q e solid fuel fired boilers are favored in large-scale applications in which steam demand allows the boiler to attain a high’ (>50%) capacity factor on an annual basis. Such large, nearly constant steam demands are typical of utility baseload power generation and large process industry installations. vt. Such applications: seek to distribute capital costs over. as high an output as possible (as a fraction of the theoretical annual Gr capacity). Economics are also more favorable for larger : : boilers which have lower capital cost per unit of capacity, as a result of economies of scale. I 4 e fluid fuel (i.e., oil and gas) fired boilers are favored to ' serve smaller, more variable steam demands, which result in lower capacity factors. Having the basic economic characteristics of a solid fuel fired boiler, the fluidized bed boiler will make few new in-roads into fluid fuel boiler dominated markets. Rather, it can be expected to find its main applications in situations which traditionally have been served by solid fuel fired boilers, in which fluidized bed combustion offers several important economic advantages over its competition: @ compared to stoker fired boilers, greater boiler efficiency; e@ ability to use all of the crushed coal supplied to it, in contrast with a stoker fired boiler which cannot generally utilize all of the fine particles produced when coal is crushed to a usable maximum particle size; @ compared to pulverized coal combustion, better suited to burning a mix of coal and materials which do not readily pulverize, such as wood or refuse derived fuel; “@e where in-plant sulfur control technology is required, substantial capital cost savings; and @ compared to coal combustion by other methods, lower emissions of nitrogen oxides. 95 A 100,000 Ib per hour fluidized bed boiler system evaluated here for use at a site in the Railbelt outside Anchorage could be expected to supply steam at a total cost of $7.55/million Btu. To summarize, fluidized bed boilers offer incremental advantages over pulverized or stoker fired boilers. Although these advantages do little to change the economic competition between solid fuel fired boilers and oil and gas fired boilers, fluidized bed combustion may offer significant benefits for Alaska if future patterns of economic development and energy use in the state give rise to site-specific demands for large quantities of steam throughout most of the year, i.e., for power generation or for industrial process use. Demand for such boilers for power generation during the balance of this century depends primarily upon whether or not the Susitna project is carried out. It also remains to be seen when boilers of economical utility scale for use in Alaska will be developed and commercially demonstrated. Requirements for such boilers in industry depend upon future decisions in the process industries to locate large facilities in Alaska. , Opportunities may also arise to use fluidized bed combustion to recover energy value from wastes. As noted above, this may contribute importantly to waste disposal, but will not significantly alter state energy use patterns. 5.2 Description of the Technology 5.2.1 Operating Principals Solid . fuel combustion can be classified into three categories, depending upon the pattern of interaction of the fuel and the air that - is used to support combustion. The first mode, the oldest, simplest, and most familiar, is fixed bed combustion, where the fuel particles are essentially stationary, with combustion air being blown over or upward through the fixed bed of particles. This may happen by natural circulation, as on a simple fireplace grate, or by forced circulation in a stoker fired boiler. Thus, this system is characterized by relatively large fuel particles which are not put in motion at all by the air which is circulating around them. The second mode is called entrained bed combustion, which is used today in large coal fired boilers. In this case, the coal is pulverized into very fine particles and burned in suspension as they move along with high velocity combustion air. This, then, is a case of very small particles whose motion is determined almost entirely by the gas stream lines. ‘ Between these two regimes--one with very large fuel particles unaffected by low velocity air and the other with very small fuel particles entrained in their own combustion air--there exists an 96 Ve intermediate regime of particle/gas interaction. Here gas passing at moderate velocities upward through a bed of particles about the size of sand will cause this bed to behave as a turbulent liquid. Such a bed is called a fluidized bed. Particles are set in motion by the gas but do not become entrained in it. They stay within the bed, with their patterns of movement determined by interaction with the gas and by collisions with other particles. : A fluidized bed combustor exhibits several characteristics which. distinguish it from other types of solid fuel combustors and which offer potential significant advantage. At their root is the thorough mixing of the material in the bed. The gas moving upward through the turbulent bed causes the particles to be mixed thoroughly, both laterally ..and vertically, resulting in near homogeneity of material _ composition throughout the bed. This highly promoted solids mixing also leads to temperature uniformity. Placement of steam tubes in the bed allows heat to be transferred from the bed to the steam tubes at very high heat transfer rates. This has the collateral benefit of controlling bed temperature by extracting a controlled amount of heat directly from the bed. Such in-bed steam raising is depicted in Figure 5.1. Some small fluidized bed boilers achieve such temperature control without in-bed steam tubes, simply by using large . quantities of excess air. This inefficient technique, suited to waste incineration, is inappropriate -in most applications where a purchased fuel is used; accordingly, this report focuses on boilers with in-bed steam tubes. Bed temperature must be controlled within an interval which has a technical ceiling and an economic floor. The technical ceiling is set by the requirement that: the coal: ash must remain solid. A higher temperature will result in melting of the coal ash, which in turn will result in fusing of the coal ash to the in-bed heat transfer surface on contact. The economic temperature floor is established by requirements for: practical levels of carbon burn-up.. The ceiling is approximately 2,000°F. The floor is ill-defined, but is definitely below 1500°F. Controlling combustion temperatures in this interval greatly reduces formation of thermal NO,. A variety of in-bed reducing mechanisms are believed to promote feduction reactions such as NO, + xCO + 1/2N, + xCO, to destroy oxides of nitrogen formed from fuel bound nitrogen. Thus, total emissions of oxides of nitrogen are sharply limited compared to those observed in conventional coal combustion. 97 STOKER FEEDER STEAM TUBE AIR Source: Reference 2. FIGURE 5.1 CONCEPTUAL ARRANGEMENT OF OVERBED-FED ATMOSPHERIC FLUIDIZED BED BOILER 98 wa as If bed temperature is controlled in the range of approximately 1,400°F to 1,700°F and if the bed contains calcium oxide, most of the sulfur content of the coal can be controlled by its in-bed reaction with the ealcium oxide to form calcium sulfate. This calcium oxide may be provided by a separate feed of limestone; alternatively, if calcium content of the coal ash is adequate, simple presence of this ash in the bed may serve for adequate sulfur control. With many if not all Alaskan coals, it is projected that ash calcium content will be more than adequate to effect 70% sulfur capture, the level required by federal law for the most stringently regulated class of boilers, i.e., those in new utility power plants. , An added benefit of the low combustion temperature is that it prevents the coal ash from melting. and limits the volatilization of the alkalis in the ash. Prevention of ash melting means that boiler components encounter crystaline ash, which is much less erosive than ash in the form of fused glass spheres. Limiting alkali volatilization reduces problems: of boiler fouling and _ corrosion. These are particularly important advantages for fluidized bed combustion .when consideration is given to burning high ash fuels such as the Beluga coal. Figure 5.2 displays schematically the flows. of materials around a fluidized bed boiler. Coal is crushed, typically to -1/4 or -1/8" and fed to the bed. Various designs call for feeding overbed via -spreader stoker or for in-bed pneumatic feeding. As needed, | limestone may be fed either with the coal or via a separate system. How fine the coal (or other fuel) is crushed varies with system design. To minimize crusher capacity requirements and carryover of unburned material out of the bed it is generally desirable to feed the material into the bed as ‘coarse as possible, subject to feed system constraints and the need to introduce no particle into the bed which could lodge between adjacent in-bed tubes. If peat were burned instead, it could be expected to be fed to the bed via systems quite similar to those used for coal. For burning wood or refuse derived fuel, overbed feeding would be preferred. If multiple fuels are to be fired, a separate feeding system would probably be required for each, owing to significant differences in the mechanical properties of these fuels. Air is driven through the air preheater by a forced draft fan. In the air preheater, the air is heated to perhaps 600°F using boiler exhaust gases as the source of heat. The preheated air is then forced upward through the fluidized bed. : In the bed, fresh fuel and_ sorbent material (if required) are intimately mixed with combustion air and the already hot bed solids composed mostly of sorbent and burned out fuel ash. This mixing 99 STORAGE AND COAL FROM DISTRIBUTION SYSTEM INET COAL FEED SILENCER SYSTEM F.D. FAN 1.0. FAN WATER AIR HEATER ae AIR HEATER! S S CYCLONE \ SPENT BED i ( \ ) 250 F WATER FROM MATERIAL BFW. PUMP BOILER FEEDWATER COOLER SYSTEM ASH TO DRY WASTE DISPOSAL Source: Reference 3. FIGURE 5.2 COAL-FIRED ATMOSPHERIC FLUIDIZED BED STEAM GENERATOR | \ \ i= produces nearly isothermal conditions, at a temperature controlled at approximately 1,550°F. The fuel devolatilizes almost immediately upon entry into the bed, with rapid volatile combustion in the vicinity of its evolution. The remnant char burns out more slowly, mixing through the bed as it burns. Combustion gas leaves the top of the bed, carrying with it the fine bed particles, which are in turn removed. from the gas in a cyclone for ‘recycle to the bed. Next,. the gas flows through a series of heat exchangers, economizers, and air preheaters to recover thermal energy and from there through a baghouse filter for removal of practically all remaining particulate matter. An induced draft fan draws it up the stack. On the steam side, boiler feed water. moves first through the economizer, where it is heated as it cools the exhaust gases. Next, it is passed upward through the walls of the combustor. These walls are themselves constructed of vertical boiler tubes, called water- walls. In many designs, the water absorbs enough heat in the economizer and waterwalls so that saturated steam is produced. This steam then flows through serpentine tubes imbedded in the bed itself, absorbing further heat to superheat the steam. In. other designs, tubes in the bed are also used for evaporation. The cold Alaskan climate makes it necessary during much of the year - to use steam (or possibly ethylene glycol) to preheat air prior to its introduction to the air preheater. This may be required in order to limit the cold end temperature on the exhaust side of the air preheater. The temperature here must be maintained at or above 270°F in order to prevent condensation of sulfuric acid, which leads to severe corrosion if it is allowed to occur. Boiler efficiency calculations and boiler. cost analyses related elsewhere in this section reflect use of such air preheating. 5. 2 -2 Technical Characteristics Fluidized bed boilers are now marketed based on two modes of construction: shop fabrication and field erection. Shop fabricated boilers are sold in sizes up to only 40,000-50,000 lb steam/hour.(5) Boilers with capacity greater than this are field erected. Units up to 1.1 million Ib/hour are now marketed, although the largest boiler built to date has a capacity of 300,000 lb/hr (a demonstrative unit, now shut down). Larger boilers which will have capacities of several million pounds of steam per hour are being developed for utility application . The mean boiler efficiency expected for these fluidized bed boilers is 84% (based on Beluga coal and a 35°F ambient temperature). Efficiency is relatively insensitive to ambient temperature, . varying to +0.5% with ambient temperature swings of +75°F. It is more sensitive to fuel hydrogen and moisture contents, declining by 20% or more if green wood (50% moisture) is fired. These patterns of variation are 101 COMPARISON TABLE 5.1 OF FLUIDIZED BED AND OTHER BOILERS Fuel feeding Fireside components Solid waste disposal Operating, administra- tive labor force Tube failure FGD Particulate removal FBC Large mechani- cal equipment, conveyors Complex, need operating data Single major discharge More No track record ‘ Accomplished in bed Yes 102 Other Solid Fuel with FGD Large mechani- cal equipment, conveyors Demanding environment Fly ash, FGD sludge Most Experience used as guide Yes Yes Oil/Gas Tanks, pipelines Relatively benign environment Not required Least Mostly due to waterside prob- lems No No typical of all solid fuel fired boilers, and especially comparable to those obtained in a pulverized coal boiler firing the same coal. However, the fluidized bed boiler enjoys the advantage of being able to burn a wider variety of fuel than the pulverized coal boiler. It generally achieves greater boiler efficiencies than stoker units fired with the same fuel, owing to higher carbon burn-out and lower excess air. Table 5.1 summarizes the salient characteristics of fluidized bed boilers and their competition. As noted, .a fluidized bed boiler is unquestionably more complicated to operate and will have more reliability problems than an oil or gas fired boiler. Its operation requires the handling of solid feeds and solid wastes, using systems far more prone to failure and requiring much more operator attention than simple oil and gas pipelines. It certainly will experience fireside damage to components, especially. within the bed themselves,. which have no analog in oil or gas fired. boilers. Fluidized: bed combustion operation necessitates solution of a solid waste disposal problem; of course, there is no analog of this problem with oil or gas firing. Design and construction of a coal fired boiler which will be easily operated and maintained requires much more attention from owner or consultant technical personnel during design and construction. Its eventual operation will require more operators, more engineering support, and more clerical support for fuel supply management than will an oil or gas fired boiler. , Comparison of operability and reliability of fluidized bed boilers with that of other solids fired boilers is less certain. ll solids: fired boilers require installation and operation of solid fuel handling and receiving facilities, particulate control systems, and solid waste disposal systems. These could be expected to cause similar if not identical problems for each class of solid fuel fired boiler. technology. The boilers differ, however, in systems which cause key reliability problems. In particular, the single largest cause of boiler shutdown, tube failure, results from design problems which differ so much from one class of boiler to another that only experience can serve as a guide. Similarly, final solid feed preparation and introduction into the combustor differ sharply among solid fuel fired _ boilers. Experience with fluidized bed boilers is sufficiently limited and the technology is sufficiently immature: that it is not possible to evaluate the comparative reliability of fluidized bed boilers versus other types, at least insofar as reliability problems are caused by systems which are unique to the fluidized bed. In situations where in-plant sulfur control is required, the fluidized bed boiler will enjoy one major advantage compared to other solid fuel fired boilers: it will not require flue gas desulfurization. Elimination of this system will improve plant reliability, decrease plant operational complexity, and simplify solid waste disposal. Collectively, it is likely that these advantages will make a fluidized bed boiler simpler to 103 operate and more reliable than any other sort of solids fuel fired boiler with flue gas desulfurization. On the other hand, in the near term, fluidized bed boilers can be expected to continue to suffer more operational problems than conventional solids fired boilers due to immaturity of design. The long-term operability and _ reliability comparison cannot be made at this time. 5.2.3 Environmental Impact Fluidized bed combustion of coal or of other solid fuels will have several distinctive environmental impacts. In most respects, these impacts will.be greater than those made by oil or gas-fired systems. Accordingly, it is appropriate to compare fluidized bed solid fuel combustion first with combustion of solid fuels in other types of boilers and second with combustion of oil and gas in- analogous systems. This comparison must be made in turn for each of several . classes for environmental impact: e air pollution, e water pollution, e solid waste disposal, and e noise emissions. In addition, fluidized bed boilers must be compared with other solid fuel boilers in terms of efficiency of resource utilization. Three classes of air pollutants emitted by fluidized bed boilers are of concern: sulfur oxides, nitrogen oxides, and particulate matter. For each pollutant, it must further be noted. which’ boilers require pollutant controls in order to achieve regulated emissions limits and what emission levels are obtained below regulated levels in cases where uncontrolled emissions do not exceed regulated limits. Fluidized bed combustion of Beluga coal will result in emissions of sulfur dioxide of about 0.13 Ib/million Btu. This is based upon the expectation that the alkaline content of the coal ash will absorb 70% or more of the coal sulfur, remaining in the bed as calcium sulfate. If this does indeed occur, then no further control of sulfur dioxide emissions need be applied in order to meet current or proposed emissions standards. By contrast, stoker fired boilers or pulverized coal boilers without flue gas desulfurization would emit as much as 0.44 lb of sulfur dioxide per million Btu of Beluga coal burned. This emissions level may readily be reduced to that obtained with fluidized bed combustion by application of flue gas desulfurization, a costly process. . . 104 Combustion of wood wastes or municipal wastes. in a fluidized bed might be expected to result in emission of nearly all of the sulfur contained in those fuels, unless limestone were added to the bed to effect SO, control. These are inherently low sulfur fuels, however, and their”combustion without sulfur dioxide control is expected to be permitted for all size facilities which might practically be used in Alaska. By contrast, combustion of natural gas results in practically no emission of sulfur dioxide. Combustion of distillate oil will result in emission of essentially all of the sulfur in the oil, typically approximately 0.05-0.2 Ib/million Btu. Oil sulfur content will be adjusted as needed to meet regulations without use of postcombustion controls. : Fluidized bed boilers enjoy a significant edge compared to other coal fired boilers in that they limit emissions of NO_ to 0.3 1lb/million Btu. Stoker or pulverized coal fired boilers firiné Beluga coal may be expected to emit roughly 0.4-0.5 Ib/million Btu. All of these figures are within federal limits. Of course, lower levels of SO, emissions are possible in oil and gas burning boilers. Also, extensive research is underway which can be expected to lower the level of emissions achieved in pulverized firing of subbituminous coals, thus eliminating this advantage of fluidized bed combustion. Particulate emissions from fluidized bed boilers would be quite high were particulate control technology not applied. Use of electrostatic precipitators or more probably baghouse filters will effectively limit particulate emissions to levels of 0.03 Ib/million Btu or less. Essentially the same levels of particulate control might be applied to other solid fuel fired systems. Any solid fuel boiler will produce solid wastes. Unlike the other solid fuel boilers considered here, however, all solid wastes produced by a fluidized bed boiler are available dry. Accordingly, no aqueous © streams contaminated by soluble species from the solid wastes may be generated and discharged. Indeed, the solid wastes produced by a fluidized bed boiler may consume wastewater (for dust suppression) generated by other on-site facilities. The quantity of solid waste produced by a fluidized boiler should be very nearly identical to those produced by other solid fuel boilers, given Alaska fuel properties. The one aspect of solid fuel utilization which tends to generate quantities of waste which vary from technology to technology is sulfur dioxide control. As mentioned above, given the very low sulfur content and high ash content of - Alaskan coal, it is expected that calcium required for sulfur control in any coal fired boiler considered would be derived from the ash. Boilers fired with natural gas or distillate oil produce no substantial quantities of solid wastes. They also produce no solid waste-related water pollution problems. 105 TABLE 5.2 BUILDERS OF CURRENTLY OPERATING COMMERCIAL FLUIDIZED BED BOILERS Company A Ahlstrom Oy Babcock and Wilcox Co. Babcock Hitachi KK C-E Power Systems Danks of Netherton, Ltd. Deborah Fluidized Combustion, Ltd. Dedert Corporation Deatche Babcock Energy Products of Idaho Energy Resources Company Fluidized Combustion Contractors, Ltd. Foster Wheeler Boiler Corporation Foster Wheeler Power Products, Ltd. Generatory Industri AB E. Green & Son, Ltd. THI, Ltd. Johnston Boiler Co. E. Keeler Co. ME Boilers, Ltd. NEI Cochran, Ltd. Struthers Wells Corporation Tampella, Ltd Wormser Engineering York - Shipley,- Inc. Source: Reference 2. Country Finland USA Japan USA Britain Britain USA W. Germany USA USA Britain USA Britian Sweden Britain Japan ’ USA USA Britain Britain USA Finland USA ~ USA 106 Commercial Boilers Operating Today Total Capacity Capacity Range Number 1000 1b/hr 1000 1b/hr 6 520 24-200 2 140 20-120 2 66 22-44 1 50 50 4 56 10-20 3 20 4-10 2 30 10-20 7 1,552 17-352 22 771 10-150 2 70 20-50 2 120 60 6 410 40-110 2 75 20-50 3 95 17-44 1 10 10 1 "69 69 14 515 10-50 1 24 24 1 80 80 6 43 4-15 2 55 5-50 © 3 103 19-44 1 70 70 12 270 7-55 28 A fluidized bed boiler should generate noise at roughly the same level and with roughly the same pattern as other solid fuel fired boiler systems, and substantially more than oil or gas fired systems. Dominated at the property line by. noise generated in solid fuel delivery and handling, these noise levels are necessarily greater than those produced by oil or gas fired boiler systems. With a boiler efficiency projected to be approximately 84%, the fluidized bed boiler offers efficiency of resource use which is comparable to that of a pulverized eoal fired boiler based. system. Although exact comparison is impossible, it is likely that a pulverized coal fired boiler without flue’ gas desulfurization will enjoy modestly greater resource utilization efficiency than a. fluidized bed boiler, while a pulverized coal fired boiler with the gas desulfurization will yield less resource utilization efficiency. A fluidized bed boiler offers utilization efficiency superior to that offered by stoker firing for two reasons: e actual boiler efficiency based on fuel fed is higher for fluidized bed combustion, owing to greater carbon burnout and lesser excess air requirements; and — e a fluidized bed boiler can utilize essentially all of a solid fuel stream; in contrast, a stoker fired boiler can utilize only those’ particles which fall within a limited size range,. with alternate use and/or discarding of particles which are too small for stoker feed. 5.2.4 Commercial Status: Table 5.2 lists organizations which are actively involved in commercialization of fluidized bed boilers, together with information on boilers sold to date. Today, fluidized bed boilers are being demonstrated up through a capacity of approximately 350,000 Ib of steam/hour. Industrial boilers with capacities up to 1.1 million lb are offered for. sale today with conventional commercial guarantees. Three sets of technical issues serve to cloud the economics and commercial potential of fluidized bed combustion and limit its current and near future application to industrial scale. The first of these concerns the interrelated issues of promotion of carbon. burn up, promotion: of in-bed. sulfur control, required calcium/sulfur stoichiometric feed ratio, and required rates of recycle of elutriated solids. Feeding of raw coal into a fluidized bed operated in what are considered to be economical gas velocity regimes (6-10 feet/second gas velocity above the bed) will not generally result in economical carbon burnout in the bed. Rather, up to .10% of the carbon may be elutriated from the bed unburned. - Early design concepts called for feeding this elutriated material to a special high temperature carbon burn up cell. Operating at up to 2,000°F, this cell would not 107 effectively control emissions of any sulfur which remained in the fuel fed to it. In the interests of limiting overall system sulfur dioxide emissions, subsequent designs have been based around the concept of recycle of elutriated solids to the primary beds. This serves to enhance sulfur dioxide control for a _ given calcium/sulfur stoichiometric ratio, while effecting economical carbon burnout. This approach has the disadvantage of increasing the quantities of solids to be handled and fed to the bed. Optimization of system design with respect to capital cost, carbon burnout achieved, calcium/sulfur stoichiometric ratio required, and _ difficulties posed by solids recycle--all subject to a constraining level of sulfur dioxide emissions--is not yet complete. The second set of issues concerns coal feeding technique. Some relatively small fluidized bed combustors employ in-hed pneumatic coal feeding. Larger industrial applications generally employ overbed coal feeding, either pneumatically or via a spreader stoker. In-bed feeding has the advantage of assuring that virtually all combustion will take place within the bed itself. It has the disadvantage that only a relatively small coal feed rate can be tolerated through any individual in-bed feed point. Its use to date has been limited to relatively small systems. Overbed feeding has the advantage of allowing much more coal to be fed to the bed by each individual feed system. A single feed system may serve 300 or more square feet of bed, compared to approximately 36 square feet for under-bed feeding. This greatly simplifies system control as system size increases. It has the disadvantage of having significant combustion occur above and near the top of the bed. Since sulfur burned in these areas will almost certainly escape up the flue, occurrence of any significant combustion in these areas will greatly aggravate sulfur dioxide emissions. The increase in sulfur emissions resulting from overbed- feeding has not generally posed regulatory problems for industrial installations. In the case of a utility boiler burning high sulfur coal regulations require control of 90% of the potential sulfur dioxide emissions. Such control would be difficult or impossible with overbed feeding. However, a boiler of useful size for the utility industry (say, 3 million lb/hr) would require use and control of several hundred individual in-bed feed lines in order to supply coal to the beds via in-bed pneumatic feeding. This is not generally considered feasible. Many in the field now believe that utility boiler development is dependent upon successful development and demonstration of a feeding system appropriate to such applications. The third set of issues retarding commercialization of fluidized bed combustion simply concern general design scale-up, particularly as it affects steam side performance. Again, these issues are primarily of importance in development of utility boilers. As fluidized bed boilers increase in size, they will exhibit different and not wholly understood 108 patterns of steam-side behavior. Accordingly, steam-side design of fluidized bed boilers of increasing size must be approached with caution, both by the vendors and the potential boiler owners. No additional constraints to use of fluidized bed combustion in Alaska are envisioned; they are basically the same constraints which inhibit its commecialization elsewhere. Alaskan fuels, in particular the very low sulfur Alaskan coals, should be well suited to its use. 5.3 Economic Implications 5.3.1 Costs Table 5.3 presents the general design specifications for three fluidized bed boiler systems to be considered here. The first, a 40,000 lb/hour boiler, is a typical example of large commercial and small-to-moderate size industrial boilers. This represents an example of a large packaged fluidized bed ‘boiler, at or near the upper end of the size range of packaged fluidized bed boilers offered by the vendors. It raises saturated steam at 300 psig, producing total steam energy of 39.4 million Btu/hour. The second boiler considered is an industrial boiler of moderate. to large scale. It raises steam at 1200 psig, superheating this steam to 925°F. These are typical steam conditions for a boiler this size which is used for steam topping cogeneration. In raising steam at these conditions, it delivers effective steam heating duty of 123.4 million Btu/hour. Unlike the 40,000 lb/hr boiler, it must be field erected. The large boiler considered. produces 500,000 lb of steam/hour at 1,650 psig and superheated to 950°F. It is typical of large industrial and small utility boilers. In Alaska such a ‘boiler would almost certainly be used in an electric utility power plant. In this capacity it would likely support net power generation capacity of approximately 75 MW. Raising 500,000 lb of steam at these conditions it delivers a steam heating duty of 615.5 million Btu's/hour. It is field erected. Table 5.3 also presents the estimated capital costs of the three boilers. It is broken down into cost of equipment and materials delivered to the site, site construction labor, site indirect costs, home office -costs, and contingency. All three cost analyses are compiled based on the assumption of installation at a site in the rail belt outside of Anchorage at which other process facilities exist. These facilities might be an industrial plant which would use the steam or simply the balance of an electric utility power plant (turbine generator, switch yard, etc.). All cost estimates are based on coal firing. 109 7. Table 5.3 . : CAPITAL COST SUMMARY TECHNOLOGY: Fluidized Bed Combustion Boilers BASIS Small Medium Large 300 psig 1200 psig 1650 psig saturated 925°F 950°F steam steam steam Location: Railbelt, outside Anchorage Year: 1982 constant dollars CAPACITY: Input (Beluga Coal): 3.4 ton /hr 10.7 ton /hr 53.3 ton /hr Output (Steam): 40,000 1b/hr 100,000 lb/hr 500,000 1b/hr ESTIMATED USEFUL LIFE: 20 years 20 years 35 years CONSTRUCTION PERIOD: 8 months 18 months 30 months CAPITAL COST: (Millions) Equipment and Materials: $3.01 $5.88 $18.60 Direct Labor: 1.75 3.75 13.38 Indirect Costs: 1.14 2.28 7.66 Home Office Costs: 0.68 1.35 4.43 Contingency : 0.99 1.99 6,61 TOTAL CAPITAL INVESTMENT: $7.57 $15.25 $50.68 Heat Duty, million Btu Steam/hr 39.4 123.4 615.5 Cost of Heat Duty, $/Btu Steam/hr $0.192 $0.124 $0.082 Source: Arthur D. Little, Inc., based on References 3 and 4. 110 aay 1 J ‘ Capital costs of boiler systems built to fire peat, wood, or refuse derived fuel would be somewhat higher for.a given boiler steam output. This results primarily from the need for greater fuel handling capability and possibly for bed and draft system design to support more combustion to evaporate fuel moisture. Examination of the cost figures and the heating duty of the various boilers indicates that the systems benefit from economies of scale. This may be seen in Figure 5.8, which presents capital: costs as a function of heating duty rather than pounds of steam raised because boiler costs are largely determined by heating duty insensitive to steam conditions within the range considered here. Comparing the 40,000 lb/hour boiler to the 100,000 Ib/hour, economies of scale are manifest in spite of the fact that the smaller boiler is shop fabricated while the larger boiler is field erected. Such substantial economies of scale are exhibited in the small size range primarily because the scale-insensitive total cost of the solids handling facilities which constitute a substantial fraction of total plant costs in the small size’ range. Economies of scale are also exhibited as size increase from 100,000 to 500,000 lb/hour, owing primarily to unit cost savings in the boiler itself. _It is not possible to specify a minimum economic size for fluidized bed boiler. Technically, the boiler itself is amenable to scaledown, but capital costs per unit capacity increase rapidly below approximately 100,000 pounds of steam per hour. Minimum economic size will be determined by application-specific considerations such as’ prices of competing available fuels and capacity factor obtainable based on the steam demand pattern. Table 5.2 also provides an estimate of the length of time required for actual field erection of each system. As with the cost figures described above, these estimates are based on construction at a site where other development has taken or is taking place. No time is allowed for site accessing work (building railroads, docks, etc.), and it is assumed that any other concurrent activities at the site do not hinder construction. Time is allowed for ground clearing, grading, and foundation work for the boiler and its auxiliaries. Table 5.4 presents a summary of operating and annual costs of the small fluidized bed combustion boiler treated in Table 5.3. This small boiler might find. applications such as raising process steam in industrial plants. Such plants are sited to be near raw materials, labor forces, and/or markets; they are not sited for proximity to boiler fuel. Therefore, such a boiler will burn a locally available waste (possibly generated on-site) and/or coal. The cost analysis presented in Table 5.4 is based on exclusive use of coal, at a delivered price. of $25.80/ton, expected to be representative of delivered coal prices in the Railbelt, recognizing that burning of available wastes may improve overall economics. 111 20 16 10 Capacity Cost, ¢/ Btu/hr 0 100 200 300 400 500 600 Heating Duty, Million Btu/hr Source: Arthur D. Little, Inc., estimated based on Reference 1. FIGURE 5.3 CAPACITY COST FLUIDIZED BED BOILERS “12 ne TABLE 5.4 OPERATING COST SUMMARY et TECHNOLOGY Small Fluidized Bed Combustion Boiler BASIS: , Location: : Railbelt outside Anchorage Year: 1982 CAPACITY: Input: 3.4 ton /hr Beluga Coal : a Output: 40,000 1b/hr saturated steam @ 300 psig OPERATING FACTOR: 0.82 ANNUAL ANNUAL OPERATING COSTS UNIT COST CONSUMPTION COST VARIABLE COSTS: Fuel: $25,80/ton 24,400 tons $ 630,000 Electricity: $92 /MWh 2,720 MWh 250,000 Solid Waste Disposal: $20/dry ton 5,800 tons 120,000 TOTAL VARIABLE COSTS 1,000,000 FIXED COSTS: Operation and Maintenance, Labor, and Materials: . 680,000 Taxes and Insurance: 150,000 . Capital Charges 720,000 TOTAL FIXED COSTS 1,550,000 TOTAL ANNUAL OPERATING COSTS: $2,550,000 TOTAL ANNUAL OUTPUT: 283,000 million Btu steam TOTAL LIFE CYCLE COST: : $9,01/million Btu steam 113 TECHNOLOGY: BASIS: Location: Year: CAPACITY: Input: Output: OPERATING FACTOR: OPERATING COSTS VARIABLE COSTS: Fuel: Electricity: Solid Waste Disposal: TOTAL VARIABLE COSTS FIXED COSTS: TABLE 5.5 OPERATING COST SUMMARY Intermediate Capacity Fluidized Bed Combustion Railbelt outside Anchorage 1982 10.7 tonne/hr Beluga Coal 100,000 lb/hr steam @ 1200 0.65 ANNUAL UNIT COST CONSUMPTION $25.80/ton 61,000 tons $92/MWh 6,800 MWh $20/dry ton Operation and Maintenance, Labor, and Materials: Taxes and Insurance: Capital Charges: TOTAL FIXED COSTS TOTAL ANNUAL OPERATING COSTS: TOTAL ANNUAL OUTPUT: TOTAL LIFE CYCLE COST: 114 Boiler 14,000 tons psig, 925°F ANNUAL COsT 1,570,000 630,000 280,000 2,480,000 1,070,000 310,000 1,450,000 2,830,000 $5,310,000 703,000 million Btu steam $7.55/million Btu steam The case evaluated here is for.a boiler operated with an annual capacity factor of 82%. This high capacity factor was chosen because it allows this small coal fired boiler to achieve steam cost parity with the distillate oil fired boiler evaluated in Chapter 4. It should be noted that very few industrial applications exist which require such constant steam loads, and that even with a constant year-round steam demand, it will be technically difficult for any coal fired boiler to remain operable during enough of the year to achieve such a capacity factor. , The boiler consumes 3.4 tons of Beluga coal per hour and 0.34 MW of electricity. On an annual basis, this corresponds to consumption of 24,400 tons of coal and 2,720 MWh of electricity. The ash content of the Beluga coal is considerable, 23.7%. Annual ash production (and disposal) total 5,800 tons. Disposal costs are estimated at $20/ton. An. example application of the 100,000-1b/hr fluidized bed boiler would be for cocombustion of coal and of refuse derived fuel (RDF) as available. The steam raised. would be used to generate electricity. A boiler this size could dispose of all of the waste generated in any Alaskan city except Anchorage. Economics are projected in Table 5.5 for a coal based system rather than for the dual fuel plant because: e the RDF supply exhibits variation, daily and _ seasonally, which will make operation of a plant dedicated exclusively to burning it difficult. The appropriate plant concept would burn RDF as available, supplementing this with coal as needed. . : e the practice traditionally adopted in the lower 48 is to transfer RDF to the user at overall cost parity with the fossil fuel being displaced; any net savings or cost accrues to the waste disposal authority. Overall cost parity includes allowance for additional. capital and operating costs. for handling a second fuel, savings/cost of disposal of RDF ash in the event that its combustion produces less/more ash than coal per pound of steam raised, ete. Thus by projecting steam costs based on coal, steam costs based on coal plus RDF are also predicted. : Extensive data on RDF supply and properties would be required to calculate the appropriate transfer price. A unit such as this could be expected to be operated at a base loaded power plant. As such, it would achieve an annual capacity factor of approximately 65%. Based entirely on Beluga coal, this boiler would use 10.7 tons of coal per hour and 1.2 MW of electricity, totalling 61,000 tons of coal and 6,800 MWh of electricity per year. A total of 14,000 tons of ash is produced annually. 115 TECHNOLOGY: BASIS: Location: Year: CAPACITY: Input: Output: CAPACITY FACTOR: OPERATING COSTS VARIABLE COSTS: Fuel: Electricity: Solid Waste Disposal: TOTAL VARIABLE COSTS FIXED COSTS: TABLE 5.6 OPERATING COST SUMMARY Large Fluidized Bed Combustion Boiler Railbelt outside Anchorage 1982 53.3 tonne/hr Beluga Coal 500,000 1b/hr steam @ 1650 0.65 ANNUAL UNIT COST CONSUMPTION $25.80/ton 300,000 tons $92 /MWh $20/dry ton Operation and Maintenance, Labor, and Materials: Taxes and Insurance: Capital Charges TOTAL FIXED COSTS TOTAL ANNUAL OPERATING COSTS: TOTAL ANNUAL OUTPUT: TOTAL LIFE CYCLE COST: 116 34,000 MWh 71,000 tons psig, 950°F ANNUAL COST 7,740,000 3,130,000 1,420,000 12,290,000 2,530,000 1,010,000 4,810,000 8, 350,00 $20,640,000 3,510,000 million Btu steam $5,.88/million Btu steam Annual costs of the large fluidized bed boiler are summarized in Table 5.6. This boiler, located in the Railbelt outside of Anchorage, would probably be sited at a steam power plant. At such a facility, it would probably be base loaded, achieving a capacity factor of 65% on an annual basis. While this power plant application is the basis of the analysis conducted here, economics are projected solely for the steam raising system. This boiler would use 53.3 tons of Beluga coal per hour. This will result in an annual consumption of approximately 300,000 tons, priced at $32/ton delivered to the plant from mines in the Beluga coal field. Due to the low heating value of this coal, it is expected that power plants which utilize it will be located in or near the coal field. Annual electricity consumption of 34,000 MWh corresponds to a system power demand at full load of 6 MW. The unit cost and resulting annual cost of this power at estimated’ as if the power were generated in a steam power plant based on this design of fluidized bed boiler burning Beluga coal. Annual ash production (annual ash disposal requirement) is 71,000 tons. Unit cost and annual cost are projected based on disposal. in the nearby coal mines. Consistent with the generic nature of this analysis, the balance of operation and maintenance costs are projected as a percentage of ‘total capital costs. Figures of 9, 7, and 5% of total capital costs are used used for the small, medium, and large boilers, respectively. These, together with capital charges, constitute total annual lifecycle steam costs of about $5.50-9.00/million Btu, being lowest for the: largest size unit. . 5.3.2 Socioeconomic Factors Direct Benefits Use of a fluidized bed boiler to supply steam will contribute more to building and diversifying the Alaskan economy than will use of an oil or gas fired boiler designed to provide the same amount of steam. Such socioeconomic benefits will manifest themselves in three principal forms: e creation of more permanent jobs for operation and maintenance ' of the boiler; @ creation of more temporary jobs for boiler construction; and e purchase of goods and services for boiler operation and maintenance which will lead: to more secondary employment in Alaska. 117 Altogether, approximately 15 full-time jobs would be created. Practice in the Lower 48 states typically results in filling the managerial positions created by promotion of engineers. These engineers typically were hired out of college with a Bachelor's Degree in mechanical engineering. Clerical positions typically are filled by college graduates with degrees in accounting. Blue collar positions typically are filled by hiring high school graduates for training within the organization. A solid fuel fired boiler requires periodic large-scale maintenance. Typically, this is conducted by skilled craftsman (steam fitters, boiler makers, etc.) on a contracted basis. These highly skilled personnel typically develop their skills within a union organization. When not engaged in boiler maintenance, they may work elsewhere in boiler construction and related activities. Construction of solid fuel fired boilers likewise produces substantially more contemporary construction jobs than does construction of oil or gas fired boilers of the same size. In the case of the 100,000 lb of steam per hour boiler considered. in the preceding section, is estimated that boiler erection would require ten times as much direct and indirect field labor and supervision as would erection of a gas fired package boiler of the same size. Indirect Benefits Operation and maintenance of a solid fuel fired boiler requires purchase of a solid fuels mined, harvested, or recovered in Alaska by Alaskans. The delivered price of such fuels as coal, peat, and waste wood typically are cost based. As such, payment for these fuels results directly in payment. to personnel involved in winning and transporting them. The relatively labor intensive nature of coal mining, wood harvesting, and peat harvest could be expected to create more jobs than oil and gas extraction. 5.4 Impact 5.4.1 Effect on Overall Energy Supply and Use Fluidized bed combustion, as other solid fuel fired steam raising systems, will have its most common application in relatively large seale, high load factor installations. In the Lower 48, typical applications would include power generation and steam raising in the process industries. In Alaska, the primary application would probably be for power generation in the Railbelt, and/or in very large industrial installations which might be built in the future. Power generation outside the Railbelt is generally conducted in response to demands which are too small and too variable to support an economically sized base-loaded power generation system which fires solid fuel.’ Commercial steam users tend to be too small and have capacity factors which are too low to make solid fuel based systems competitive with oil or gas fired boilers. 118 In these favored applications, fluidized bed combustion will offer the same advantages as other solid fuel fired systems: e opportunity to replace relatively large quantities of premium liquid or gas fuels, making them available for higher value uses, including possible shipment out of the state; e@ job creation through stimulation of coal production in Alaska; e® mitigation of Alaskan environmental problems by providing a preferred means of disposal for the bulk of municipal solid waste generated in the Anchorage and possibly the Fairbanks areas; and e lower steam or power costs to be paid by customers compared to those’ required to support an oil or gas based system (assuming that oil and gas pricing is consistent with world oil prices). . In addition, fluidized bed combustion offers several advantages over other solid fuel fired boiler technologies: e inherently lower rates of emission of nitrogen oxides; ‘ @ simplified one step control of emissions of. sulfur dioxide incorporated directly into combustion, alternatively leading to lower sulfur dioxide emission rates for lower system capital and operating cost; e simpler and more economical adaptation to firing of solid fuels such as municipal solid waste, wood waste, etc; and e dry handling of all solid waste, if desired. Fluidized bed combustion will not fundamentally alter the comparative advantages of use of a solid versus a liquid or gas fuel for steam raising. When and to the extent that it is available as an option for solid fuel use, it will offer improved environmental performance and, compared to conventional coal combustion systems with flue gas desulfurization, superior system economics. Then the true significance of these modest advantages for Alaska will depend upon the future prospects for deployment of solid fuel fired steam raising capacity in Alaska. 5.4.2 Future Trends Fluidized bed boilers may. find use in the same three classes of applications discussed for the more conventional stoker and pulverized _ coal fired boilers in Chapter 4: 119 @ moderate to large industrial boilers; e coal fired utility boilers; or e opportunistic waste utilization. In these applications they will offer advantages enumerated above over their conventional counterparts. Fluidized bed boilers can be considered to be proven industrial technology today in the smaller size ranges. Such boilers could be expected to satisfy any steam demand in Alaska except those associated with future power plants in the Railbelt. The potential for utility application of fluidized bed combustion depends upon timely demonstration at larger scales. Although fluidized bed boilers are now offered commercially with rated capacities which cover much of the range of interest for use for power generation in the Railbelt outside of Anchorage, they are not demonstrated at all above a capacity of 300,000 pounds of steam per hour, with the largest commercial unit operating today being 200,000 lb per hour. Their demonstration in the near future in the 1,000,000 pound per hour size range outside of Alaska is doubtful. Accordingly, fluidized bed boilers cannot be considered a prudent candidate for use in large scale steam raising in the Railbelt outside Anchorage in the late 1980's and possibly not in the early 1990's. If further coal fired capacity is built in this area subsequent to the early 1990's, fluidized bed combustion could be expected to offer a very attractive alternative to conventional boilers with flue gas desulfurization. Whether or not coal fired generating capacity will be built in the Railbelt during the mid- to late-1990's depends both on load growth and on whether or not the Susitna hydroelectric project is implemented. 120 REFERENCES Davy McKee, Inc., Coal Methanol Feasibility Study: The Beluga Methanol Project, prepared for the Cook Inlet Region Inc., and Placer Amex, under DOE Grant DE-FG01-80RA-50299, 1981. Makansi, J. and R. Schwieger, "Fluidized-Bed Boilers," Power, August, 1982, pp. S1-S16. Foster Wheeler Energy Corporation, private communication. Cogeneration Technology Alternatives Study (CTAS), Final Report, Vol. IV - Heat Sources, Balance of Plant, and Auxiliary Systems, prepared by Power Systems Division of United Technologies Corporation for National Aeronautics Space Administration, Contract DEN3-30 for the U.S. Department of Energy, January 1980. : Private communication. Cogeneration Technology Alternatives Study (CTAS), Volume 6, Part 1, Coal-Fired Noncogeneration Process Boiler, Section A, prepared by the General Electric Company for the National Aeronautics Space Administration, Contract DEN3-31, for the U.S. Department of Energy, May 1980. “121 6.0 COMBUSTION TURBINE COMBINED CYCLES 6.1 Introduction and Summary 6.1.1 Technical Overview A gas turbine uses the energy available in hot high pressure gas to produce work in the form of shaft power by expanding in a gas turbine. Typically, this high temperature, high pressure. gas is - produced by burning a liquid or gaseous fuel in high pressure air using a portion of the total work produced to compress this air. A thermodynamic system which produces net work in this fashion (i.e., by compressing the gas, heating it, and then expanding it to yield work) is called the Brayton cycle. Exhaust from a gas turbine system still contains substantial amounts of heat, typically at temperatures around 1000°F.(1) This’ heat can be used to raise steam, which in turn can be used to drive a steam turbine to produce additional shaft power via a Rankine cycle. A system: which uses both gas turbines and their exhaust heat. to drive steam turbines combines the two thermodynamic cycles; hence, it is called a combined cycle system. The principal’ stationary application of gas turbines alone is to produce electricity. In such applications, the shaft power developed is used to drive a generator. The other important stationary application of such turbines is for pumping/compression power on long distance pipelines. Essentially the only application for combined cycle systems is for power generation. The gas turbine is the most desirable power generation technology for many applications. This results from several advantages which it has over any other power generation system: @ it offers convenient increments in generating capacity (tens of megawatts) with minimal capital cost; e it offers generating capacity with minimal operator attendance; e it offers generating capacity with very short construction times; and e it offers maximum siting flexibility due to its compact size, its use of relatively clean fuels, and minimal utility needs (in particular, no need for a large source of water). Use of combustion turbines in the Lower 48 states has been limited by their need for a premium fuel and their relatively low power generation efficiency (i.e., high heat rate). 123 .Combined cycle power plants retain ‘most of the advantages of simple cycle gas turbines, while achieving much lower power plant heat rates. They require somewhat greater capital investment per unit capacity, some additional operator attention, and do require a source of cooling water. However, they still require less of each than does a conventional steam-electric plant of the same capacity. 6.1.2 Alaskan Perspective Gas turbines account for approximately 64% of currently installed electric utility generating capacity in the State of Alaska.(2) The most recent capacity additions include two combined cycle systems, one owned by the Municipality of Anchorage, the other by the Chugach Electric Association.(3) Future fuel-fired capacity additions as required will probably be dominated by combined cycle generation. This may be accomplished either by addition of grassroots combined cycle plants or by retrofitting steam cycle systems to existing gas turbine generation systems. The domination of power generation capacity by gas turbine and combined cycle technologies. This is particularly true in the Anchorage area, where electricity demand is sufficiently great. to warrant installation of large combined cycle systems and where natural gas is available at low prices. This contrasts markedly with the power generation mix which predominates in the Lower 48, which is dominated by coal and nuclear-based steam-electric generating plants. The greater role of the gas turbine in Alaska is appropriate for three reasons. First, natural gas is available at very low prices for use in power generation in the Anchorage area, the area where most electricity demand is concentrated. Thus, lower quality fuels such as coal offer no fuel cost advantage. Second, gas turbine and combined eycle systems are available in capacity increments which are compatible with the needs with relatively small electric utility systems found in urban Alaska. Third, gas turbine and combined cycle systems require a minimum of construction and operating labor. They are designed to make maximum use of shop fabricated components, which can be shipped from the Lower 48 for easy installation, and require minimal operator attention. Thus, in an environment characterized by relatively high labor costs and high demand for available specialized skills such as that in Alaska, gas turbines and combined cycle technologies offer advantages in Alaska. 6.1.3 Significance of the Technology Both simple cycle gas’ turbine and combined cycle systems are based on mature, well understood technology. Both provide very low cost power in those parts of Alaska where natural gas is readily available. The lowest cost option for baseload or intermediate load power would be combined cycle power generation, with life cycle power costs of approximately 4¢/kWh. , 124 Combined cycle power generation, in particular, can be expected t play an increasing role in thermal electric power generation in th interconnected portion of the Railbelt in Alaska. This may occur as : result of installation of new, grassroots combined cycle systems or o: retrofitting steam cycle equipment in existing gas turbines. In this area, only very large increases in the real price of natural gas would make any other thermal electric power generation system economically competitive with combined cycle systems. Barring such increases in natural gas prices, the factors determining future deployment of combined cycle power generating systems will be electric power demand growth and the decision whether or not to build the Susitna project. Susitna would proably preclude any significant additional thermal electric generating capacity in the electrically interconnected portion of the Railbelt well into the next century. If the Susitna project is not built, perhaps 500 MW of new thermal electric capacity will be required over the balance of this century. Most or all of this will be expected to be filled by combined cycle systems. 6.2 Description of Technology 6.2.1 Operating Principles A gas turbine converts the energy contained in high temperature, high pressure gas into mechanical energy. In the case of the aircraft (jet) engine, this energy is in the form of thrust to propel an airplane. For stationary turbines, the output is shaft power which may be used to generate electricity or to provide large-scale direct mechanical drive (e.g., as required for pumping on oil and gas pipelines). Most practical modern gas turbine generating stems produce their high temperature, high pressure gas by burning a liquid or gas fuel in high pressure air. This hot, high pressure gas passes through an expander turbine at high velocity. The heart of the turbine consists of a rotating shaft with blades, called "buckets," mounted radially outward from the shaft. The hot gas moves at high velocity along tortuous paths parallel to the axis of the turbine shaft through the openings between the buckets. As it moves, it imparts work to these buckets, causing the turbine shaft to rotate; thus, shaft power is produced. The hot, high pressure gases expand and cool as they impart work to the buckets. The high pressure air used to combust the fuel is provided by a compressor turbine. This turbine is typically mounted on the same shaft as the expander turbine and consumes a portion of the shaft power produced by the expander turbine. A practical gas turbine system is, of course, designed with gas temperatures and pressures selected so that the gross shaft power produced by the expander turbine exceeds that consumed by the 125 compressor turbine. This is accomplished by heating the high pressure air prior to expansion by burning fuel in it, adding thermal energy which the expander turbine can.extract mechanically. This system, which produces net work by compressing the gas in a turbine compressor, heating that gas, then expanding that gas through an expander turbine, is called a Brayton cycle. Gas turbine exhaust, though cooled significantly from the temperatures achieved in the combustor, still contains substantial usable heat, typically at temperatures up to 1000°F when operated at or near full load. In many gas turbine applications, where the turbine is used to supply, needed, reliable electricity at minimal capital cost, these exhaust gases are simply vented to the atmosphere, In other types of installations, various uses are made of this heat. The most important use is to raise steam either for heating or power generation using highly specialized equipment which differs somewhat from the oil and gas-fired boilers described in Chapter 3. Fuels Four distinct classes of fuels may be considered for use in gas turbines: e natural gas; e distillate oil; e residual oil; e low and. medium Btu gases produced via partial oxidation of solids (principally coal). Natural gas and distillate are the most easily used in gas turbines, since they require no specialized pretreatment prior to use. Residual oils, on the other hand, require special pretreatment to remove trace metals and other corrosion-causing impurities prior to being fired in gas turbines. For both technical and economic reasons, any of these fuels are reasonable candidates for use in either simple gas turbine or combined cycle power generation. Low or medium Btu gases from coal would only be contemplated with combined cycle power generation, not with simple gas turbines. This is because the relatively high capital cost and low fuel cost of the coal gasification system itself dictate that it be used in an application which will have a high capacity factor and high efficiency. In such baseload’ cases, combined cycle power generation offers more economical power generation, since the high fixed costs are distributed over more electric power output. 126 7s Use of coal gasification in conjunction with combined cycle power generation presents unique problems and offers unique opportunities. Problems center around coal gasifier operability and capital cost, and the need to clean the gas thoroughly prior to introduction into the combustor. Opportunities exist for achieving very high overall power plant. thermal efficiency by integrating the gasifier with the combined power cycle power generation system. Plant Design Concepts Figure 6.1 displays block flow diagrams which represent a simple . combustion turbine generator system and a combined cycle power generation system. In the simple cycle system (Figure 6.la) air is drawn into the compressor turbine and compressed for introduction into the combustor, where it is mixed with the fuel. : If residual fuel oil is used, it may require special treatment prior to introduction into the combustor, to remove trace materials which might damage the turbine. When natural gas or distillate fuel oil is used, as is typical in Alaska, no such treatment is required. In the combustor, fuel burns with the pressurized air yielding a hot, high pressure gas. State-of-the-art turbine design calls for combustor -exit temperatures of approximately 2000°F, which are controlled. at this level by use of an appropriate mix of fuel and air (substantially more air than is required simply to combust the fuel).(1,4) The temperature limit is determined by the capability of the first stage turbine blades to withstand the mechanical stresses which they incur as the turbine rotates. Elaborate systems are used to cool these blades to temperatures below the actual gas temperature; nevertheless, use of gas temperatures significantly above 2000°F will result in premature blade failure due to stress. Expansion of the hot combustor exhaust through the turbine yields “enough shaft power to drive the compressor and an electric generator. The hot exhaust is discharged to the atmosphere through a relatively short stack, typically not more than 75 feet high. Use of so short a stack is possible because the exhaust contains only very low concentration of air pollutants and because, with the high exhaust temperature, the gas is quite buoyant and will naturally rise to much greater heights as it mixes with the surrounding air. Figure 6.lb depicts a combined cycle power generation system. As ean be seen; it is essentially identical to the gas turbine generating system up to the point of discharge of gas turbine exhaust. In this ease, this exhaust is passed through a heat recovery steam generator. where, it is cooled to approximately 350°F. The heat extracted from the gas turbine is used to raise steam, typically at approximately 900 psig and 850°F under full load conditions, which in turn drives the steam turbine to generate additional electricity. (4) 127 Air Compressor Turbine Compressed Air Combustor and Expander Turbine Fuel Treatment (Residual Fuel Oil Only) Fuel Generator Electricity FIGURE 6.1a GAS TURBINE POWER GENERATION SYSTEM Air Compressor Turbine Compressed Exhaust Air Fuel Combustor Exhaust and Expander — Turbine Fuel Treatment {Residual Fuel Oil Only) Heat Recovery Steam Generator Shaft Power Steam Turbine Generator Generator Electricity Electricity Source: Arthur D. Little, Inc. FIGURE 6.1b COMBINED CYCLE POWER GENERATION SYSTEM 128 These steam generators are shop fabricated and take advantage of the cleanliness of the exhaust with closely spaced steam tables which minimize rates of convective heat transfer. (Such close tube spacing is not feasible if the heat-bearing gas contains ash.) Thus, steam generator cost and special requirements are minimized. 6.2.2 Technical Characteristics Unit Size Table 6.1 summarizes key technical characteristics of simply cycle gas turbine generators and of combined cycle generating systems. Both systems are available in broad capacity ranges. Simple cycle gas turbines are marketed today with capacities ranging from’ less than 1 MW to approximately 85 MW. Specialized uses are found throughout the size range, owing to light weight, compactness, and low capital cost. In most locations, they are of interest for electric utility power © generation in capacities greater than approximately 20 MW because _ these larger size units are more efficient that smaller ones. Also, in the smallest sizes, gas turbine generators are unable to compete with diesel systems in. applications where weight and volume are not a problem. Combined cycle units are available in a very broad size range. Large units usually include multiple gas turbines and heat recovery steam generators with a single steam turbine. This is done to take advantage of economies of scale in the steam turbine and its small efficiency improvement as size increases. Throughout the size range, roughly 70% of generating capacity may be attributable to the gas turbines, with the balance attributable to the steam turbine. Thermal Efficiency The full load design heat rates presented in Table 6.1 are based on natural gas firing of gas turbines located at sea level with an ambient temperature of 35°F. The corresponding thermal efficiencies are also displayed. These heat rates differ significantly from those to be expected in a substantially warmer climate. For example, the full ‘load design heat rate of the simple cycle gas turbine is approximately 93% of that which might be expected if the ambient temperature were 85°F.(1) The heat rate exhibited by the combined cycle generating system, however, is essentially the same as that which would be expected in an 85°F environment. This results despite improved gas turbine performance because substantially less heat: is available for use in steam power generation. 129 TABLE 6.1 TECHNICAL CHARACTERISTICS OF GAS TURBINE GENERATOR SYSTEMS Available Unit Size Range Unit Composition Full Load Design Heat Rate (Higher Heating Value Basis) Full Load Design Efficienty Time from Cold Start to Full Load Simple Cycle Gas Turbine __Generator_ Up to 85 MW One gas turbine generator 10,600 Btu/kWh 32% 20 minutes 130 Combined Cycle Generating System 60-600 MW One to six gas turbine generators, one to six heat recovery steam generators, one steam turbine generator 8,400 Btu/kWh 412% 2.5 hours Pod Heat rates also vary significantly with variations in load. The gas turbine heat rate would be approximately 20% higher when operated at 50% of rated load.(1) Combined cycle generating system heat rate is determined not only by variations in the total amount of power to be generated, but also by philosophy of operation at reduced load (cutting all generator back versus removing some individual gas turbine generators from operation altogether); therefore, no analogous general statement can be made concerning these plants. Operability Both simple cycle gas. turbine and combined cycle generating. systems have established records. of reliable ‘operation. Gas turbine generators may operate with little or no continuous. on-site- attendance. Their reliability and ease of start-up has led to their use as sources of large-scale emergency power. Rotating equipment requires fairly sophisticated, though infrequent maintenance; many parts can only be fabricated by the original equipment manufacturer. However, some components which require frequent replacement, such as the combustors, can’ be and are fabricated by the utilities themselves. This is practiced currently by the Municipality of Anchorage. : Combined cycle generating systems are of necessity more complex than simple cycle gas turbine generators. They require more operator attention and technical support, due in part to the presence of a steam cycle. However, they typically are based on more than one gas turbine (two gas turbines are included in the combined cycle systems currently operated by the Municipality of Anchorage and by the Chugach Electric Association). Proper attention to and scheduling of preventive maintenance can, in most cases, prevent the simultaneous shut-down of all gas turbine generators within a single combined cycle system. 6.2.3 Environmental Issues This section considers three classes of pollutant emissions: e air pollutants; e water pollutants; and e noise. A gas-turbine based generating system fired with an _ essentially sulfur-free fuel, such as natural gas, will emit one air pollutant in concentrations which might cause concern: nitrogen oxides (NO_). The gas turbines considered here, with their turbine inlet temperatures of approximately 2000°F, can be expected to emit only very small quantities of oxides and nitrogen, typically less than 0.3 131 pounds of NO_ as NO,/million Btu fuel fired. (1) This emission rate can be reducéd by about half by injecting water or steam into the combustor, but at the expense of increased generator heat rate. (5) If a sulfur-bearing fuel, such as distillate or residual oil is burned, essentially all of the sulfur in the fuel will be emitted to the atmosphere as sulfur oxides. Emissions of these gases are governed by limiting the sulfur content of the fuel. Fuel oils are available with sulfur contents from less than 0.1 wt % to more than 1 wt %. An oil will be chosen to meet whatever local regulations require. Operation of a simple eycle gas turbine will result in negligible discharge of water pollutants. A combined cycle adds a steam cycle system and its usual water effluents: : e boiler blowdown (to remove residual water contaminants from the system); e cooling system blowdown; and _@ steam side boiler cleaning wastes. Of these, the first two are produced continuously. Extensive work has been done to develop methods of mitigating the environmental impacts of these discharges. They cause far less concern than effluents in coal-fired power plants where waste streams have had direct contact with the coal and/or the coal ash. Boiler cleaning wastes are produced as a result of periodic chemical ‘ cleaning of the boiler. They are quite toxic. Typically their disposal is arranged with the assistance. of the chemical firm engaged for the cleaning procedure. If no deliberate steps are taken to limit noise emissions from gas turbine generating systems, they can constitute a substantial nuisance in populated areas. Current standard designs include muffling equipment which will reduce noise to 59 dBA at a distance of 400 feet. Additional noise reduction may be achieved if required. © Noise emissions from combined cycle plants are expected to be dominated by those from the gas turbines. This is because no other system within these plants will produce nearly so much noise as the gas turbines themselves. Combined: cycle generating plants offer very efficient conversion of fossil fuels to electricity. Current technology allows efficiencies of up to 41%. Advances in gas turbine technology can be expected to improve this rate two ways: : e by allowing for more efficient gas turbine operation; and 132 os @ by operating with higher exhaust temperatures, allowing use of a higher temperature, higher pressure, more efficient steam generating turbine cycle. The simple cycle gas turbine generator necessarily makes less efficient use of its fuel. . Nevertheless, it enjoys economical application in situations where capital costs must be minimized. 6.2.4. Commercial Status The gas turbine generator and the steam turbine generator are both fully mature technologies. Worldwide, both have accumulated over 100 million machine hours of operating experience. Approximately 50 combined cycle generating systems have been deployed worldwide. Current U.S. vendors of gas turbines and combined cycle generating units are General Electric and Westinghouse. Table 6.2 summarizes electric utility generating capacity in Alaska. Gas turbines predominate, accounting for 64% of total capacity. (2) Although used at several locations around the state, they are most important in the Anchorage area where the size of the interconnected’ electric power system is sufficiently great as to allow use of large, efficient gas turbine generators. Their use in this area is also facilitated by the availability of inexpensive natural gas. Gas turbine generators do not play a significant role in smaller cities or in the Bush due to the relatively low efficiency and relatively high unit capital costs of small gas turbine generating systems. Two combined cycle systems are in service today: one owned by the Municipality of Anchorage, the other by the Chugach Electric Association. They are not accounted as combined cycles in Table 6.2; rather, their steam turbines (one per unit) are accounted to steam power generation and their gas turbines (two each) to gas turbine power generation. Adoption of combined cycle power generation even in the Anchorage area has been slow. It has repeatedly been suggested that growing demand for electricity here should be served by retrofitting heat recovery steam generators and steam turbine generators onto existing simple cycle gas turbine generators, such as those listed in Table 6.2. Three classes of considerations have effectively served to prevent such retrofits to date: — e Water availability. If a water-based. condenser cooling system is to be used, a large source of water must be available close to the plant site. Since simple gas turbine generators require virtually no water, the older gas turbines were sited at locations with little or no available water. 133 Prime Mover Gas Turbine Diesel Steam Turbine Hydro Total TABLE 6.2 ELECTRIC UTILITY GENERATING CAPACITY INSTALLED IN ALASKA 1980 Installed Capacity, MW 825 234 101 123 1,283 134 % of Total Installed Capacity 64 18 8 10 100 atic r r e Land area combined cycle generating systems require substantially more land area than do simple cycle gas turbine generators. A plant such as Municipality of Anchorage Plant #1 would be difficult to retrofit for this reason. e Air pollution control regulations. Older sources of air pollutants are regulated less stringently than are new sources. Regulatory distinctions between "new" and "old" sources include provisions for considering an older system as a "new source" if a substantial new capital investments is made in it. By this definition, the retrofit of a steam cycle onto simple cycle gas turbine generators would cause an old unit to be regulated as if it were new. Research is being conducted in two distinct areas which may change future patterns of use of gas turbines and combined cycle systems: e@ development of coal gasification and gas cleaning systems for direct fueling of gas turbines; and e@ development of turbine blade materials and/or design concepts which will allow use of higher combustor temperatures. — Work in the first category focuses on raw gas cleanup and on capital cost reduction. The concept will soon be tested with Texaco gasification at Southern Edison's Coolwater Station. As mentioned previously, combustor temperatures are limited by the effect of that temperature. on the capability of the hottest rotating blades to withstand their own cetrifugal stresses during operation. These blades are hollow and are cooled by passing air through them; thus, they experience temperatures which are substantially lwoer than the 2000°F state-of-the-art combustor temperature. Nevertheless, blades will fail prematurely if temperatures are increased further. Work is now underway to develop new means of blade cooling (water cooling, transpirational cooling of porous blades, etc.) to to develop ceramic blade materials which can withstand higher temperatures. 6.3 Economic Implications 6.3.1 Costs Table 6.3 presents the general design specifications for a 250 MW natural gas fired, combined cycle power plant. Such a plant could be expected to include three gas turbine generators operated with combustor temperatures of about 2,000°F and combustor pressures of about 135 psig. Exhaust from the gas turbines is used to raise 850°F steam at 900 psig in heat recovery steam generators to drive a steam turbine. Such a plant might prove to be an attractive candidate for addition to existing generating capacity in the Anchorage area. 135 TABLE 6.3 CAPITAL COST SUMMARY TECHNOLOGY Combined Cycle Power Plant, 2000°F combustor temperature, 135 psig combustor pressure, 900 psig 850°F steam turbine throttle conditions BASIS Location Railbelt outside Anchorage Year 1982 constant dollars CAPACITY . Input 2100 Million Btu/hr Natural Gas Output 250 MW Electricity ESTIMATED USEFUL LIFE 30 years CONSTRUCTION PERIOD 2 years CAPITAL COST, $ MILLION Equipment and Materials 109 Direct Labor 49 Indirect Costs 33 Home Office Costs 23 Contingency _32 246 TOTAL CAPITAL INVESTMENT 136 Table 6.3 also predicts the estimated capital cost of the complete combined cycle power plant. It is broken down into the cost of equipment and material delivered to the site, construction labor, site indirect costs, home office costs, and contingency. The costs are compiled based on the assumption of installation at a site in the Railbelt outside of Anchorage at which no other process facilities are located. Note that costs of equipment and materials constitute more than half of specified capital investment (i.e., total capital investment minus contingency) because the plant is constituted primarily of a few shop fabricated pieces of equipment which require minimal field erection effort. Total system cost and estimated time to construct are based on an undeveloped site, not including allowance for site accessing work (building roads, pipelines, etc.). Time and funds are allowed for ground clearing, grading, and foundation work. Capacity cost as a function of net rated generating capacity is shown in Figure 6.2. Economies of scale are slight and, beyond approximately 100 MW, are limited primarily to steam turbine system and auxiliaries. Table 6.4 projects the total annual cost of owning and operating this 250 MW plant based on a capacity factor of 65%. This capacity factor is chosen to simplify comparison with steam cycle power generation as treated briefly in section 3.5. It lies between the capacity factors characteristic of a base loaded unit and an intermediate loaded unit. This unit will consume approximately 2.1 billion Btu of natural gas per hour, 12 trillion Btu annually. Operating costs, together with capital charges, property taxes, and insurance constitute total life cycle power costs of $.040/kWh. Comparison with power costs projected in other chapters indicates that this is the least expensive power generation option available in the area immediately surrounding Anchorage within which natural gas is available at a price of approximately $1.70/million Btu. 6.3.2 Socioeconomic Factors Direct Benefits Operation and maintenance of a 250 MW combined cycle power plant will create approximately 25 full-time jobs. Most of these will be essentially blue collar positions filled by hiring high school graduates for training in the plant. These employees can be expected to perform all necessary tasks for plant operation and most tasks required for plant maintenance. ‘Construction of a combined cycle power plant will require a minimum of direct and indirect site labor. As noted previously, this results directly from plant design, which makes maximum use of shop 137 Capacity Cost, $/kW 1100 1000 900 800 0 100 200 300 Generating Capacity, MW Source: Arthur D. Little, Inc. FIGURE 6.2} CAPACITY COSTS AS A FUNCTION OF POWER GENERATING CAPACITY: COMBINED CYCLE GENERATING UNIT 138 400 OPERATING COST SUMMARY TABLE 6.4 TECHNOLOGY Combined Cycle Power Plant BASIS Location Railbelt outside Anchorage Year 1982 constant dollars CAPACITY Input 2,100 Million Btu/hr natural gas Output 250 MW Electricity OPERATING FACTOR 65% Operating Costs Unit Cost Consumption Annual Cost VARIABLE COSTS Natural Gas $1.70/Million 12,000,000 $20.3 million TOTAL VARIABLE COSTS we $20.3 FIXED COSTS Operation and Maintenance Labor and Materials 8.3 Taxes and Insurance 4.9 Capital Charges 23.4 TOTAL FIXED COSTS $36.6 TOTAL ANNUAL OPERATING COSTS $56.9 TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST 139 1,424,000 MWh 4.0¢/IWh fabricated gas turbine generators and heat recovery’ steam generators. Altogether, it is expected to require only approximately one third as much construction labor as a coal fired steam electric power plant with the same rated capacity. Indirect Benefits Operation of a natural gas fired combined cycle power plant will require purchase of natural gas extracted in Alaska. Payment for such gas will have limited benefits in terms of job creation in Alaska, due to the limited labor requirements of oil and gas extraction. 6.4 Impact 6.4.1 Effect on Overall Energy Supply and Use Gas turbine generators and combined cycle generating systems play a vital role in the Alaskan economy by generating the bulk of the electricity consumed in the state. For electric utility power generation in Alaska, simple cycle gas turbines have several key advantages. e At large scale they offer lower capital costs and simple operation and maintenance than diesel generator systems. At ’ these scales, they offer thermal efficiency nearly as great as diesel. e They can be installed in much shorter periods and at much lower capital costs than steam electric generating capacity. e They can be deployed close to demand centers, with minimal siting constraints. e They offer simplicity of operation, maintenance, fuel procurement, permitting, etce., generating power, while making minimal demands on the owner/operator. The chief disadvantages of the simple cycle gas turbine are that it requires a premium fuel (fuel oil or natural gas) and _ exhibits relatively high heat rate. The first of these disadvantages is of little economic importance in the Anchorage area, due to availability of an low cost natural gas. Power plant heat rate can be improved to levels below those achievable with any other existing electric power generation system by adopting use of combined cycle generation technology instead of simple cycle gas turbine generators. Retrofitting of steam cycle equipment on existing gas turbines in the vicinity of Anchorage could yield quite substantial savings of natural gas. 140° 6.4.2 Future Trends Existing gas turbine generating capacity in the Anchorage and Fairbanks areas substantially exceeds current requirements. If Susitna is built, this electrically connected portion of the state will require little or no additional thermal electric generating capacity well into the 2lst century. . If additional electric generating capacity is needed (of particularly importance if Susitna is not built), it can be supplied most economically bv installation of combined cycle generating systems.. These units could be dedicated to baseload operation, while using the existing gas turbines to continue to serve intermediate and peak load demands. The principal alternatives to such combined cycle units would be coal-fired steam-electric generation. However, comparison of the costs projected here and in Chapter 4 would indicate that combined cycle generation is far. more. economical. This conclusion could change if natural gas prices increase at rates much faster than ‘those of coal. A potentially attractive alternative is presented by the existing gas turbine generators. Depending upon site-specific considerations, it may be possible. in some locations to retrofit steam cycle equipment to the existing gas turbines. Such retrofits are currently inhibited by regulatory policies which would reclassify the gas turbines as new sources of air pollutants if combined cycle equipment is retrofitted to them. This is the case despite the fact that retrofit of the combined cycle equipment will not decrease gross emissions of air pollutants from the site. (By improving plant thermal efficiency, -it will actually decrease emissions of air pollutants per unit of electricity generated.) Amendment of these regulations to facilitate such retrofits of steam cycle systems to existing gas turbines could effectively provide Alaska with an inexpensive and highly efficient source of new electric generating capacity. 141 REFERENCES General Electric Company, private communication. State of Alaska Long Term Energy Plan, Department of Commerce and Economic Development, .Division of Energy and Power Development, 1982. H. Nikkels, Municipality of Anchorage, private communication. Technology Assessment Guide, prepared by the Electric Power Research Institute, EPRI P-2410-SR, May 1982. Pacific Gas and Electric, private communication. Cogeneration Technology Alternatives Study (CTAS); Vol. 4, prepared by the General Electric Company for the National Aeronautics and Space Administration, Contract DEN 3-31, for : the U.S. Department of Energy, May 1980, 142 . 7.0 TRANSPORTATION FUELS FROM NATURAL GAS 7.1 Introduction and Summary 7.1.1 Technical Overview The emphasis of this chapter is chemical transformation of natural gas to produce fuels for transportation use. In all, six important portable liquid fuels can be manufactured from natural gas: : @ compressed natural gas (CNG) e liquefied natural gas (LNG) e liquefied petroleum gas (LPG) e methanol e@ gasoline e methyl tertiary butyl ether (MTBE) The first three involve physical processes, the latter three chemical processes. Compressed natural gas is simply the light fractions of natural gas (mostly methane) compressed and stored at ambient temperature. It is finding application as an automotive fuel for fleet vehicles. Fleet testing of: this fuel is being conducted today in the Anchorage area by ENSTAR Gas. The more significant technical issues regarding this fuel concern its automotive utilization rather than in its production, which are discussed in Chapter 30. For long-range large-scale transport, the lightest fractions of natural gas can be liquefied. to produce liquefied natural gas. This is now. being practiced at the Cook Inlet. However, this is. a mechanical cryogenic: refrigeration process, involving no chemical conversion. LPG is finding increased uses in motor fuel. It is used for fleet vehicles in several major cities, and has long been used as a fuel for short range industrial vehicles. It is produced by liquefying the heavier fractions of natural gas (principally propane and butane) by condensing them at elevated pressure. Fuel grade methanol can be produced by reacting natural gas with steam to form carbon: monoxide and hydrogen, followed by catalytic synthesis of methanol... After distillation to remove low boiling impurities, water and higher alcohols, the latter stream is blended back into the purified methanol. This methanol can be used directly 143 as a fuel for modified gasoline engines, modified diesel engines, or gas turbines. Its performance as an automotive fuel may be improved by blending of up to 15% gasoline with the methanol to aid in cold starting. Alternately, a small amount of methanol can be blended into gasoline (a few weight percent) as an octane enhancer. M-Gasoline is a proprietary process to produce synthetic gasoline by step-wise conversion of undistilled methanol to dimethyl ether, olefins, and finally to aromatics and branched paraffins. The product is gasoline-like material which performs with an octane number (R+M)/2 of. 90.(1) Since it does not have all of the properties normally associated with gasoline, it is identified in this chapter by the trade name M-Gasoline. : Methyl tertiary butyl ether (MTBE) is manufactured by liquid phase reaction of methanol with. isobutylene over ion exchange resin catalyst. The product is a high octane blending component used in unleaded gasoline. : In the long term, fuel grade methanol appears the most attractive natural gas based liquid fuel for displacing gasoline use in Alaska due to its proven technology, cost of manufacture and reduction in air emissions. It may also find other economical applications in the state, particularly as a fuel for small to medium scale power generation. In the short term, M-Gasoline appears to be more a economical fuel. However, its production is technically unproven on a commercial scale. MTBE is also economically attractive, however, it would have limited impact on Alaskan gasoline consumption since it is a gasoline additive, not a bulk constituent and its production could be limited by the availability of isobutylene. oO Production of methanol and/or its derivatives M-Gasoline and MTBE would allow high value use to be made of Alaskan natural gas in: the near future. Such high value uses could include: e Displacement of use of gasoline in Alaska. Up to 5,000 barrels of motor gasoline are brought into Alaska from the Lower 48 daily.(2) Alaskan methanol could displace all of this if used as a motor fuel neat or blended with 15% gasoline. This methanol could also be used to synthesize M-Gasoline, displacing imported gasoline while using the existing distribution systems and vehicle stock. Finally, it could be used to stretch Alaskan gasoline supplies by use in synthesis of the blending ingredient MTBE. 144 e Displacement of use of light distillate fuels in Alaska. — Methanol could be used as a fuel for gas turbines and modified diesel engines for power generation. In the future, it may also be used as a highly efficient fuel for fuel cell power plants. e Export to the Lower 48 and Japan. Methanol and its derivatives could be expected to find uses in these distant markets similar to those described above for use in Alaska. Use of Alaska's natural gas to produce methanol and its derivatives must still overcome economic hurdles. Any such plant built in Alaska would probably be located at a site where limited infrastructure and local skills availability would increase its capital, operation, and maintenance costs to levels well above those to be expected in the Lower 48, ' Also, in many cases, the synthesized liquids must still be transported great distances to markets. This will be particularly true for liquids synthesized on the North Slope from that region's gas resources. Liquids produced here could only be shipped by sea during a few months of the year; their shipment by pipeline to the Railbelt would require construction of a very expensive pipeline. . 7.1.2 Alaskan Perspective Presently, there are no methanol, M-Gasoline, or MTBE production facilities in Alaska. A 7500 ton per day facility is being planned by the Cook Inlet Region, Inc. and Placer Amex Inc. (San Francisco). The plant would utilize 8.7 million tons/yr of subbituminous coal from the Beluga field (rather than natural gas). The product would be - exported to the U.S. West Coast.(3) . The price and availability of natural gas in Alaska (4) make production of methanol or M-Gasoline here an opportunity which is unique in the U.S. Even though there are presently many methanol production facilities operating on natural gas in the Lower 48, no new facilities are being planned. The primary reason is that the current price of natural gas makes it uneconomical compared to foreign locations, where natural gas is available at prices comparable to those prevailing in Alaska. , Lack of a source of isobutylene in the Cook Inlet area may effectively prevent economical production of MTBE there. It might be produced eventually using butane from the North Slope as the source of isobutylene. 145 FUEL TABLE 7.1 METHANOL AND M-GASOLINE ECONOMICS Manufacturing Cost Storage, Transportation, and Distribution Cost Federal Tax State Tax Total Cost at Pump Volumetric Mileage Equivalent to One Gallon of Gasoline Mileage Equivalent Cost to One Gallon of Gasoline Methanol 0.28 0.13-0.40 0 0 0.41-0.68 1.82 $0.75-1.24 ($/Gal) Gasoline Blend M-Gasoline 85% Methanol 15% Gasoline 1.19 0.42 0.77 0.13 0.13-0.40 0.13 0.12 0 O° 0.13 0 0 1.44 0.55-0.82 0.90 1.00 1.62 1.00 1.44 0.89-1.33 0.90 146 7.1.3 Significance of the Technology There are no technical limitations to production of fuel grade methanol from natural gas in the Anchorage area. The main barriers to its broad use to displace gasoline in Alaska are likely to be in the manufacture or conversion of automotive vehicles to run on this fuel, and in the storage and distribution of the fuel. Conversion of methanol to M-Gasoline eliminates these problems and appears to offer some economic advantage, especially in the short term. However, its application raises questions of conversion technology immaturity. MTBE synthesis. raises no fundamental technical issues. Its economical production in Alaska will depend upon development of an inexpensive supply of isobutylene. . Fuel grade methanol could be produced in Alaska from natural gas at a pump selling price of $.82 for equivalent mileage to a gallon of gasoline (Table 7.1). The analysis is based on use of fuel grade methanol as a blend with 15% gasoline to achieve acceptable cold starting performance; $0.40 per gallon storage, transportation, and distribution cost for the blend; and no federal or state excise tax. Similar analysis yields a pump price for a gallon of M-Gasoline of $0.90 per gallon. In the long term, if methanol becomes more widely used, _ storage, transportation, and distribution costs would drop to close to that for gasoline, $0.13/gal. In this case, fuel grade methanol would be economically competitive with M-Gasoline; both would be more economical than gasoline based on petroleum. : 7.2 Description of Technology 7.2.1 Operating Principles 7.2.1.1 Fuel Grade Methanol Methanol can be manufactured from natural gas in a two-step process consisting of steam reforming followed by methanol synthesis (Figure 7.1). Steam reforming converts natural gas to a synthesis gas composed primarily of carbon monoxide (CO), carbon dioxide (CO,), and hydrogen (Hj). This synthesis gas is compressed and then reacted over a catalyst to produce methanol. In the steam reformer, methane, the major constituent of natural gas, reacts as follows: 147 Natural Gas Purge Gas Fuel Oil Feedstock Refined Methanol Fuel Grade Methanol Boiler Feedwater Flue Gas Heat Recovery To Atmosphere Steam Reformer Process Gas Cooling Compressor ‘Compressor Desulfurization Wastewater Recycle Synthesis Distillation Source: Arthur D. Little, Inc. FIGURE 7.1 SIMPLIFIED METHANOL PROCESS FLOWSHEET . . CH, + H,O + CO+ 3H, (1) Other hydrocarbons undergo similar transformation by: Cc +nH,O + nCO+ (2n+ 1) Hy (2) Hon 42 2 The gaseous products may also undergo the water gas shift reaction: CO+H,O = CO,+H, (3) Since the overall reaction is endothermic, heat must be supplied. This is accomplished by burning natural gas and vent gas from the subsequent methanol synthesis step to heat the process natural gas and steam as they pass through tubes packed with catalyst. This gas combustion takes place inside the reformer, a refractory lined vessel, through which the catalyst-filled reformer tubes pass. This reformer catalyst is readily. deactivated by: sulfur in order to protect the catalyst, essentially all sulfur is removed from the gas prior to introduction into the reformer. An excess amount .of process steam must also be added to the process gas to drive the reactions to the right and prevent formation of carbon (soot) according to the following reaction: > 2° C + HO (4) Formation of this free carbon in the catalyst pores causes physical breakdown of the catalyst... It also reduces overall methanol yield. CO +H A number of factors must be considered in optimizing operating temperature and pressure in the reformer including final methane content of the synthesis gas, pressure requirements for the synthesis gas, temperature limitations on the reformer tubes, and process steam requirements. Optimum reformer operating conditions are. 1600-1650°F and 290 to 365 psig for a 2750 ton/day facility. The preferred reformer vessel configuration is based on firing fuel gas at the top. The advantages of this design are as follows: e most of the heat is added at the top of the reformer tubes where the reforming reactions are proceeding most rapidly; e the flue gases are forced downward through the furnace and exit the bottom by the induced draft fans, thereby reducing installation cost of the convection section; @ it allows for a modular design which occupies less area and reduces heat losses because less surface area is exposed to atmosphere, resulting in 5% to 10% higher thermal efficiency. 149 T a Oo Natural Gas = Boiler Feedwater Steam 4 Feed Steam Flue Gas Heat Recovery Atmosphere Steam Reformer Synthesis Autothermal Reformer Process Gas Cooling Desulfurization fos Feedwater Purge Gas Fuel Grade Methanol Source: Arthur D. Little Inc. FIGURE 7.2 LURGi STEAM/AUTOTHERMAL REFORMING Compressor Compressor —_— " Methanol Synthesis (’ Recycle Gas Refined Methanol pf _ : Wastewater Sidewall fired burners allow more uniform control of temperature, but at the expense of installing more burners and more controls. The idewall design reduces the risk of direct flame impingement on. the tubes and is also capable of part load operation under natural draft conditions. This aspect would reduce reformer unavailability due to failure of an induced draft fan. Lurgi offers another reforming approach. The Lurgi synthesis gas production system combines steam reforming similar to that described above with autothermal. reforming (partial oxidation) as shown in Figure 7.2. Half the process gas is compressed and then fed through a steam reformer. The temperature is run lower (1380 to 1440°F), but the pressure is higher (580 psig) than normal for steam . reforming, The methane concentration in the gas leaving the steam reformer is significantly higher (20 volume %). This reformed gas is then mixed with the rest of the compressed process gas and fed to the autothermal reformer. Here pure oxygen is injected with some steam to convert the methane to carbon oxides and hydrogen. The scheme processes a synthesis gas, with-.a composition that is chemically correct for production of methanol. In this respect it is a major improvement on conventional steam reforming, which. produces one mole of excess hydrogen per mole of carbon oxides. This excess hydrogen must later be purged from the methanol synthesis system. Other. advantages of the Lurgi system are a 25% reduction in synthesis gas compression requirements and lower temperature in the conventional reformer. The disadvantage is the need for an air separation plant and oxygen compressors. The hot synthesis gas leaving the reformer is cooled by exchanging heat to raise steam and to preheat other liquid streams. Final cooling is by exchange with cooling water. Air cooling is also an option and particularly attractive in cold climates such as Alaska's. The reformer flue gas is also cooled before venting out the stack by raising and superheating steam, preheating process gas, and preheating combustion air. Process and waste heat recovery: schemes vary widely between licensors, but the end results are similar. . The cooled synthesis gas is then compressed to 1160 to 1450 psig. This is done to enhance the rate of methanol formation in the synthesis reactor. Methanol synthesis gases consisting of carbon monoxide (CO), carbon dioxide (CO,), and hydrogen (Hy) combine to form methanol according to the following reactions: Reaction Heat of Reaction CO + 2H, + CH,OH -39.1 x 10° Btu/Ib mole (5) CO, + 3H, 7 CH,OH + H,O -21.3 x 10° Btu/Ib mole (6) 151 Both reactions yield heat energy and are favored by high pressure and low temperature. The development of selective copper based catalysts has allowed pressure requirements to drop to 725 to 1450 psig. Temperatures must be kept above 400°F in order to maintain catalyst activity. Since the methanol reactions are limited by equilibrium, a large excess of hydrogen is fed to the reactors to drive the reaction of methanol production. This is accomplished by recirculating unreacted synthesis gas, from which methanol has been condensed and to which additional synthesis gas has been added. Two distinct methanol synthesis reactor designs are available. Imperial Chemical Industries (ICI) uses a reactor separated into a number of catalyst beds (Figure 7.3). As a synthesis gas passes down through each bed, conversion to methanol generates heat which raises the temperature of the gas. Cooler synthesis gas is injected between beds via a series of distributors and serves to cool the gas before entering the next bed. Lurgi offers a reactor design which consists of a series of vertical, catalyst-packed tubes (Figure 7.4) connected to tube sheets at each end. The tubes are surrounded by boiler feedwater (BFW) in the shell. As reaction proceeds down the catalyst tube, heat of reaction is removed by vaporizing the BFW into steam. This method of heat recovery is more efficient because it is done at a higher temperature. This allows. a more constant reaction temperature for the synthesis gas as it proceeds through the reactor and results in a larger fraction of the synthesis gas to be converted to methanol on each ‘pass through the reactors. With either reactor design, some of the unreacted synthesis gas must be purged to remove inerts (mainly methane) and excess hydrogen. This purge is used as fuel gas in the steam reformers. The methanol synthesis reactor product is not pure methanol. It contains water, higher alcohols (ethanol, propanol, etc.) and other. low boiling impurities. This mix is called crude methanol or wild methanol. All of these impurities are removed by distillation to produce chemical grade methanol. To produce a fuel grade methanol, the higher alcohols (fusel oil) removed in the distillation are blended back into the chemical grade methanol. These higher alcohols tend to alleviate the vapor pressure and phase separation problems when blending methanol with gasoline and improve cold starting of spark ignition engines fueled with neat ’ methanol. (5) 7.2.1.2 M-Gasoline | The Mobil M-Gasoline process converts methanol to a high-octane, gasoline-like material containing substantial quantities of aromatics. 152 Source: Uhde GmbH. QUENCH GAS” CATALYST | FIGURE 7.3 ICI METHANOL REACTOR 153 Synthesis IN in Methanol Out J Source: Arthur D. Little, Inc. FIGURE 7.4 CONCEPTUAL DIAGRAM OF FLOWS IN DIRECT STEAM-RAISING (LURGI) METHANOL REACTOR 154 Steam Out Catalyst Boiler Feedwater In The chemistry involves catalytic dehydration of methanol to olefins followed by recombination reactions’ producing a mix of higher molecular weight hydrocarbons consisting of a mix similar to that found in motor gasoline today. The methanol feed to this process does not require purification, thus eliminating the need for the distillation step in the methanol synthesis. Mobil has experimented with both fixed and fluidized bed designs in their pilot plant operations. The advantages of the fluid bed over the fixed bed are superior heat transfer characteristics providing simplified heat removal and constant produce quality due to. continuous catalyst regeneration capability in the fluid bed design. However, the fluidized bed requires a second reaction step to produce gasoline blend stock (Figure 7.5), which results in higher overall gasoline yields. In this second step; called alkylation, some of the butanes and propanes in the fluidized bed reactor are synthesized. The fixed bed design offers lower capital investment and higher direct yields of gasoline blend stocks. : 7.2.1.3 Methyl Tertiary Butyl Ether (MTBE) MTBE is produced by reacting methanol with isobutylene under mild conditions, according to the following reaction: C,H, + CH,0OH > Cc aH if Isobutylene Methanol : MTBE The liquid phase reaction uses a readily available ion-exchange resin catalyst (Figure 7.6). .The process consists of a fixed bed reactor system with cooling to remove the exothermic heat of reaction. . The reactor effluent passes to a debutanizer for recovery of MTBE product. The debutanizer overhead, consisting mostly of impurities introduced into the system with the isobutylene, typically contains 0.5 wt. methanol. If desired, the residual methanol in the debutanizer overhead can be reduced to less than 50 ppm with additional equipment. The MTBE product, containing a minimum of 97% MTBE with the remainder essentially all methanol, can be blended directly into gasoline. The blending value of MTBE depends on the octane and composition of the gasoline basestock as shown below: MTBE Blending Octane Values Unleaded Volumetric Basestock Blending Octane Octane. of MTBE in Basestock RON _MON RON , MON 100 90 “11 97 95-100 86-90 , 113-116 98-99 95 86 118 _ 101 155 Alkylation Unit Gasoline Fluidized . Bed Reactor Gasoline Water High-octane gasoline Methanol Catalyst Source: Arthur D. Little : FIGURE 7.5 MOBILE M-GASOLINE PROCESS Butane Methanol Isobutylene/ Butane Mixture Reactor MTBE Distillation Column Source: Arthur D. Little, Inc. FIGURE 7.6 MTBE PROCESS 156 Thus, blending 5% MTBE into a basestock with research octane number (RON) 95 and motor octane number (MON) 86 would yield blend octane numbers of: RON = 0.95 x 95 + 0.05 x 118 = 96.15 MON = 0.95 x 86 + 0.05 x 101 = 86.75 7.2.2 Technical Characteristics Historically, methanol plants have tended to become more economical with increasing size. However, above capacities of about 2,200 ton/day capital investment per unit of production remains essentially - constant. The largest. world scale single train units, at 2,750 ton/day, are still under construction. A. 3,850 ton/day design is presently being offered with hopes to increase the size in the near future to 5,500 ton/day, but no significant economy of scale is expected. (4) The only commercial M-Gasoline facility, under construction in New Zealand, will have a capacity of 14,600 bbl/day. World scale MTBE facilities are on the order of 100,000 tons/year. The Alaskan climate will have only minor impacts on plant conceptual and detailed designs. Economic consequences of these impacts will be mixed. For instance, compared to a U.S. Gulf Coast location, an Alaskan plant would require substantially more freeze protection and insulation. Also, due to the low ambient temperatures, higher quality materials of construction must be specified to avoid brittle fracture. On the other hand, some savings will result because the low ambient temperatures would make air cooling economical throughout the plant. Plant construction in the Cook Inlet will pose no more severe problems than have been successfully addressed elsewhere. For example: e@ methanol plants have been built and operated successfully in Alberta, Canada; and e methanol plants are planned for Tomsk and Gubakha, USSR. These climates are roughly comparable to the Cook Inlet area. (6) Operability Methanol plants are relatively simple to operate. Top-fired reformers are the easiest to control. The Lurgi combined conventional steam reforming/autothermal reforming scheme is also simple to operate, but the auxiliary air separation units and air compressor add significant complexity. The main factors affecting synthesis unit operability are complexity of equipment and controls, ease of startup/shutdown, and catalyst changeout. The Lurgi reactor is more complex than ICI because of its integral waste heat boiler, but its controls are simpler. 157 Startup of the Lurgi reaction is also the easier. On the other hand, catalyst changeout is simpler for the ICI reactor. However, this is of lesser importance than ease of normal operation or startup/shutdown since catalyst is changed annual (or less often) during a scheduled shutdown. Even though there is limited experience with commercial MBTE units, their operability is expected to be comparable to methanol plants. The same is not true for M-Gasoline. Because the technology has not yet been demonstrated commercially, it must be considered to be more complex to operate than either methanol or MTBE. Reliability The reliability of process equipment, services, and the operators determine the time any plant is available for production. The more reliable the process, the smaller the daily capacity of a plant must be in order to achieve the same annual production. , The process equipment for a world scale methanol plant is reliable enough for a typical plant to be available 340 days per year on the . U.S. Gulf Coast. In Alaska, the climate, the relative scarcity of skilled labor and spare parts, and the lower reliability of electric power are expected to result in plant availability equivalent to 310 days/year at full operating output. MTBE plants appear to be essentially as reliable as methanol plants. The immature M-Gasoline techology can be expected to incur operating problems which will reduce plant reliability to that for methanol synthesis. 7.2.3 Environmental Issues The primary environmental considerations for methanol or methanol ‘based chemical plants are air, water, and noise pollution. No significant solid wastes are generated. The major gaseous effluent is the reformer flue gas at 170,000 scf/ton of methanol. Major components in this stream will be nitrogen, carbon dioxide, and water vapor. Approximately six pounds of oxides of nitrogen would be emitted per ton of methanol produced.(7) Sulfur content will depend on levels in the natural gas feed, but can easily be reduced by passing all the natural gas through the feed desulfurizer. Flash gas from let down of the process condensate is vented to atmosphere at 1,600 scf/ton of methanol. This stream contains mostly hydrogen and earbon dioxide, but also contains carbon monoxide and natural gas. Methanol will be. vented from the storage tanks at a maximum rate of 60 scf/ton methanol while filling. Emissions can be minimized by use of a nitrogen blanket and/or internal floating roofs. 158 The major liquid waste will be the.aqueous stream from the refining column bottoms. This stream, consisting of 0.3% methanol, will be generated at the rate of 0.2 ton per ton of methanol and will require biological treatment to reduce BOD levels to below 20 ppm. Other liquid effluents are primarily the result of water treatment for cooling and boiler feedwater. MTBE manufacture will not generate any significant air or water emissions beyond those associated with production of the methanol feed. Methanol to M-Gasoline conversion eliminates the aqueous effluent from the methanol plant, but creates a different aqueous effluent in the M-Gasoline facility, containing 3,700 ppm of formic acid, 3,600 ppm of acetone, and 440 ppm of other hydrocarbons. It is generated at the rate of 1.67 ton per ton of gasoline. About half of this effluent is recycled to reform natural gas. The rest is treated by steam stripping followed by chemical oxidation. No significant air emissions are generated by the M-Gasoline process, apart from those. produced in synthesis: of. the methanol feed. Primarily emissions are from product storage tanks. Noise levels at the plant boundary can easily be limited to 50-55 dBA for all three facilities.This is far below the OSHA limit of 85 dBA, for . an eight hour exposure. Above this level hearing protection is required. As the price of natural gas has increased in recent years, significant improvements in thermal efficiency has been achieved in methanol production to levels as low as 25.1 million Btu of natural gas/ton of methanol are reported. No further significant reduction in natural gas consumption are possible due to thermodynamic limitations. For a world scale methanol plant using conventional steam reforming overall thermal efficiency is about 62% to convert this methanol to M-Gasoline lowers the thermal efficiency to 56%. Use of methanol in motor vehicles will produce less air pollution than gasoline. Emissions of oxides of nitrogen, carbon monoxide, and ‘reactive hydrocarbon would be reduced substantially. The anticipated reduction in air pollution is a primary reason for a major test program in California. (8) 7.2.4 Commercial Status A breakthrough in methanol technology occurred in 1966 when ICI developed a copper based synthesis catalyst capable of operation at low pressure (725-1450 psig versus over 4500 psig previously). Since that time most new plants have been built using the low pressure concept. Today all the plants are built this way. 159 TABLE 7.3 CAPITAL COST SUMMARY TECHNOLOGY : ‘BASIS: Location: Year: CAPACITY: Input (Natural Gas): Output (Methanol): ESTIMATED USEFUL LIFE: CONSTRUCTION PERIOD: CAPITAL COST: Equipment and Materials Direct Labor Indirect Costs Home Office Costs Contingency TOTAL CAPITAL INVESTMENT: ICI Low Pressure, High Efficiency Railbelt outside Anchorage 1982 26.0 million Btu/ton methanol 2,750 tons/day 20 years 4 years $124.5 million 32.9 32.2 18.9 16.9 $225.4 million 160 Data summarizing modern methanol production capacity operating or under contract to construct is shown in Table 7.2. There are no commercial M-Gasoline plants, although a 14,600 bbl/d facility is under construction in New Zealand. Approximately a dozen MTBE units are in operation around the world with a total production of about half a million tons/yr of MTBE. Most have started up in the last two years. . Another three-quarters of a million tons/year of capacity is planned or under construction. Other than modifications to adapt to the climate discussed in Section 7.2.2, there are no constraints to construction of methanol or M-Gasoline facilities. However, for MTBE. a supply of isobutylene must be available. Since the cheapest source of isobutylene is from refinery processing operations not now conducted in Alaska, its availability in the state may pose the strictest limitation on implementation of MTBE in the near future. 7.3__Economic Implications 7.3.1 Costs Capital investment for a world scale, high efficiency methanol plant based on the ICI low pressure methanol synthesis technology is summarized in Table 7.3. This design consumes approximately 26.0 million Btu of natural gas/ton of methanol. Previous Arthur D. Little work indicates that this design becomes economical only when gas prices exceed $2/million Btu (1982). With current Alaskan natural gas prices at $1.70/million Btu, it may be argued that a less efficient plant should be used. However, when natural gas price escalation in real terms (constant 1982 dollars) over the life of the project is considered, a high efficiency design could be economical. _Further, the high efficiency design assures optimum use of natural gas to conserve this resource. If desired, a savings in capital investment of approximately 7% could be realized by building the low cost design which would consume approximately 27.3 million Btu of natural gas/ton of methanol. , For a plant using the Lurgi methanol synthesis technology and conventional steam reforming, capital cost and natural gas eonsumption would be comparable to the ICI low cost design. For the Lurgi synthesis. and combined autothermal and conventional steam reforming, capital costs are comparable to the ICI high efficiency design but natural gas consumption per ton of methanol produced would be approximately 3.5% lower. However, as far as we know, no such methanol plant has been built using the hybrid reforming approach, so these estimates are not based on actual experience. 161 TABLE 7.2 MODERN (LOW PRESSURE) METHANOL SYNTHESIS CAPACITY WORLDWIDE . Process Licensor Number of Plants ICI 39 Lurgi 16 Source: . Reference 6. 162 Total Capacity (tons/day 46,000 14,500 Annual operating cost is split fairly evenly between variable and fixed cost (Table 7.4). Natural. gas dominates the variable cost even when using the high efficiency design and when charged at the modest price for natural gas in Alaska. It is apparent that if the natural gas price increases in real terms over the life of the plant, it will totally dominate the operating cost. Power consumption is rather low at 45 kWh. per ton of methanol; with appropriate design modifications, it could be eliminated completely by cogenerating power within the plant. Boiler feedwater and catalyst costs are also minimal. Catalyst replacement in the reformer and synthesis reactors is required periodically. 7.3.2 Socioeconomic Factors. To operate a 2,750 - ton/day methanol plant would require approximately 50 operators, 40 mechanics (another 200-300 subcontract maintenance personnel would be required primarily at the annual turnaround) and 20 management administrative personnel. Local ‘hiring could fill up to eighty of the operating, maintenance, and administrative positions. The operating and maintenance positions would require six to twelve months of training depending on prior | chemical plant experience. Existing capability to provide any of the equipment and materials for construction and maintenance is essentially nil. However, should a plant be built, some local manufacturing or distribution capability might spring up to supply some of the maintenance materials. On the front’ end, to supply the natural gas for the facility would require increased production capacity with the resulting increase in employment. Use of methanol as an automotive fuel in Alaska would create employment opportunities for modifying automotive vehicles, servicing these vehicles, and storage and distribution of the fuel. 7.4 Impact 7.4.1 Impact on Overall Energy Supply and Use Fuel methanol substitution for gasoline in automotive vehicles would afford a 10% savings in energy over gasoline use due to the increased iuel. economy; i.e., miles per million Btu.(4) However, on a volumetric basis, methanol contains only 49% the equivalent energy of gasoline, so to provide the same driving range, fuel storage requirements would nearly double (1.82 gal methanol/gal gasoline). Methanol/ gasoline (3) and MTBE/gasoline blends would improve fuel octane, which could be credited as energy savings in gasoline blend stock production in the refinery but will not improve automobile mileage. M-Gasoline would not improve mileage either. 163 TABLE 7.4 OPERATING COST SUMMARY TECHNOLOGY: BASIS: Location: Year: CAPACITY: Input (Natural Gas): Output (Methanol): OPERATING YEAR: Unit Operating Costs Cost Variable Costs: Natural Gas $1.70/million Btu ICI Low Pressure, High Efficiency ‘Railbelt Outside Anchorage 1982 (constant dollar) 26.0 million Btu/ton methanol 2,750 tons/day 310 days Unit Annual Cost Consumption (million $) 26.0 million Btu/ton 37.68 Power , $0.092/kWh 45. kWh/ton 3.57 Boiler Feedwater $0.20/m? 0.9 m/ton 0.16 Catalyst $1.75/tonne methanol -- 1.36 TOTAL VARIABLE COSTS 42.77 Fixed Costs: Operating Labor $54,600/yr 50 operators 2.73 Maintenance Labor $54,600/yr 40 mechanics 2.18 Maintenace Material and Services . _ -_ 4.51 Overhead -- -— 2.46 Marketing $5.40/ton methanol , -- 4.65 Taxes and Insurance = - 2.46 Capital Charges 9.5% of Capital Investment 21.41 TOTAL FIXED COSTS 40.40 TOTAL ANNUAL OPERATING COSTS: 83.17 TOTAL ANNUAL PRODUCTION: 853,000 tons TOTAL LIFE CYCLE. COST: $97.50/ton $0.28/ gallon 164 M-Gasoline would allow 100% substitution for gasoline on a Btu basis. Fuel methanol would only allow 85-90% substitution, since 10-15% gasoline would be required to improve cold starting, especially in the Alaskan climate. Near-term substitution of fuel methanol will probably be limited to fleet vehicles. Conversion cost can run _ from $1,200-1,800 per vehicle depending on type.(9) .To redesign and build a new car capable of operation on fuel. methanol would add 5% to the cost in 1983, but could be eliminated by 1987.(4) - Fuel methanol distribution and storage is complicated by its affinity for water and. ability to loosen dirt and scale. Water in methanol used for blends can cause phase separation at levels as low as 0.1 wt.%.(10). This level decreases with a -decrease in temperature, making the problem more severe in Alaska than in the Lower 48. Water in fuel grade methanol actually improves its octane rating.(11) Entrained scale and dirt can plug fuel lines. Limited initial availability of methanol will favor use in fleet vehicles to centralize storage and distribution. . Methanol may find other applications in Alaska as well. It can be used as a gas turbine fuel which yields lower NO, emissions than natural gas or distillate oil. It can displace diesel fuel as the basis ‘of electric power generation in the Bush, either directly as a fuel for diesel engines or more efficiently as a fuel for fuel cell power systems. In the latter applications its use would serve to reduce state dependence on distillate fuels imported from elsewhere. 7.4.2 Future Trends Production of methanol, M-Gasoline, and/or MTBE from Alaskan natural gas would provide new ways to develop this resource for world markets. It would have the added benefit in Alaska of reducing or eliminating gasoline imports for the foreseeable future, despite gasoline demand growth. Methanol might also be used to displace distillate fuels in Bush power generation and to reduce air pollutant emissions associated with operation of gas turbines. The technology for methanol production is well established and should pose no major problems for implementation in Alaska. The main obstacles in the future will be economic. If natural gas price escalates in real terms to a level. comparable to crude oil ($5.50/million Btu), methanol from natural gas will lose its price advantage over gasoline derived from crude. oil or possibly methanol derived from coal. Also unless the auto industry makes a commitment to assembly line production of methanol fueled vehicles, the conversion cost per vehicle will deter consumer investment. A short term alternative would be for the state of Alaska to provide some funding to cover this cost. For instance in California, the state allows a $1,000 state income tax credit for modifying an automobile for methanol use. Initially, the state may also have to subsidize the storage and distribution system to ensure adequate supply and availability. 165 Timing to design and construct a world scale methanol plant is four years based on Cook Inlet gas. To develop and commercialize vehicles with high compression engines suitable for operation on neat methanol would require about five years. (4) Design and construction of an M-Gasoline plant likewise would require about four years. To benefit from New Zealand operating experience, this design and construction process should not begin until that plant has operated for approximately four years. This implies an earliest possible on-line date of 1989-90. Design and construction of an MTBE plant would likewise require approximately four years. Such a plant requires a source of butanes. Deployment of an MTBE plant in Alaska must await development of the natural gas resource on the North Slope, the only substantial source of butanes in Alaska. 166 | 10. 11. 7.5 REFERENCES S.E. Voltz et al., Development Studies on Conversion of Meth- anol and related Oxygenates to Gasoline, prepared by Mobil R&D Corporation for ths U.S. Energy Research and Development Administration under Contract No. EX-76-C-01-1773, November 1976. State of Alaska Long Term Energy Plan, Department of Commerce and Economic evelopment, ivision of Energy and Power Development, 1982. Davy McKee, Inc. Coal Methanol Feasibility Study: The Beluga Methanol Project, prepared for the Cook Inlet Region, Inc., an Placer Amex, under DOE Grant DE-FG01-80RA-50299, 1981. Confidential memo. World Bank, Emerging Energy and Chemical Applications of Methanol: Opportunities for Developing Countries, . Lurgi GmbH, private communication. Confidential communication. P.W. McCallum, et al., "Methanol/Ethanol: Alcohol Fuels for Highway Vehicles," Chemical Engineering Progress. August 1982, 52-59. G. Parkinson, et al, "California Puts Methanol Plan into High Gear," Chemical Engineering, November 1, 1982, 30-33. J.L. Keller, et al., Methanol Fuel Modification for Highwa Vehicle Use, prepared by Union Oil Company for the Us. D.O.E. under Contract No. EY-76-C-04-3683, 1978, W.H. Kampen, "Engines Run Well on Alcohols," Hydrocarbon Processing, February 1980, pp. 72-75. 167 8.6 LARGE-SCALE SOLID FUEL GASIFICATION ; 8.1 Introduction and Summary 8.1.1 Technical Svervisn * Gasification is: a generic term applied to processes which produce predominantly gaseous products, or intermediates. by thermal conversion of solid fuels, such as coal; peat, and wood. These carbon-bearing solids are reacted with a hydrogen source, usually steam, in the presence of heat to. yield a mixture of mainly carbon monoxide and hydrogen plus other gases. such as methane, carbon dioxide, and various contaminants. The gas mixture may be purified and adjusted in composition so that it can be used directly as a clean | fuel or to synthesize fuels, such as methanol or substitute natural gas (SNG), or chemicals such as ammonia. .Producing a gas from a solid fuel creates a fuel which is .essentially as versatile and inexpensive to use: as natural gas and, hence, one which is directly substitutable for natural gas in many of its fuel.and chemical applications. To the extent that natural gas and petroleum products compete for energy markets, so too can synthetic gas be considered a potential substitute for oil. Since petroleum, and natural gas prices and availability are importantly impacted by the world oil supply and demand situation and the potential disruptions to it, solid fuel gasification has received much attention in the United States and elsewhere to provide an alternative to the conventional energy forms and to help dampen the potential impact of any future disruptions. 8.1.2 Alaskan Perspective Alaska has substantial solid fuel resources in terms of coal, peat, and wood. However, due to relatively low natural gas prices in the Anchorage area and sparse infrastructure to handle solid fuels and/or their gaseous derivatives elsewhere in Alaska, large-scale solid fuel gasification has’ not been an economic alternative. As a result, no such plants have been built. Recently, Placer Amex, Inc. and Cook Inlet Region, Inc. have studied the feasibility of locating a 25,000 ton/day coal-to-methanol plant near the Beluga coal field west of Anchorage.(1) The 7,500 tons/day of methanol produced would be transported to California to be marketed there as a utility fuel and possibly as an automotive fuel. The market value of methanol would: apparently be more. favorable there than in many other locations because, as a relatively clean-burning fuel, methanol in these applications would ease the burden of meeting the stringent California emission limits.. . 169 In concept, the Beluga project is a means of exploiting one of Alaska's significant reserves of solid fuel by mining it, upgrading its value through gasification, and marketing it out of state. Again, because favorably priced natural gas is currently available, there is little incentive to attempt to compete directly with it for in-state markets, particularly in the Anchorage area. In populated areas outside Anchorage, where natural gas is not as readily available, gas from solid fuels can potentially be produced at a cost competitive with conventional fuels. But the infrastructure for gas distribution is lacking, and this approach would consequently suffer economic disadvantages. As a result, it would appear that for the foreseeable future, while conventional fuel sources remain reasonably priced, large-scale solid fuel gasification would be mainly limited to synthesis of commodities. for which sufficient value can be obtained for Alaskan resources in out-of-state markets. The socioeconomic. benefits of a major project in Alaska would be significant. A two-plus billion: dollar project such as Beluga would involve the temporary employment of a large construction labor force, development and operation of a substantial mining effort, and a permanent plant operating and maintenance force of about one thousand. 8.1.3 Significance of the Technology Since the late 1800's numerous solid fuel gasifiers have been developed and operated: worldwide. Initially, the gaseous fuels were distributed in pipelines for domestic heating and lighting. Later, new uses were found in chemical applications (methanol: and ammonia). During World War II, the Germans utilized coal gasification to synthesize liquid fuels for the war effort. After the war, cheap natural gas captured many of the markets previously served by coal gas, especially in the United States. The rapid escalation in world oil prices in the 1970's and the temporary shortages of natural gas in parts of the United States during the middle of that period, have rekindled interest in gasification technology as a means of stabilizing fuel prices by basing them on domestic, non-OPEC resources. As a result of the early experience with the technology and the recent development efforts to modernize it, gasification is technically feasible today, at a wide variety of scales.: Reliable equipment is available to process from a few tons per day of solid fuel to more than one thousand tons per day. Other comparably sized gasifiers are in various stages of development, some quite advanced. As a general rule, small gasifiers are uncomplicated. They are used to produce fuel gas and are close-coupled to boilers, heaters, diesel generators, etc. Large gasifiers, although they sometimes find similar applications, are frequently a part of a more complex scheme to produce chemicals or synthetic fuels. As such, large-scale 170 od gasification involves the integration of several subsystems to prepare the fuel for gasification, gasify it, purify the raw gas, provide steam and power, control emissions and, if necessary, convert the gas to other products. Many of these subsystems display significant economies of scale and, as a result, large-scale systems can involve several gasifiers which collectively are designed to convert as much as several thousand tons per day of fuel. This allows further economies with regard to the supporting fuel supply operations (e.g., coal mining, or peat and wood harvesting) as well as the necessary product transportation systems. Countering these benefits are the need to raise larger sums of capital, and the technological and project risks commensurate with very large projects. Based on Alaska-specific data prepared for the Beluga project, it appears that methanol can be produced from coal for about $0.80/gallon (equivalent to about $12.30/million Btu). A somewhat simpler version of the same project might produce clean fuel gas alone at a cost- of $7.80/million Btu. The fixed plant costs for these projects would be about $2 billion and $1.5 billion, respectively. Although methanol and fuel gas produced from coal at the above values cannot compete: with some currently available sources in. Alaska, large-scale gasification may find application in the~ next several years in producing commodities for. export to locations where prevailing energy values may be. higher. ‘ 8.2 Description of the Technology 8.2.1 Operating Principles Gasification Reactions The major constituents in solid fuels are carbon, hydrogen, and oxygen. It is this carbon and hydrogen which account for the fuel's energy content. Impurities such as sulfur, nitrogen, and minerals are usually also present. The primary thrust of solid fuel gasification is to react the carbonaceous solid fuel with a hydrogen source, usually steam, to produce a mixture of combustible gases by the following reaction: CHO, + (1 - y)H,O + heat +CO + (1-y + DH, (1) Since the. gasification reaction is. endothermic (i.e., requires heat), additional energy must be supplied to enable the reaction to progress. The heat may be supplied by adding a hot medium such as molten salt, lime, a hot reactant, or hot ash to the -gasifier vessel.. However, in most systems the preferred approach is to combust a portion of the fuel within the gasifier itself using either air or oxygen. When oxygen is used, the chemical reaction is: ,x_y x CHO, + (1+ Z 704 > co, + 5 H,O + heat (2) 171 While these irreversible reactions involving solid fuel and gas are taking place, several reversible reactions are occurring among the gasification products. The two most important are methanation: - CO + 3H, = CH + HO + heat (3) and the water gas shift: CO + H,O + CO, + H, + heat (4) As these reactions proceed to the right, they liberate energy which partially offsets the need for combusting a portion of solid fuel (equation (2)). Because both these reactions are exothermic, they are favored at lower temperatures. Thus, the actual composition of the product gas is importantly determined by the operating temperature of the gasifier. Temperature is controlled by the quantities of oxygen and steam which are fed to the gasifier. Oxygen via equation (2) liberates a considerable quantity of heat, tending to increase the temperature. Steam promotes equation (1), a consumer of heat, while at the same time acting itself as a heat sink, needing to be heated from an inlet temperature of several hundred degrees Fahrenheit up to the gasifier temperature, which can vary from 1500°F to over 3000°F, depending on the technology selected. Some gasifiers accomplish the gasification reaction at temperatures which are low enough to allow the inorganic impurities in the feed to remain in solid form. The ash fusion temperatures for most solid fuels, particularly coal, are usually above 1900-2000°F. Thus, gasifiers which avoid gas melting must operate at temperatures of 1800-1900°F or less. Gasifiers which run hotter than the ash fusion temperature of. the fuel are said to be run in a slagging mode. Although handling molten slag is sometimes more complicated than dry ash, the major benefit is that high temperature tends to promote gasification reactions at more rapid rates and also enables the gasifier to process a fairly wide variety of solid fuels. High temperature gasifiers do not produce tars and oils, since these compounds are destroyed at high temperatures. However, low temperature gasifiers frequently do produce significant quantities, particularly in moving bed gasifiers where the tars and oils evolve in the cooler upper zones before they are fully heated to gasification temperature. These troublesome materials cause obstructions in downstream equipment and contaminate process streams with which they come into contact. The technology for handling these problems is available but adds somewhat to the cost of the plant. 172 uf Solids Feeding For atmospheric pressure gasifiers, a variety of solids feeding techniques have been employed including pneumatic conveying, screw feeders, and other conventional mechanical systems. However, many potential applications of solid fuel gasification require that the pressure of the gas be substantially above atmospheric pressure. For example, synthesis of methanol and other chemicals and fuel is ordinarily accomplished at elevated pressure. Operating the gasifier at elevated pressure eliminates the need for a compressor and the associated compression energy demand, but it requires special consideration of feeding solid fuel to the gasifier and removing solid ash particles from it. Two generic approaches have been employed for: maintaining a pressure seal within the gasifier: slurry feeding and lock hoppers. . Slurry feeding involves grinding the solid ‘fuel into particles which can be dispersed in an appropriate liquid, usually water or oil. In this way, the solids can be pumped as a part of the slurry in order to achieve the pressure level needed to flow into the reactor. Despite the relative simplicity of this approach, slurry feeding does have drawbacks. One drawback is that the slurrying liquid must be vaporized within the gasifier, an operation which consumes energy which would otherwise be available to support gasification. Another drawback, the use of oil as a slurrying medium, introduces the added costs of oil as a gasifier fuel. Lock hoppers, which are used for feeding of solids to a pressurized ' gasifier, are depicted in Figure 8.1. Solids flow by gravity from an atmospheric pressure feed hopper into a lock hopper. When the lock hopper is sufficiently full, the top valve closes and an appropriate gas is injected to raise the pressure to a level consistent with that of the gasifier. Then the bottom valve is opened to replenish the fuel in the pressurized feed hopper from which the fuel is metered ‘ directly into the process. A critical aspect of lock hopper design is that the valves which pass solids withstand erosion and yet seat well to provide a good pressure seal. Materials of Construction Most gasifier parts are exposed to operating temperatures considerably higher than can be withstood by commonly available materials of construction. As a result, these parts must be protected by one (or both) of two methods: refractory lining and water jacketing. Refractory linings, which involve brick or castable ~ materials, insulate the structural metallic components from the hot gasifier temperatures. Appropriate refractory materials are usually developed especially for a given gasifier design based on its need to withstand temperature and erosive and corrosive effects of mineral 173 _ x ry Solid Fuel Atmospheric Pressure Hopper Lock Hopper Pressurized Feed Hopper To Gasifier Lock Hopper Charging Source: Arthur D. Little,Inc. Solid Fuel To Gasifier (b) Lock Hopper Pressurization FIGURE 8.1 Solid Fuel Inert Gas To Gasifier (c) Lock Hopper Discharging LOCK HOPPER OPERATION Solid Fuel Gasifier (d) Lock Hopper Depressurization faneiceety matter and gaseous components within the gasifier. With water jacketing metal can be exposed directly to high temperature conditions because jacket water cools the metal by absorbing heat in the form of pressurized boiling water, which is ordinarily collected and used for steam in the process. Gasifier Configurations Most gasifiers are categorized by the. nature of the solid and gas interaction which effects gasification. These are: moving bed, entrained bed, and fluidized bed modes. These modes are analogous to the solid fuel combustion modes of stoker firing, pulverized fuel-firing, and fluidized bed combustion, respectively. The Lurgi gasifier (Figure 8.2) typifies the moving bed - gasifier, which is sometimes referred to as a fixed bed. Coal is fed at the top of the vessel and slowly passes downward through the gasifier into successively hotter zones. At the top, where the fuel begins to heat, moisture evaporates and the fuel becomes drier. Next, the dried fuel begins to pyrolize, and some of its volatile oils and tars evolve as vapors. In the next zone the carbon-rich materials react with water vapor in accordance with equation (1) to produce the major gasification products. Finally, in the bottom of the gasifier, residual char is burned with oxygen on a rotating grate to produce the heat needed in the upper zones of the gasifier, leaving a residual dry ash which is removed through a lock hopper. Because of the importance of good gas distribution with the moving bed, care must be exercised in operation and fuel selection for moving bed gasifiers. For example, fuel dust (smaller than about 1/8 inch) cannot be tolerated as it tends to fill the interstices between the larger fuel particles; therefore, fines must be utilized for another purpose, such as steam raising, or discarded. Caking coals which swell as they heat up to gasification temperature sometimes cannot be used because they cause bridging of fuel particles which hamper proper flow of solids through the bed. , A recently developed version of the Lurgi gasifier permits slagging of ash particles at the base of: the gasifier. This development of the British Gas Corporation (3) increases gasifier capacity and overall conversion efficiency, while it eliminates the need for a grate. In an entrained bed gasifier, an example of which is the Koppers-Totzek gasifier (Figure 8.3), powdered fuel particles pass through the gasifier along with the gases.(4) Because of the intimate mixing between gas and solids at high temperature (about 3000°F), reaction times are considerably shorter than a moving bed gasifier, and good carbon utilization results. In the so-called two-headed Koppers-Totzek gasifier, powdered fuel enters the burner where it is mixed with oxygen and steam in a high temperature flame, and the 175 SCRUBBING COOLER GAS ISVUGEOUDLOGAUEONUGSOUGOOESUOGUKOOUULIGUQUSUOHOUGEHE| “WATER: JACKET ASH LOCK Source: Lurgi Corp. FIGURE 8.2: MOVING BED GASIFICATION (LURGI) 176 ar ; ee En . Ret a) s 01) : a ‘ a Een renrase . eos Varg I) oe : AND OXYGEN ‘ COOLING WATER : IN BOILER FEED WATER . ! Source: KBW Gasification Systems, inc. FIGURE 8.3 ENTRAINED BED GASIFICATION (KOPPERS-TOTZEK) bad 177 Raw fuel gas product and fines as N A Ky Tsay TYSSNSSSSSESS aN Freeboard Ss SESS DSSS SSS SESS Gasifier bed SoSD Auxiliary _, Steam Combustor Char - ash separator Ash agglomerates Coal and transport gas Source: Westinghouse Electric Corp. FIGURE 8.4 FLUIDIZED BED GASIFICATION (WESTINGHOUSE) 178 . reactions move rapidly to completion as the gas is passed upward and out of the gasifier. Because of the high temperature the ash melts and tends to coalesce and drop to the bottom of the gasifier where it is collected, quenched, and removed. The slag droplets which leave with the gas require some quenching of the gas with water to cool it below the ash fusion temperature. This must be done in order to prevent adhesion of slag to downstream equipment. Fluidized bed gasifiers such as that developed by Westinghouse (Figure 8.4) also rely on good solid-gas mixing.(5) A fluidized bed is one where solid-.particles of fuel and ash become buoyant in the stream of gases which pass vertically upward through them. The velocity of the gas is maintained such that the particles are retained _in the bed rather than entrained in the gas. Because the action of the solid particles tends to be rather turbulent in a fluidized bed, the heat transfer is very efficient. The bed becomes nearly isothermal and, as such, its temperature is readily controllable. Advantage is taken of this fact in the Westinghouse and some other fluidized bed gasifiers by controlling the temperature in the bottom zone such that the ash particles become sticky and agglomerate, causing them to fall to the bottom of the gasifier for removal. Gas Processing Each of the generic gasifier configurations described above produces a raw fuel gas which requires further treatment for most applications. These steps usually include gas cooling , gas cleaning, and acid gas removal. Gas cooling is necessary because there are -presently no commercially available systems for subsequent cleaning of the raw gas which -operate at gasifier temperatures. Gas cooling is usually accomplished in one of two ways: by quenching with a direct water stream or by a waste heat boiler. In water quenching, liquid water is contacted directly with the gas such that the water vaporizes, absorbing much of the thermal energy contained in -the gas, thereby reducing its temperature. In a waste heat boiler, water is evaporated to produce steam inside steel tubes; this steam may be utilized for any of a number of purposes within the gasification plant. Gas cleaning refers to the removal of liquid oil and tar particles, if present, and solid ash from the gas, usually by. scrubbing with water. When gas cooling is accomplished by the direct . quench method, it is usually economical to clean the gas simultaneously by scrubbing it with the quench water in a single vessel. Acid gas removal involves removal of sulfur-containing gases and sometimes carbon dioxide by use of scrubbing with an appropriate solvent. Numerous proprietary processes for acid gas removal are available. Selection. among them is based upon the inlet gas 179 composition, the required reduction in acid gas composition, the temperature and pressure at which the system will operate, and the cost, which is affected by each of the preceding factors. 8.2.2 Technical Characteristics The efficiency of a gasifier is usually considered to be a ratio of gross combustion energy contained in the raw gas to the total combustion energy contained in the fuel fed to the gasifier. Since both these measurements are ordinarily made at room temperature, this efficiency is frequently called the "cold gas efficiency." Cold gas efficiency taken alone can be a very misleading measure, since it only considers the gasifier proper and not the fully integrated system of which it is a part. For example, cold gas efficiency ignores: ° the thermal energy contained in the gasifier effluent gas which can be converted to useful form, e.g., steam; e the pressure energy contained in the gasifier effluent which, depending on downstream needs, may eliminate the need for a compressor; and e the energy required to produce oxygen for use in the gasifier. ‘Further, the efficiency of a gasifier varies from one fuel to another depending mostly on the moisture content and heating value of its energy. This is because with a wetter fuel, a larger fraction of the fuel must be combusted in order to liberate the heat necessary to vaporize the moisture. As an ‘example of this, consider two gasifiers operated at similar temperatures: Texaco and Koppers-Totzek.(4,6,7) The Texaco entrained bed gasifier, which is ordinarily water slurry fed, will have a poorer cold gas efficiency and higher oxygen consumption (per unit of fuel) than a dry fed Koppers-Totzek gasifier operating on the same fuel. Countering this, however, is the fact that Texaco is capable of operating at much higher pressures than Koppers-Totzek and therefore balances the drawbacks with other benefits when integrated into an overall fuel conversion process. . Large-scale fuel gasification is a very capital-intensive operation and, as such, is usually intended to run at full capacity nearly all the time. As a consequence, it is not common to think of a gasifier as a load following device. However, if the throughput of the gasifier must be reduced, most designs are capable of operating at about 50% of the design throughput or below without a substantial loss in efficiency. More commonly, if turndown is required, multiple parallel 180 . | r i i gasifier modules will. be installed so that one or more can be taken out of service, leaving the remaining gasifiers in operation at their full throughput. : . Solid fuel. gasifiers are typically manufactured in standard vessel sizes. Such standardization economizes on engineering requirements and enables shop fabricated vessels. to be transported readily by common carrier. For plants designed to process more fuel than a single gasifier's capability, multiple modules are employed. In this way, the gasification section proper does not exhibit economies of scale, but such economies would exist in other areas of the plant, for example, coal handling and storage, oxygen plant, acid gas removal, and utility and offsite systems. Solid fuel gasification plants are relatively complex integrated systems which must be well designed, maintained, and operated in order to be reliable. Frequently, the modular approach is used to advantage in enhancing reliability by spreading the reliability risk among several operating units. An extra gasifier module is often installed to help carry the operating load when another gasifier must be removed from service for either preventive or forced maintenance. As an example of the complexity of a typical integrated coal’ gasification project, the major process flows for the Beluga coal-to-methanol project are depicted in Figure 8.5. Each block is representative of a process subsystem which itself is comprised of numerous pieces of operating equipment procured separately from different fabricators and interconnected in the field. During startup, each subsystem is tested, calibrated, and debugged independently. Once operated as a complete process, the plant may still take several months to achieve full capacity. During normal operation, the plant must be monitored closely by trained operators to ensure efficiency and reliable, safe operation. Operators today are assisted by sophisticated instrumentation and control systems which ease much of this burden automatically. A significant step in most large-scale gasification processes is the air separation plant which produces relatively pure oxygen for use in the gasifiers. Although the use of oxygen is not a firm technical requirement in most applications, it occupies a smaller volume than air and results in a smaller volume of product gas.. These factors offer the following advantages for oxygen over the direct use of air as. oxidant in the gasifier: , e smaller oxidant feed system and, if necessary, compression requirements; e@ smaller gas cooling and cleaning requirements; 181 ost Ain Steam and Power Generation 300 Steam Chat Coal Waste Receving — [Cost Gai Heat Particulate and sification Recovery Removal Preparation & Cyclone Ash ] aa Source: Reference 1. FIGURE 8.5 Ty? psutter Raw CO Shift Acid Gas cos Gas Compression Hydrolysis Removal Steam Make-up Gas Compression Purge Gas Reformer Methanol Synthesis ES Methanol Distillation Product Sto He FLOW SCHEMATIC — METHANOL FROM COAL e if necessary, smaller product gas compression requirements; and . . e@ no need to. remove nitrogen diluent. from the gas prior to its use for chemical or fuel synthesis. Weighted against these considerations, the substantial investment and power requirements for the oxygen plant is usually -- though not always -- justified. The instance where the product gas is to be utilized directly as a fuel is the one where air gasification is most likely to be used. The large quantity of nitrogen in the product dilutes its energy concentration to about 100-150 Btu/cubic foot (low-Btu gas) compared to 250-300 Btu/cubic foot (medium-Btu: gas) for most oxygen’ gasification options. In contrast, the energy content of methane, the predominent constituent of natural gas, has a heat content of about 1010 Btu/cubic foot. Substitute natural gas (SNG), sometimes called high-Btu gas, is rich in methane and can be synthesized from medium-Btu gas. In the Anchorage area where natural gas is relatively inexpensive and elsewhere in Alaska where little gaseous fuel infrastructure exists, it appears that neither the low- nor the high-Btu option would be competitive with the alternatives in the near term. : 8.2.3 Environmental Issues There are three major sources of atmospheric emissions from a gasification facility. e Fugitive dust will arise from the handling and storage of fuel to the facility and, if not collected wet, the ash byproduct. e The tailgas from the acid. gas removal system must be incinerated and released to the atmosphere after removal of all but a trace of sulfur-bearing compounds from ‘the stream. e If direct coal combustion is required for steam and power generation and for coal drying, there will be a stack gas of composition similar to that of coal- fired industrial or utility boilers. Liquid wastes would emanate from a variety of sources. The type and extent of treatment of wastewater is very much a function of the gasification technology employed, the fuel utilized, and the processes employed in converting the fuel gas to marketable products, if any. A major source of such wastewater is the raw gas quench system where many of the impurities in the raw gas are collected by. the water. These include ash, chloride, sulfur compounds, ammonia, and 183 sometimes organic compounds such as oils, tars, phenols, and alcohols. Waste- water treatment is a significant concern in gasification plants and it contributes to the overall cost of processing. Most gasifier licensors offer process packages based on their experience to purify wastewater to applicable standards before they are released. Solid waste is principally the ash byproduct of the gasification process and any solid fuel combustion steps within the process. The main concern in ash disposal is the potential leaching of hazardous substances from the disposed ash into the surface and groundwater. Proper disposal of such ashes can involve: @ onsite disposal in lined ponds; e stabilization and landfilling; e@ mine disposal; or: . e in the case of. wood, disposal on the forest floor as a soil conditioner and nutrient source. Sources of noise in a plant will be typical of any large industrial project involving the operation of heavy machinery. However, most rotating equipment such as compressors, pumps, turbines, and the like are designed to operate relatively quietly at a steady level. The major noise audible outside the plant boundary will be the solid fuel handling equipment. Depending on the solid fuel utilized, the product(s) produced and the gasifier selected, the overall efficiency of a large-scale gasification facility will be about 40-70%. Since this is the efficiency of producing one fuel from another, it may, on the surface, not appear very attractive. However, the solid fuel without conversion is generally more expensive to transport and utilize and is less versatile with respect to end uses than the products produced via gasification. Further, using the products usually results in lower emissions than using the solid fuels directly at the point of end use. But in many instances, this benefit may be exported from Alaska. 8.2.4 Commercial Status Commercial-scale coal gasification began in earnest in the early 1900's to produce gaseous fuels for pipeline distribution to end users in urbanized areas where coal was available... This so-called town gas was later displaced from such applications as cheaper natural gas became available. In the 1930's and 1940's coal gasification in Germany was an important source of synthetic liquid fuels for the war effort. This technology is still used for purposes as ammonia and methanol production and for synthetic liquid fuel production. Since 184 ™" ~ 1955, coal gasification has been practiced in South Africa to produce synthetic liquid fuels as a means of minimizing its vulnerability to disruption in crude oil supply. Table 8.1 summarizes the major large-scale gasification technologies which are commercially available or in advanced development today. Three large-scale gasification projects are currently in construction in the United States, two using Texaco gasification, one using Lurgi. The products of these plants will be electricity, methanol and chemical derivatives, and substitute natural gas. 8.3 Economic Implications 8.3.1 Costs The costs of building and operating a large solid fuel gasification facility are difficult to quantify accurately because they are very dependent on coal properties, site conditions, design philosophy, gasification technology, and end product(s). . In this section, costs are presented for two coal gasification schemes which could be implemented in Alaska, one to produce clean fuel gas for any of a> variety of applications, and one to produce the same fuel gas and to synthesize methanol from it. Each set of costs is based on available data for the Beluga project, (1) updated to 1982 costs. and supplemented by other data as necessary. The process flowsheet for the methanol scheme has already been shown in Figure 8.5 and is that contemplated for the Beluga project. The flowsheet for the clean fuel gas scheme is essentially the same except that the gas compression, methanol synthesis, and related steps have been omitted. This also results in some energy savings in terms of feed coal which does not have to be burned in the power plant, since both electricity and steam are saved by omission of these process steps. Further, since the energy loss associated with methanol synthesis is eliminated, the fuel gas scheme shows a considerably better coal-to-product efficiency, 60% versus 42% for methanol. : Coal gasification processes are capable of better overall efficiencies than these. However, in the case of Beluga, site conditions indicate a relatively lower coal cost and relatively higher capital costs for a given plant than in most Lower 48 locations. _ Therefore, the optimal plant design has apparently attempted to economize on capital at the partial expense of operating efficiency. Tables 8.2 and 8.3 summarize the capital investment for the two options which range to over $2 billion in 1982 dollars. The operating cost estimates shown in Tables 8.4 and 8.5 indicate life cycle costs for methanol of $.80/gallon (about $12.30/million Btu) and for fuel gas of $4.43/million Btu. 185 98T LARGE~SCALE SOLID FUEL GASIFICATLON PROCESSES Operating Conditions Status of Development, Selected Ins . Pressure Temperature Size) __ Technology Gasifier Type Atmospheres _ °F . tons/day Status . Owner Location Lurgt Moving Bed 30 1,800 300 Commercial Britsh Gas Corp. ScotJand . 1,000 Commercial SASOL South Africa 1,000 Commercial - American Natural Resources North Dakota Construction BGC/Lurgi Slagger Moving Bed 30 2,000 350 Demonstration British Gas Corp. Scot Land 700 Demo-construct ion British Gas Corp. Scotland Koppers-Totzek (2) GKT Entrained Bed 1 3,000 1,000 Commerc ial AESCL Ltd South Africa Texaco Entrained Bed 50-90 2,600 150 Demonstration RuhrKohle W. Germany 200 Comme rcial Tennessee Valley Authority Alabama 900 Comm' 1-construct fon Tennessee Eastman Tennessee 1,000 Demonstration — S. California Edison Callfornia Construction Shell Entrained Bed 40 2,800 150 Pilot, Shell W. Germany Winkler Fluidized Bed 1 1,600 400 Commercial Neyvelt India Winkler-High Temp. Fluidized Bed i1.5-10 2,000 20 Pilot Rheinbraun W. Germany Wes t inghouse Fluidized Bed 10-20 1,800 30 Pilot West inghouse Pennsylvanta 1,200 Demonstration — SASOL South Africa Construction Peatgas Fluidized Bed 35 1,700 50 Pilot IGT LLlinvis Source: Arthur D. Little, based on industry Literature. Notes: (1) Size is approximate, per gasifier; may vary with fuel properties. (2) Koppers-Totzek and GKT are versions of the same technology offered by Koppers, Pittsburgh and Krupp, West Germany, respectively. 7-_—_o eee TABLE 8.2 CAPITAL COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD CAPITAL COST Equipment and Materials Direct Labor Indirect Costs Home Office Costs Contingency TOTAL CAPITAL INVESTMENT Coal Gasification Railbelt outside Anchorage 1982 constant dollars 25,000 tons/day Beluga coal 7,500 tons/day methanol 20 years 4 years $1,280 million 434 252 184 $2,150 million TABLE 8.3 CAPITAL COST SUMMARY TECHNOLOGY _ Methanol from Coal BASIS Location Railbelt Outside Anchorage Year 1982 constant dollars CAPACITY Input 22,600 tons/day Beluga coal Output 7,720 million Btu/hr. fuel gas ESTIMATED USEFUL LIFE 20 years CONSTRUCTION PERIOD 4 years CAPITAL COST Equipment and Materials $942 million Direct Labor 319 Indirect Costs 185 Home Office Costs 135 Contingency . - TOTAL CAPITAL INVESTMENT $1,581 million 188 TECHNOLOGY BASIS Location Year CAPACITY Input Output OPERATING FACTOR TABLE 8.4 ' OPERATING COST SUMMARY Coal Gasification Railbelt outside Anchorage 1982 constant dollars 25,000 tons/day Beluga coal 7,500 tons/day methanol 85% (310 days/year) : Annual Operating Costs Unit Cost _ Consumption Annual Cost VARIABLE COSTS 7 Fuel (coal) $25.80/ton 7,750,000 $200.0 million Catalysts, Chemicals & Other Fuel 10.0 TOTAL VARIABLE COSTS 210.0 FIXED COSTS Labor Operating $54 ,600/man-year 300 16.4 Maintenance . $54,600/man-year 600. . 32.8 Maintenance Materials . “53.2 Taxes and Insurance 43.0 Capital Charges 204.3 TOTAL FIXED COSTS 349.7 TOTAL ANNUAL OPERATING COSTS $559.7 TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST ' 2.325 million tons $240.70/ton $0.80/gallon 189 TABLE 8.5 OPERATING COST SUMMARY TECHNOLOGY Coal Gasification BASIS Location Railbelt outside Anchorage Year 1982 constant dollars CAPACITY Input 22,600 tons/day Beluga coal Output 7,720 million Btu/hr OPERATING FACTOR 85% (310 days/year) Annual Operating Costs Unit Cost Consumption Annual Cost VARIABLE COSTS Fuel (coal) $25.80/ton 7,006,000 $180.8 million Catalysts, Chemicais & Other Fuel 4.2 TOTAL FIXED VARIABLE COSTS . . 185.0 * FIXED COSTS Labor Operating $54,600/man-year 180 16.4 Maintenance $54 ,600/man-year 440 — 24.0 Maintenance Materials 39.1 Taxes and Insurance 31.6 Capital Charges 150.2 TOTAL FIXED COSTS 254.7 TOTAL ANNUAL OPERATING COSTS $439.7 TOTAL ANNUAL OUTPUT 57.44 x 10 million TOTAL LIFE CYCLE COST $7.65/million Btu © 190 — ie Figure 8.6 shows the effect of economies of scale on the production cost of fuel gas and methanol from coal. In both cases, a doubling of plant capacity has a benefit of lowering the production cost by 6-9%. For both coal and peat gasification, a larger scale of operation implies a larger mining or harvesting operation, and a deposit of sufficient size. Usually, the gasification plant will be sited within a mile or two of the resource, almost regardless of the size of the project. For wood, however, the forest represents a more dispersed resource. A larger area of forest must be accessed to harvest enough wood for a larger gasification plant, i.e., wood must be cut at distances farther from the plant. Since such distances may extend fifty miles or more for a large plant, the increased transport cost will diminish or even overwhelm the available economies of scale. 8.3.2 Socioeconomic Factors Operation and maintenance of a 7,500 ton/day methanol-from-coal plant would require several hundred permanent operators and mechanics including contract maintenance personnel required primarily at the annual maintenance turnaround. Local hiring could fill the vast majority of these positions which would involve six to twelve months of training depending on the extent of prior chemical and coal plant experience. Existing capability in Alaska to provide any of the equipment and materials for construction and maintenance is extremely small. However, in response to a plant being built, some local manufacturing or distribution capability might develop to supply ‘some of the maintenance materials. The development of a coal gasification industry in Alaska, alone or in combination with other energy-related process industries, would establish a continuing need for local contract maintenance and design services and, hence, further employment opportunities. A similar potential exists for construction and operation of supporting manufacturing capacity for items such as pipe, structural and reinforcing steel, cable, and various civil materials. Operating a solid fuel based plant requires purchase of solid fuels mined or harvested in Alaska. The delivered. price of such fuels as coal, peat, and. wood typically are cost based. As such, payments for these fuels largely flows through as payments to personnel involved in mining, harvesting, and transporting them. The relatively labor-intensive nature of coal mining, wood harvesting, and peat harvesting could be expected to create more jobs than oil and gas extraction, per unit of fuel energy. 191 Fuel Gas Cost $/MBtu Methanol Cost $/gal 10.0 9.0 8.0 7.0 6.0 5.0 5 10 15 20 Coal Input, 1000 tons/day a. Fuel Gas from Coal 1.25 1.00 0.75 0.50 5 10 15 20 Coa! Input, 1000 tons/day b. Methanol from Coal Source: Arthur D. Little, Inc., estimates based on Reference 1. FIGURE 8.6 ECONOMIES OF SCALE FOR COAL GASIFICATION 192 25 25 8.4 Impact 8.4.1 Effect on Overall Energy Supply and Use As a general rule, solid fuels are more expensive to transport and to utilize in conventional combustion applications than are natural gas and petroleum-based fuels. Consequently, solid fuels are typically utilized where there is sufficient economic incentive, i.e., where their costs are much less than for natural gas and oil. In: this sense, coal, peat, and wood are frequently. considered as substitutes for liquids and gases. In many energy markets, the infrastructure and end-use hardware are incapable of accommodating solid fuels directly. In such instances, solid fuel gasification offers a mechanism for synthesizing products which can be readily substituted for conventional fuels. either (1) by virtually replicating their physical and chemical properties as in the case of synthetic gasoline (from methanol) or substitute natural gas (high methane gas); or (2) by enabling direct substitution with little or no end-use hardware modifications as in the case of methanol for spark ignition engines or clean fuel gas for combined cycle power plants. As long as it is envisioned that Cook Inlet gas will remain available at relatively low prices, it does not appear that the role of large-scale gasification there will benefit the local energy picture. Alternatively, solid fuel gasification could be deemed a method to synthesize more readily transportable fuels (e.g., methanol) which could be moved to benefit other areas of. Alaska. However, the same product(s) can be synthesized from the inexpensive Cook Inlet gas--again to the extent that it remains available--and moved to other Alaskan destinations at a lower delivered price. Hence, as long as natural gas is priced competitively with solid fuels, the latter cannot prove a cost-effective substitute via large-scale gasification. Instead, in the near term, gasification is envisioned as a means of benefiting Alaska by exporting its solid fuel resources to other markets. For example, the Beluga project is predicated not on selling its methanol product within Alaska, but exporting it to California, where it can be competitive with other premium value fuels, in part because lower atmospheric emissions will result form its use in peak load power plants and possibly in automobiles. 8.4.2 Future Trends In the future, as natural gas prices rise relative to Alaskan coal and peat resources, the latter will have an opportunity to compete more effectively for in-state markets, in part through gasification. Similar forces in world markets will tend to stimulate demand for the same synthetically derived products. 193 Development and demonstration of gasification technology is continuing mainly in the United States, West Germany, and Great Britain. These activities are likely to effect evolutionary improvements in the technology and slowly bring about modest improvements in economics through lower equipment costs, enhanced reliability, and higher efficiency. : These changes will be far less dramatic in moving the implementation of technology forward than the probable changes in world economic and energy balances. . 194 REFERENCES Davy McKee, Inc., Coal to Methanol Feasibility Study: The Beluga Methanol Project, prepared for the Cook Iniet Region, Inc. and Placer Amex, DOE Grant DE-FG01-80RA-50299, 1981. Cornelius, G. and. H. Vierrath, Lurgi Kohle and Miheralotechnik- Gmbh, Methanol and Electric Power from U.S. Coals Using Lurgi Gasification, presented at Coal Technology > oth International Coal Utilization Conference, Houston, Texas, December 1982. "U.S. Coal Test Program on BGC-Lurgi Slagging Gasifier," Final Report, EPRI AP-1922, prepared by British Gas Corporation, London, England for Electric Power Research Institute, August 1981. "Coal Gasification Plant Conceptual. Design, Volume _ II," ‘ Koppers-Totzek Process, Project 5672-N, prepared by C F Braun & Company for Tennessee Valley Authority, October 1980. "SASOL to Test Westinghouse Fluid Bed Gasifier," Coal ReD, Volume 4, Number 14, August 12, 1982, p. 4. "Coal Gasification Plant Conceptual Design, Volume III," Texaco Process, Project 5672-N, prepared by C F Braun & Company for Tennessee Valley Authority, October 1980. "Screening Evaluation: Synthetic Liquid FUels Manufacture," Final Report, EPRI AF-523, prepared by The Ralph M. Parsons Company, Pasadena, California for Electric Power Research Institute, August 1977. 195 9.0 COAL SLURRY PIPELINES 9.1 Introduction and Summary 9.1.1 Technical Overview Pipelines have long been a standard and integral part of the domestic transmission system for transporting energy related products. For . example, one of the advantages of oil and gas is the ease with which they can be moved over long distances by pipeline, an environmentally and economically attractive transport mode. : By comparison, the traditional means of transporting coal over land by rail and/or truck is cumbersome. Therefore, in the United States today a major emphasis is being placed on employing slurry pipelines to provide a useful option for the long distance transport of coal. In addition, slurry pipelines provide coal in a form that is usable for rapid loading and unloading of ships in the coal export market. A slurry pipeline is one that carries a solid material suspended in a liquid medium. The idea of using’ pipelines to transport solids in slurry form is not new -- it was conceived over 80 years ago. But until recently their use has been limited to relatively short lines, primarily within the mining industry. Commercial slurry pipelines can be found throughout the world, operating in Tasmania, South Africa, Japan, and North and South America, transporting a variety of commodities including: iron, copper, and = zine concentrates; limestone; kaolin; and various mine tailings. (1) To date, however, only two coal slurry pipelines have been built in the United States, both using a water medium to move coal from a single source to a single user. The first, the now idle Consolidation Coal Company line (108 miles) extending from Cadiz, Ohio, to Cleveland began operations in 1957. The second, and currently the largest operating slurry system in the world, is the Black Mesa pipeline. This system began operations in 1970 and transports 4.8 milion tons of coal per year over a distance of 273 miles from Peabody Coal's Black Mesa Mine near Kayenta, Arizona, to Southern California Edison Company's 1500 MW Mohave Power Station, located near the southern tip of Nevada. While the details of slurry pipelines would vary depending on site and situation specific conditions, the transportation of coal via a water slurry system would involve three major components (Figure 9.1): slurry preparation; pipeline transport; and _ slurry dewatering. Slurry preparation involves assembling coal from a mine or group of. mines at a single location where coal cleaning, pulverizing and mixing with an equal weight of water takes place. The prepared slurry is pumped from agitated storage tanks into a steel pipeline through which it is transported with the assistance of reciprocating positive 197 Coa! Supplier 4 Pipeline System Slurry Preparation wf F-< c— Coal Cleaning Water Supply Pipeline System o¢ Coal Buyer we _ Zea aS, a nn Dewatered Coal Dewatering Plant ‘Source: John M. Huneke, Testimony betore the House Commitise on interior and Insular Aftairs a Wem Barges ‘On Coal Sturty Pipeline Legislation, Washington. D.C... Sept. 12. 1978. FIGURE 9.1 COAL-WATER SLURRY PIPELINE SYSTEM 198 Be displacement pumps. The intervals at which the pump stations are located (generally every 60 to 100 miles) depends on the terrain along the pipeline route, pipe size, location accessibility, and other design considerations. At the downstream end of the pipeline, the coal-water slurry is received in agitated storage tanks, from which it is fed into a dewatering facility. Dewatering is accomplished by natural (gravity) settling, vacuum filtration, or by centrifugation, followed by chemical flocculation. After final thermal drying, the coal can be stored, transported further by other modes, or utilized directly. The primary advantages or potential benefits of coal-water slurry pipelines compared to the traditional truck and/or unit train modes of transport are: e very high reliability whose operation is unaffected by inclement weather and ground conditions; e generally lower total capital costs than railroads when comparing new installations of either mode of transport; e relative insensitivity to price inflation due to _ their capital-intensive nature; and e environmental advantages over railroads and other modes of surface transportation (trucks, etc.), since slurry pipelines, as cross-country oil and gas pipelines would be buried underground and therefore operate silently and out of sight. 9.1.2 Alaskan Perspective Coal is a major.energy resource in Alaska (Table 9.1), but the only significant coal mining ongoing is the Nenana field near Fairbanks to supply two local electric power. generating units. If coal is to be mined more extensively for use in the state or for export to world markets, appropriate infrastructure will be required. At present, the only existing means is the Alaska Railroad which is properly located to move Nenana coal to the south coast for end use or trans-shipment. But a sizable coal movement would more than double the current’ revenue tonnage of the entire railroad and some significant new investment would be required. Coal mined at Beluga or the Northern Fields would require completely new infrastructure to move the product. A potentially appropriate means of providing dedicated transport for Alaska coal is a slurry pipeline. Compared to other modes, slurry pipelines are reliable, environmentally benign, and frequently less 199 Northern Fields Nenana Field Jarvis Creek Field Susitna Field Matanuska Field Bering River Field Herendeen Bay Field Chignik Field TOTAL a Nena Broad Pass o S| Susitna Fret => wT fe TABLE 9.1 ALASKAN COAL RESERVES AND RESOURCES FROM EIGHT MAJOR FIELDS IN MILLIONS OF TONS Proven Indicated Reserves Reserves 235.0 49,000 - 120,000 861.6 6,000 0.3 13 - 76 275.0 2,700 - 10,200 6.6 108 - 130 0 0 0 10 - 100 0 100 1.37 billion ‘57.9 billion 200 Hypothetical Reserves 330,000 8, 700 0 27,000 149 36 - 1,000 300 300 366 billion 4 as expensive than railroads, although enhancement or extension of the other modes would offer the indirect benefit of improving the infrastructure for other uses. : . Water availability and right of way issues may be less of a problem in Alaska than in Lower 48 where these non-technical issues have retarded progress. 9.1.3 Significance of the Technology The two major oil supply disruptions of the 1970's have influenced a shift toward the use of coal as a primary energy source. Various options for integrating coal into the energy utilization structure of the nation has been receiving increasing attention in many sectors. One sector is transportation. The major options available for coal transport include the following: (2) uy .. : e Railroads. Railroads are the dominant factor in coal transportation today and are expected to be so for the foreseeable ‘future.’ At present, for. large-scale transport, train load ‘volumes and unit train concepts. are usually employed where coal is a commercially important commodity. The Alaska Railroad would require investment in new capacity, as a minimum, and possibly an extension to reach future coal mine locations. : e Water Transport. Barging is an economical method of coal transport and is used wherever practicable. It may be of limited value in the Interior and many coastal locations of Alaska where winter freezing prohibits traffic. _ Additionally, barging is often an extremely important supplement for rail (rail/barge) transport in several areas of the country. In the years to come, ocean transport would become a key factor’ in the overall U.S. coal transportation network, since it is anticipated that the export of coal from the United States is likely ‘to increase substantially, both from the East and West Coasts, potentially including Alaska. e Slurry Pipelines. Water slurry pipelines are commercial today and several new ones have been proposed. However, opposition in the form of eminent domain legislation and water rights is a major impediment to the growth of water slurry pipelines in the western part of the Lower 48. Non-water slurry pipelines including liquid carbon dioxide are under study and development and may become commercial in. the next few years. 201 e Truck Transport. Truck transport of coal is often used for short distances. Trucks usually carry the coal mined for relatively short distances to tipples. (A tipple is a structure used to tip loads of coal onto rail cars and barges, and possibly in the future to processing facilities for slurry transport.) In addition to trucks, conveyor belts are also used for transporting small amounts of coal over relatively short distances. e Minemouth Power Plants. This concept involves large power plants at the minemouth and the use of extra high voltage (EHV) transmission of a major product of coal utilization -- electric power although coal export from Alaska cannot be practically effected this way. This concept was proposed for some large facilities in the western United States, but environmental opposition to large power plants’ in the minemouth areas has deterred its implementation to date. One potential long-term benefit of slurry pipeline transportation over truck and/or rail is the relative escalation rates of the costs for each transportation mode. Slurry pipelines are capital-intensive in that approximately 50-70% of their annual operating cost is due to capital related charges. For example, of the estimated $19.75/ton it would cost to move Nenena coal to the south coast, about $9.50/ton would be for capital-related cost. In contrast, unit trains are highly labor-intensive -- as much as 80% of unit train operating costs are due to operating and maintenance labor and materials, and the remaining 20% due to capital related charges.(1) Since capital costs are not vulnerable to inflation, while operating .costs escalate over time, the cost of unit train transport will rise at a more rapid rate than that of pipelines, even though pipelines may be more costly initially. 9.2 Description of Technology 9.2.1 Operating Principles The flow of slurries (mixtures of solids and liquids) in pipelines differs from the flow of homogeneous liquids in several important ways. With liquids, the complete range of velocities is possible, and the nature of the flow (whether it is laminar, transition, or turbulent) can be characterized from data on the physical properties of the fluid and pipe system itself. Characterization of slurry flow, however, is- not as simple. The reasons for this are twofold: first, superimposed on the properties of the liquid are the properties of the solid particles which must be accounted for; second, a range of slurry behavior is possible of which the two extremes are homogeneous and heterogeneous flow. 202 ie Homogeneous flow exists when the solid particles making up the slurry are uniformly distributed throughout the liquid medium. It is more nearly encountered in slurries of high solids concentrations and fine particle sizes (e.g., sewage sludge, fine limestone slurries, etc.).(3) Heterogeneous flow exists when the solid particles are not evenly distributed. In horizontal flow, concentration gradients exist vertically in the pipe, even at high velocities. In other words, the liquid and solid phases, to a large extent, retain their own separate identities. Heterogeneous slurries, of which phosphate rock systems are typical, tend to be of lower solid, concentrations and larger particle sizes than homogeneous slurries. Many slurries encountered commercially, however, are of a mixed type, incorporating both homogeneous and heterogeneous properties. This is true of coal-water slurry pipeline systems, in which the finer particle size fractions of eoal join with the carrier liquid (water) to form a homogeneous slurry, while the coarse size fraction of coal becomes suspended heterogeneously in this slurry. The transportation of coal via a coal-water slurry pipeline involves three major components: (1) slurry preparation; (2) transmission through the pipeline; and (3) dewatering of the coal for use as a boiler fuel, for storage, or for transloading to another mode of transport. Since the Black Mesa pipeline is the only operating coal-water slurry pipeline in the United States, it serves as an appropriate model on which to focus the following discussion of coal-water slurry transport (Figure 9.1). Slurry Preparation Plant Coal is received at the slurry preparation plant by conveyor belt which delivers 2 x 0 inch powder into three elevated raw coal bins. Each bin feeds directly into a process train consisting of an impact crusher, a rod mill, a sump, and centrifugal sump pump. Impact crushers reduce the coal to less than 1/4 inch by dry crushing, while downstream rod mills pulverize the coal by wet grinding to less than 8 mesh. A slurry of coal and water is formed in the rod mills where water is first introduced. Slurry is pumped from the rod mill sumps by centrifugal pumps into one of four large storage tanks. These are open top tanks with mechanical agitators to maintain slurry suspension. During normal operation one tank would be supplying the pipeline with slurry, another would be full and awaiting slurry quality test results, a third would be filling with slurry, and a fourth would be a spare. The coal-water slurry is then transferred from these storage tanks by way of centrifugal charge pumps into suction of the transmission system's high pressure pumps. 203 Transmission System (Pumps/ Pipeline) The Black Mesa line is an 18-inch diameter capable of transporting 660 tons of coal per hour (5 million tons of coal per year). At 50% solids by weight, the slurry flow rate is about 4,200 gallons/minute, which corresponds to a velocity of nearly 6 feet/second.(4) If lower delivery rates are required, it is accomplished, to some extent, by reducing the flow rate and further by inserting water "slugs" between batches of slurry. Transit time for the Black Mesa pipeline is three days, with a line fill of 45,000 tons of coal. The line extends for a total distance of 273 miles, with a 16% maximum slope limitation. The line is extensively telescoped with wall thicknesses ranging from nearly .47 inch where -pressure is high. at pump station outlets to .22 inch in low pressure areas at the inlets to pump stations. The spacing of pump stations depends on the terrain, line size, and the economic balance between multiple pump stations and high pipeline design pressures. Generally, stations are spaced at 60 to 100 mile intervals. In small diameter pipelines (less than 10 inches) pump station spacing will tend to be shorter because of higher friction losses. . : The Black Mesa pipeline required four pumping stations, with the largest available piston installed pumps with electric motor drives and hydraulic couplings for speed control. Three stations have three pumping units each plus one a spare, and the fourth has four units plus one spare. In addition, each station is equipped with a dump pond capable of holding the upstream line fill in case of pump failure or other emergency. Each station also has a water pond for flushing out the downstream section of line if an emergency should require it. Dewatering Plant From the pipeline, slurry is directed into large’ tanks equipped with agitators. For the Black Mesa system, slurry is withdrawn from the tanks by centrifugal pumps and transferred to a battery of forty centrifuges. . Centrifugation removes nearly 75% of the water, resulting in a wet cake which is then directed to 20 bowl mills for drying and further pulverization. The dried, pulverized coal is then conveyed pneumatically to power plant boilers. Effluent from the centrifuges containing about 6% solids is pumped to clariflocculator tanks where chemical additives are introduced “to collect the coal fines. The underflow (18% solids) is pumped directly into the boiler. Clariflocculator overflow, the bulk of the original slurry water is sent to a lime soda ash softener and, in the case of the Black Mesa pipeline, is used as cooling tower makeup at the Mohave power plant. (4) 204 9.2.2 Technical Characteristics Except for the availability.of water, slurry pipelines have no inherent geographical limitations.(5)- In fact, the route of pipelines is not generally restricted by the terrain to be crossed, and the pipe can follow the most direct route compatible with access.* In this way, - slurry pipelines can have a distinct advantage over railways. However, routing and terrain are reflected in construction costs, pipeline pressure, and required pumping power. Coal/water slurry pipelines require a considerable supply of water, a problem which has delayed the inception of new slurry pipelines. - In the United States, most of the proposed coal-water slurry pipelines originate in the western coal fields where water supplies are not readily available. One solution to this is to use a closed-loop pipeline system, where a parallel pipeline is laid to be used to return the water for reuse in the slurry preparation plant. Such a system has the advantage of eliminating the need to treat large quantities of water whether for use in some process or discharged to the environment. However, the economics of closed-loop systems are not advantageous over. long distances but may be suitable for short distances. Water availability does not appear to be a problem in much of Alaska. ae Typically, slurry pipelines are made of commercially available mild steel and wall thicknesses. However, the selection of pipe grade, specific wall thickness, and corrosion/erosion allowance must be done with care. Although the external corrosion of the slurry pipeline would not differ from that of oil or natural gas pipelines in a buried environment, the rate of internal metal loss can be accelerated by the. presence of solids. , Presently, all commercial long distance slurry pipelines are buried. The primary reason-for this is that the mass production techniques developed for oil and natural gas pipeline contruction apply equally as well to slurry pipelines. In addition, the earth above a buried pipeline is usually adequate to protect the line from freezing. Under flowing conditions, freezing is generally not a-problem in a buried pipeline. However, if the system is -shut down during low temperatures, pipeline freezing and rupture could occur. Hence, the depth (and cost) of a water-slurry pipeline is importantly determined by the expected maximum depth of the frost penetration, a factor * However, it must be kept in mind that a 16% maximum slope limitatation must be adhered to in the design of the pipeline, so as to reduce the probability of the solid portion of the slurry sliding to the bottom of a steep slope. If this did occur, the potential for causing a plug in the line would be high. 205 which can be costly in Alaska. In addition, burying a line in permafrost requires special protection of both the environment and the pipeline. The importance of careful mechanical design of a slurry system cannot be overemphasized. Pipelines have many advantages over other modes of transportation, but in the absence of a high operating factor (over 95% reliability), many of these advantages would be lost. For example, one important way of ensuring a high availability is to provide spare pumping capacity, thereby minimizing unnecessary shutdowns due to pump breakdown or wear. Also, an adequate supply of spare parts susceptible to wear by abrasive slurries is mandatory. These would include valves, valve seats, plunger or piston packing plunger sleeves or cylinder lines, usually with quick replacement features. Control of slurry particle size and density is important primarily to meet the hydraulic design limits of the pipeline. If particles are ground too fine, pumpability may be good, but the slurry may be difficult to dewater at the terminal. If they are too coarse, an overly heterogeneous slurry results, requiring higher pumping velocities (and power costs) and causing increased wear rates. Consequently, once the size consist of slurry pipeline system is decided upon, it must be rigorously maintained throughout the system operating life. (6) To maintain slurry pipeline operability and stability, slurries are usually designed to have "soft settling" characteristics. This means that when the system is shut down for any reason, the slurry does not pack solid and can be resuspended. This is.an important design consideration, since all systems will ultimately experience unplanned shutdowns which will require restarting the system. 9.2.3 Environmental Issues Adverse environmental impacts from a coal-water slurry pipeline system in Alaska would be minimal. Typically buried with some above ground sections of piping, a coal-water slurry pipeline would be generally out of sight. If the pumps are electric-motor-driven, they would be relatively quiet and have no air emissions. If diesel-driven, some particulate and SO, emissions would result. Water consumption is significant, but the water used for transporting the coal can be clarified and returned to the environment with little . adverse consequence. The risks of spills from slurry pipelines is slight. As a relatively inert mineral, coal is nontoxic to plant and animal life. (3) 206 There is some concern, however, regarding disturbance to the permafrost along the pipeline route. If heat from construction activities or the actual operation of the pipeline thawed sections of the permafrost, resultant subsidence, slumping, and the establishment of new drainage patterns could result. In addition, disruption of the permafrost regime could cause secondary effects of frost heaves which could subsequently dislodge and rupture a buried pipeline. Proper preventive measures (insulated working pads, pipe insulation, etc.) can avoid permafrost damage. (7) 9. 2 -4 Commercial Status The technology for long distance slurry pipeline transport is now commercially available. A summary of the major commercial ; slurry pipeline systems throughout the world is presented in Table 9.2. In addition to the slurry material being transported, this table indicates the location of the slurry pipeline system, transport distance, annual throughput (capacity), and year in which the system began operation. As is evident from. this table, only one of the 21 operating systems (throughout the world) transport coal. The major commercially transported minerals via a slurry pipeline are limestone, copper concentrate, and iron concentrate. The first coal slurry pipeline ever constructed in the United States was built by the Consolidation Coal Company to provide an alternative transportation system to the existing railroads and thus introduce a competitive element into the transport of coal. The pipeline was completed in 1957, linking a coal mine on the Ohio River in Cadiz to a power plant approximately 108 miles to the north, near Cleveland. The pipeline was designed to transport 1.3 million tons per year of coal in a 10-inch diameter pipe. During its six years of operation, it moved. over 7 million tons of coal before. being shut down in 1963. This pipeline was mothballed after introduction of the unit train resulted in reducing haulage costs on railroads. While termed a commercial system, the Ohio pipeline is sometimes more realistically termed a demonstration-type pilot plant because it was too small to achieve sufficient economy of operation, and it experienced considerable coal dewatering difficulties. But, the basic design of the Ohio pipeline was largely followed in the engineering/design of the second coal-water slurry pipeline, the: Black Mesa pipeline. Constructed, owned, and operated by Black Mesa, a wholly-owned ‘subsidiary of Southern Pacific Pipe Lines, it links the Black Mesa coal mine in Arizona with the Mohave power plant in Nevada. This was designed to transport approximately 5 million tons. of coal per year in an 18-inch diameter pipeline over a distance of 273 miles. Since its startup in mid 1970, the system has been operating successfully. If the recent emphasis on shifting towards the use of coal as a primary energy source continues, however, much of the expansion in slurry pipeline volume will be due to large coal slurry pipeline 207 TABLE 9.2 SUMMARY LISTING OF MAJOR COMMERCIAL SLURRY PIPELINE SYSTEMS Pipe Capacity Startup Distance Diameter (million Date Slurry Type System/Location (miles) (inches) tons/yr) (year) Coal Consolidation 108 10 1.3 1957 Ohio Black Mesa "273 18 4.8 1970 Arizona Limestone Calaveras 17 7 1.5 1971 California Rugby 57 10 1.7 1964 England Trinidad 6 8 0.6 1959 Copper Bougainville 17 6 1.0 1972 Concentrate West Irian 69 4 0.3 1972 KBI Turkey 38 5 1.0 NA Pinto Valley 11 4 0.4 1974 Arizona Iron Savage River 53 9 (2.30 - 1967 Concentrate Tasmania Waipipi (Land) 4 8 1.0 1971 New Zealand Waipipi 1.8 12 1.0 1971 (Offshore) Pena Colorada 30 9 1.8 1974 Mexico Las Truchas 17 8 1.5 1976 Mexico Sierra Grande 20 8 2.1 1976 Argentina , Samarco 245 20 12.0 1977 Brazil x Chongin 61 NA 4.5 1975 No. Korea Kudremukh 44 18 7.5 1980 - India : Gilsonite American 72 6 0.4 1957 Gilsonite Copper Japan 44 12 0.6 1968 Tailings Nickel Refinery Western Mining 4.3 4 0.1 1970 Tailings . Phosphate Valep 74 9 2.0 1978 "NA - Information ‘not available Source: Reference 6. 208 systems. Table 9.3 lists the coal-water slurry pipelines currently planned to be constructed in the United States. From this table, it is evident that these six proposed and/or planned coal-water slurry pipelines would increase the total coal-water slurry pipeline system operating capacity (in the United States) by well over an order of magnitude. Alternatives to water have been considered as a transport medium for coal slurries, especially in areas where water is scarce. These alternatives include crude or refined petroleum products, methanol, and liquid carbon dioxide (CO,). The idea to use crude and residual oil is not new. It was explored by the U.S. Bureau of Mines and others in the early 1960's for possible application where synergy of © moving two commodities could be achieved.. Even by itself, however, residual oil is not easily moved over long distances, and slurry pipelines employing such as a transport medium would require heating and insulation. The use of methanol and liquid carbon dioxide as coal slurry transport medium is also being researched. Both liquids can be made from coal at mine mouth plants. In addition, CO, can be obtained from coal-fired power plant stack gases. In the’ case of methanol (suggested by Chieftain Development of Canada (9)), the medium could be sold for chemical or fuel use after its use for transporting coal. Liquid CO, (suggested by Arthur D. Little, Inc.(9)) offers two potential options for the utilization of CO,. First, it could be used in a closed-loop system so that it would” only be necessary to produce makeup CO,. Secondly, in an open loop system, the CO could be marketed “for enhanced oil recovery from depleted fields located near the pipeline terminus. On advantage both methanol and liquid CO, would have over water, in the Alaskan environment, is that both of these fluids would be less suceptible to freezing than water. In addition, coal-liquid co, slurries offer the following advantages over coal-water slurries: e Liquid CO, has very low viscosity compared to water (about 1/15 to 1/30); hence, friction losses in the pipelines would be much lower. Hence, the energy for transport and horsepower requirements are lower (for the same throughput). e The slurry medium can be loaded to a greater extent with powdered coal. For example, water slurry pipelines are usually limited to a maximum of 50% coal (by weight). Studies ‘on the use of liquid. CO, as the transport fluid appear to indicate that liquid co, can be loaded to. about 70% to 80% by weight and still exhibit a lower viscosity than a 50% coal-water slurry. Such higher loading permits the use of smaller pipelines for the same capacity. 209 OT? TABLE 9.3 PROPOSED AND/OR PLANNED COAL/WATER SLURRY PIPELINES v Capacity Distance Pipe Diameter Terrain Name (million tons/year) (miles) (inches) Factor Note Gulf Interstate - NW Pipeline 10 800 30 2.2 Single Delivery . (proposed) Alton Pipeline* 11.6 183 28 2 ETSI Pipeline 37.5 1378 42 1.1 Multiple Delivery Texas Eastern” 22 1260 36 1.2 Single Delivery Houston Natural Gas Co. San Marco Pipeline 15 760 22 1.2 Multiple Delivery Coalstream 15-45 1500 14-36 1.5 Multiple Delivery #al1en-Warner Valley Energy buy tex Source: Reference 8 Systems e@ Carbon dioxide can be separated from coal at the terminal almost completely, using flashing methods supplemented by fabric filters; this is more. efficient than the separation of coal from coal-water. slurries. In contrast to oil and natural gas pipelines, coal-water slurry pipelines require significant quantities of water for operation. Unfortunately, most of the nation's coal deposits (with the exception of Alaska) occur in the water-seare regions of the West. The. major share of the nation's presently identified and potential coal reserves are in Wyoming, Montana, Utah, Colorado, New Mexico, North Dakota, and South Dakota where there has been historically a heavy and keen competition for available fresh water. Further, projected population -inereases to the year 2000 indicate that this situation is likely to be strongly accentuated. Consequently, coal slurry pipelines would compete directly. with other potential users of potable water. (10) 9.3 Economic Implications 9.3.1 Costs The only currently operating coal slurry pipeline in the United States is the Black Mesa. Capital costs and operating experience developed there are basic to any current coal-water slurry pipeline cost analysis. Consequently, this section provides a ‘magnitude-type capital and operating cost estimate for a pipeline of the dimensions of the Black Mesa pipeline in 1982 in the Railbelt. The 18-inch, 273-mile line capable of carrying 5 million tons per year is typical of a possible line from the Nenena field to the Cook Inlet. Since the Black Mesa is a successful commercial coal-water slurry pipeline system for a one source, one market model, its design constitutes a system that may be applied as common to other coal-water slurry pipelines. However, even though capital and operating cost multipliers to obtain Alaska-specific costs from "typical costs" in the contiguous states were used, extreme caution should. be employed in the use of these estimates because they cannot adequately account for such site-specific input variables as: actual pipeline distance; terrain difficulty factor; elevation changes (to determine booster station spacing requirements associated with different routes); water. supply facilities, climatic conditions, construction _ site accessibility, number of river and road crossings, etc.; and labor availability. The Alyeska pipeline is a graphic example of this where cost overruns resulted because of delays for easement needed for rights-of-way, ‘environmental regulations, work stoppages, material delays, severe climatic conditions, and unanticipated cost inflation. Therefore, these estimates should not be substituted for more detailed estimates prepared from a site-specific engineering/cost evaluation. 211 ’ TABLE 9.4 ” CAPITAL COST SUMMARY : TECHNOLOGY Coal/water slurry pipeline a BASIS | Location Railbelt ! ‘Year 1982 constant dollars CAPACITY Input 5 million tons/year 273 miles “ ESTIMATED USEFUL LIFE 30 years CONSTRUCTION PERIOD 3 years . atl CAPITAL COST Equipment and Materials Direct Labor Indirect Costs Home Office Costs Contingency TOTAL CAPITAL INVESTMENT $84.7 million 224.6 53.2 a 17.3 Cos 113.9 5 $493.7 million 212 : The magnitude-type of capital cost estimated presented in Table 9.4 includes the following major process areas and associated equipment: e Slurry preparation plant - coal. handling conveyors, coal storage bunkers, impact crushers, rod mills, slurry holding tanks (with agitation), water wells, and water reservoir and piping. e Pipeline and pumps - main line (273 miles), 4 pumping stations (piston pumps with electric motor drives), 4 emergency slurry dump ponds (1 per pump station), and 4 emergency flushing ponds (1 per pump station). e dewatering facility - permanent slurry storage pond, slurry holding tanks, centrifuges, pulverizers; and flocculating tanks. : : : The capital cost estimate for this hypothetical coal-water . slurry pipeline system includes direct cost of materials and installation plus indirect costs for engineering, management of construction, contingency, and startup costs. The capital cost of such a coal-water slurry pipeline system, in constant .1982 dollars would be nearly $500 million. Of this total, excluding contingency, direct labor accounts for nearly 60%, while equipment and materials contributes nearly 25% of the total. The magnitude-type operating cost estimate presented in Table 9.5 includes: variable costs associated with power and_ water requirements; fixed costs for operating and maintenance labor, maintenance materials, taxes, insurance, and capital charges. The total annual operating cost is estimate at $98.7 million per year. This translates to a total life cycle cost of $19.75/ton of coal transported, of which nearly 50% is due to capital charges, while power and water costs account for 28% more. 9.3.2 Socioeconomic Factors The potential employment effects of building a slurry pipeline are felt in two distinct ways: (1) the short-term effect during actual pipeline construction; and (2) the long-term effect during the operation of the pipeline. During the construction of a pipeline there would be a need for road. and airport construction and maintenance, increased surveillance, protection and enforcement efforts, increased education expenditures and other such effects associated with the expected influx of workers and their dependents. All such activities would create temporary new jobs.(7) : Expanded economic activity generated by both pipeline construction and coal mining activities would be of benefit to Alaska. However, the operation of the pipeline itself would not be very labor-intensive. 213 TABLE 9.5 OPERATING COST SUMMARY TECHNOLOGY Coal/water slurry pipeline BASIS . : Location ‘Railbelt Year : 1982 constant dollars CAPACITY 5 million tons/year DISTANCE 273 miles OPERATING FACTOR 0.95 : : Annual ' Operating Costs Unit Cost Consumption Annual Cost VARIABLE COSTS oe 8 Electricity 9.2¢/kwh 2.05 x 10 kwh 18.9 Water _ $6.60/kgal 1.3 x 10 kgal 8.6 TOTAL VARIABLE COST $27.5million FIXED COSTS Labor Operating 99 ,200/year 140 man-years 13.8 Maintenance , , _ Maintenance Materials 2.0 Taxes and Insurance 8.5 Capital Charges 46.9 TOTAL FIXED COSTS $71.2million - TOTAL ANNUAL OPERATING COST $98.7 TOTAL ANNUAL OUTPUT : 5 million tons/year TOTAL LIFE CYCLE COST $19.75/ton coal 214 r. The more than 100 new permanent jobs for operations and maintenance would be concentrated at the slurry preparation plant near the mine, the pump stations, along the route, and the deslurrying plant at the point of end use and/or trans-shipment. 9.4 Impact 9.4.1 Effect on Overall Energy Supply and- Use The enlargement of coal's role as a major energy source in Alaska will be determined primarily by its price per comparable unit of usable energy, which includes production, transportation, and other related costs -- to. other energy sources such as oil and gas. Coal is a relatively bulky commodity, and transportation costs represent a significant portion of the delivered price of coal. Consequently, transportation considerations are important for implementing state and national coal utilization policies, therefore, moving large amounts of it from mine to user by rail, slurry pipeline,. highway, and water will present a major challenge to the transportation sector. Even though railroads are the dominant mode of land transport for coal, slurry pipelines have been the subject of intense investigation for the movement of coal and other materials. But despite: the technical status of slurry pipeline technology two conditions must be met for it to be feasible and competitive: (1) a large quantity of coal (over one million tons/year) must be moved between stable, but distant locations for long periods of time; and (2) the grinding of the coal necessary for transport via a slurry does not conflict with the ultimate end use for which that coal is required. In the case of coal for steam raising, such a conflict does not exist.(2) The main advantages of slurry pipelines over unit trains (presently the dominant mode for land-based. coal transport) include the following: e very high reliability unaffected by weather or ground conditions; e lower total costs than railroads at least when comparing new > installations of either mode of transport; e lower life cycle costs. The system is capital-intensive. Once installed, labor and other variable costs are a smaller percentage of the total annualized cost compared to railroads. It is estimated that pipeline operating costs escalate at 1% to 2% per year, while rail operating costs escalate to 4% to 12% or more per year. Life cycle costs are thus much lower; 215 e environmental superiority to railroads and other surface transportation, particularly when comparing new installations of either kind. 9.4.2 Future Trends Slurry pipelines also offer the advantage of delivering a constant flow of coal over a long period of time in the United States, while other modes of transport (unit train and/or truck) deliver coal in large single-block quantities. Similarly, consensus seems to indicate that slurry pipelines are a more reliable mode of transport, much less subject to intermittent failures than unit train operations. Additionally, slurry pipelines are usually constructed underground and consequently may cause lesser environmental distractions than do rumbling unit trains. If the recent emphasis on shifting towards the use of coal as a primary energy source continues, much of the anticipated expansion in slurry pipeline volume may be due, in fact, to large coal pipeline systems. At the present time, six coal-water slurry pipelines are proposed and/or planned. The annual throughput capacities for these six proposed pipelines range from a low of 10 to 45 million tons of coal-year; transport distarices for these same proposed lines range form just about 200 to 1500 miles. Consequently, if all six of these proposed and/or planned coal-water slurry pipeline systems do come to fruition, the total coal-water slurry pipeline system operating capacity (in the United States) would increase by well over an order of magnitude. Alternatives to water as a medium for transporting coal include crude or refined petroleum products, methanol and liquid carbon dioxide , (CO,). The sue of crude and residual oil as a coal slurry transport - medium has been studied since the 1960's. In contrast, the use of methanol and liquid CO, are relatively newer alternatives being evaluated. For example, “a development program for employing liquid co, as a coal transport medium has been underway for nearly four years and currently has an operating pilot plant. Each of these options has unique economic advantages which, tailored to local circumstances, may aid the slurry concept in capturing a larger share of land-based transportation than water slurries alone. 216 10. REFERENCES Polenske, K.R., Western Coal Transportation Unit Trains or Slurry Pipelines, Final Report, prepared for Department of Transportation, Washington, D.C., August 1976. Santhanam, C.J., "Potential Applications of Coal/Liquid CO, Slurry Pipelines," presented at ASCE International Convention, New York, New York, May 11-15, 1981. : Wasp, E.J., et -al., "Solid-Liquid Flow - Slurry Pipeline Transportation (Series on Bulk Materials Handling Vol. 1 (1975/77), No. 4." ‘Montford, J.G., "Operating Experience of the Black. Mesa Pipeline," presented at 5th International Technical Conference on Slurry Transport, Lake Tahoe, Nevada, March 26-28, 1980. Rieber, M. and S.L. Soo, Comparative Coal Transportation - Costs: An Economic and En ineertn Analysis of Truck, Belt, Rail, Barge, and Coal Slurry and Proumanic Pipelines (Volume » prepare or Unite tates epartment of the Interior, Bureau of Mines and the Federal Energy Administration, Final Report, August 1977. Hanks, R.W. and T.C. Aude, "Slurry Pipeline Hydraulics and Design," presented as course notes at Lake Tahoe, Nevada, March 24-25, 1980. U.S. Department of the Interior (Special Interagency Task Force for the Federal Taks Force on Alaskan Oil Development), "Final Environmental Impact Statement Proposed Trans-Alaska Pipeline (Volume 1 & 2)," 1972. Rieber, M. and S.L. Soo, Coal Slurry Pipelines: A Review and Analysis of Proposals, Projects, and Literature, prepared for. Electric Power Research Institute, Palo Alto, California, EPRI EA-2515, Project 1219-5, Final Report, August 1982. Financial Times Business Information Ltd., "Coal Transport," 1982. : U.S. Congress, Office of Technology Assessment, A Technolo Assessment of Coal Slurry Pipelines, OTA-E-60, March 1978. 217 10.0 OVERVIEW OF DISPERSED ENERGY SYSTEMS 10.1 Introduction The term "dispersed energy system" is applied to those energy systems which produce power, fuels, or heat at the local level in order to. serve the needs of single home, small industries, or remote settlements. This is in contrast to central energy systems, such as electric utilities which generate the energy at a central location and then distribute it to a multiplicity of end users. Central energy svstems have advantages of economies of scale which result in the producing lower cost energy services than their smaller, distributed counterparts. For example, both the capital and mainte- nance costs of large, central, diesel generators are lower than for small diesels such as those used in the Bush regions of Alaska. For this reason, central energy systems dominate wherever there are sufficiently large population centers to justify them. But despite their frequently higher overall costs, dispersed energy systems offer several advantages including: e increased incentives to use locally available energy resources, especially renewables such as solar energy, wood, wind, and hydro; e smaller environmental and socioeconomic impacts than large central facilities; e ability to use small increments of power generation which may be infeasible at larger scales; and e independence of large central energy production and distribution companies. In Alaska, dispersed energy systems, particularly in the form of diesel generators, are already in widespread use. The low and diffuse population of the state, combined with the high costs of energy distribution infrastructure could make the role of these systems increasingly important as their technical and economic performance levels improve. Figure 10.1 shows the primary dispersed options which are: small scale hydro diesel engine wind power systems small scale gasification photovoltaic power units bottoming cycle engine Stirling engines fuel cells END USE APPLICATION 7 . ENERGY CONVERSION SYSTEM RESOURCE . g RURAL RESIDENTIAL WIND WIND TURBINES REFRIGERATION SOLAR n 8 PHOTOVOLTAICS 2 i & 77 HYDRO WATER TURBINES to woo to So COMMUNICATION . STIRLING ENGINES PEAT al COAL = INTERNAL COMBUSTION REMOTE COMMERCIAL COMMUNICATIONS ENGINES a a 2 DIESEL FUEL ‘ F < . " —, @ ~ om 3 FUEL CELLS NATURAL GAS be VILLAGE/UTILITY POWER GENERATION ' ‘ an - an FIGURE 10.1 TYPICAL DISTRIBUTED ENERGY OPTIONS So eo OR Te a eS md Aside from small scale gasification, the dispersed energy options are aimed at producing electricity. Even gasification would produce a fuel which would, in most cases, be close-coupled to a power system such as a diesel driven generator or fuel cell. The reason is that in rural locations most of the energy is needed by residences in the form of thermal and electrical energy (the few institutional and commercial buildings have larger but analogous needs). Production of thermal energy is carried out almost exclusively within the residential building using purchased fuel oil or gas in water heaters, boilers and/or furnaces (which in this report are considered separately as end-use systems in Chapter 19). Production of electricity; sometimes carried out at the end-user level, is usually done at the community level to aggregate the needs of several end users. The dispersed technologies divide into two generator categories: e systems which use commercial fossil fuels (gas, diesel, etc.) for their operation--i.e., this category is exemplified by the diesel engines, although Stirling engines might also use a multiplicity of commercial fuels as well as locally available biomass fuels; e systems which convert renewable energy resources into useful energy forms--these include photovoltaics (solar energy), wind power units, gasifiers and Stirling engines (wood and peat), hydro power, and wind power. Apart from. the baseline diesel engine, the primary thrust of such developments is directed toward those items which can use locally available energy sources for their operation. In most cases the associated energy conversion technologies are still in a state of rapid transition, improving with time. For example, photovoltaic power units have only been extensively used for terrestrial applications for only the last five years. The technical performance and. economics of many of these systems, therefore, are projected to improve as the result of ongoing R&D programs and ‘increased commercial experience. Systems which use renewable energy resources (wind, solar, wood, and hydro) for their operation have the distinct advantage of reducing the need to transport and store commercial fuels to rural areas. Their use does, however, tend to be more site and application specific than fossil fuel fired systems. For example, wind energy systems can only be effectively operated in areas with average wind velocities in excess of 12 mph. This would probably tend to restrict their use to selected coastal regions and a few interior locations with sufficiently high wind regimes. 221 S20 qa) (2) G3) (4) (5) (6) @) (8) TECHNOLOGY Small Scale Hydro Diesel Engines Wind Power System Fuel Cells Small Scale Gasifiers Photovoltaics Bottoming, Cycle Engines Stirling Engines FUNCTION Convert energy in flowing water to electricity Convert diesel fuel into mechanical or electrical energy Convert energy in wind into mechanical or electrical energy Convert clean fuels such as gas and methanol into elec- tricity using membrane processes Convert solid fuels, such as wood, coal, and peat, into a more readily used gaseous form Convert sunlight into elec- tricity Convert waste heat from diesel engine exhaustion or industrial processes into mechanical power Convert a wide variety of fuel forms (gas, diesel fuel, wood, etc.) into mechanical or electric power TABLE 10.1 DISPERSED ENERGY SYSTEMS OVERVIEW TYPICAL COMMERCIAL CAPACITIES STATUS 5 kW to 50 MW Early Commercial 3 kW to 1000 kW Mature 1 kW to 2000 kW Early Commercial 3 kW to 1000 kW Developmental Early Commercial 0.1 kW to 1000 kW Early Commercial 50 to 200 kW Early Commercial 3 kW to 100 kW Developmental ALASKA APPLICATIONS Smaller systems (5 kW to 100 kW) for supplying power to small communities. Higher end of capacity range for connecting to grid Provide power in rural areas for single resi- dences and whole communities Small systems (10 ki!) for supplying power to single or groups of residences having no access to grid power Larger systems (up to 2000 mW) located in particularly windy areas for converting to the grid, thereby reducing need to run thermal power units. Possible replacement for diesel generators in rural areas having advantage of very high efficiency (i.e., low fuel use) and high reliability , Provide clean, easily used, fuel in rural communities and for industry utilizing locally available feedstocks (peat, wood, or coal) Remote communications and safety equipment Small scale power (200-500 watts) for remote homes and camps . Decrease fuel use in diesel power systems serving the larger rural communities. Provide power in rural areas for single residences and whole communities. - Early Commercial - Equipment available from several suppliers but technology still in evolution stage. - Equipment widely available and accepted as standard practice. - Equipment has been demonstrated but is still not available on a commercial basis. - Mature - Developmental Similarly, wind, solar, and hydro resources must be utilized when they are available, stored in some form for future use, or lost. . Storage frequently would be in electric batteries. While this frequently enables reliable operation without dependence on other systems for backup, it adds to the cost. Alternatively, the renewable resource may be utilized in the "fuel-saver mode" where it is combined with another system, usually diesel, to reduce the needs for transported fuel and its subsequent storage. For example, photovoltaic power units can be- used to provide seasonal (summer) power with small diesels taking on most of the load in the winter. This combination would significantly reduce fuel consumption and reduce the annual maintenance requirements on the diesel generator (due to decreased hours of operation). Each of the dispersed options has unique operating characteristics, input energy requirements (wind, diesel fuel, etc.) and potential applications in Alaska. Table 10.1 summarizes some of the salient technical features of each. As indicated, only diesel engines would be considered a mature technology and is the only one in widespread use in Alaska. Several other technologies have been implemented on a limited basis in Alaska and are in sufficiently widespread use worldwide that systems can be installed on a commercial basis with a high degree of confidence in their technical performance characteristics. These include photovoltaics, wind power units, small scale hydro, and small scale gasification. The remaining technologies are under development by major firms in the U.S. and their performance potential is demonstrated through both government and private sector programs. These systems show good promise for effective use in Alaska, but will require some combination of technical development and applications demonstration to bring them to a commercially accepted status for Alaskan conditions. These technologies include fuel cells, Rankine bottoming cycle engines, and Stirling engines. . ‘ The following sections briefly. describe the operation, applications, and status of these options. 10.2 Diesel Engines The diesel engine is an internal combustion, compression ignition, reciprocating engine. Fuel is injected directly into the cylinders of the engine (Figure 10.2) and burned, to convert that fuel energy into mechanical: energy. This mechanical energy can be used directly or it can be converted into electrical energy through the use of a generator. . 223 Fuel Intake Injection "Exhaust INS Piston Cylinder C Cc & (a) (b) (c) (d) Air Air Combustion Hot Gas Intake Compression Power Exhaust Source: Reference 1 FIGURE 10.2 FOUR-STROKE DIRECT INJECTION ENGINE 224 Applications for diesel engines generally demand high efficiency, good reliability, low maintenance, and long life. Typical applications range from small units for pumping water and stand alone power generation to automotive and total energy prime mover systems up to utility scale, multi-megawatt, low speed diesel - engines for central electric power generation. Diesel engine-generators are designed and manufactured in many countries of the world. There are millions of hours of operational experience with diesel engines and generator sets, and as a result the state-of-the-art of this technology is mature. Diesel-generators are the single most predominant. technology for electric power generation in Alaska by numbers of installations with a total capacity of just under 250 MW. Units fall into three basic size ranges: large units (500 kW and up) for use in central power stations, medium size units (50 to 300 kW) often used in Alaskan village utilities and school systems, and small sets (2 to 4. kW) for individual household installations.. Diesel engines are usually designed to operate at about 25-35% efficiency at full load. At part load, down to about 35-40% of load, about 5-6 efficiency points may be lost. Efficiency also tends to deteriorate somewhat between maintenance shutdowns. For 1981, 120. diesel-generators operated by the Alaska Village Electric Cooperative averaged nearly 22% efficient year-round. One of the primary disadvantages of operating diesel generators in remote areas of Alaska is the difficulty of access to a suitable fuel supply source and qualified mechanics for maintenance .and overhaul of these units. As a result, greater than expected fuel use and unscheduled maintenance can result in very expensive air lifts from maintenance and fuel centers. In some cases, duplication of equipment is provided so that users will not go without electricity for long periods, albeit at considerable added expense. Probably the group with the most’ experience in diesel generator application in Alaska is the Alaska Village Electric Cooperative (AVEC). It presently operates 130 generator sets from 50 to 600 kW in size in villages through Alaska. Fuel and maintenance are arranged for members through AVEC's central facility in Anchorage. The diesel-generator will continue to be an important factor in the Alaskan energy economy for electric power generation for some time to come. With its established fuel distribution, and maintenance/service infrastructures, it is most likely that the diesel-generator would at most be supplemented and not supplanted, by other competing technologies, such as renewable resource (photovoltaics, wind, etc.) and Stirling engines run with solid fuels. 225 BOILER Hot Gas or Liquid FIGURE 10.3 Cooling Water ' CONDENSER {Liquid Heating) a ~ TYPICAL FLOW DIAGRAM FOR A RANKINE CYCLE ENGINE 226 10.3" Rankine Bottoming Cycle Engines One way to maximize the role of the diesel engine in Bush Alaska is to extract more power without utilizing more fuel. - One way is to use the hot exhaust gases from the diesel or (industrial process) to operate a lower temperature power system based on the Rankine cycle (similar to that used in conventional steam power plants). A schematic of such a system is shown in Figure 10.3. These engines, often referred to as Rankine Bottoming Cycle (RBC) engines can be prepackaged in integrated units whereby the key subsystems (turbine, feed pumps, controls, generator) are mounted on a single skid. Other subsystems such as the condenser and boiler are also predesigned to facilitate site installation. They use a number of different working fluids including water, toluene, and commercial refrigerants. The. selection of working fluid depends on system capacity, operating temperature, and compatibility with manufacturers equipment. . RBC engines. are most commonly available in capacities from 300 kW to 1 MW. However, Rankine engines using the same working fluids have been built with capacities ranging from 3 kW to 150-kW for solar, flame fired, and geothermal applications. Thus, smaller RBC engines could be made available if sufficient demand developed for such systems. The largest potential for RBC engines in Alaska would probably be to generate power using the hot exhaust gases from diesel-generators as the heat source. When used in such applications, RBC engines can generate 15-25 percent additional power with no increase in fuel consumption. They could, therefore, be utilized: to reduce fuel consumption at the larger diesel fuel generating plants located throughout the state. Presently, four firms in the United States offer RBC engines on a commercial basis and several others can provide custom units. Due, in part, to support. from the U.S. Department of Energy, over a dozen units are now operational in the U.S. using both diesel exhaust gases and industrial process reject heat as the energy source. Most of the outlying population centers such as Bethel, Nome, and Kodiak generate most of their power with diesel generators. RBC engines could be used to both reduce fuel consumption at these larger facilities and to expand the peak generating capacity. In such applications, the RBC engines would typically reduce fuel consumption by 20 percent. As indicated previously, most commercially available RBC engine systems are in the capacity range from 300 kW to 1 MW. Other than larger regional centers few locations in Alaska have sufficient base loaded diesel generating capacity to warrant RBC installations of this size (requiring in excess of 1500 kW of base 227 Source: Reference 1 FIGURE 10.4 — Gas—> Reduction | Ln | B= — || ;eombustion 4 | | | | SCHEMATIC DIAGRAM OF UPDRAFT GASIFIER 228 + loaded _ diesel). Therefore, the extensive use of RBC _ engine technology in Alaska will require developing prepackaged engines of lower capacity ( 100 kW) consistent with the relatively. modest size of village size diesel-generators. 10.4 Small Scale Gasification One of the limitations of the diesel engine in rural Alaska is’ its dependence on petroleum fuels which must be transported in relatively small quantities over long distances. One means of adapting locally available wood and peat fuels to existing diesels would be to gasify these materials so that the gas can be burned directly by the diesel. - In solid fuel gasification (Figure 10.4), the fuel is reacted with an oxidant--usually air in the case of small-scale systems--to produce a mixture of carbon monoxide and hydrogen. Water also participates in the reaction and is supplied as part of the wet or partially dried fuel, or alternatively as steam. The gas, containing tars and dust entrained from the gasifier, is cleaned by scrubbing with water. . Gasification has been practiced for over 100 years, but relatively cheap oil and gas have largely displaced this technology. Since the energy crises of the 1970's, there has been renewed interest in the technology. Numerous demonstration and commercial systems have _ been plagued. by operating problems with primarily material handling equipment. Therefore, considerable operating and maintenance effort have been required, a distinct disincentive for use of gasification at remote Alaskan sites. Another disadvantage of. gasification relates to its need for solid fuel. While a gasifier would be sited to take advantage of an available resource, the resource ‘may be expensive to obtain. For example, driftwood is thought to be free for many small rural uses, but the quantity needed to operate an 80 kW diesel would be over 500 tons per year. The wood gathering operation would be very time consuming, although probably not as tedious as for coal mining or peat harvesting. Numerous manufacturers supply gasifiers worldwide. About twenty small-scale (less than 3.5 tons per hour) systems have been installed in the United States in the past few years. Among them was a demonstration in Anchorage sponsored by the Alaska Division of Energy and Power Development. This unit also suffered many of the operating problems characteristic of these systems. 10.5 Fuel Cells As indicated in Figure 10.5, a fuel cell is an electrochemical device that converts the chemical energy of a fuel directly into electricity. As such, it is a primary source of electrical energy, not a storage 229 CURRENT COLLECTOR FUEL 1N meen OXIDANT IN some CURRENT COLLECTOR Source: Reference 2 FIGURE 10.5 FUEL- HYDROGEN CONTAINING ANODE - OXIDATION REACTION r—~ ELECTROLYTE -ION CARRIER 230 L_ CATHODE- REDUCTION RE ACTION OX!DANT-OXYGEN CONTAINING emp §=FUEL OUT LOAD emg = OXIDANT OUT A7SO6I30} CONCEPTUAL DIAGRAM OF A PHOSPHORIC ACID FUEL CELL device, that is well suited to a decentralized energy system. The characteristics that make it an attractive candidate are modular construction, good load following capabilities without loss of efficiency, automatic startup and shutdown, low-pollutant emissions, and relatively quiet operation. : The major limitation in the application of existing fuel cell technology _ (based on a phosphoric acid electrolyte operating at 375°F) is the ~ requirement for a clean fuel that can be used to generate hydrogen. The fuel most commonly used in. fuel cells is natural gas. However, fuel cells designed for use with methanol or naphtha are also in the advanced development stage. Dependence on these fuels is a drawback similar to that of diesels. Hydrogen is the only fuel that can be oxidized in the fuel cell itself at acceptable power levels. A primary incentive for the development of fuel cells is their very high efficiency. which is about 35 percent for present day phosphoric acid units. . More advanced fuel. cell systems are projected to have efficiencies in excess of 45%. These efficiency levels are considerably in excess of that from more conventional power systems. : Fuel cells could be used for many applications in Alaska including remote power units, total energy, and grid connected distributed power centers. Their highly modular construction, high efficiency, and excellent part load performance characteristics make their use in remote settlements particularly attractive. In such applications they would require only about one-half the fuel of that now used by. generators. The ready access to high temperature reject heat would also facilitate the implementation of total energy systems in schools or closely spaced residences. : In rural applications the most likely fuel based on present technology would be methanol -- probably derived from the plentiful natural gas supplies in Alaska. Methanol could be distributed via the same system as now used for diesel fuel. A future alternative will be to operate fuel cells from low-Btu gas produced from gasifiers which in turn use locally available fuels (wood, peat, etc.) as the energy input. However, the operating problems of current generation gasifiers would. negate some of the benefits envisioned for fuel cells. The .commercial and economic acceptance of the fuel cell is difficult to project. It faces a chicken-or-egg problem. Manufacturers cannot justify the investments required to produce fuel cells in quantity until the sales are sufficiently large. However, the high sales volumes cannot be achieved until large-scale production begins to lower the hardware costs. 231 Source: Reference 3 FIGURE 10.6 IMPLIFIED ILLUSTRATION OF THE STIRLING ENGINE CYCLE 232 10.6 _ Stirling Engines A versatile, developmental engine which has_ several potentially attractive advantages compared to the diesel engine is the Stirling engine. A Stirling engine is a closed cycle system containing a working gas (most commonly air, helium, or hydrogen) which is alternately heated (during an expansion stroke) and cooled (during a compression stroke) to produce power (Figure 10.6). The source of heat is generally a flame fired furnace although other heat sources such as solar energy or nuclear energy can (and have) been used. Since the heat input to the gas is via a heat exchanger external to the engine (as compared to an internal combustion engine where the fuel is burned in the’ cylinder) this furnace can, in principal, burn any fuel. As a result, the Stirling engine has a wide multi-fuel capability which could make them particularly attractive for use in . Alaska. Any given engine, for example, could use diesel fuel, heating oil, gas or gasoline which could significantly reduce logistics problems of transporting and storing fuels in rural areas. Perhaps more importantly, Stirling engines can be designed to use wood, peat, or coal energy resources which may be locally available thereby decreasing fuel cost and making | maximum use of labor and material -resources in the State. Stirling engines can be built over a wide power range from 1 kW up to several thousand kW.* . However, most attention.is now directed toward automotive applications ( 50-100 kW), small generators and heat pump drives (3 kW), and remote, biomass fired, engines ( 1-5 kW). By far the largest development program is directed toward the automotive engine. The advantages of the Stirling engine for this and other applications. are a multifuel. capability, high. thermal efficiency, low noise and vibration, and low exhaust emissions. These advantages will also be important in many applications in Alaska. The Stirling engine is still in the advanced development stage, with about a half dozen firms in the U.S. participating in the development process. If these programs are successful, Stirling engines in the 1 kW to 100 kW power range would become available for use in Alaska in a 2-5 year time frame. However, the state has recently shelved plans to test a small biomass (wood) fired Stirling engine to demonstrate the potential for these engines to serve a range of applications in Alaska in those regions where locally. available fuel (wood, peat, etc.) is available. * One of the artificial heart pump programs uses a very small Stirling engine of a few watts capacity which shows the wide scaling capability of this engine configuration. 233 Impulse Turbine & SS RSSSSS er Reaction Turbine Source: Reference 4 FIGURE 10.7 TURBINES ao Stirling engines could be used as drives for. electric. generators and total energy systems wherever diesel systems are now used. In these applications the Stirling engines would have the previously mentioned advantages of using the most readily available fuel, high efficiency (leading to lower fuel cost), and low noise and emission levels (making the Stirling engine option particularly attractive when used close to or in inhabited buildings). However, perhaps the largest potential for Stirling engines will be in those applications using locally available fuels for their operation. . Such applications would include generating power for single residences and small settlements in most areas of Alaska. Such power units could often be combined with providing heat for single or multiple buildings due to the ease with which the reject heat from the engine can be extracted. 10.7 Small Scale Hydro Small scale hydro systems are usually defined as those with capacities of less than 5 MW. Within this range one may define several very appropriate capacity subdivisions: e Microhydro: 1-150 kW e Minihydro: 150-700 kW e Small hydro: up to 5 MW All hydropower units use the energy contained in flowing water to operate a turbine which, in turn, can drive a generator for the production of electric power.. The design of the system depends primarily on the capacity and the head available (i.e., drop in elevation between the turbine system inlet and outlet). In general, systems with a low head (10-50 ft) to moderate head (50-300 ft) use some form of reaction turbine while those with a higher head use an impulse turbine. Figure 10.7 shows turbines of each type. The effective use of hydropower units requires having a water source capable of being developed to give a reasonably consistent flow of water all year round. The site preparation required to install the hydropower unit will vary greatly by site and capacity. In sites where the terrain is relatively flat it may be necessary to construct a small dam to result in sufficient head for a reaction turbine unit. In more mountainous areas (Southeast), many sites may be usable by diverting a portion of a stream flow through a vertical drop by means of a length of piping without the need for significant site construction. : Small-scale hydropower systems are now generally ‘available as prepackaged units consisting of the turbine-generator, control package, and associated housings. Experience with some systems 235 - 98% eas 1a e: Sebel a LSE Ky te, Vertical Axis Wind Turbine Horizontal Axis Wind Turbine - (VAWT) (HAWT) FIGURE 10.8 TWO TYPES OF WIND TURBINE GENERATORS indicates almost maintenance free operation, particularly with the use of modern control systems which can be used to effectively regulate system operation when unattended. Over the last five years there has been an increasing interest in the use of small scale hydropower resources. This has resulted in several firms entering this field with prepackaged units in the 1 kW to 5 MW power range. Alaska is well endowed with hydropower resources ranging from small mountain streams in the Southeast to larger rivers throughout the Bush regions. In fact, most settlements. in Alaska are in close proximity to a flowing water resource. However, many do not have a sufficient flow or fall of water to permit economical development of a small scale system. Others may be too distant from the load to justify construction of a power line. Further, the severe climate in Alaska complicates the use of small scale hydro for two reasons. First the freezing of the water courses can greatly reduce the water flow available and secondly, the ice can damage the hydropower equipment. These factors are very site specific and require careful attention in both design and maintenance. 10.8 Wind Power Systems Wind turbines are now available over a wide power range from 200 watt battery chargers to multi megawatt units having blade diameters as large as football fields. Most of these systems use blades similar to airplane propellers to extract energy from the wind as they rotate on a horizontal axis. Other models, however, have vertical-axis blades in configurations similar to egg beaters (Figure 10.8). Both configurations have proponents and are now part of the commercial practice. , The amount of energy produced by a wind generator is directly proportional to the swept area of the blades and to the cube of the wind velocity. As a result, a wind generator operating in a location with 8 mph average winds produces only 30 percent that operating in a nearby area with 12 mph winds. In general, wind power systems are considered economically viable only in those locations with annual average wind velocities of 12 mph or above. At that:wind speed, the energy output of a wind generator is roughly 8 watts per square foot of swept area. Therefore, to generate 1 kW (minimal needs. of a residence) a blade diameter of about 12 ft is required while a- small community size system (100 kW) would require a blade diameter of about 120 ft. Similarly to solar energy, the availability of wind energy is uncertain and does not necessarily match the demand. As a result, remote stand-alone systems (not connected to- the village grid) will often require battery storage so that electricity can be used on demand by the users. If the wind system is connected to the village diesel it would be operated in the fuel saver mode. 237 ow Solar Cells Make Electricity Pre SILILON, /" Fyrt SILICON —_—> CUHHINE 1toap——— Source: Solarex Product ‘Literature. FIGURE 10.9 ‘OPERATION OF PHOTOVOLTAIC CELLS 238 Wind generators are highly modular in their construction. As such, capacities of almost any size can be built up by using a multiplicity of smaller units. In areas with high wind speed such “wind farms" could be connected to a central fuel fired generating capacity (diesel power units, etc.) and -used to save fuel. In such. systems, no storage is required since the fuel fired system guarantees power output equal to the demand. Most commercially available equipment is in the capacity range from 200 watts to 100 kW. Also, several large aerospace companies now have developed multi-megawatt units designed to be operated in parallel with large central utilities. Several firms in Alaska distribute and install equipment and have a familiarity with the special requirements imposed by Alaska weather and applications. Alaska's coastal regions have average wind velocities which appear to be highly favorable for the use of wind power systems. A significant percentage of the population in Alaska therefore live in or near areas. with reasonably good wind potential. . Therefore, that wind power. could be used over a wide range of applications from small (200-1000 watt) units to provide essential power needs of individual homes to community size (50-500 kW) units to supplement the power output from diesel generators. 10.9 Photovoltaic Power Units As indicated in Figure 10.9, photovoltaic (solar cells) power units can directly convert solar radiation into electricity. The solar cells ‘which undertake this conversion process use thin layers of processed semiconductor materials which are encapsulated in weather tight panels. The material now used in commercially available panels is single crystal silicon, which is widely used in the electronics industry. A wide range of alternative and potentially lower cost materials is also being investigated including amorphous silicon and gallium arsenide. The amount of electricity generated by a photovoltaic array is directly proportional to the incident solar energy. Commercially available panels convert. 9-13 percent of the solar energy into electricity. Under clear sky midday conditions, about 10 watts of power are generated per square foot of panel area. Therefore, to generate 1 kW of electricity under highly favorable solar conditions requires about 100 square feet of panel area. The power output of a photovoltaic array varies greatly during the day and on a seasonal basis depending on solar availability. As a practical matter, therefore, most applications require that energy storage be incorporated into the system design to allow for nighttime and cloudy weather operation. Storage is usually in the form of a heavy duty battery with associated charging ‘and dischanging control 239 systems. Further complicating the system design may be the need for an inverter system to convert the DC output of the photovoltaic array (and battery storage) to AC power in order to operate conventional appliances. The need for the inverter can be eliminated by using appliances (lights, etc.) which can operate on DC power. Such appliances are becoming increasingly available in order to decrease the cost of using photovoltaic power units throughout the world. Presently, six firms in the U.S. are manufacturing commercial photovoltaic panels. Most of these firms also have developed system packages consisting of the panels, battery storage, and controls in order to facilitate their use. About a dozen small units (10-100 watts) have been installed in Alaska; primarily to operate remote safety and communication equipment. Several small companies in Alaska now distribute photovoltaic equipment and provide design services to adapt this equipment to the very. special needs of Alaska applications. The solar -availability in the more favorable areas of Alaska- (Anchorage, Fairbanks) is about half that in the better locations in the Lower 48. Due to persistent high cloud cover, the solar availability in the Southeast is so low’ (about two-thirds that in Anchorage) as to preclude all but the smallest scale applications in that area. Due in part to the relatively low availability of solar energy in Alaska, most applications over the next decade of photovoltaics will be where small amounts of power (10-1000 watts) are required to serve critical energy needs in areas without access to grid power. These applications include the operation of remote signal and_ safety equipment. such as railroad crossing lights. As the cost of photovoltaics decreases, the range of application could increase to include providing minimal power needs to remote individual residences or seasonal camps where the cost (and inconvenience) of operating inefficient small diesel power units are particularly high. 240 REFERENCES Reed, T.B., ed., Biomass’ Gasification Principles and Technology, Noyes Data Corp., Park Ridge, N.J., 1981. Institute of Gas Technology, Handbook of FUel Cell Performance, DOE Report, Contract No. EC-77-C-03-1545, Chicago, Mlinois, 1980. . Postma, N.D., et al., "The Stirling Engine for Passenger Car Application," preprint, Society of Automotive Engineers Meeting, June 18-22, 1973, SAE Preprint No. N 730648. ~ Alward, R., et al., “Micro-Hydro Power: Reviewing an Old Concept National Center for Appropriate Technology," Butte, Montana, 1980. 241 11.0 DIESEL ENGINES 11.1 Introduction and Summary 11.1.1 Technical Overview The diesel engine, patented in Germany by Rudolph Diesel in 1892, is an internal combustion, compression ignition, reciprocating engine. The diesel is a very well known engine, used throughout the world in a wide variety of sizes and applications. These applications range from small units of 3 to 5 horsepower for pumping water and stand-alone power generation to automotive and total energy prime mover systems up to utility scale, multi-megawatt, low speed diesel engines for central electric power generation. : Diesel engines are most often used in applications where high efficiency is important, where reliability and low maintenance is critical to successful operation and where long life is needed. These applications typically require an engine with self-contained and self- standing characteristics. In terms of mechanical power production, the efficiency of diesel engines varies from 20 to 40%, when properly operated and maintained, depending upon the size and design. For electric power generation, the engine shaft is connected to an electric generator and the mechanical power of the shaft is converted into electric power by the generator. Today diesel-generators are designed and manufactured in many countries around the world. There are millions of hours of operating experience with diesel-generator sets and as a _ result the state-of-the-art of this technology is mature. There is much research and development to improve and broaden the application of diesel engine technology and hardware. A more important issue from the standpoint of most Alaskan diesel engine users is the operation of the diesel engine in their environment Therefore, this chapter focuses more on using a diesel engine and the options available to optimize its performance. 11.1.2 Alaskan Perspective Diesel-generators are the single most predominant technology for electric power generation in Alaska by numbers of installations; units fall into three basic size categories: large units (500 kW and up): for use in central stations, medium size units (50 to 300 kW) often used in Alaskan village utilities and school systems and small sets (2 to 4 kW) for individual household installations. Small diesel-generator sets can be quite compact and efficient. For example, a small 3.5 kW Lima-Lister diesel-generator set has approxi- mate dimensions of 45" x 24" x 32", weighs almost 700 lb, and can 243 achieve efficiencies of approximately 25%; as another example, a skid-mounted 85 kW Caterpillar set has approximate dimensions of 100" x 48" x 61" weighs almost 3500 lb, and operates at approximately 30%. Relatively small size, high efficiency and good reliability of operation are the primary reasons for the predominance of this technology in Alaska. . One of the primary problems of operating diesel-generators in remote areas in Alaska is the difficulty of access to a fuel supply source and qualified mechanics. for maintenance and overhaul of these units. ‘Maintenance personnel have to be dispatched by airplane from major cities and towns to the remote locations. This can take several days especially in bad weather. As a result it is necessary to - provide 100% standby capacity for many applications. This duplication of equipment at each site leads to a considerable added capital . investment. In Alaska the Alaska Village Electric Cooperative (AVEC) is presently operating electric power systems in a number of villages. AVEC is an organization affiliated with the Rural Electrification Administration (REA) and is independent of the State of Alaska. It was established to develop and maintain electric power systems in Alaskan villages with the aid of 2% loans provided by REA. -Presently it operates some 130 diesel-generator sets, ranging from 50 kW to 600 kW in capacity throughout villages’ in the Bush region. The smallest community served by AVEC has a population of 100. AVEC has the responsibility for purchasing equipment, installing generation and distribution facilities, as well as operating and maintaining them.in: AVEC villages. It is a non-profit corporation, owned and directed by its member owners. ‘ AVEC has one central warehouse for spare parts.in Anchorage. Field repairmen carry a modest number of common replacement parts. When a call from an AVEC village is received, it is generally possible to dispatch one of the maintenance personnel in the immediate area to the site. There are usually three diesel engines in each village; units are not paralleled as unattended parallel operation could lead to system instabilities. .One of the objectives is to sectionalize distribution systems within each village so that if one section goes down the rest of the load remains supplied. AVEC is currently carrying out tests in Anchorage on fuel efficiency of diesel-generators. The ultimate goal is to develop an optimization schedule for the best possible engine fuel efficiency. This can be accomplished by developing consumption charts, to determine the optimum phase-out schedule for a unit. 244 11.1.3 Significance of Technology Data on installed generation capacity in Alaska (1,2,3,4) indicate that total installed diesel electric power generation capacity (excluding industrial and military generation) is approximately 228,000 kW. Of this total approximately 25,000 kW is installed in villages with a population of 250 or less, and 55,000 kW in villages with a population between 250 and -500. Clearly, the diesel engine is of critical importance in Alaska, and its technical feasibility is adequately demonstrated. Diesel engines typically cost from $35-40/kW for light duty, automotive type diesels. Small 3-4 kW residential diesel-generator sets cost approximately $1,625/kW and larger 150-300 kW sets could cost anywhere in the $230/kW to $325/kW range, depending upon the accessories and control systems supplied. Transportation and installation costs are additional and vary according to distance from Anchorage and type of installation. For fuel costs of $1.50/gallon to $2.00/gallon and engine efficiency of 24%, cost of energy ranges from $0.15/kWh to $0.20/kWh for fuel alone and up to $0.40/kWh for total delivered electricity cost. The diesel-generator will continue to be an important factor in the Alaskan energy economy for electric power generation for sometime to come. With its established fuel distribution, and maintenance/service infrastructures, it is most likely that the diesel-generator would at most be supplemented and not supplanted, by other competing technologies, such as renewable resource (photovoltaics, wind, etc.) and Stirling engines run with solid fuels. 11.2 Description of Technology 11.2.1 Operating Principles The diesel cycle engine is an internal-combustion reciprocating engine.(5,6) Figure 11.1 illustrates the basic diesel cycle. It is similar to the standard Otto cycle or spark ignition engine, which is used in most lawn mowers and automobiles. Both engines generally use a piston/cylinder configuration. The piston is mechanically linked to a crank shaft for extraction of the energy that is liberated during the combustion process. During the first stroke of a four-stroke design, the piston is moved downward in the cylinder to draw air for combustion into the cylinder. The second stroke consists of the piston moving upward, thereby compressing the air. At the instant when the piston is at the end of this stroke, the fuel charge is injected. The third stroke occurs after ignition when the high temperature products of combustion force the piston downward. This is the only stroke 245 | \ (a) (b) (c) ‘Intake Compression Power Source: Reference 6. FIGURE 11.1 FOUR-STROKE.CYCLE ENGINE 246 during which energy is extracted from a four-stroke engine. The final stroke is the expulsion of spent exhaust gases from the cylinder by movement of the piston upward in the cylinder. The principal difference between diesel and Otto cycles is the manner in which the fuel-air mixture is ignited. In the Otto cycle engine, ignition is accomplished: by means of an electrical spark at the appropriate time in the cycle. In the diesel engine, the compression stroke itself. produces a-sufficiently high temperature to cause ignition of the fuel in the combustion chamber. The diesel engine, like the Otto,’ can operate on either a two-stroke or four-stroke cycle, depending on the design requirements of the application. In power generation applications, the crankshaft is connected to the rotor of an electric generator and the rotational energy is converted into electrical energy by the generator. 11.2.2 Technical Characteristics Diesel engines used for electric power generation applications in Alaska can be categorized in three basic size ranges: e Small; 3-25 kW; typically used ‘for individual or small. groups of residences or light commercial applications; generally from one to four cylinders, often an air cooled engine in lower end of the range. e Medium; 25-500. kW; representative of school, village, commercial or small scale industrial use; often a four, six or eight cylinder, water cooled (perhaps turbo charged) engine. e Large; 500-2500kW; representative of town utility or heavy industrial size engines for stationary applications; could be a medium speed, twelve or more cylinder engine. Table 11.1 lists the characteristics most often used to describe the operation of a diesel-generator. The following paragraphs discuss each with emphasis on the baseline operation in an Alaskan setting. Possible operational and equipment modifications are discussed subsequently. Fuel Requirements Diesel engines require fuel which is relatively clean, and free of contaminants such as water and dirt. A diesel engine will run on a variety of fuels such as Diesel No. 2, kerosene, and home heating oil. For operation in cold Alaskan climate however, Arctic No. 1 is 247 TABLE 11.1 CHARACTERISTICS OF DIESEL-GENERATORS Fuel Requirements - ‘Type and quality of fuel needed. Efficiency - Portion of fuel energy converted to shaft output. Size and Weight - Important for shipping, housing, etc. Cost - Capital, fuel, operation and maintenance. Maintenance Schedule - Frequency and skill level requirements. Operating Life - Operating time available before overhaul. Reliability - Potential for uninterrupted service. Load Following Capability - Ability to meet time varying load. Emissions - Constituents of exhaust gases. Noise and Vibration - Qualitative environmental characteristics. Reject Heat - Quantity and temperature of fuel energy given off as heat. Operator Requirements - Need for personnel to monitor and perform simple maintenance. 248 generally used as it has a pour point of -45°F.* These fuels are generally called distilates, which is reflective of the processing by which they are produced in petroleum refineries. Efficiency As noted above, the diesel engine is the most efficient heat engine commercially available for the conversion of liquid or gaseous hydrocarbons to shaft power. Design engine efficiencies can range from 25-40+% regardless of the engine size. However, operation under field conditions often leads to lower overall efficiency due to extended low or part-load operation. idling, performance deterioration between maintenance intervals, and other non-ideal operating conditions. An important characteristic of the diesel engine is the ability to maintain its efficiency through a range of load factors. Figure 11.2 illustrates the efficiency of a 325 kW diesel engine. As indicated, down to 45% of full load, engine efficiency is quite constant. This characteristic is important in non-grid connected electric power generation applications since load tends to vary widely with time of day and season. Below 45% of load, however, efficiency drops off rapidly until it reaches zero at idle where fuel is being consumed but no shaft power is being extracted. Size and Weight Two areas in which the diesel engine does not typically perform very well compared to combustion turbine or conventional spark ignition engines are engine size and engine weight per unit of power delivered. However, in stationary power generation applications, these are not particularly critical limitations since the engine is seldom transported. Table 11.2 indicates typical weights and dimensions for diesel-generator sets appropriate for Bush applications. The small sets generally fit into a shed of dimensions 8' x 6' x 6' and can sit on the floor of the building. The medium size units often are skid-mounted and fit into small trailer or shipping container housings and are fastened to the structure of the building with no special foundation. Large units (1000 kW and up) require special foundations due to their weight and vibration. *Pour point is a measure of the temperature at which the fuel will flow as a liquid. 249 0gz Efficiency (%) 75 125 175 225 275 325 375 Generator Output (kW) Source: Specifications of Cummins KTA-1150-GC Diesel, Reference 8. FIGURE 11.2 FUEL UTILIZATION EFFICIENCY DURING PART LOAD OPERATION OF A DIESEL WITH 325 kW NET OUTPUT POWER TSé MANUFACTURER AND MODEL NO. Lister/Lima STl Lister/Lima ST2 Lister/Lima HR3 Caterpillar 3304 Caterpillar 3306 Caterpillar 3406 Cummins NTA-855-GS Cummins KTA-1150-GC TABLE 11.2 SELECTED CHARACTERISTICS OF DIESEL-GENERATOR SETS APPROPRIATE FOR BUSH APPLICATIONS PRIME POWER NUMBER OF EFFICIENCY AT DIMENSIONS OF WEIGHT OF OUTPUT (kW) CYLINDERS FULL LOAD THE SET LxWxH THE SET kWh/gal (%) ) Gn.) (ibs) 4.5 1 9.8 24.1 46 x 24 x 32 660 9 2 9.9 24.4 52 x 24 x 32 925 25 3 10.9 26.8 58 x 26 x 43 1,675 85 4 12.1 29.8 101 x 48 x 61 3,380 105 6 12.8 31.5 122 x 48 x 63 4,400 175 6 13:5 33.2 134 x 48 x 71 5,700 275 6 13.0 32.0 129 x 39 x 57 6,050 325 6 13.0 32.0 140 x 49 x 77 8,985 (1) Assuming 138,700 Btu/gal. (2) Dry, uncrated. SOURCE: References 8, 9, and 10. (2) The foundation is particularly important in Arctic applications due to the care required to not damage the permafrost which may cause settling. An example of such a system used in Bethel consists of nine feet of gravel directly on the permafrost plus two inches of insulation covered with a fourteen inch pre-stressed concrete slab. The operation of four 2.1 megawatt diesel-generators, each having an approximate weight of 70,000 lbs has not yet generated any cracks in the flooring or settling of the structure after six years of operation. Cost First cost for a diesel engine must be weighed against other factors impacting future costs of the system, namely operating efficiency, reliability, and life. Costs are discussed more fully in Section 11.3.1. Maintenance As with any piece of mechanical hardware, a diesel engine will require periodic maintenance. Table 11.3 lists typical maintenance items and the frequency with which they must be accomplished for an example diesel engine. As indicated later, there are various options available to the diesel engine operator to extend this interval for maintenance on certain components. Maintenance of AVEC-run generators fall into four categories: Preventive, including oil and filter change, pressure and temperature check, etc., is carried out by local village people under agreement with AVEC; the agreements are generally in the form of a performance incentive whereby AVEC pays the village governments on the basis of their performance -- as evaluated by AVEC -- such as changing the oil in a timely manner or not. manning a generator when not needed. Elementary Corrective, including fan belt replacement, water pump replacement, etc., is also encouraged to be carried out locally. Substantial Corrective, including major repairs and overhauls are handled by an AVEC maintenance crew consisting of 6 to 8 mechanics in the field and another 4 to 6 in the shop in Anchorage. Special Maintenance, such as electrical control board trouble- shooting, is done by outside servicemen. Operating Life An important aspect of diesel-generator operation is long operating life, or more specifically, life between major overhauls. Table 11.4 is a summary of what is required at overhaul for four different diesel 252 é&e TABLE 11.3 GENERAL MAINTENANCE REQUIREMENTS Every 250-1000 hours of operation Every 3000 hours of operation (approx. annual) Every 6000 hours of operation Every 12,000 hours of operation - Fuel filter change - Oil and filter change ~ Service air filter - Check operating temperature - Coolent flush and change - Tune up: .» compression test . check timing - check governor - adjust valves . vteplace rocker gasket - Replace injector nozzles - Remove and replace cylinder head - Install new valve guides and seats - Replace or reface valves - Replace and adjust rocker arms - Clean, inspect, and replace if necessary: . gaskets .- valve seats ». valve springs . adjusting screws . lock nuts « push rods . valve rotors 253 TABLE 11.4 SELECTED DIESEL ENGINE CHARACTERISTICS AND RECOMMENDED MAINTENANCE Lister Perkins Waukesha Caterpillar Engine Model ST-1A 4,108 VR 2000 3,408 Description Cooling Air Liquid Liquid Liquid Derated hp 8 20 69 225 Recommended Oil Change 250 1,000 250 1,000 Maintenance Adjust valves 10,000 2,500 2,000 1, 750 Schedule Fuel Injector 10,000 10,000 10,000 10,000 - (hrs. ) Overhaul Top Overhaul 5,000 10,000 6,000 10,000 Schedule Major Overhaul . 25,000 20,000 10,000 30,000 (hrs. ) Ultimate 100,000 100,000 100,000 100,000 Economic Life 254 a { } I! CT engines and how many hours this is apt to represent for each engine. Times between overhauls vary from 10,000 hours for the small engine to more than 100,000 hours for the large engine. In practice the lifetimes for large diesel engines are nearly unlimited, because all the parts are- replaceable and can be overhauled many times. Generally, economics dictate whether to overhaul or replace a unit. , Reliability A very important characteristic for an engine. which is not run continuously is how reliably it will start. One study (7) indicated that the average rate of failure to start is one in every 650 attempts; 80% of these failures were due to electrical faults in the starting system; only random mechanical failures were noted. For continuous operation, worldwide failure rate is reported to be one "stoppage" in every 2,500 hours, the majority of the forced outages being associated with the fuel system failure. ‘ Load Following One strength of the diesel-generator in non-grid systems is its ability - to follow rapidly changing load. -The relatively sophisticated fuel control system of a diesel engine permits this sort of load following behavior with a minimum of additional equipment. For example, an average diesel operating at 40% of rated load can experience an increase up to 90% with a maximum generator output voltage dip of 20%; it will recover to full steady state voltage within 4 to 5 seconds. (8,9) Emissions The exhaust gases of diesel engines contain hydrocarbons, carbon monoxide, and nitrogen oxides. The amount of each source varies considerably with the engine, fuel system design, and operating temperature.(5) They .are discussed’ in more detail in 11.2.3 Environmental Issues. Noise and Vibration A diesel engine tends to be a rather noisy and vibration prone engine, due to the violent nature of the combustion process. While the use of a water jacket (for water coolant), vibration isolation mounting systems and insulated enclosures can help eliminate much of the noise, these all can add significantly to the cost of the engine system. Another way of reducing the perceived noise level is to locate the generator systems away from inhabited areas as_ is presently done in most villages’ in Alaska. However, for engines equipped with waste heat recovery, this could raise the cost of distributing reject heat energy from the engine systems to the end users. 255 Heat Recovery Through the use of a recuperator in the exhaust gas stream and heat exchangers to cool circulating liquid (in liquid cooled systems) and lubrication oil, usable heat can be recovered from the engine. In a typical well-maintained diesel engine, only 35% of the fuel input energy can be extracted in the form of useful shaft power under proper operating conditions. About 30% is contained in the exhaust gases at temperatures as high as 800°F and a roughly equal amount is available from the lubrication and cooling systems, at 180°F for atmospheric pressure cooling systems, and 220-250°F for pressurized systems. Another 5-10% is lost in form of radiation and is generally not recoverable. Approximately one-half of heat energy in the exhaust gases and in the cooling system can be practically recovered. Operator Requirements © Typically, a small diesel-generator will operate unattended. A medium size village power system generally will have someone assigned to it part-time whose responsibility is to monitor engine condition and operation and perform simple maintenance tasks. For large scale utility size systems, generally at least two or three persons will be on site full-time to monitor the operation of the plant. Operational and Equipment Improvements There are a number of operational and equipment improvements that might be made to diesel-generator installations to optimize their performance in a given application. These improvements fall into one of two areas: modifications to the operation of the diesel engine, and alterations or modifications in the equipment itself. The following are some selected examples. Operational Modifications to Diesel-Generators 1. Waste Heat Recovery/Bottoming Cycle Engines One of the most commonly practiced methods of optimizing the performance of a diesel engine is to capture its waste heat for use in other applications. As indicated above, this can have a 30 to 40% improvement in the overall fuel utilization efficiency of a diesel engine. Recent studies (11,12) and discussion with AVEC indicate that at present only a few diesel generation sets are equipped with waste heat recovery equipment. Another option is to use the same reject heat to operate a bottoming cycle engine. In a bottoming cycle engine, normally rejected energy is converted to shaft output to augment that of the diesel engine. These cycles can convert up to 15-25% of the captured reject heat, depending on the system design, into useful work output. 256 2. Load Management In load management, the objective’ is to lessen the spread between peak demand and average or minimum demand for power in a given application. Thus, in a village setting where the peak load might be three to four times higher than the average load, various discretionary loads in a village might be arranged to operate at times when the diesel engine is not being utilized to its full capacity. This allows the use of a smaller diesel engine, which will in turn, be operating closer to its rated capacity and thus at higher efficiency. In cases where there are multiple engines operating concurrently, the loads can be further divided into essential and non-essential loads. If one of the units shuts down unexpectedly, the other engine(s) can carry on but only with the essential loads. This permits the use of multiple, smaller engines. , 3. Peak Shaving With this concept, the diesel-generator set is used during off-péak periods to charge a storage system. This would most likely be batteries or possibly a thermal storage system for operating a bottoming cycle. When the load approaches a peak the set continues to operate, but additional power is drawn from the storage system ‘(batteries in the case of electricity storage or a Rankine cycle engine in the case of thermal storage) to meet that peak. This in both cases. allows the. diesel engine to operate more efficiently, -at its rated capacity. In certain applications, this could allow the set to be shut down during periods of low demand, with minor loads being carried by the storage system and providing a reserve power for the diesel restart. 4. Fuel Saver Another potential method of reducing fuel consumption § in diesel-generator systems is the combination of the diesel-generator with a fuel. saver, renewable energy technology, such as a photovoltaic or wind system. In this configuration, the output from the renewable resource system is used to assume all or most of the load of the diesel-generator during daylight hours or during windy periods, thereby saving fuel. The diesel-generator is still available at reduced load or idle to assume full load should the renewable resource not be available. 5. Insulated Enclosure Enclosure of the diesel-generator in. an insulated box will both lower the outside noise level attributable to the diesel engine and will contain heat inside the enclosure. This permits the engine to be shut down and started up later. with a reduced danger of freezing the lubrication oil or the cooling. system a liquid cooled engine. 257 6. Blending of Spent Lubricating Oil with Fuel =, Another: approach to effectively increasing the fuel efficiency of the engine is to filter and blend small quantities (typically 5%) of the oil that is drained from the engine lubrication system during routine maintenance. In a typical village application and a medium sized. (150 kW) engine, the amount of lubricating oil available from routine oil changes will only be adequate for 3 to 4% blending. This has the added benefit of eliminating the spent oil disposal problem. Diesel-Generator System Equipment Modification 1. Multiple Small Units While in some cases there is incentive to operate a single large diesel-generator there are applications where substantial fuel savings will be realized with the use of multiple small units. Load swings can be followed by shutting down or starting up one or more units, allowing those operating to run efficiently near their rated capacity. In addition, reliability through redundancy can be achieved with - fewer total units. . For example, for 400 kW of peak capacity, one — i would need at least two 400 kW units (total 800 kW), if only one bee engine were to operate at any time. However, if four 100 kW engines were to be used, it might prove acceptable to install two additional 100 kW (total 600 kW) units, perhaps with one rotated out of “4 operating sequence for maintenance and one available for backup. ” - Thus 200 kW of generator capacity are eliminated, while maintaining 7 good backup provisions. The economic disadvantage of this mode of 1 operation is the requirement for sophisticated auto-paralleling ; equipment and for small unattended units. , — 2. Multi-Fuel Capabilities . wa Some larger diesel engines are now designed to run on gaseous fuels \ as well as liquid fuels. In these engines, a gas charge is admitted to the cylinder and at the appropriate instant a small quantity of diesel fuel is injected into the compressed gas causing ignition. In most eases, only 10% of the energy output is attributable to the diesel fuel and the remainder comes from the gaseous charge. These engines typically operate with efficiencies comparable to those operating 7 strictly on diesel fuel and are generally only appropriate where the 1 gaseous fuel is available at a cost less than that of the diesel fuel per i Btu. 3. Turbocharging and Turbocompounding Most modern diesel engines currently use turbocharging. Turbocharg- ing is the extraction of energy. from the exhaust heat stream to increase the amount of combustible mixture introduced into the 258 _ engine, thereby increasing power output. Turbocompounding is the extraction of this same waste energy from the exhaust gases, but adding most of that energy directly to the output shaft of the diesel engine. Both of these concepts allow increased output from a given size engine system, without deterioration of efficiency and performance. Therefore the weight of an engine with turbo charging or turbocom- pounding will be less than the weight of a conventional diesel engine of the same power output. This can be particularly important in mobile applications. 4. Adiabatic Diesel Cycle Like any thermal engine, the diesel cycle operates more efficiently at higher combustion temperatures. One of the limitations of the diesel engine is the temperature at which it can operate. This limitation is material based and considerable research is underway into new materials such as ceramics. Through the use of high temperature components and the absence of a cooling system, the adiabatic diesel cycle is theoretically capable of attaining efficiencies of over 50%. Thus far only a few adiabatic diesel engines have been operated for short time periods, so it appears that this technology is at least five years from commercial availability. 5. Electronic Fuel Injection One of the methods of controlling exhaust emissions is the careful control of the injection of fuel into the diesel cycle. Electronic systems have the potential for increasing the precision with which the fuel can be injected, thereby drastically reducing exhaust emissions and making the engine more suitable for use in congested areas. 6. Exhaust Catalizers, Filters and Combustors Another method of cleaning up exhaust emissions is through the use of catalytic converters, filters and units to complete the combustion of unburned hydrocarbons in the exhaust gases. As with the adiabatic diesel cycle and electronic fuel injection these are not likely to be offered within the next five to ten years in commercially available diesel cycle engine systems, especially since diesels already meet most applicable environmental regulations. 11.2.3 Environmental Issues The principal environmental issue associated with the diesel engine is the rejection of the products of combustion to the atmosphere. The composition of exhaust gas products depends upon engine and fuel 259 TABLE 11.6 SUMMARY OF DIESEL ENGINE SALES IN THE 0-400 hp RANGE IN THE U.S. IN 1980 CAPACITY RANGE APPLICATION SALES IN THOUSANDS OF UNITS 0-50 : Lawn and Garden Tractors, Welders, 130 Compressors, Gen. Sets 50-100 cars, Trucks, Tractors, Gen. Sets 155 100-150 Cars, Trucks, Tractors, Combines, Gen. Sets 277 150-200 Tractors, Combines, Cranes, Gen. Sets 129 200-250 Trucks, Tractors, Combines, Gen. Sets 114 250-300 Trucks, Buses, Gen. Sets 54 300-400 Trucks, Gen. Sets 34 SOURCE: Compiled by Arthur D. Little, Inc. 260 system design, as well as engine operating temperature. Categories of emissions include carbon, hydrocarbons, carbon monoxide, and oxides of nitrogen as summarized in Table 11.5. | f TABLE 11.5 TYPICAL EMISSIONS SOURCES FROM DIESEL ENGINE Hydrocarbons 300-3000 ppm Carbon up to 5000 ppm me NO up to 3000 ppm ' Carbon monoxide ‘ under 1000 ppm Incomplete flame propagation results in unburned combustible gases in the chamber, leading to hydrocarbon emissions’ in the exhaust. Carbon monoxide (CO) results from rich-mixture combustion in the chamber where there is a _ deficiency of. oxygen .for complete f combustion. Nitric oxide (NO), the primary oxide of nitrogen in the i exhaust gases, forms upon chemical combination of nitrogen and oxygen from air at the. combustion temperatures. | ! Carbon smoke and odor are also emitted from diesels. The principal kai source of carbon (soot) is pyrolysis in the combustion chamber with _ rich fuel mixtures. It is believed that certain low-concentration | products of oxidation and the presence of unburned fuel fractions are the chief sources of odor in the diesel fuels. | Noise is another environmental concern which can affect the siting of a diesel engine. Methods of reducing noise from a diesel are discussed in 11.2.2 Technical Characteristics. 11.2.4 Commercial Status Diesel engine technology is extremely well established and mature. \ There are a number of manufacturers competing in a variety of Fi! different market segments. Table 11.6 is a summary of diesel engine : sales versus engine size in the U.S. for 1980, for the size range, ~ 0-400 horsepower (0-300 kW). As indicated, there is a relatively ' even distribution of different size engines. Table 11.7 is a list of : selected manufacturers and packagers of diesel-generators. 11.3 Economic Implications 11.3.1 Costs Cost categories associated with diesel generator sets consist of capital cost for equipment and installation, and costs for operation of the 1 units including fuel costs and maintenance costs. 261 TABLE 11.7 SELECTED DIESEL-GENERATOR MANUFACTURERS AND PACKAGERS (High Speed Diesel-Generator Sets in the 3 kW to 1100 kW Range) Output Power Range Manufacturer/Packager ; (kw) AVCO-Lycoming 3-30 2603 Reach Road : 1! Williamsport, PA 17701 Lt Caterpillar Tractor Co., Industrial Division 55-930 100 N.E. Adams Street : Peoria, IL 61629 Culler/Detroit Diesel Allison : 40-1000 P.O. Box 82100 fo Burnaby, Canada Cummins Engines Co., Inc. , 120-675 Columbus, OH 47201 Deutz Corporation © . 2-200 _ 7585 Ponce-de-Leon Circle Doraville. GA 30340 i Fermont Division ‘ 13-1100 mF 141 North Avenue . \ Bridgeport, CT 06606 Isuzu Motors Limited . 7-50 22-10 Minami-oi 6-Chome Shinagawa, Tokyo, Japan Kohler Co. 41000 | Kohler, WI 53044 / “) Libby Power Systems 7.5-900 om 1648 Oakland Avenue Kansas City, MO 64126 ad Lister Power Plant _ 4-100 aa 555 East 56th Highway . ! Olathe, KS 66061 ONAN 3-750 2 1400 73rd Avenue, NE . : Vy Minneapolis, MN 55432 ; Stewart & Stevenson 12-1100 a4 P.O. Box 1637 4516 Harrisburg Blvd. ; Houston, TX 77043 . 7 262 _ Capital costs for equipment vary over a wide range depending upon the size of the units and accessories supplied. Furthermore, distance of the installation site from Anchorage usually determines the transportation and installation costs of the unit. Tables 11.8-11.10 summarize the costs for three engine sizes typically used in Bush applications. A small 4 kW diesel-generator set with battery system, battery maintenance, and control panel may be purchased in Anchorage for approximately $6500. Transportation, contingency, and installation costs add another $2,300 to the purchase price. This is equivalent to the total installed cost of $2,200/kW. These small units have an efficiency of approximately 24% and at full power consume about 0.1 gallon of fuel per kilowatt hour of electricity generated.. At $1.80/gallon, the fuel cost of generation alone is 18¢/kWh. Within the same size category, the following equipment costs are applicable for units delivered at Anchorage: Unit Size Approximate Total Cost Unit Cost (kW) / Cr_O.B. Anchorage) ' (F.0.B. Anchorage) 4 : $6,500 © $1,600/kW 8 8,000 1,000 12 9,400 780 33 15,000 450 Equipment capital cost is greatly reduced on $/kW. basis for units in the 100 kW and larger range. For example, the delivered cost of a 150 kW diesel generator set complete with muffler, flex pipe, lube and jacket water heater, battery system and maintainer, and control panel is approximately $41,000 or $273/kW. Transportation costs to the site and installation costs of diesel generator units also vary over a wide range depending on the remoteness of the site, whether the installation is new or an add-on, and the degree of sophistication and automation desired for the installation. According to AVEC (2) the cost for installing new diesel generator sets in the 100 kW to 500 kW range with minimum necessary equipment is approximately $200/kW. This figure including the cost for fuel storage tank, generator shed, day tank, various fuel transfer pumps as well as per diem and ‘cost of technicians and laborers doing the installation. However, if it is desired to have diesel generators installed in individual trailer-type compartments with fire suppression equipment, remote cooling systems and more sophisti- eated electronic control, the installation cost can be as high as $500/kW. Table 11.9 shows an approximate breakdown of capital costs for a 150 kW unit for minimum standard installation total installed cost amounts to $484/kW. 263 TECHNOLOGY : BASIS: Location: Year: CAPACITY: Input: Output: TABLE 11.8 CAPITAL COST SUMMARY DIESEL-GENERATOR SET Bush Region/Individual Household Installation 1982 Constant Dollars Diesel fuel - 0.1 gal/kWh at full capacity Electric power - 4 kW capacity ESTIMATED USEFUL ECONOMIC LIFE: 100,000 hours CONSTRUCTION PERIOD: 2-3 days CAPITAL COST: Equipment and Accessories: ‘1 $6,500 Direct Labor: : 1,000 Indirect Costs: Home Office Costs: Transport Costs ° 500 Contingency (10%) 800 TOTAL CAPITAL INVESTMENT: $8,800 (1) Basic unit: (equipped with low oil shutdown, fuel filter and electric starter) Accessories: (includes battery, battery maintenance, control panel) 264 TABLE 11.9 CAPITAL COST SUMMARY TECHNOLOGY: DIESEL-GENERATOR SET BASIS: |: Location: Bush Region/Village Utility Year: 1982 constant dollars : CAPACITY: Input: Diesel fuel -'0.072 gal/kWh at full capacity. Output: Electric power - 150 kW capacity i ESTIMATED USEFUL LIFE: 10 years \ CONSTRUCTION PERIOD: Add-on 4-5 days we New 10-15 days | CAPITAL COST: Equipment and Accessories: $ 41,000 Direct Labor: ‘?) 10,000 Ke Installation Materia1: (2 10,000 { Indirect Costs: Home Office Costs: Transport Costs: 5,000 7 Contingency (10%) 6,600 ; TOTAL CAPITAL INVESTMENT: $ 72,600 ft (1) LG Basic Unit Accessories: (muffler, flexpipe, lube and jacket, water heater, battery system and battery maintainer control panel, overspeed shutdown). (2) The installation type is the minimum standard set by A.V.E.C. and is not the modular installation in trailer type housing with fire suppression and a other protections recommended by A.V.E.C. 265 TABLE 11.10 CAPITAL COST SUMMARY TECHNOLOGY: DIESEL-GENERATOR SET BASIS: Location: Bush Region/Village Utility Year: 1982 constant dollars CAPACITY: Input: Diesel fuel: 0.074 bal/kWh at 275 kW output. Output: Electric Power - 300 kW capacity. ESTIMATED USEFUL LIFE: 10 years CONSTRUCTION PERIOD: Add-on - 4-5 days New - 10-15 days CAPITAL COST: Equipment and Accessories (1) $ 70,000 Direct Labor: 20,000 Transport Costs: 15,000 Indirect Costs: Installation Material: 15,000 Home Office Costs: Contingency (102%) 12,000 (1) TOTAL CAPITAL INVESTMENT: $ 132,000 (1) Capital cost reflects minimum installation requirements and not, modular installation with fire suppression equipment, etc. recommended by A.V.E.C. 266 Data from AVEC show that the average heat rate of some 120 diesel generators operated in 44 villages was 15,714 Btu/kWh (efficiency of 21.7%); at $1.80/gallon this translates into an average fuel cost of 20.4¢/kWh, : Operation and maintenance costs are designed to cover routine maintenance .and periodic overhauls described in Table 11.4. Estimates indicate an approximate cost in the range of 6 to 8¢/kWh for O&M costs. This includes the salary of the plant operator, maintenance mechanic's per diem charge, cost of spare parts and delivery charges for spare parts both for routine maintenance and major overhauls ($15,000-$20,000 per overhaul). Distribution cost of the energy is approximately 12-16¢/kWh for village utilities. Table 11.11 summarizes the operating cost summary of a 150 kW size diesel-generator in a village application. 11.3.2 Socioeconomic Factors Diesel generator installations in rural areas requires, a comprehensive fuel distribution system, qualified technicians, mechanics, and operators as well as regional centers for spare parts for effective operation and maintenance of these units. These infrastructure, personnel, and distribution network are already in place to serve the needs of rural Alaska. Similarly, the management and administrative support offered by AVEC is well established. 11.4 Impact 11.4.1 Effect on Overall Energy Supply and Use Diesel generator technology is presently one of the principal sources of electric power generation in Alaska, and in all likelihood will remain the dominant source of electric power in rural Alaska in the foreseeable future. Therefore, technical and operational improvements that could lead to fuel use optimization will have a great impact on the state-wide fuel supply and demand. Fuel consumption data (1,2,3,4) indicate that in 1981 approximately 36 million gallons of diesel fuel were consumed for generating electric power by diesel generators. This does not include fuel used by the industrial or military diesel installations. Further analysis shows that the average efficiency of all diesel generators in Alaska (excluding industrial and military installations) in 1981 was 24%. Clearly any technical or operational steps that would lead to an increase in the efficiency of power generation will have substantial impact on the state-wide fuel supply and use. For example, if the average efficiency of all diesels is increased to 30%, an annual fuel savings of 7.2 million gallons (based on 1981 consumption) would be realized. This also means that 7.2 million gallons less fuel must be transported and stored annually. , 267 Table 11.11 OPERATING COST SUMMARY TECHNOLOGY : Diesel-Generators BASIS: Location: Bush Region/Village Utility Year: 1982 CAPACITY: 150 kW - Continuous Input: Diesel fuel: 0.098 gal/kWh average Output: Electric power: 150 kW capacity OPERATING FACTOR: 30% capacity factor ANNUAL OPERATING COSTS UNIT COST CONSUMPTION VARIABLE COSTS: Fuel: $1.80/gal 38,800 gal TOTAL VARIABLE COSTS: FIXED COSTS: Operating Labor: Maintenance: Taxes and Insurance: Capital Charges: @ 9.5% TOTAL FIXED COSTS: 53% TOTAL ANNUAL OPERATING COSTS: TOTAL ANNUAL OUTPUT: TOTAL LIFE CYCLE COST: 268 ANNUAL {| cost hd $ 69,800 _ $ 69,800 - $ 10,900 [~ 14,600 mt 2,200 m 6,900 $ 34,600 i $104,400 394,200 kWh $ 0.265/kWh 1g Data (1,2,3,4) shows that the total installed diesel generator capacity in Alaska (except for industrial and military installations) in 1981 was approximately 228,000 kW. Of this total some 26,000 kW were installed in “communities with a population of less than 250, and approximately 33,000 kW were installed in villages with population between 250 and 500. Diesel fuel used by, these two classes of communities for electric power generation in 1981 is approximated at 6.3 million gallons. Since most of these communities are located in remote areas, improvements in efficiency of operation will have substantial impact on fuel shipments and storage requirements. 11.4.2 Future Trends A potential source of substantial fuel utilization improvement for diesel engines is the use of waste heat in exhaust gases and jacket water for domestic hot water and space heating applications. This technology is under serious consideration in several communities in Alaska. As discussed in Section 11.2.2, the overall: fuel utilization may be increased to as much as 70%. ©The application of diesel generators in total energy systems will further reduce the require- ments for heating oil and this presents another major source of fuel savings in future years. Through load management techniques, it may be possible to lessen the spread between peak demand and averdge or minimum demand and thus operate smaller diesel generators at higher part-load conditions (60% or higher). Technological and design improvements in engine design consist of applications of units with turbocharging and turbocompounding. This- will allow increased output from a given size engine system without deterioration of efficiency and performance. It is anticipated that. as older diesel units are retired, the replacement units will incorporate these advanced design features. The use of combination of diesel generators with renewable electric generation technologies, particularly wind turbines, presents real promise for Alaska. Several demonstration projects are presently underway in Alaska and the results will determine the feasibility of this method of electric power provisions. 269 10.: 11. 12, REFERENCES Alaska Power Administration, Alaska Electric Power Statistics, 1960-1981. August 1982. Alaska Village Electric Cooperative, Anchorage, Alaska, Personal ‘Communications and Interviews, September 1982. Division of Electric Power Development, Department of Commerce and Economic Development, Anchorage, Alaska, 1981 Community Energy Survey, 1982. Division of Electric Power Development, Department of Commerce and Economic Development, Anchorage, Alaska, State of Alaska Long-Term Energy Plan. Report prepared by Booz, Allen & Hamilton, 1982. Baumeister, T., Editor-in-Chief, Standard Handbook for Mechanical Engineers. McGraw-Hill Book Company, 7th Edition. Gill, P.W., et al. Fundamentals of Internal Combustion Engines. The U.S. Naval Institute. Shearer, A.B., "The Historical Development and _ Future Prospects of the Diesel Engine," Diesel Engineering, 72 (790): 155, 1976. . Cummins Engine Company, Inc., 100 Allied Drive, Dedham, Massachusetts 02026. Private Communications, November 1982. Caterpillar Tractor Company, 100 N.E. Adams, Peoria, Illinois 61629. Private Communications, November 1982. Lister Diesels. R.H. Mitchell Co., Inc. 76 Main Street, Fairhaven, Massachusetts 02719. Private | Communications, November 1982. Northern Technical Services, Reconnaissance Study of Ener Requirements and Alternatives, Report prepared for Alaska Power Authority, July 1982. Acres American, Reconnaissance Study of Energy Requirements and Alternatives, Report prepared for Alaska Power Authority, July 19827 270 { ' A Q Boe 12.0 PHOTOVOLTAICS 12.1 Introduction and Summary 12.1.1 Technical Overview A photovoltaic (PV) system produces DC electric power directly from sunlight using semiconductor materials. These systems do not have any moving parts or complex machinery, as do solar power systems based on thermal processes, such as those using concentrated solar energy and Rankine cycle power plants. The electric power generated is directly. proportional to the amount of solar insolation intercepted by the PV system and is, therefore, directly proportional to the area:of solar cell material and the solar insolation available. Since PV systems contain no moving parts, they are particularly suit- able for use in remote area applications, where the major alternatives are operation of engine driven generators or extending the grid, both of which are often particularly costly. There are several thousand small PV systems installed worldwide to serve such: applications as corrosion protection of pipelines and operation of remote communication equipment. There are at least 40 establishments in the United States either doing research and development, or manufacturing PV cell arrays and other components required for these systems. Complete system packages are available from over a dozen manufacturers. 12.1.2 Alaskan Perspective A large portion of rural Alaska is currently without access to grid supplied electricity. In many locations, electricity is generated by small diesel generator sets with critical fuel supply and maintenance considerations. Photovoltaic power units can be suitable for supply- ing small amounts of power in these rural areas to displace a portion of that fuel, depending on the type of load. PV systems are highly reliable and almost maintenance free, which makes them particularly attractive to satisfy the needs of isolated groups of homes, seasonal fishing or trapping camps, communication equipment, and specialized applications such as the operation of warning lights at railroad crossings. Even at presently high costs, small (10-100 watt) photovoltaic power units are already in limited use in Alaska for small power needs in remote areas for such functions as railroad crossing lights and mountain top signal equipment. The use of photovoltaics in these special applications will probably expand as implementing organizations gain experience with their use. 271 How Solar Cells Make Electricity NTYEE SILICON J‘ LZ r— CUKKENT FL YRE SILICON Typical I-V Characteristic Typical Spectral Response (Quantum Yield) of Solarex Solar Cell of Solarex Terrestrial Solar Cells sister 1 sun | Current bea : 1 ‘ 4 Wavelength. Nanometer Voltage 45V Vv Source: Solarex Product Literature. FIGURE 12.1 OPERATION OF PHOTOVOLTAIC CELLS 272 12.1.3 Significance of the Technology Photovoltaic panels now cost between $7,000 and $13,000 per kilowatt of peak capacity (attainable during optimal solar conditions); system costs (including battery storage, controls, etc.) range from $15,000 to $25,000 per peak kilowatt. Panel costs are projected to decrease to $2,000-$4,000 per peak kW in the next few years and possibly lower ($1,000-$2,000 per peak kW) thereafter, with corresponding system costs of $3,000-$6,000 per peak kilowatt, as the technology of their manufacture improves. Despite the relatively low availability of solar energy in most regions of Alaska, photovoltaic power systems might become economically viable for residential power generation, in specialized Bush applications if PV system costs decrease in accordance with the above industry projections. This is primarily due to the very high cost of delivered fuel and the high cost of maintaining a _ small engine-generator set in remote areas. The potential impact of photovoltaic power in Alaska in the foreseeable future will be primarily to serve some of the needs of remote settlements and reduce the onerous costs associated with delivering fuel and maintenance in these areas. However, given the rather low solar availability in all areas of Alaska, it is doubtful that photovoltaic power will be competitive with grid connected power or larger scale diesel power generation (50-500 kW) in the foreseeable future. 12.2 Technology Description 12.2.1 Operating Principles Solar cells are constructed by processing thin sheets of semiconductor materials such that they produce free electrons when illuminated by solar energy. The semiconductor wafers have attached electrical grid structures to capture these electrons and utilize them in the form of an electrical current in an external electrical circuit. The voltage at which electricity is produced ranges from 0.4 to 0.6 volts depending on the material and its processing. This process is illustrated in Figure 12.1. The efficiency with which solar energy can be converted to electricity by the photovoltaic process depends on the atomic structure of the material and typically range from 10-20% for most semiconductor materials of practical interest. There are numerous photovoltaic material combinations producible from several fabrication processes which are being investigated by private, government, and academic organizations. However, most solar cell panels now in use for both terrestrial and space applications utilize single crystal silicon as the photovoltaic semiconductor material, due to its demonstrated high reliability, long life, and relatively high 273 Table 12.1 THE THEORETICAL AND PROJECTED EFFICIENCIES OF SFLECTED PHOTOVOLTAIC MATERIALS Theoretical Panel 1 Material Efficiency (4) Efficiency (%) Silicon-Crystalline 22 9-14 Silicon-Amorphous 15 7-117 Gallium Arsenide 24 12-16 Cadmium Sulfide 17 7-11 SOURCE: References 2,3,4, and 5 1. Panel efficiencies take into account both imperfections in the solar cell material itself and losses of light through the protective glass covers, resistance losses in the wiring, and unused space between cells. 2. Projected, not yet attained in practice. 274 efficiency. In the near term most solar cells will probably continue to be made from single crystal silicon, using steadily improving material purification and cell fabrication techniques to lower cost for the next 5-10 years. Single-crystal silicon, which, is also commonly used throughout the computer and electronics industry. has been the preferred photovoltaic material for more than a decade. Its theory, technology, and application are well developed, and array efficiencies of 9-13% are routinely realized in commercially available panels. The major thrust of present single-crystal, silicon solar cell development is to reduce material preparation cost by such means as directly forming thin ribbons of silicon ready for use as solar cells so that the conventional process of sawing and polishing wafers from bulk single-crystal material can be avoided. Alternative solar cell materials which are farthest along in their development are amorphous silicon, gallium arsenide and cadmium sulfide. The primary incentive for developing these alternatives is that they can be effectively utilized in very thin films (2-10 microns), which is consistent with low cost manufacture. Although there are numerous R&D programs sponsored by both government and industrial organizations to improve the performance of these alternative mate- rials, none of them is expected to replace silicon before the mid to late 1980's. Table 12.1 summarizes the efficiency characteristics of selected photovoltaic material options. For long life and reliable operation, all photovoltaic cells require some form of protective covering (referred to as encapsulation). Aside from protection against dirt and moisture, the active region of the cells must be kept free from contact with various atmospheric pollutants which can migrate through the semiconductor and deteriorate performance. Some solar cells such as cadmium sulfide are particularly dependent on a glass-faced, hermetic enclosure for a useful life. But, even for less sensitive silicon and gallium arsenide, this type of completely hermetic encapsulation will probably be required to provide the 10-20 year cell life desired. 12.2.2 Technical Characteristics A photovoltaic power system usually consists of a photovoltaic array, electronic controls, and energy storage, the nature of which depends on the application of the system. A schematic of a typical photovoltaic power system is shown in Figure 12.2. The primary energy conversion unit is the photovoltaic array which produces direct current (DC) electricity. This DC energy can be applied directly to a load (i.e., DC motor, water pump, lights, etc.), converted to alternating current (AC) for use with conventional AC equipment, or stored for later use. 275 BATTERY SOLAR CELL ARRAY “CONTROLS py POWER SYSTEM on a N ‘ ~ _ - ~. “4 ‘. LIGHT ae WATER PUMP REFRIGERATOR . FIGURE 12.2, CONCEPTUAL PHOTOVOLTAIC POWER SYSTEM 276 . she ELECTRICAL __. LOADS a Photovoltaic Modules The photovoltaic array consists of the photovoltaic panels, mounting structure, and all electrical wiring and hardware to carry the elec- trical output from the modules to the load or other conversion or storage modules in the total power system. ‘ ‘ Photovoltaic panels consist of solar cell wafers which have been mounted on a rigid frame, provided with electrical interconnects, and encapsulated in a weather resistant enclosure. Individual cells within . a panel are connected in series arrangements to result in electrical a output typically at 6 to 24 volts. Commercially available panel sizes js typically range from 1 ft? to 4 ft? with peak output levels of 10 to 40 7 watts per panel. i Energy Storage Batteries . Due to the wide variations in solar energy availability in Alaska and - the need for energy under no-sun conditions, electrical energy t storage (batteries) is normally used in conjunction with a PV system. Since these storage batteries may undergo deep charge and discharge « eycles as often as once a day, heavy duty lead-acid batteries are | . required for these applications (such as batteries used in electrically - powered industrial fork lift trucks). ‘ : | f DC to AC Conversion The most efficient way to use the power generated by PV panels in | j small systems is in the DC form in which it is generated. Most of the ' major needs for electric energy such as pumping, refrigerating, and lighting can be provided by 12 volt DC appliances which are becoming is increasingly available. It is possible to add an inverter to the system which converts DC 7 . supply. to AC (120V) power in order to be able to use regular house- fo hold appliances. . Inverters developed for photovoltaic applications | using highly reliable solid state. electronics are available from over a dozen manufacturers. ’ Performance 4 Typical overall conversion efficiencies of photovoltaic panels using i single crystal silicon are in the 9-14% range. The power output of a oe solar cell is directly proportional to the level of incident insolation and the area of solar cell array being illuminated. Solar cell and : panel performance are typically quoted for insolation levels of 1000 i? W/m? (approximately 320 Btu/hr-ft?), corresponding approximately to | clear sky conditions at noon during the gummer. A 10% efficient array would produce approximately 100 W/m” (10 watts/ft?) at design conditions.” Thus, in order to generate 1 kW of electricity under. ho 277 82 2000 Anchorage 1500 2 Fairbanks —E | f- \ ff fF fee UN UN fee 2s 58 an, 1000 Juneau ow —-——-— + cp 5 os 3 £ 500 Jan Feb Mar Apr May June July Aug — Sept Oct Nov Dec Month Source: Reference 7. FIGURE 12.3 SOLAR: AVAILABILITY IN THREE ALASKAN LOCATIONS high insolation peak conditions, approximately 100 square feet of photovoltaic panels are required. Since output varies with the level of insolation, capacity is stated in peak kW (kWp) which corresponds to about 100 square feet of panel area. The salient issues regarding the technical performance of photovoltaic power units in Alaska are the relatively low solar insolation on an annual basis and the very high seasonal variability of the solar resource (Figure 12.4). To compare Anchorage insolation levels with the more nearly ideal conditions of Phoenix. Figure 12.3 shows the monthly and annual variations in output which would be associated with identical PV power systems located in the two cities. Total solar resource estimated for Bethel, Matanuska, and Fairbanks over an extended period indicate less than a 5% variation on an annual basis between these locations. Thus, data for Anchorage can be _ considered representative of a broad cross-section of Alaska. As indicated in Figure 12.3, the system in the Alaska generates only half the electric energy output on an annual basis as that in Phoenix. As a result the photovoltaic array area required to serve a given load in Alaska is roughly twice that in favorable locations in the Lower 48 states. However, the resultant increase in system costs might still be more justified in remote areas where the cost of supplying small elec- tric needs by small generators is particularly high. Furthermore, the seasonal variation in output in the Alaskan system is very large (ratio of maximum-to-minimum of 2.5) which complicates its design and makes it necessary to have an engine-generator backup for many systems of practical interest and year-round operation. 12.2.3 Environmental Issues The environmental impact of photovoltaics is largely associated with the manufacturing process.(8) The actual deployment of the arrays themselves should have little impact on the local environment, partic- ularly if they are mounted on existing structures or future buildings which satisfy other local needs. They produce no exhaust or chemical emissions, no noise, and no significant thermal output. However, the manufacture of the arrays will require mining of selected materials and subsequent chemical and electronic processing. These processes and their energy requirements may _ have environmental consequences that are difficult to quantify at this time. The manufacture of solar cells will most likely be done in large integrated facilities in the Lower 48 and would, therefore, not directly impact on environmental issues in Alaska. 279 Power Output Wh/day 1100 Phoenix, AZ 1000 900 800 700 600 Anchorage, Alaska 500 400 300 200 100 J F M A M J J A Ss oO N D Source: Arthur D. Little, Inc. estimates based on Climatological Data. FIGURE 12.4 TYPICAL 220 WATTS PHOTOVOLTAIC POWER SYSTEM OUTPUT PERFORMANCE COMPARISON By 280 wd 12.2.4 Commercial Status There are currently more than 40 companies involved in the research, development, and manufacture of PV materials and equipment in one form or another. A list of the most active manufacturers of commercially ready PV modules is given in Table 12.3. Several of these manufacturers listed are either wholly or partially owned by major oil companies which suggests that industry foresees large potential markets for PV’ systems as their costs decrease. There are several firms in Alaska which are now distributing products from these companies and assisting in the design and installation of systems. * ‘ Currently there are several thousand PV_ systems’. operating worldwide, primarily to supply minimal (10-500 watt) power needs in remote regions for such critical functions as navigation aids, cathodic protection of pipelines, and communication equipment. It is estimated that there are 12-24 small installations currently operating in Alaska for similar applications. Table 12.4 indicates several applications which are either now. operat- ing in Alaska or might represent potential applications as the cost of photovoltaics in Alaska: and worldwide are primarily for providing small amounts of power in rural areas where the cost of conventional alternative (small diesel generators, grid extensions, battery packs, etc.) are particularly high. The annual worldwide sales of PV modules for 1981 was estimated at about 5,000 kilowatts of which more than 50% went to the foreign markets to serve remote application often under similar conditions as found-in Alaska. The annual sales of PV have grown at approximately 50% per year for the last 3 years. The major companies leading the industry in sales are Arco Solar, Solarex, and Solar Power Corporation. Several of the firms offer pre-engineered systems whereby the photovoltaic panels, controllers, battery pack, and. inverters are provided as an integrated system. Significant use of photovoltaics in Alaska will probably require that such integrated systems be designed for typical Alaskan applications: and consistent with Alaskan solar availability conditions. * Names of photovoltaic panel distributors and installers are available from Alaska's Division of Energy and Power Development, Energy Extension Service, Anchorage. : _ 281 Table 12.3 SELECTED MAJOR: SOLAR FLAT PLATE PHOTOVOLTAIC MODULE MANUFACTURERS SILICON SOLAR CELLS Arco Solar Inc. 20554 Plummer Street Chatsworth, CA 91311 Solarex Corporation 1335 Piccard Drive Rockville, MD 20850 Solar Power Corporation 20 Cabot Road Woburn, MA 01801 Solec International, Inc. 12533 Chadron Avenue Hawthorne, CA 90250 Solavolt, Inc. 5005 East McDowell Road Phoenix, AZ 85008 Photowatt International, Inc. 2414 W. 14th Street - Temper, AZ 85281 Applied Solar Energy Corp. 15251. W. Son Julian Road P. 0. Box 1212 City of Industry, CA 91749 Mobil Solar Energy Company 16 Hickory Drive Waltham, MA 02154 CADMIUM SULFIDE SOLAR CELLS Photon Power, Inc. 10787 Gateway West El Paso, TX 79935 SES, Inc. Traler Industrial Park Newark, DE 19711 282 no / ‘Table 12.4 - : POTENTIAL APPLICATIONS FOR PHOTOVOLTAIC POWER SYSTEMS IN ALASKA we ;: (Example List) Residential Minimal Power for Rural Homes (lighting, pump, refrigerator, etc.) , Commercial Sector Radiation Samplers Visibility Monitors ! Noise Monitors , ~ Navigational Beacons ian Weather Monitors : Particulate Sensors mR : Meteorological Sensors a Military Ammunition Security i Electric Fence io Intrusion Detectors Cathodic Protection mm Power Line Towers an Submarine Cables uO Forest Lookout Towers a Transportation Sector ~ Beacons Buoys : Anemometer rd Moving Target- Indicators Radar Beacons 4 Aircraft Arresting Systems Astronomical Monitor Flash Beacons Remote Instrument Platforms Starpex Beacon Railroad Crossing Lights 7 Forest Lookout Towers Repeaters, Special Purpose i! Venting System, Sanitation Miscellaneous yt * Special Functions . | Seasonal Fishing Camps. (lights, radio, etc.) Season Cold Storage (for hunters) | Scientific Monitors Source: Reference 9. ‘ * Needed primarily during summer months when solar availability is | - high. ue 283 12.3 Economic Implications 12.3.1 Costs Generalized Costs The major cost element of a photovoltaic array today is the cost of the photovoltaic panels. The mounting structure, panel wiring, installation, and miscellaneous parts currently represent less than 20% of the array cost. As the cost of the panels decreases due to the development of lower cost cell technologies and panel fabrication techniques and the increase in production, the cost of the remaining components of a photovoltaic array will become relatively more important. Government agencies and industrial organizations are both working to reduce the cost of these components by developing standardized components and installation techniques. These efforts are being conducted in parallel with those efforts associated with the photovoltaic panel and materials themselves. Presently, the cost of PV panels purchased in large quantity (tens of kW of peak capacity) are approximately $7,000-13,000 per peak kilowatt. For single panel purchases, costs are approximately $12,000-17,000 per peak kilowatt. As indicated earlier, the major reduction in panel cost will occur primarily due to the reduction in solar cell costs resulting from lower cost materials and cell fabrication techniques. In addition, the cost of panel fabrication is also expected to decrease because of lower cost assembly techniques and panel materials, automated assembly and increased production rates. A reasonable projection for mid to late 1980's panel costs is $2,000-5,000 per peak kilowatt. The U.S. Department of Energy and some industry participants project even lower panel costs of $1,000 per peak kilowatt after 1986. Heavy duty battery systems designed for long life (excess of 5 years) under daily operation cost in the range of $150-300 per kWh of storage capacity. The cost of battery storage can be a sizable percentage of overall system costs and, therefore, the cost optimization of system design usually involves trade offs between the reliability with which power is available from the photovoltaic system (high availability implying large battery storage capacity) and the cost of the battery storage subsystem. This issue could be particularly important in Alaska due to the aforementioned large seasonal variations in solar availability. The cost of solid state, self-controlling, inverters in the lower capacity range of primary interest (1-20 kW) is in the $500-1,000/kW range. This type of inverter is required for stand-alone systems independent of the utility grid which would be of primary interest in Alaska. Inverters designed to be frequency controlled by utility 284 fo supplied power cost approximately half that of the self-controlled designs. Such inverters are of interest primarily in high solar availability regions of the Lower 48 where grid connected systems are under consideration. Table 12.5 provides an overview of the costs associated with photovoltaic power system components. Figures are presented for present practice as well as projections for the mid to late. 1980's based on industry estimates of cost reductions in photovoltaic panels. The installed system cost range is now approximately $12,000-23,000 per peak kW of photovoltaic array capacity depending on battery storage requirements and the need for inverters. This system cost may drop to $5,500-12,000 per peak kW by the mid 1980's assuming significant reductions in panel costs. In general, the lower end of the cost range applies to larger systems (over 10 kW) while the higher end to smaller systems (1 kW) which would be of most near-term interest in Alaska. Illustrative Example To illustrate system design,. operating parameters, and economics a PV power system for an existing house or camp in the Bush with its own dedicated diesel generator has been selected as an example application. This potential class of application may become economically increasingly viable in particularly remote areas as the cost of photovoltaic panels decrease. Power is required for such functions as lighting, refrigeration, appliances operation, and pumping water. The design of a photovoltaic power unit to serve a specific load depends on a number of interrelated factors such as the daily and seasonal variations in the load profiles, solar insolation charac- teristics, battery storage system performance, and acceptability of using backup power sources. Several reports and brochures are available (16,17,18) describing. how to size critical system components given the details of the electric load, solar resources, and backup power system characteristics. This very approximate system sizing is intended only to provide a preliminary indication of photovoltaic array size and the probable need for backup power for this class of application. Typical electricity requirements in small residences range from 3 kWh to 6 kWh per day. However, a recent study (13) of energy needs in remote settlements indicate that these electricity loads could be lowered to about 1 kWh per day assuming the use of low wattage fluorescent lights, small (6 cubic foot). refrigerators, and high efficiency pumps and appliances, the power draw would probably increase significantly during the winter months when lighting would be required for most of the day. 285 Table 12.5 APPROXIMATE UNIT COST BREAKDOWN OF PHOTOVOLTAIC POWER SYSTEM COMPONENTS (1982 Dollars) 1982 Mid to Late 1980's Photovoltaic Panels ($/kWp)! 8,000-15,000 2,000-5 ,000 Battery Storage ($/u)2 : 1,400-2,800 1,400-2,800 | : Inverters & Controls (kW) 500-1 ,000 500-1,000 | Installation (s/Kwy? 2000-4 ,000 1,500-3,000 | COST RANGE $ 11,900-22,800 $ 5,400-11,800 SOURCE: Arthur D. Little estimates based on References 8, 9, 10, and ll, | 1. Panel costs of $8 to $15 per peak watt in 1981 decreasing to $2 to $5 per peak watt by mid 1980's. 2. Assumes fixed cost of $850 (wiring, control, electric switches, etc) and $100 to $200 per kWh, 5-10 kWh of battery capacity per kW of PV capacity (References 10 and 11). 3. Assumes $2 to $4 per square foot for mounting and wiring (Refe- rence 12). 286 For illustrative purposes it is assumed that the electricity loads are reduced to about 1 kWh per day by careful selection of appliances in order to reduce the relatively high capital costs associated with the photovoltaic array and supporting equipment. Array size and initial cost would increase almost directly with electricity requirements and would, therefore, be about 3 times higher for a system to serve a dwelling with a more typical 3 kWh per day electricity requirement. Since array area and initial cost increase approximately with system output requirements, the costs per unit output would be relatively insensitive to system size for such residential applications. In a photovoltaic power system about 30% of the electricity produced by the photovoltaic array is lost during the process of battery charging and discharging. Therefore, for the example system to deliver 1 kWh of electricity to the home, the photovoltaic array must generate about 1.3 kWh of electricity. A peak kilowatt of photovoltaic capacity generates a kilowatt only within one hour or so of noontime. At other times during the day, it produces considerably less as the angle between the sun and the photovoltaic array deviates from 90°. As a result, over the course of the day during the summer each kilowatt of photovoltaic array capacity produces about 5 kWh of electricity at the latitude of Anchorage. Therefore, to satisfy a significant portion of the assumed residential load during these months, about 260 peak watts of photovoltaic array capacity would be required. In order to assure that the battery storage units can be charged for cloudy weather operation and to increase output during marginal months, a total capacity of 350 peak-watts is assumed (about 35 square feet of solar cell panel area). Clearly, even this capacity will not be sufficient to satisfy a large portion of the load during the winter months when solar availability is very low. To satisfy the winter loads with photovoltaics would require installion of about three times the capacity needed during the summer months, with correspondingly large increases in cost. The approach taken here is, therefore, to assume a design which handles most of the load for about 6-7 months of the year and to count on extensive use of small diesel generators during the winter months. The benefits to be derived from this system approach will be: e greatly reduced use of diesel fuel on an annual basis; and @ a reduced number of operating hours on the diesel generator leading to longer engine. life and extended maintenance intervals. The specifications for the illustrative system design are indicated in Table 12.6. As indicated, the system assumes the use of 2 kWh of heavy duty batteries which is sufficient for about two days of operation without solar availability. 287 Table 12.6 ILLUSTRATIVE SYSTEM DESIGN SPECIFICATIONS Solar Collector Array - Collector Area - Number of Panels - Capacity - Tilt Angle - Output Battery Storage - Battery Jype - Capacity - Operating Voltage Other Controls 35 ft? 15 350 peak watts 50° 24v DC Deep Draw, Lead-Acid 2 kWh . 24v DC (System) - Battery Charger/Penetration Device - Circuit Breakers * Assuming typical panel size of 12 in. by 24 in. kk Sufficient for about 2 days operation with minimal solar availability. 288 The assumed system would on average satisfy most of the load during six or seven months of the year. Even: during these months, however, use of the engine-generator would be needed during — prolonged cloudy periods (longer than two days with the storage assumed). From October through January, over half the power needs would have to be provided by the engine generator. The estimated annual useful output of the photovoltaic battery system would be about 260 kWh per year. The cost of the small residential system outlined in the previous section is summarized in Table 12.7. As indicated, the cost of the system would presently be about $6,400 (about $18,000 per peak kW) on an installed basis decreasing to about $4,000 (about $11,000 per peak kW) by the mid to late 1980's. By that.time, highly modular pre-engineered systems could be available which would allow relatively straightforward installation by the purchaser. This is consistent with the way small diesel-generators are now sold and installed and reflects the high level of practical, hands-on capabilities which are normally developed in rural settlements. It is estimated that at least 8-10% can be saved through self-installation. , Table 12.8 summarizes the operating costs for the photovoltaic power unit and indicates effective electric energy costs of about $2-3.25/kWh, depending on he date of installation. These cost numbers are very high compared to utility power or community size diesel power plants. However, the benefits of the photovoltaic unit for remote households without access to centrally supplied. power include: a number of factors including reduced fuel consumption, longer back-up engine life, and reduced engine maintenance require- ments. : , 12.3.2 Socioeconomic Factors There are already about 3 firms in Alaska distributing photovoltaic equipment. The increasing use of photovoltaic power systems for both individual use (homes, camps, etc.) and for communications and safety equipment would lead to increasing activity in the state by local firms in the design of systems tailored for Alaskan applications. and the distribution of systems and components. However, these activities are incremental on those already well established for systems for which photovoltaics would probably substitute. 12.4 Impact 12.4.1 Effect on Overall Energy Supply Photovoltaics offer several potential advantages for Bush power generation. They are reliable and operate only on renewable solar energy. And they are applicable for individual households or groups of them. 289 Table 12.7 _ CAPITAL COST SUMMARY TECHNOLOGY: Photovoltaics BASIS: Location: Anchorage Year: 1982 constant dollars CAPACITY: Output: Peak: 0.35 kWp Annual: 330 kWh ESTIMATED USEFUL LIFE: 20-30 yrs. CONSTRUCTION PERIOD: 1 day to 1 week CAPITAL COST: Equipment and Materials: Direct Labor: TOTAL CAPITAL INVESTMENT: * Solar Cell Array (Includes supports and wiring) Controls (Includes battery charge) Storage Battery Systems (2 days storage) (a) 1982 $ 6,400 500 $ 6,900 4,700 500 1,200 - 290 Mid 1980's $ 3,600 400 $ 4,000 1,900 500 1,200 (a) Assumes $13/p-w in 1982 and $5/p-w in mid 1980's. Table 12.8 OPERATING COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Output Peak Annual OPERATING COSTS Operations and Mainte Taxes and Insurance Capital Charges (9.5% TOTAL ANNUAL OPERATIN TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST 1 mance ) iG COST Photovoltaics Anchorage 1982 constant dollars 0.35 kWp 330 kWh ANNUAL COST 1982 Mid 1980's 118 72 118 72 603 380 844 524 260 kWh 260 kWh $3.25/kWh $2.02/kWh 1a ssumed to be 2% per year of equipment costs in replacement parts Routine collector cleaning and battery (faulty panels, etc.). checks done by owner at 2 assumed to be 2% of init no cost. ial cost. 291 Although the cost of photovoltaics is projected to move down rapidly over the next few years however, the overall economics do not appear favorable for the remainder of this decade for any significant penetration into applications beyond very specialized ones already in place in Alaska and perhaps a few others. Further technological advances will be required for photovoltaics to have a large impact. 12.4.2 Future Trends Photovoltaics is one of the most rapidly changing energy technologies with several dozen companies participating in R&D on new materials and processes to lower the cost of panels. As a result, the industry and the U.S. Department’ of Energy expect that the cost of photovoltaic panels will decrease dramatically over the coming decade and that the present method of fabricating cells with single crystal silicon will be replaced with either new materials or with more . efficient means for fabricating single crystal silicon wafers. This lowering of panel costs will increase the range of applications where photovoltaic power is competitive. As a result, both manufacturers and local distributors will develop an expanded line of prepackaged systems to efficiently address these applications and to lower the cost of system installations. 292 10. 11. 12. 13. REFERENCES Solar Energy Comes Down to Earth, Communications Magazine, April 1976, pg. 2. . Wolf, M., "A New Look at Silicon Solar Cell Performance," Energy Conversion, New York, Pergamon Press, 1971, Vol. II, pp 63 ff. Carlson, D.E., et al "Amorphous Silicon Solar Cells" Applied Physics Letters 28 (11):671, June 1, 1976. Hovel, H.J. "Solar Cells" Semiconductors and Semimetals Edited by R.K. Willardson and A.C. Beer, New York Academic Press, 1975, Vol. II, pp 71 ff. Rothwarf, A. Theoretical Prospects of the CdS-Cu,S Solar Cell, Prepared by Delaware University, Institute of Energy Conversion for National Science Foundation, March 1976, NTIS Report No. PB 252 409. Photovoltaic Power Systems for Rural Areas of the Third World, NASA-Lewis Research Center, Cleveland, OH, October 1980 (presentation charts). Alaska Solar and Weather Information Fact Sheets, Division of Energy & Power Development, Alaska Energy Extension Service, Anchorage, Alaska. Proceedings of the DOE Annual Photovoltaics Program, April 28-30, 1980, CONF-8004101. Cost of Photovoltaic Energy Systems as Determined by Balance-of-System Costs, Lewis Research Center, 1978 NASA TM-78957. . . : Parametric Analysis of Residential Grid-Connected Photovoltaic Systems with Storage, March 1980, SAND 79-2331. Regional Conceptual Design and. Analysis Studies for Residential Photovoltaic Systems, Vol. II, May 1980, SAND 78-704012. "Integrated Residential Photovoltaic Array Development," Quarterly Report No. 3, DOE/JPL 955894-3. Market Definition Study of PV Power for Remote Villages in the United States, DOE/NASA-49--80/1. : 293 14, 15. 16. 17, 18, Reconnaissance Study of Energy Requirements and Alternatives, Northern Technical Services and Van Gulish Associates for the Alaska Power Authority, July 1982. Market Definition Study of Photovoltaic Power for Remote Villages in the United States, C. Ragsdale and P. Quasbie, February 1980, DOE/NASA/0049-80/1. Designing Small Photovoltaic Power Systems, Monegan Publication No. M111, May 1981. Photovoltaic Stand-Alone Systems, Preliminary Engineering Design Handbook, August 1981, DOE/NASA 0195-1. "Solar Photovoltaic Application Seminar: Design, Installation, and Operation of Small Stand-Alone Photovoltaic Power Systems," U.S. Department of Energy, July 1980, DOE/CS/32522-T1. 294 s a 13.0 RANKINE BOTTOMING CYCLE ENGINES 13.1 INTRODUCTION AND SUMMARY 13.1.1 Technical Overview Rankine bottoming cycle (RBC) engines can use the hot exhaust gases of diesel engines and hot waste heat streams from industrial processes to generate power with no increase in fuel consumption. These engines operate on the same basic thermodynamic cycles as conventional steam power plants. _During operation the hot exhaust gases are used to vaporize the working fluid of the RBC engine in a boiler of special design. The hot, pressurized vapors then are used to operate a turbine which, in turn, drives an electric generator. The low pressure vapors are returned to the liquid state in a condenser (air or water cooled) and then pumped into the boiler thus completing the cycle. Engines can be designed to operate over a wide range of temperatures from hot engine exhausts (700-1000°F) to'as low as 300°F reject heat from industrial processes. Engines have been made in capacities from as low as 3 kW up to 2 MW. However, most commercial activity is presently in the power range from 300 kW to.1 MW. The optimum working fluid, selected from among refrigerants, hydrocarbons and water, depends on engine capacity, hardware configuration, and operating temperature level. When using hot diesel engine exhaust gases as the energy source, RBC engines provide an additional 15-25% power output without any increase in fuel consumption. For this reason they have been used to improve the fuel efficiency of several diesel power plants in the U.S., Japan, and Europe. They have also been used as a means of using hot waste heat streams in industrial processes to generate power for in plant use. 13.1.2 Alaskan Perspective Far more commonly so than in the Lower 48, Alaska utilizes diesel engines for generating power in rural areas. . These engines vary in size from 3 kW units to serve a single remote home to village units of 50-200 kW. In addition there are several larger units with capacities in excess of 1000 kW serving larger communities. RBC engines could be used to decrease significantly the fuel consumption of many of these facilities, thereby decreasing both operating costs and reducing the need. to transport and store diesel fuels. -Such applications in Alaska might be particularly attractive for both technical and economic reasons. The availability of either cold air or cold water for condenser cooling both increases the efficiency and lowers the cost of the RBC engines. The high delivered cost of diesel fuel to rural 295 Cooling Water ' CONDENSER Cooling) (Liquid Heating) REGENERATOR BOILER EXPANDER Output Hot Gas or Liquid FIGURE 13.1 TYPICAL FLOW DIAGRAM FOR A RANKINE CYCLE ENGINE 296 areas makes the additional power generated by the RBC of particularly high value thereby decreasing the payback period of the additional capital investments. RBC engine systems can be designed over a wide power range and can be retrofitted onto existing diesel power units. .There are economies of scale associated with RBC engines which will tend to limit their practical application to capacities in excess of about 50 kW so that their use would be most commonly considered where the diesel capacity is in excess of about 250 kW. 13.1.3 Significance of the Technology The technology of RBC engines has been developed over the last twenty years by over a dozen firms in the U.S. and elsewhere. This has led to the installation of over a dozen units of 100-1000 kW capacity using diesel engine exhausts and industrial waste heat streams as energy sources. Over half these installations have involved partial funding by the U.S. Department of Energy in order to. accelerate the commercialization of this potentially important technology. The cost of RBC engine systems depends on a number of factors including operating temperature level, capacity, and working fluid. Equipment costs for 100-1000 kW systems are in the $1,500-3,000 per kW range when coupled with a primary diesel generator capable of several times the power output of the RBC operating on exhaust heat, the RBC's total cost is roughly the same as the diesel itself. Smaller ‘systems have been demonstrated but are not offered on a commercial basis primarily due to a lack of demand for in the Lower 48. 13.2 Description of Technology 13.2.1 Operating Principles In a Rankine cycle system, high pressure working fluid passes through a heat exchanger where it is vaporized ‘by a heat source. The vaporized working fluid is then expanded through a turbine to extract the shaft power output of the system. The low. pressure vapor leaving the turbine passes through a condenser and finally through a feed pump, which raises its pressure prior to re-entering the heat exchanger. ‘ For an RBC, these ' principles are illustrated in a Rankine. cycle engine diagram in Figure 13.1 and a typical pressure-enthalpy diagram shown in Figure 13.2. In the process: @ heat is transferred in a heat exchanger (referred to as a vaporizer or boiler) from the hot exhaust gas stream so that the working fluid vaporizes (process 1-2) at a high pressure level; 297 400 300 200 Pressure (PSIA) 100 FIGURE 13.2 ' ! ! | ! | 1 BOILER \ ' Liquid Pre Heat ! (Heat Input) ! U Le Exchanger — ‘ 6) with (3—4) @ LIQuiD Feed Pump Expander {Work Input) ; MIXED VAPOR (Work Output) AND LIQUID Condenser Vapor Cooling @)— Exchange 9 —e! i with (6-1) | (5) (Heat Removal) U -300 —200 —100 0 100 Enthalpy (Btu/Ib) TYPICAL PRESSURE-ENTHALPY DIAGRAM FOR A RANKINE CYCLE USING TOLUENE 298 rc m e the high pressure vapor is expanded, usually through a turbine, to obtain shaft power (process 2-3). The expander could also be a rotary screw or other positive displacement device. This shaft power is usually converted to electricity by a generator; e the hot -- but low pressure -- vapor at the exit of. the expander is passed through a heat exchanger (process 3-4) in order to preheat the working fluid entering the boiler (process 6-1). This heat exchanger is called.a regenerator, improves cycle efficiency with most organic fluids; e the vapor then passes ‘through a water or air cooled heat exchanger where it is condensed (process 4-5) back to the liquid state; and e a feed pump raises the liquid pressure (process 5-6) to force it through the regenerator (process 6-1) and back to the vaporizer (point 1) where the cycle is completed. ; Many of the components of organic Rankine cycle engines are standard commercial items. For. example, the condenser and regenerator are often shell-and-tube counterflow heat exchangers. The feed pump might be hydraulic fluid pump modified to account for the low viscosity of refrigeration fluids. _ , The key component within the system that is. of a unique nature is the expander, used to convert high-pressure vapor into mechanical work during its expansion from boiler to condenser pressure. For relatively large systems (in excess of 100 kW), turbines are generally the most appropriate expander type. However, for smaller units, positive displacement expanders have several advantages, including low operating speeds, high efficiency, and good _ part-load performance. The heat exchangers acting as boiler, regenerator, and condenser are by far the largest components, based. on physical size. This is indicated by Figure 13.3 which shows the layout of a 600 kW system. These heat exchangers will also tend to dominate commercially avail- able system costs, particularly in larger production quantities. The Rankine bottoming cycle engines are usually provided in a small number of pre-packaged subassemblies in order to minimize site installation requirements.. The core subassembly (referred to as the "Power Conversion Module" in Figure 13.3) consists of the turbine expander, feed pump, regenerator, and condenser unit. The boiler must be custom fitted to the application and depends on the -nature (temperature level, flow rate, cleanliness, etc.) of the hot waste heat stream being utilized. 299 00¢ DIVERTER VALVE HEAT DUCTING Source: Sundstrand Energy Systems. FIGURE 13.3 RANKINE BOTTOMING CYCLE INSTALLATION CONVERSION MODULE “pescmm Working Fluid Considerations . Several common refrigeration fluids, such as R-114 and R-113, are candidates for lower-temperature applications (up to 400°F), while water and organic fluids with high thermal stability, such as toluene, and F-85 (Fluorinol) are used for higher temperature applications. The characteristics of several working fluids which are now being used are summarized in Table 13.1. There are relatively modest variations in efficiency between working fluid options at any given temperature, and therefore, working fluid choice is usually determined by considerations such as the need for a regenerator, line size requirements, compatibility with expander configuration, and safety. At any given power and temperature level there is no unique choice of working fluid and manufacturers select working fluids. based in large part on their compatibility with their specific equipment capabil- ities. 13.2.2° Technical Characteristics The efficiency of the system depends on both the ideal thermodynamic efficiency of the cycle (assuming 100% component efficiencies and zero pressure loss in the heat exchangers) and the actual efficiencies of the major system components, such as the expander and feed pump, and the operating temperature levels. Figure. 13.4 indicates the effect of operating temperature levels in system efficiency assuming component efficiencies commonly attained in commercial equipment. As this figures indicate, system performance tends to improve as the peak cycle temperature increases and as the condenser temperature _ decreases. This latter point, in particular, favors the use of such systems in Alaska. For example, in Fairbanks, average ambient air temperatures range from -11°F in January to 54°F in August. As such, condenser temperatures of 80-120°F can be readily attained using air cooled condensers with low fan power parasitic losses. The property that efficiency improves as the boiler temperature in- creases is common to all thermal power systems. In the Rankine cycle, it manifests itself in the fact that as the temperature increases, the pressure in the boiler can be increased so that the pressure drop (and therefore the potential to extract power) across the turbine expander increases. In a diesel engine about 50-75% of the energy content of the fuel shows up as hot exhaust gases at temperatures of 700-1000°F, 20-30% as useful work, and the remainder as engine reject heat (either via a 301 Table 13.1 TYPICAL WORKING FLUIDS FOR RANKINE CYCLE ENGINES AND THEIR CHARACTERISTICS | , PROPERTIES SAFETY WORKING SUITABLE SOURCE CRITICAL CRITICAL MOLECULAR FLUID - TEMPERATURE °F TEMPERATURE PRESSURE WEIGHT FLAMMABILITY TOXICITY STABILITY °F psia - Water 450-1000 705 3208 18 None - None Stable Toluene 200-790 605 596 92 High Medium Stable up to 700° Fluorinol-85 200-700 . 440 715 100 ' Medium High . Stable up Ss : to 650° 8 Freon R-113 200-450 353 495 187 None Low Stable Freon R-12 150 234 600 121 None None Stable SOURCE: References 1, 2, and 3. cooling jacket or by radiation and convection with air cooled engines). As indicated in Figure 13.4, if these exhaust gases are used to heat a working fluid to 500-700°F, the RBC. engines can convert 20-30% of the energy content of the hot exhaust gases into useful shaft power. As a result, the RBC engines can provide 15-25% additional. power over and ‘above that being generated by the diesel power plant with no increase in fuel consumption. The output of a diesel power plant will usually vary significantly over the course of the day. This, in turn, will require the RBC engine to . vary accordingly since the volume of hot exhaust gases available for its operation are proportioned to diesel engine output. RBC engines have excellent part load characteristics with little degradation in efficiency, even at 50% of capacity. This is one reason RBC engines were considered to improve the efficiency of long haul trucks which have large (10 to 1) variations in power output requirements. The working fluid and critical mechanical components in a RBC engine operate within a closed cycle (similar to vapor: compression air conditioners). As such, they have the potential for high reliability and low maintenance operation. In fact, small (1-2 kW) flame fired Rankine cycle power units are used as power sources for the emergency valves on the Alaska pipeline due to the demonstrated high reliability of properly designed systems. Experience with large RBC engines for heat recovery applications is somewhat limited with most systems having been installed within the last four years. This limited experience does verify, however, the promise of these systems for high reliability with limited attention required by operating personnel. As indicated in Figure 13.3, the RBC engine requires routing the hot exhaust gases through a heat exchanger (vaporizer) with associated diverter valves and manifolds. In addition, the critical components of the RBC engine must. be complicated by the existing physical layout of the plant. For example, there may be no easy access to the hot diesel engine exhaust. In general, therefore, new installations, where space can be allocated for a possible future expansion are usually easier to retrofit at a later date. Because of the nature of these systems, safety considerations (other than those associated with any piece of rotating equipment) are mainly limited to working fluid properties. None of the organic working fluids is completely free of problems associated with flammability and toxicity. Most of the low cost fluids such as toluene are flammable, and may also be moderately toxic and carcinogenic. Fluorinol appears to be one of the safest of the high temperature working fluids since both toxicity and flammability levels are quite low. 303 Engine Efficiency Expander Efficiency Feed Pump Efficiency Condenser Temperatures 100 300 500 700 900 1100 Peak Operating Temperature OF Source: Arthur D. Little, Inc. FIGURE 13.4 RANKINE CYCLE ENGINE EFFICIENCY — VARIATION WITH PEAK OPERATING AND CONDENSER TEMPERATURE 304 1300 » rt The fluorocarbon refrigerant fluids being considered for organic Rankine cycle engine applications are non-flammable and have low toxicity. One of the more serious safety considerations with the refrigeration fluids could) be the products of their thermal decomposition. These products can include phosgene, which is highly toxic. This consideration may be a serious one in some waste heat recovery applications, particularly if the boiler is in a hot flue-gas stream. : 13.2.3 Environmental Issues The use of RBC engines will reduce the need to burn diesel fuels and to transport such fuels to the point of use. As a result, their use will tend to reduce air pollution resultant from diesel engine and transport system operation. Environmental concerns are principally due to working fluid considerations and effects associated with the condenser cooling system. With some refrigerant working fluids, atmospheric ozone layer .degradation due to possible release of working fluid gas is of concern if such engines come into widespread use on a worldwide basis. The RBC engines must reject heat to the environment either via an air cooled condenser or by means of a water cooled condenser. 13.2.4 Commercial Status Table 13.2 provides a list of manufacturers of prepackaged Rankine cycle engine systems. About four firms offer standard systems on a commercial basis while others can provide custom designed systems to fit specific applications. Equipment is available primarily in the power range from 300 kW to about 2 MW. Diesel engine manufacturers are presently not involved in the packaging of RBC engines. However, the RBC engine manufacturers, themselves, will. often assist in the design of the system package (interface heat exchangers, interface controls, etc.) as part of providing the RBC engine system. Presently about 12 systems in this power range are being operated as RBC engines in the U.S. Example applications include power for municipal power plants in Beloit, KA, Eaton, MI, and Homestead, FL (using hot diesel engine exhausts as the energy source), and for industrial waste heat recovery of a ceramic plant in Ferguson, KY, and at a California refinery. Most of the systems have been installed over the last four years or so. Thus operational experience is still somewhat limited. Experience to date, however, indicated that systems have generally operated within performance and reliability expectations. In addition to these larger systems, over 50 Rankine engine systems with capacities of -8-100 kW have been built for a wide range of solar and fired generator applications. Many of these systems would also be adaptable for use with diesel engine exhausts as the energy source if a demand for such units developed. 305 Table 13.2 SUPPLIERS OF RANKINE CYCLE ENGINE SUPPLIER WORKING FLUIDS POWER OUTPUT (kW) Thermo Electron Corporation Water, Flourinol 400-15, 000" Sunstrand Corporation Tolerance 600 Mechanical Technologies, Inc. Water, R-11 500-2 ,500 Whiting Corporation” MCB 300-600 Barber-Nichols Company Toluene, Isobutane, R-11 15-5007 1. Higher end of range with water only. 2. Division of Wheelabrator-Frye, Inc.; distribute units manufactured by ORMAT of Israel. 3. Primarily used to date in solar and geothermal applications; could be adapted for heat recovery functions. 306 The U.S. Department of Energy provided partial funding for seven of the systems installed to date and has, therefore, been a significant factor in accelerating the commercialization of these systems. There is, therefore, a significant technical base of experience with RBC engines which could be adapted over a wide power range to improve the efficiency of community and industrial size diesel generating plants in Alaska. : 13.3 Economic Implications 13.3.1 Costs The equipment costs for commercially ready RBC engines varies from $1,000 to $1,600 per kW in the power range from 300 kW to 1000 kW. Lower capacity units in the 100-300 kW range could be provided but at somewhat higher costs ($1,500-$2,000 per kW) since system costs associated with controls and electric switchgear are rather insensitive to capacity. The cost of installation varies significantly depending on such factors as the nature of the heat source and requirements for site preparation (provision of. support pads, expanding building space, etc.). Manufacturers indicate that they range from 25 to 50% of equipment costs in applications with ready access to the heat - source and higher in those retrofit applications where providing. space and access for the RBC equipment is complicated by the layout of existing facilities. It is difficult to estimate the maintenance cost because of limited commercial experience. However, considering the type of machinery employed, and based on projections by company representatives, it appears that maintenance costs will be in the order of 5-10% of the total system installed cost per year. Typical maintenance and repair items could include expander shaft seals, boiler cleaning, and replacement of control modules. In general, the diesel generating capacity in Alaska is distributed throughout the state in rather modest capacity increments of 100 kW or less to 1000 kW with a few’ locations such as Bethel, Nome, and Chitka having total capacities in excess of 2000 kW usually using a multiplicity of diesel generators. Where more than one diesel engine would operate at a given time, the exhaust gases can be manifolded to operate a single RBC engine. . In order to make the results applicable to a larger number of potential sites, the example here assumes the use of a 100 kW RBC engine which could be used wherever diesel capacity exceeds 500 kW. Engines of this capacity are lower than units in commercial operation today but could be produced using established technology. Because a unit of this size cannot benefit from available economies of scale, a cost of $2,000 per kW was assumed for equipment. 307 Table 13.3 CAPITAL COST SUMMARY TECHNOLOGY: Rankine Bottoming Cycle Engine BASIS: Location: Bush Diesel Power System Year: 1982 constant dollars CAPACITY: 100 kW ESTIMATED USEFUL LIFE: CONSTRUCTION PERIOD: CAPITAL COST: Equipment and Materials: Direct Labor: Indirect Costs: Contingency: TOTAL CAPITAL INVESTMENT 20 Years 6 Months to 1 Year $230,000 30,000 20,000 _20,000 $300,000 308 As indicated in Table 13.3, the total investment cost of this system . would be about $300,000 site preparation, installation, etc. - The relatively modest portion of the cost associated with installation labor results from the fact that the RBC engines are prepackaged units which help to reduce the level of on-site labor, a particular benefit in remote locations. Table 13.4 summarizes the operating costs and economics of the RBC assuming a capacity factor of 25%. The cost of energy from the system is estimated. to be about. 19.4¢/kWh which is competitive with the fuel cost of diesel generators © (18¢/kWh, assuming diesel/generator efficiencies of 25% and fuel costs of $1.80/gallon). The net cost of power generation from the combined diesel/RBC engine power system would, therefore, appear to have potential for lowering the use of diesel fuel in remote applications in a cost-effective manner. — 13.3.2 Socioeconomic Factors The introduction of RBC engines into a small. power generating facility significantly adds to overall thermal/mechanical complexity of the equipment mix. But it is probably reasonable to assume that existing operating and routine maintenance labor at these facilities will be trained in the required maintenance functions for the RBC engines so that no new personnel are required. : Major overhauls and equipment replacement for the RBC engines would require establishing an infrastructure within the state of specially trained personnel with the required spare parts inventories. The commercial introduction of RBC engines into Alaska will therefore, probably require manufacturers' identifying a minimum market size which justifies their presence in the state. The individual components within an RBC engine (heat exchangers, turbine expanders, etc.) would be -manufactured by _ existing industries in the Lower 48. However, with sufficient demand, it is possible that a local industry would package these components in systems tailored for the capacity ranges, climatic variations, and maintenance capabilities of rural Alaska. 13.4 Impact’ 13.4.1 Effect on Energy Supply and Use In 1981 electric power produced at the community level by diesel generators consumed in excess of 6 million gallons of diesel fuel annually. The introduction of RBC engines into Alaska could, therefore, have a significant effect in reducing overall consumption of diesel fuel in the state. Much of the diesel capacity is in the more 309 Table 13.4 OPERATING COST SUMMARY TECHNOLOGY : Rankine Bottoming Cycle Engine BASIS: Location: Municipal Diesel Power System Year: 1982 constant dollars CAPACITY: Input: Not Applicable Output: 100 kW OPERATING FACTOR: 25% OPERATING COSTS: Labor: $ 5,000 Maintenance Materials: 5,000 Taxes and Insurance: 4,000 Capital Charges: @ 9.5% 28,500 TOTAL ANNUAL OPERATING COSTS: $42,500 TOTAL ANNUAL OUTPUT: 219,000 kWh TOTAL LIFE CYCLE COST: $ .194/kWh 310 Ree uv remote villages where the costs of transporting and storing diesel fuels are particularly. onerous. The beneficial statewide benefits of using RBC. engines are, therefore, far in excess of the diesel fuel displaced. The combination of a diesel/RBC is one of the most efficient power plants which can be built and, in larger systems, efficiency could exceed 40%. This could improve the economics of on-site power generation systems and postpone the need for grid extensions. As such, if RBC engines become established as standard commercial practice, their use could impact on the optimum generating mix in the State of Alaska (and elsewhere). 13.4.2 Future Trends The commercial acceptance of RBC engines is only now beginning in the United States and . elsewhere. Present emphasis of most manufacturers is on larger systems (over 500 kW) for waste heat recovery in large industries and to supplement power at large diesel plants (in excess of 3 MW). ‘The industry is,. however, well positioned to provide RBC engines of lower capacities for use in conjunction with diesel generating plants smaller than 1 MW. This will increase the range of applications in areas with more modest sized dispersed generating capacity in Alaska, developing countries, and island economies. The ongoing demonstration program of DOE and industry financed pilot projects are providing information for improving the design and reliability of RBC engine systems. As a result, the technology of RBC engines is improving. However, the basic heat transfer and mechanical equipment technologies associated with RBC engines are quite well known and no major technical breakthroughs or cost reductions can be expected. : 311 10. 11. 12. REFERENCES Rankine Bottoming Cycle Safety Analysis, Vector Engineering, Inc., DOE/ET 15455, February 1980. Flammability Studies on Fluorinol 61 and Fluorinol 85, Hazards Research Corporation, September 1974, Environmental Readiness Document, Transportation Programs, Department of Energy, April 1980. "Marine Diesel Bottoming Cycle May Save Fuel," Automotive Engineering, July 1979. Saini, H., et ‘al., Combined Diesel-Organic Rankine Cycle Powerplant, 12th IECEC Conference, 1977. Rhinehart, H., Bottoming Cycle Gives More Power With Some Fuel, Public Power, July-August . Reinemer, V., Municipals Recover Waste Heat, Public Power, July-August 1977. Patel, P. and E. Doyle, Compounding The Truck Diesel Engine With An Organic Rankine Cycle System, Automotive Engineering congress, Detroit, Michigan, paper 761343, February 1976. Organic Rankine Cycle Engine Technology In Japan, Department of Energy, October 1979. Kresinski, J. et al., Predictin The Performance and Cost of ORC Waste Heat Recovery Systems, Argonne’ National Laboratories, 1981. "ORC's Converts Steel Mill Waste Heat To Electricity," Power Magazine, June 1982, p. 123. Product Literature - Thermo Electron Corporation - Sundstrand Corporation - Whiting Corporation - Mechanical Technologies, Inc. - Barber-Nichols, Inc. 312 14.0 STIRLING ENGINES 14.1 Introduction and Summary 14.1.1 Technical Overview The Stirling engine was first commercially introduced in 1818 by Robert Stirling and was based on a low pressure hot-air design.(1) A relatively large, inefficient, low power density. engine, it was overshadowed by the more efficient steam engine in the mid 1850's and later by the internal combustion engine. In the late 1930's, Philips Research Laboratories needed a small, quiet, reliable, external heat engine to power remote radio installation, and interest in the Stirling engine was rekindled. (2,3) Modern day, high power density, Stirling engines have been under development for over 40 years by organizations in Sweden, The Netherlands, and the United States.’ However, the interest level in the Stirling engine as a clean, efficient power converter has increased dramatically in. the last 10- 18 years as a result of rising world oil prices and concern for the environment. The Stirling engine is an external combustion, reciprocating, engine which converts thermal energy to mechanical energy by the alternate heating and cooling of the working gas contained in the engine. It has a number of potential advantages which have provided the incentive for these development programs, including: e multi-fuel capability which allows operation with a wide range of liquid and gaseous fossil fuels, as well as non-conventional or solid fuel heat inputs, such. as solar energy, coal, biomass, nuclear, and thermally stored energy; e high thermal efficiency which in theory can approach that of a Carnot engine, which is. the thermodynamic theoretical mechanism achievable; @ low emission levels in fuel fired applications because of the controllable external combustion process; this advantage is particularly important in vehicular and closed environment applications (mines, etc.); e low noise and vibration resulting from the use of mechanically balanced mechanisms, the absence of valves, and relatively low operating speeds; e high reliability and long life resulting from outwardly simple mechanical configurations and relatively fewer moving’ parts than other reciprocating heat engines; and 313 @ good part load and variable speed characteristics which are important for vehicular and some generator applications. Most of the above advantages have been demonstrated in operating hardware. For example, efficiency levels in excess of 35 percent have been achieved as part of the automotive program and engines have. been operated using solar, isotope, and thermal storage heat inputs. The primary funding for Stirling engine development in the United States has been for automotive applications due primarily to the engine's high efficiency and low emissions potential. However, applications requiring unconventional heat inputs such as biomass and solar energy are often proposed as promising options. For example, studies (4,5) indicate that Stirling engines combined with high concentration ratio parabolic dish concentrators are an attractive solar power option and that simple hot air, biomass fired Stirling engines are suitable for water pumping and electric power generation in rural areas of developing countries that have a good supply of biomass - materials. 14.1.2 Alaskan Perspective There are no commercially significant. applications of Stirling engines in Alaska at this time, due to the non-commercial stage of development of the engine. Two of the attributes of the Stirling engine which have the most potential benefit in Alaska are: e the ability to. use a variety of locally available fuel forms including coal, municipal: wastes, low-Btu gases, wood wastes, and peat as well as more conventional fossil fuels; and _ e the possibility of achieving very high efficiency (in the 40-48 percent range). ° Both these issues would tend to reduce the dependence of rural areas on expensive liquid fuels. 14.1.3 Significance of Technology The Stirling engine has demonstrated a number of potential operational advantages over more conventional engines such as the diesel or Otto cycle engines. However, current technology Stirling engines still have not demonstrated the life and reliability required to address the applications for which they are being most actively considered. It is not certain that the technical reasons for the lack of demonstrated reliability can be successfully addressed for many mass market or non-specialty applications, while still maintaining the 314 other required attributes. For example, high efficiency requires high operating temperature, which often dictates expensive materials and can adversely impact long-term reliability. Since the Stirling engine is not a commercially available engine per se, reliable cost data are not available. Projections for engine generator costs range from $600/kW for derated automotive engines made in large scales, up to $2,800/kW for robustly constructed, hot air, biomass fueled engines of the type which are most likely to find near-term use in Alaska. Future work on Stirling engine technology will likely proceed along parallel paths of continued development of both high and low power density engines and demonstration of life and reliability goals for critical components and engines. The future of the Stirling engine in Alaska will likely start with the small biomass fired units and progress to higher power density, derated automotive style units later. 14.2 Description of Technology 14.2.1 Operating Principles ‘ The basic principle of Stirling engine operation is illustrated in Figure 14.1. First, a cool volume of gas, entrapped by a piston, is compressed (Figure 14.1la) and then heated by an external heat source (Figure 14.1b). As the temperature of the gas increases, its pressure also increases and the. piston is driven downward to turn the crankshaft (i.e., to produce. mechanical work). After expansion (Figure 14.1¢c), the gas is cooled by an external cooling source (Figure 14.1d). Its pressure decreases and the cycle is repeated. Since the pressure during the hot expansion is much higher than during the cool compression, there is a net work output from the engine. The complete cycle takes place in one revolution of the crankshaft as opposed to two revolutions required by conventional four stroke combustion engines. Exchanging the heating and cooling .sources is a cumbersome process. Stirling replaced the alternating use of hot and cold sources by adding a displacer piston to move the working gas instead of the heat source. This mechanism allows the heating source to be stationary at one end of the cylinder and the cooling source to be stationary at the other end (Figure 14.2). A regenerator located between the fixed heating and cooling sources, enables efficient cycle performance by storing the heat that is otherwise wasted during the constant volume cooling process and returning it during the heating phase. Figure 14.2a shows the cooled gas being compressed by the power piston as in a conventional internal combustion engine. In Figure 14.2b, the compressed gas is 315 Source: Reference 6. FIGURE 14.1 SIMPLIFIED ILLUSTRATION OF THE STIRLING ENGINE CYCLE 316 HEAT INPUT REGENERATOR ISPLACER x : ISTON HEAT REJECTION Source: Reference 6. FIGURE 14.2 ILLUSTRATION OF THE IDEAL STIRLING ENGINE CYCLE WITH DISPLACER PISTON 317 ste PRESSURE CONSTANT TEMPERATURE HEAT ADDITION DURING EXPANSION HEAT EXTRACTED FROM REGENERATOR CONSTANT TEMPERATURE HEAT REJECTION DURING COMPRESSION VOLUME FIGURE 14.3 STIRLING ENGINE THERMODYNAMIC CYCLE HEAT ADDED TO REGENERATOR . i being heated and its pressure increased because the displacer piston is moving a portion of the gas into the upper or hot part of the displacer section. The pressure increase acts on the power piston. In Figure 14.2c, the hot, high-pressure gas, is expanding. Figure 14.2d shows the displacer piston moving upward to force the working gas through the regenerator where the gas pressure is reduced as its temperature is lowered. The power piston is now ready to repeat the compression stroke, thus completing the cycle. The ideal Stirling engine cycle, shown in Figure 14.3, consists of: e@ an isothermal (constant temperature) expansion during which heat is added to the cycle; e@ a decrease in temperature and pressure at constant volume which results from moving the gas from the hot zone to the cold zone, during which heat is transferred into a regenerator matrix; e an isothermal compression during which heat is rejected from the gas; and e@ an increase in temperature and pressure at constant volume which results from moving the gas from the cold zone to the hot zone, during which heat is transferred from the regenerator matrix back into the. gas. In an ideal Stirling cycle the efficiency is the same as the Carnot efficiency. In principle, therefore, it can achieve the highest efficiency possible for a heat engine. Several loss mechanisms reduce the efficiency of practical Stirling engines from the ideal,. these include: e@ mechanical losses in the pistons and the drive mechanisms, such as friction from seals, viscous losses is lubricants, etc.; e flow losses (pressure drops) resulting from the cyclic motion of the gas between the hot and ‘cold spaces and, in particular, in the regenerator matrix; ’ @ deviations from ideal isothermal compression and expansion processes; e conduction heat transfers between the hot and cold ends of the engine cylinders; e less than perfect regeneration of heat during the constant volume processes; and . 319 - Crank shaft -Air compressor Diagram of a crank drive as previously used in small hot-air engines. The forked connecting rod of the piston is driven directly by the crankshaft. The displacer connecting rod is actuated by means of a rocker arm, which is driven from a point on the piston connecting rod. Source: Reference 2. FIGURE 14.4 STIRLING ENGINE WITH CRANKSHAFT/ROCKER ARM DRIVE 320 @ non-ideal relative motion between the power and displacer pistons because of the use of real drive mechanisms. The following is a brief discussion of several important areas in Stirling engine design: temperature, configuration, sealing, and power control. Temperature As with any heat engine, the efficiency of a Stirling engine increases with increasing heat-input temperature levels. Most Stirling engines to date have been limited to working gas temperatures of about 1300°F, in part because of high working pressures (2000-3000 psi) required to achieve high power: densities for automotive propulsion applications. At these temperatures, engine efficiencies have typically been in the 30-35 percent range. If higher temperature alloys and/or ceramic heaters can allow operation at temperatures. of 1500-1800°F, efficiency levels of 42-48 percent. may be obtainable, assuming conservatively. that a. constant percentage of Carnot efficiency is obtained. In order to operate at high efficiency, Stirling engines require an efficient, high temperature, combustion/heat transfer system. Overall combustion system efficiencies of. 85 percent are normally required to- achieve high engine efficiencies. Heater head designs must be consistent with both the need to achieve high heat transfer rates and to minimize engine void volumes. These somewhat conflicting requirements have led to complex heater head. designs. The above set of operational characteristics impose severe material and stress problems in the combustion/heat transfer system, particu- larly during cyclic or non-steady state “operation in high power density systems. Configuration In an ideal Stirling cycle engine, the power piston and the displacer would move in a fixed relationship to one another (90° out of phase) in a series of discrete movements.. In practice, all engines use mechanisms which allow both the power and displacer pistons to move in a sinusoidal motion, which is intended to approximate the movement required of an ideal engine. In many of the earliest engines, this motion was developed by using simple crankshaft arrangements that put the motion of the power and displacer pistons 90° out of phase. A diagram of an early hot-air engine developed by Philips is shown in Figure 14.4, In this arrangement the power and displacer systems are contained. in a common cylinder and the relative motion between the two pistons is maintained by a crankshaft/rocker arm arrangement. 321 SINE WAVE MOTION DEVICE Source: Reference 6. FIGURE 14.5 DOUBLE-ACTING STIRLING CYCLE ENGINE 322 Po The type of configuration represented by Figure 14.4 has several drawbacks, including: e large physical volumes; @ unbalanced motion; and @ requirement for a pressurized crankcase. The rhombic drive was developed to be a compact and balanced drive system, and was the primary tool in early Stirling engine development work. It was still too large and complex for use in vehicular propulsion systems. A more recent Stirling engine development, the double-acting piston, results in an engine better suited for high power density (automotive) use. Philips has found it is possible to construct an engine of four separate interconnected cylinders, as shown in Figure 14.5, and to control the motion of the. pistons by a device which phases them at 90° intervals. With this type of construction, each piston serves as both a power piston and as a displacer for the adjacent cylinder, thus, the name "double acting". Arranging the four double-acting piston configurations into a cylinder forms a relatively compact engine in which the four separate heating sources are grouped into one common source. The four-cylinder engine can be perfectly balanced and has four torque impulses per revolution similar to an eight-cylinder internal combustion engine. Probably the best known double-acting engine is the 4-95 (formerly P-40) engine of United Stirling of Sweden which has been the basis of the NASA Automotive Stirling Engine Program.(7) A much more recent development in the Stirling engine field is the concept of the free piston Stirling engine. In this engine, the kinematic or mechanical linkages are replaced by carefully designed gas springs and working gas spaces. During operation, the pistons oscillate within their respective cylinders at their resonant frequency but without mechanical connection to one another. By careful selection of the mass of pistons, the "spring" constant of the gas springs, the resistance offered to gas flow in passages, regenerators, etc., one can adjust both the frequency and magnitude of resonant. oscillation as well as the phase angle between the two free pistons. Figure 14.6 illustrates the basic .configuration of the free piston engine, as well as its major components. The lack of a mechanical output form, such as a crankshaft, allows these configurations to be hermetically sealed. In designs built to date, either a linear alternator has been made a part of the power piston and energy taken out in the form of electricity or a hydraulic pump is attached to the power piston with diaphragms and hydraulic seals, to pressurize the fluid for energy extraction. Sealing There are basically two types of seals used in the kinematic Stirling engine; piston seals and shaft seals. 323 Piston Gas Spring & Gas ’ Bearing Compressor Bearings \ Monolithic Finned . Heater Head Alternator Armature Alternator Stator Displacer Displacer Gas Spring Source: MTI product literature. FIGURE 14.6 FREE PISTON STIRLING CYCLE ENGINE 324 = The piston seal generally takes the form of a series of piston rings that keep the working gas from passing by the piston, rather than going through the cooler, regenerator, and heater circuit as designed. These rings must operate without lubrication in a relatively warm part of the engine. Leakage past. these seals results in reduced engine performance and although they do not seal the entire working pressure of the engine from the atmosphere, they have been a source of considerable. design difficulty. Typical seal materials used to date have been engineering plastics such as Teflon, Rulon, etc. The shaft seal performs two functions: e sealing the high pressure gas (up to 2000 psi and above) inside the engine; and . e preventing the lubricating oil contained in the crankcase for the drive train from entering the engine and fouling the heater head, piston seals, regenerator, etc. The two approaches tested most to date are the Roll Sock seal of Philips (Figure 14.7a) and the sliding shaft seal used by both Philips and United Stirling (the Leningrader oil pumping ring seal, Figure 14.7b). Neither has reliably demonstrated the length of life required in stationary power generation or other long life application. It is important even in the relatively short lived automotive applica- tion that the goals for seal life be reliably achievable since, in present configurations, these seals are not easily replaced as in a, routine maintenance procedure. Two approaches to piston seals are being developed for free piston Stirling engines: sliding, and clearance: or gas gap seals. Seals in free piston engines are not required to seal against the same pressures as in a double acting kinematic engine, due to fundamental differences inthe configuration of the engines. However, the life requirements are similar as can be the temperature of operation. Additionally, changes in sliding seal friction due to seating or wearing-in of the seal can have a more’ pronounced affect on the performance of free piston engines than of kinematic engines, due to its sensitivity to slight changes in phase-angle between pistons. There are basically two types of clearance seals under investigation: hydrodynamic and hydrostatic. In the hydrodynamic bearing, gas is drawn into the gap or clearance by the viscous forces of the gas and the relative motion of the parts. This type has the advantage of not using any high pressure bleed gas and therefore having the potential for low leakage losses past the bearing. 325 KEGULATING VALVE (a) Rollsock Seal Subsystem (Philips Laboratories) (b) 041 Pumping Ring (Philips Laboratories) Source: Reference 2. FIGURE 14.7. | SHAFT SEAL CONFIGURATIONS 326 However, to result.in acceptably low gas~ leakage rates the gap between the pistons and cylinders must be on the order of .0001" which. implies machining tolerances of 0.00001". It has not yet been demonstrated that such: precision equipment can be made consistent with the cost constraints of consumer products. Test engines using clearance seals have been operated for periods of up to 10,000 hours and preliminary results suggest that the technical performance of the seals appear to be attractive, if practical cost.and thermal distortion issues can be resolved. : : Hydrostatic bearings work by supplying gas under pressure to the bearing. surfaces, whether or not the surfaces are in motion relative to one another. Tolerances can be ‘at least an order of magnitude less stringent than with the. hydrodynamic seals, but the losses associated with the use of. the pressurized gas can become large, if the clearance gap is excessive. Clearance gaps of 0.001-0.003 inches are currently under test in free piston Stirling engine systems and, based on initial test results, appear to be operating satisfactorily. Power Control Present kinematic Stirling engines use variable working gas pressure to control power output. at a constant operating temperature. This mode of control requires an auxiliary working gas storage volume, compressor, sensors, and check valves to implement. In parallel with the pressure control system, the combustion system operation must be controlled (variable blower speeds, fuel inputs, etc.) so as to maintain constant temperature operation at all loads. Free piston engines can be even more difficult to control since working gas pressure is not changed. Generally some form of load change on the power piston is used although other methods are being investigated. These relatively complex control arrangements require the simultaneous, reliable operation of many sensors and components which can often be difficult to achieve for extended periods. 14.2.2 Technical Characteristics Kinematic Engines Most of the following commentary on the performance characteristics of kinematic Stirling engines is based on the 4-95 engine (Figure 14.8) and its more recent derivatives, since this is the engine receiving the most development effort and for which information is in the public domain. A less well publicized but potentially important development for Alaska is the recent fabrication and initial testing of a small (5 kW) biomass fired, hot air Stirling engine by Sunpower Inc.(5) Most modern kinematic Stirling engine programs are concerned with achieving 327 Fuel-injector Turbulator Combustt Preheater ombustor -——————— Heater Cylinder P i i 2 Piston Regenerator Piston rod Piston rod seal Cross head Drive shatt Connecting rod Crank shaft Source: Reference 8. FIGURE 14.8 UNITED STIRLING OF SWEDEN 4-95 (P-40) ENGINE 328 7 maximum efficiency with minimum weight and volume (due to an emphasis on automotive type application). This generally dictates high temperatures, high gas pressures and the use of helium or hydrogen as the working gas. This, in turn, places emphasis on special (and often costly) materials and what might be referred to as a “high technology" design. Hot air Stirling engines, on the other hand, do not use a high pressure gas, and tend to be more robustly: built (for long life) from more available (and hence lower cost) materials. As a result, they are maintainable at a field level, do not use special seals and have lower heat flux requirements making them ‘suitable for use with biomass or other low heat density fuels. Their attributes are made at some sacrifice in peak efficiency; typical biomass hot air engine might operate at 20 percent efficiency. What follows is a list of characteristics often used to describe the operation of an engine. The following section will discuss each of ’ these characteristics as they apply to the Stirling engine. Efficiency Fuel requirements Emissions Noise and vibration Startability Load following capability Size and weight Operating life Maintenance schedule Reject heat Efficiency Stirling engine efficiencies of 30 percent have been consistently demonstrated on test ‘and demonstration engines for over 15 years. (3,9) More recent tests with the MOD-I engine and solar version of the 4-95 have demonstrated efficiency levels in excess of 35 percent (including burner and parasitic power losses).(10,11) Projections by NASA suggest potential for automotive engines with efficiencies in the low 40 percent range and in the high 40 percent range for certain stationary applications of automotive derived engines which do not have the losses associated with a combustor subsystem, i.e., Solar, nuclear, and thermal storage. The efficiency capabilities of Stirling engines are, therefore, well demonstrated, and further improvements might be expected, as higher temperature heater head materials are developed. 329 K Fuel Requirements % The fuel flexibility of the Stirling engine in dispersed energy system applications is matched only by the Rankine cycle engine alfernatives. However, simple Rankine cycle engines using water or organic ‘working fluids do not have nearly the efficiency of Stirling:.engine systems, in addition to having a range of other characteristics that make their use in dispersed energy systems less - attractive, particularly in Alaska's cold climate. The A.S.E. MOD-I engine has been operated on most liquid. motor fuels (gasoline, diesel, etc.) of near-term interest.(12) Earljer work on Stirling engine generator sets (GPU-3, developed: .at. GM) demonstrated the capability of Stirling engines to switch between a wide range of liquid fuels used in military field service.(13) Philips operated a biomass fired engine. using a heat pipe heat transfer system (14) and more recently, Sunpower, Inc. has operated a simple hot air Stirling engine directly fired by charcoal and rice husks. (5,15) mo The basic multi-fuel capability of a Stirling engine has, therefore, been extensively demonstrated, particularly when using liquid and gaseous commercial fuels. The ability of large, high temperature, high efficiency Stirling engines to utilize coal has been the subject of several studies funded through the Argonne National Laboratories. These studies did not identify insurmountable technical barriers to the operation of the Stirling engine in this fashion. However, no experimental work has been done to date, in order to resolve issues associated:.with high temperature, high power density coal fired systems (fouling, corrosion, slagging, etc.). : Emissions As discussed in Section 14.2.3, the Stirling engine has very, favorable emission characteristics. 3 « Noise and Vibration Stirling engines have no valves, can be fully balanced, gnd use a continuous .combustion process in their operation. As a regult, they have low operational noise levels and minimum vibration. These attributes have been demonstrated on several engines inckiding the MOD-I and the V160 engine of Stirling Power Systems (SPS). Both these engines demonstrated noise levels of less. than g70 dB--a comparable diesel can be over 90 dB (16) -- and very low mechanical vibration levels. This characteristic makes the use of such engines in applications requiring low noise levels (heat pumps, indoor vehicles, etc.) particularly attractive. 330 “e Ba oae Startability Stirling engines require a number of auxiliaries for starting. In the automotive application, the engine has two starter motors, one for auxiliaries including the blower, atomizing air compressor and _ hydraulic oil pump and another to operate the starter. The high energy flux of the combustion chamber/heater head subassembly is the source of considerable design difficulty. However, this high energy flux also results in allowing the engine to become operational very quickly from a cold start. Thus, there is no reason that the Stirling engine would be less satisfactory than diesel or Otto cycle internal combustion engines in terms of time to start and develop power. Tests on the automotive Stirling engine indicate the engine will be operational within 15 seconds at 80°F and 40 seconds at -20°F which compares well with an automotive diesel (from 20-60 seconds). (7) Part Load Operation In the Stirling engine, power output is proportional to the pressure and the swept volume of the engine. Therefore there are two common suggested methods of controlling the power output. One is to adjust the volume that is swept and the other is to adjust the mean pressure of the engine, usually in conjunction with a change in speed. Most Stirling engines to date are controlled by adjusting the gas pressure and rpm of the engine. These systems use a compressor and a storage bottle with check and control valves, in addition to short circuits to adjust the output of the engine. In most cases, the heater head temperature is maintained constant by varying the fuel flow to the combustor. Alternative drive mechanisms can be made to vary the displacement of the engine with variable stroke drives through the use of swashplates. These systems hold promise’ to improve the load following characteristics of the Stirling engine and decrease _ its complexity, but are not yet demonstrated in large sizes. The part load operation of an engine is particularly important for vehicular propulsion applications. It can also be important in other applications such as rural electricity generator sets where loads fluctuate daily and seasonally. Figure 14.9 shows a.curve of the efficiency of the MOD-I Stirling engine as a function of speed and pressure.(10) As indicated, the part load efficiency is excellent. Furthermore, the high efficiency is accompanied by very good torque characteristics. The part load characteristics of Figure 14.9 were accomplished mainly by changing the working gas pressure to change power at constant operating temperatures. Similarly good part load characteristics are expected to occur when using a variable displacement arrangement to vary output.(17,18) 331 Engine Efficiency, percent Mod | Stirling Engine System Data 50 Hydrogen Working Gas 720°C Heater Head Temperature 50°C Cooling Water Temperature 40 Mean Working Gas Pressure 15 MPa 30 10 0 1000 2000 3000 4000 Engine Speed, rpm Source: Reference 10. FIGURE 14.9 PART LOAD ENGINE EFFICIENCY CHARACTERISTICS 332 5000 System Size and Weight Stirling engines developed to date have usually been somewhat heavier and larger than conventional engines, particularly in applications utilizing light duty internal combustion engines. This is due, in part, to the fact that most engines developed thus far have been demonstration type engines. It would appear that the continuous combustion process described above will result in smaller engines, since the power is generated smoothly without rapid transients in the engine. In addition, high working gas pressures, high, continuous, heat flux rates, and short stroke, tend to make the size of the mechanical portion of a double acting Stirling engine (piston and drive train) smaller than for an internal combustion engine of similar capacity. For example, the. displacgment volume of the automotive Stirling engige is approximately 30 in’ (.38 in‘/hp) versus more than 100 in” (1 in"/hp) for a typical conventional engine in the same power range and application class (automotive). However, Stirling engines require a relatively large and complex external combustion system which is not is required by an internal combustion engine. This combustion system includes the blower, heater head heat exchanger, air preheater system, insulation and associated controls. As suggested by Figure 14.8, the combustion system is a major factor in determining engine size and weight. The addition of heat pipe systems in connection with the heater head design is not likely to decrease the volume, since at some point the volume will be limited by that surface area needed to input the required amount of heat to operate the engine based on the heat transfer coefficients on the flame side in the combustor system. (19) The heat pipe might, however, allow 4 different configuration to permit more effective space utilization. There are a number of drive configurations: which, if successful, could improve the specific power output for the Stirling engine. However, at this stage these are still design programs with no - working prototypes on which to base accurate estimates of engine size. It should be noted that the incremental size and weight of Stirling engines relative to conventional engines appears to be relatively modest and will not, in itself, be a major barrier for Stirling engine use in most of the applications under consideration. Life and Maintenance The Stirling engine has good potential for. low maintenance. The lubricating oil is not in contact at any time with combustion products, therefore eliminating one of the major maintenance items. There are no valves or complex, precise injection or ignition systems to need 333 maintenance. Potential reliability problems unusual to the Stirling engine include unlubricated piston seals, rod seals, the heater head, and the gas compressor systems, in engines that use that method of power control. Overhaul items include bearings, rings, seals, and heater heads in the kinematic systems. For these same reasons, Stirling engines have good potential for long life. However, this potential is as yet undemonstrated in kinematic engines. Although free piston engines have operated continuously for some period of time (about 500 hours) (15) very little long-term, total life testing has been performed on either design. This is probably due to the fact that the application for which ‘the Stirling engine is currently being considered most intensively is the automotive application, which has a relatively shorter life requirement than stationary engines. The main impediment to long life indicated so far in Stirling engine development programs is heater head design and seal life for both piston seals and rod seals But most engines built to date have been experimental in nature and, therefore, life and maintenance data is still quite limited. . Heat Recovery Approximately 85 percent of the waste heat in a Stirling engine must be rejected through a radiator (in air cooled systems) as compared to only about 60 percent in an internal combustion engine.(10) Since Stirling engine performance is more sensitive to cooling temperature than in the internal combustion engine, there is. incentive to operate the engine at as low a temperature as practical. There are two sources of heat from the Stirling engine: e Coolers: Stirling engines use coolers for heat rejection from the cycle during compression. These coolers should operate at as low a temperature as possible, typically from 80 to -100°F, , ‘ e Exhaust gases: Even with air preheaters, the exhaust gases from the Stirling engine combustor can reach temperatures of 400°F. A portion of this exhaust gas heat could be used to heat water to temperatures in excess of 300°F, thereby increasing the range of applications for this heat source. — Free Piston Stirling Engine Free piston Stirling engines either have demonstrated or could readily demonstrate many of the major operational advantages which have been verified on kinematic engines, namely: 334 high efficiency; low emissions; low noise and vibration; and multi fuel capability In fact, at low power levels (1-10 kW), free piston Stirling engines may have a higher efficiency than their kinematic counterparts, since there are no mechanical shaft seal losses. Without a mechanical drive train, the free piston machinery may be even quieter than kinematic equipment. The above attributes have been reasonably well demonstrated in equipment tested at Sunpower, General Electric and MTI. However, despite the superficial simplicity of a free piston Stirling engine, there are still difficulties in transferring this simplicity into mechanically reliable hardware. Although generally designed to be hermetically sealed to avoid the need for a shaft seal and contamination problems, free piston Stirling equipment still require a high temperature heater head, regenerators and piston seals for their operation Therefore, three of the critical design problem areas in kinematic engines are also present with free piston equipment. There is no fundamental difference in the heater head and regenerator requirements, between the kinematic and free piston designs. Therefore, the comments made earlier regarding heater head and regenerator design apply equally here. With free piston equipment, two approaches are being pursued: for sealing the working gas, sliding seals similar to those used in kinematic equipment; and clearance seals. The sliding seals in free piston equipment are subjected to a similar environment as in kinematic equipment and, therefore, success in developing long lived seals would be generally applicable to both kinds of equipment. Clearance seals are now used in free piston equipment (refrigeration cycles) developed for aircraft and space use. They have the advantage of resulting in no contacting surfaces and, therefore, engines using such seals should achieve long life operation. The absence of a shaft seal on a free piston engine is a significant reliability advantage. However, to date this advantage has been somewhat counteracted by problems in the complexities associated with extracting power from a free piston machine. With linear alternators, the side forces (perpendicular to the stroke) generated by imperfect geometry and windings can create loads that can cause large wear rates in sliding bearings. Clearance bearings are not quite as affected by this problem, but are correspondingly more costly and difficult to make, as well as sensitive to operate. 335 TABLE 14.1 SUMMARY OF EMISSION GOALS FOR PASSENGER AUTOMOBILES COMPARED WITH STIRLING ENGINE PERFORMANCE (gms/mile 336 HC co. NO x ' STIRLING ENGINES City 0.25 _ 3.21 0.84 Highway 0.004 0.40 0.61 FEDERAL GOALS 0.41 3.40 0.40 CURRENT AND PROPOSED FEDERAL EMISSION STANDARDS Precontrol-Actual 8.8 87.0 3.6 MANDATED 1980 0.41 7.0 2.0 1981 0.41 3.4 1.0 1982 0.41 3.4 1.0 POSSIBLE 1983 0.41 3.4 1.0 1985 0.41 3.4 1.0 1985 0.41 , 3.4 1.0 SOURCES: Environmental Protection Agency and Reference 22. PARTICULATES 0 .087 0.008 0.20 0.6 0.6 0.2 0.2 4 i As a result of the above factors, the reasons for shut down in free piston equipment has, to a great extent, paralleled that of kinematic equipment in: heater heads; piston seals (sliding or clearance) ; mechanical or electric power tranducer failures; “and miscellaneous instrumentation. Due to the limited amount of operational experience, there is not a statistical breakdown of these shutdown modes. 14.2.3 Environmental Issues One of the primary motivations for the continued development of the Stirling engine is environmental in nature, specifically: e. low exhaust emission characteristics; e very low noise; and e high efficiency, hereby maximizing utilization of scarce and/or expensive resources. Since the combustion process takes place outside the working volume of the Stirling engine, the process is more easily controlled and can be carried out with less violent transients and more steadily than in an internal combustion engine. This greatly simplifies control of the air/fuel mixtures and results in reduced emissions. The MOD-I engine has been extensively tested over automotive driving cycles and has resulted in emission levels as summarized in Table 14.1. These emission levels are well below those required of present and projected regulations and below those of competitive IC engines. Stirling engines have consistently demonstrated such low emission levels, in systems with good air/fuel ratio control and adequate combustor design, thus making it one of the lowest emission engines available. Since the engine has a continuous combustion process, a balanced mechanism, and no exhaust valves, it is a very quiet engine. It is expected that it will be operable in a number of environments that are not open to the diesel engine without substantial muffling of the exhaust and engine noises through insulated enclosures, etc. This facilitates the placement of engines close to inhabited areas also making the distribution of waste heat less difficult. Since the engine has the potential to operate efficiently it should make effective use of fuel resources, thereby holding down the need for fuel as well as reducing exhaust gas emissions. 337 1930 Philips Laboratories Starts Develop- ment of Remote Power Unit 1940 Air Cycle Machine @15% 7 Developed 1950 Stirling Refrigerator: Revived Interest in Power Cycles 1960 GM License Started 1970 1980 FIGURE 14.10 HISTORICAL DEVELOPMENT OF STIRLING TECHNOLOGY 338 © > United Stirling , . Sweden License MAN-MWM Licenses C6) < © > poe Li * © GM License Joint Program Y or oe Terminated 4 w/Beale MTI © MTI License © () | BEALE F PSE 14.2.4 Commercial Status The closed cycle, regenerative Stirling engine concept was initially developed by Robert Stirling in 1816. In the late 1800's and early 1900's several thousand hot air engines based on a simplified version of the Stirling cycle were manufactured.(3) These relatively heavy and inefficient engines could operate on coal, charcoal, or wood to produce power for water pumping and remote applications. Virtually all hot air Stirling engines were eventually replaced by more efficient and compact steam power systems. The modern era of Stirling engine development began in the mid 1930's at Philips in Holland.(1,2) The initial effort was to develop a low power .(0.2 kW) generator for powering radios in remote areas. Later developments emphasized vehicular propulsion systems. Over the ensuing 45 or more years, over 8 major companies have been involved in Stirling engine development: programs. However, as indicated in Figure 14.10, there have been only two primary developers of kinematic Stirling engines Philips and United Stirling. All other developments, including those in the United States, are the result of a complex series of license and joint venture agreements between the companies involved. The most extensive recent, well documented experience with kinematic Stirling engines has been achieved with those based on the 4-95 (P-40) engine designed by United Stirling of Sweden. Various modifications of this engine have evolved over a period of seven years. It is the basis of the MOD-I engine, being developed as part of the DOE/NASA Advanced Automotive Stirling Engine (ASE) Program, and a solar operated engine, which recently initiated testing as part of the JPL solar thermal/parabolic dish program.(5,6) This engine, shown in Figure 14.8, is a four-cylinder, double acting configuration which can produce 40 kW (55 hp) under design conditions. As of January, 1981, 21 engines of. this basic design had been built and over 13,000 hours of operation had _ been accumulated.(4) As indicated above, however, the accumulated hours of operation have usually entailed unplanned shutdowns and the reliability of critical subsystems (primarily seals and heater head) are not fully demonstrated. . The evolution of free piston Stirling engines started with the work of Dr. William Beale at the University of Ohio in about 1962. This early work .resulted in small scale prototype engines (fractional horsepower) which demonstrated basic operating principles of this form of engine. The potential advantages of this type of engine became more widely recognized in the early 1970's. This resulted in larger companies taking an interest in its development primarily for heat pump an solar applications, which were and still are of great interest to the gas industry and the DOE. As indicated in Figure 14.10, this resulted in 339 free piston Stirling engine programs being initiated at General Electric and Mechanical Technology, Inc. (MTI) in the 1972-75 time period based in large part on Beale's technology developments. As a practical matter, free piston Stirling engines have only been investigated for commercial applications over the last 8 years and total expenditures have been on the order of $10 million -- a very modest amount for a new engine development program. The cumulative operating time on free piston engines in a size range of practical interest is still less tan 5000 hours with the longest operation of a single engine being about 500 hours. This longer term operation has been accomplished both at Sunpower on a 1 kW electric machine and at General Electric on a 3 kW Stirling/linear compressor device for integration into a heat pump system using a conventional vapor compression cycle. Even though Stirling engines have been under development for over 40 years, it is estimated that less than 500 kinematic engines have been built. and fewer than 50 free piston engines have been built. As a result, ther are very few key personnel and organizations with extensive Stirling engine experience in developing more advanced engine systems. More detailed overviews of the history of Stirling engines can be found in References 1, 2, and 3. 14.3 Economic Implications 14.3.1 Costs Perhaps the most developed cost projections to date are for the automotive Stirling engine being developed by NASA (20) for the biomass fired, hot air power system. It is very likely that these costs will bracket actual engine costs for a number of categories and will be used as examples in the following discussion. A number of general studies have been performed in an attempt to quantify the cost of Stirling engine systems when produced in quantities sufficient for automotive use. They were conducted by both automobile manufacturers and independent companies that conventionally perform cost analyses for the automobile industry. All studies seem to agree that the automotive Stirling engine would be 25-100 percent more expensive than the same size internal combustion engine currently considered for automotive use, assuming the same production rates and power outputs. Therefore, most developers of Stirling engines have assumed that they would be premium cost engines and that this premium would be justified by their superior efficiency, emission, multi-fuel, and noise characteristics. The primary reason for the assumed incremental cost 340 of Stirling engines (as compared to internal combustion engines) is their requirement for an efficient, high temperature combustion/heat transfer system, larger radiator requirements, and possibly more’ complex controls. Unfortunately, there has been relatively limited detailed cost estimation of Stirling engines. The most up-to-date study was undertaken by Pioneer Engineering in support of the automotive program.(20) This study indicated that MOD-I Stirling engine would cost in 1984 roughly $1815 when produced in large numbers (300,000) as compared to a cost of $940 for a gasoline internal combustion engine including emission and catalytic equipment. Other studies on automotive (21) and _ stationary power (19) applications have also assumed a cost differential for Stirling engines as compared to the most likely alternatives. Engines designed for automotive use are generally designed with low first cost, high power density, and relatively short life in mind. Recent studies indicate that, for stationary power applications (where long life is important but high power density and low cost are not so critical), a reduction in operating speed and rather small reductions in operating pressure and temperature would result in an engine better suited to long life and low maintenance operation. ‘For the purposes of a lower bound cost estimate, it is assumed that the 60 kW MOD-I style engine is derated to 15 kW, but that the cost remains approximately the same for the basic engine, but including a generator. Thus, the derated automotive style engine might be expected to cost $120/kW or 5-8 times what a conventional spark ignition engine for automotive use costs, assuming the same production rates. This does not include fuel tank, generator mounting frame, enclosure, controls, batteries, etc., which add another $170/kKW to the cost of the system. Based on the above assumptions, the derated automotive-derived Stirling engine might be expected to cost $4300 in equipment, $2000 for installation labor, and $2300 for transportation and miscellaneous expenses, but not including distribution. This would be a total of $8600 for the system as described. Cost estimates for the hot air type Stirling engine are not quite as well publicized; however a recent study of Stirling engine applications addressed the issue of biomass driven Stirling engine costs. (22) Since fuel costs for such an engine are much lower than for diesel systems one can pay significantly more for the engine and still have lower delivered energy costs. Engine costs as high as $1800/kW still - result in attractive application of these engines. Table 14.2 is the capital cost estimate summary for the 4 kW biomass fired hot air Stirling engine. These costs are based on $1800/kW (including a generator) in Anchorage and manufacture of less than 10,000 units per year. The engine assembly cost of $8700 includes 341 TABLE 14.2 CAPITAL COST ESTIMATE SUMMARY TECHNOLOGY: Biomass Fired Hot Air Stirling Engine BASIS: Location: Bush Region/Individual Household Installation Year: 1982 Constant Dollars CAPACITY: Input: wood, wood chips, coal, peat, as well as conventional fossil fuel Output: 4kW of shaft power ESTIMATED USEFUL LIFE: 50,000 hours CONSTRUCTION PERIOD: 2-3 days . CAPITAL COST: Equipment and Materials: $8,700 Direct Labor: 1,000 Transport Costs: 500 Contingency: - 10% 1,000 TOTAL CAPITAL INVESTMENT: ‘$11,220 342 TABLE 14.3 OPERATING COST SUMMARY oo TECHNOLOGY : Biomass Fired Hot Air Stirling Engine _ BASIS: ; Location: Anchorage Year: 1982 CAPACITY: ‘ Input: Locally available solid fuels : Output: : 4 kW of shaft power OPERATING. FACTOR: 4000 hours/year, full load _ _ ANNUAL’ ANNUAL | OPERATING COSTS UNIT COST CONSUMPTION cost vt VARIABLE COSTS: i! Fuel-Wood: $30/Ton 18 Tons - $ 540 1 TOTAL VARIABLE COSTS: | 540 AL! . , . i FIXED COSTS: Maintenance: : _ 2,200 Capital Charges: @ 9.5% . 1,070 TOTAL FIXED COSTS: 3,270 || TOTAL ANNUAL OPERATING COSTS: $3,810 )< TOTAL ANNUAL OUTPUT: 16,000 kWh ‘| TOTAL LIFE CYCLE COST: $ 0.24/kWh | I \ Al | | uo 343 | | | ' q | \ | { the basic engine at $7200 (generator, starter, basic controls, etc.) and $1500 for accessories (small battery for. starting, control panel, charging circuit, etc.). Total installed engine cost is estimated to be $11,220. The primary benefit of the Stirling engine in rural areas is the ability to utilize a variety of fuels, even those in solid form. This reduces the burden of fuel transport and expense if local supply of fuels (such as wood, peat, coal) are suitable for consumption in the Stirling engine. Table 14.3 is an oferating cost summary for the example of a biomass fueled Stirling engine of 4 kW of output, assuming a locally available fuel is available. Under the specific conditions, the engine would consume from 12 to 18 cords of wood per year or 18 to 24 ton of dried peat. As indicated, energy cost. of $.24/kWh are achievable, compared to $.35/kWh with small diesel engine of the same size and approximately $2/gallon fuel costs.(22) This cost would fall commensurately with reduced engine costs, which would be brought about through increased production. 14.3.2 Socioeconomic Factors At this time, electric power generation from diesel engines is the only source of electric power in most remote villages in Alaska. However, fuel costs and distribution difficulties, as well as maintenance considerations indicate that if an engine were available that could burn locally available fuels and not requiring sophisticated . maintenance, it would find use in many of Alaska's remote settlements. The Stirling holds promise of being such an engine. Increased employment opportunities in Alaska from Stirling engine use would fall in the categories of installation of systems, maintenance and overhaul of the equipment and the gathering, preparation, and distribution of locally derived fuels. The use of locally available fuels for power generation would relieve the burden of transporting and storing fossil fuels, with increases in safety and self-reliance of individuals and communities. The ‘Stirling engine would also permit power generation where logistically it was impractical to run a diesel generator but local fuels are suitable for the use of a Stirling engine generator. : 14.4 Impact 14.4.1 Effect_on Overall Energy Supply and Use In 1981, approximately 36 million gallons of diesel fuel were consumed in Alaska for electric power generation in a total of 228,000 kW of diesel electric capacity. Of this total, about 60,000 kW is installed in 344 villages of 500 population or less. It is estimated that this group consumed 6.3 million gallons of fuel in 1981. If, through the introduction of Stirling engines run on local fuels, this consumption can be reduced by 20 percent, this’ would save approximately 1.2 million gallons of fuel valued at about 2 million dollars delivered. The introduction of a successful Stirling engine in rural areas of Alaska would clearly have a substantial beneficial impact, particularly if the reject heat from-the engine can be used to displace oil normally used for heating purposes. 14.4.2 Future Trends Future trends in Stirling engine development will likely include the development of both kinematic and free piston designs. The hot air, solid fuel fired engine discussed above will have the nearest term application in Alaska; however, larger, higher power density engines currently under development may also find final application in Alaska if current development programs are successful. 345 10. 11. 12. 13. REFERENCES Meijer, R.J. "Philips Stirling Engine Activities," International Automotive En gineering Congress, Detroit, Michigan, January 11-15, ’ arrendale, ennsylvania, Society of Automotive Engineers, 1965. Meijer, R.J., The Philips Stirling Engine, Eindhoven, The Netherlands, Philips Physical Laboratories, 1966. . Walker, G., Stirling Engines, Clarenson Press, Oxford, 1980. W. Percival and H. Nelwing, "First Phase Testing of Solar Thermal Engine at United Stirling", presented at Annual - Parabolic Dish Solar Thermal Program Review 1981, sponsored by _Jet Propulsion Labs. "Design of a Low Pressure Air Engine for Third World Use." Postma, N.D., et al., "The Stirling Engine for Passenger Car Application," preprint, Society of Automotive Engineers Meeting, June 18-22, 1973, SAE Preprint No. N 730648. Automotive Stirling Engine Development Program Technical Program, Technical Progress Report for Period: January 1 - March 29, 1980, MTI Report No. 80ASE E129 QT8, NASA CR-159851, June 1980. United Stirling Product Literature on P40 Engine. Kelm, G., et al.,"Test Results and Facility Description for a 40 Kilowatt Stirling Engine," NASA Lewis Research Center, NASA TM-82620, June 1981, Dowdy, M., “Automotive Stirling Engine Development Program - MOD-I Stirling Engine System Performance," presented at Automotive Technology - Development Contractor Coordination Meeting. MTI Report No. 81 ASE 224 PR 17, October 1981. "Stirling Power Conversion Unit Continues to Meet Goals," PDTS Highlights, Vol. 2, No. 2 by Jet Propulsion Laboratories for DOE, April 1982. Communique from NASA Lewis Research Center in Presentation to DOE. Percival, W., Historical Review of Stirling Engine Development in the United States from 1960-1970, EPA Contrast No. T= ES-00505~ NASA CR-121097. 346 14, 15. 16. 17. 18. 19, 20. 21. 22. Meijer, R. and C. ‘Spigt, "The Potential of the Philips Stirling Engine for Pollution Reduction and Energy Conservation." presented at Second Symposium on Low Pollution Power Systems Development, organized by Committee on the Challenges of - Modern Society of the NATO, November 1974. Personal communications with Sunpower, Inc. "V160 Stirling Engine...For a Total Energy System," Stirling Power Systems, presented at 5th International Symposium on Automotive Propulsion Systems, April 1980. Meijer, R. and B. Ziph, "Evaluation of the Potential of the Stirling Engine for Heavy Duty Application," Stirling Technology, Inc., NASA Contract No. NAS3-22226, May 1981. Vatsky, A. et _al., A Conceptual Study of the Potential for Automotive-Derived and Free-Piston Stirling Engines in 30-500 kW Stationary Power Applications, Mechanical Technology, Inc.,. MTI Report No. 81TR38, NASA Report No. NASA CR-165274, May 1981, Design and Development of Stirling Engines for Stationary Power Generation Applications in the 500-3000 Horsepower Range, Volume 1. Advanced Mechanical Technology, Inc. for Argonne Laboratories, Report No. 15207-T2, Volume 1, September 1980. Mather, K. and M. Siometokosku, Automotive Stirling Engine Development Program - Stirling Engine Manufacturability, Mechanical Technology Inc. MTI Report No. 81ASE222PR15, October 1981. Kaminiski, H., Manufacturing Cost Study of the 4-215 Stirling Engine, Pioneer Engineering and Manufacturing Company, Contract No. DEN 3-99, MOD-I, February 1980. t, Arthur D. Little, Inc., based on onging work. under NASA (Lewis Research Center) Contract No. DEN 3-254. 347 15.0 SMALL-SCALE HYDRO POWER 15.1 Introduction and Summary 15.1.1 Technical Overview A hydroelectric generating station converts the energy available from naturally flowing water into electrical energy. Since the water is continually replenished by natural precipitation and no fuel is required, hydro-based energy is a renewable resource. The water is harnessed by a dam or. diversion, which directs water into a conveyance system. such as a penstock or tunnel and then to a turbine where the energy in water is transferred to the mechanical rotation of a turbine. Coupling this rotation to an electric generator produces electricity. Once past the turbine, the water returns to the stream or river essentially unchanged. Since fuel is not required to generate electricity, as in a diesel or other fossil-fuel generating station, the hydroelectric generation of electricity is not affected. by rising fuel costs. Once the structures are constructed to contain and transport the water to the turbines and house the turbines and generators,; and the appurtenant equipment is installed, the only additional costs are for operation and maintenance. Yet even the operating and maintenance costs of a hydroelectric’ station are small compared to fossil fuel plants. Small-scale hydro, therefore, offers a cost-effective and environmentally. acceptable alternative to other sources of electricity generation. To encourage new hydroelectric power generation projects in the Untied States, a variety of federal programs have been instituted to provide incentives such as grants and loans for the development of operating facilities. Such programs have dealt with small-scale micro and mini hydro facilities, which characterize hydro installations by their capacities. This chapter focuses on such small hydro units with a maximum capacity of 1,000 kW (1 MW), which can supply the electricity peak demand of up to 200 homes. For present purposes, small-scale hydro plants are designated as micro (0.5-150 kW), mini. (150-700 kW), and small scale (700-1,000 kW) in order to cover three representative capacity ranges of less than 1 MW. . 15.1.2 Alaskan Perspective According to the National Hydropower Potential Study, Alaska has a large hydro potential of 33.5 million kW, but only 0.3% of this total is developed.(12) A large portion of the remaining 99.7% potential is attributed to large-scale hydro developments. The small-scale hydro (ess than 1 MW) of Alaska's real potential is not exactly known. It is unlikely that small-scale hydro will supply a significant proportion of Alaska's energy needs, but these systems may contributed to 349 energy independence, decentralization, and reserve security of remote areas. There are numerous small hydro sites in Alaska that could supply electricity throughout the year. Small hydro in these areas appear to be very practical. Since turbine manufacturers have always been responsive to market needs and have provided technical experience and training for unit servicing to site owners, competition is expected to provide favorable economic alternatives to site owners for turnkey projects. : 15.1.3 Significance of Technology Equipment manufacturers are beginning to standardize hydroelectric equipment and simplify civil. works, especially for plants with capacities less than 1 MW. The physical layout of a hydroelectric project, small or large, low head or high head, is a function. of the site's natural conditions such as topography and hydrology. All hydroelectric plants have three basic components: a water intake; a waterway to convey to the turbine; and the generating equipment. Construction of both impulse and reaction-type turbines is almost always technically feasible, but the energy characteristics of a site, the efficiency of the system, and the economics dictate’ its practicality. The seasonal variation of stream flow and the gross head determine the capacity (in kilowatts) and the energy generation over time (in kilowatt hours). The capital cost for a 1 MW smali-scale system would be in the vicinity of $3 million in a rural Alaskan location. With an allowance for operation and maintenance, the power cost would average less than 7¢/kWh, very attractive in comparison with conventional village diesels and other alternatives. 15.2 Description of Technology 15.2.1 Operating Principles A hydroelectric plant converts the energy available from ‘running water (low-head plant) or falling water (high-head plant) into mechanical and subsequently electrical energy. The physical layout is a function of the site's natural ee such as topography and hydrology. All hydroelectric plants have three basic components in common: (1) a water intake; (2) a waterway to convey to the turbine; and (3) the generating. equipment. Figure 15.1 illustrates these features for a low-head plant layout and Figure 15.2 illustrates these features for a high-head plant. The water intake may draw the water directly from the river or from behind a small dam. 350 Source: AB Botors-Nohals. FIGURE 15.1 ‘Thrust and guide bearing Control equipment Generator SMALL SCALE HYDRO PHYSICAL LAYOUT 351 solace iC PSTREAM 7 ms TIQK OR srRNe P: 7 ’ / S SYSTEM Poors / “ GENERATED ELECTR:CITY 7 FOR DISTRIBUTION SHUT- OFF VALVES GENERATOR TURBINE Source: Noyes (1980), Adapted from: Independent Power Developers’ brochure “Hydroelectric Power”. FIGURE 15.2 HIGH HEAD INSTALLATION 352 The waterways can be divided into two major categories: those with free surface flow and these with flow under pressure. A combination of these two cases is used quite frequently with the free surface flow waterway section usually preceding the under-pressure section. Recently, some successful attempts have been made to eliminate the intake and the waterways of a small-scale hydroelectric scheme. These new designs incorporate self-contained generating equipment, i.e., "packaged units," that can be installed directly. into the river, brook, or canal and produce electricity. The turbine is the basic mechanical. component of a hydro plant that converts the energy of the water into rotating mechanical energy. There exist two basic types of turbines: impulse and reaction. The energy delivered to both types of turbines results from the momentum of the water through a dynamic force exerted on the runner or rotating element. In an impulse turbine a free jet of water impinges on the revolving element of the machine that is called a "runner," which rotates freely (Figure 15.3). Impulse turbines are commonly known as a waterwheel or Pelton wheel, and are typically employed at high-head installations (above 60 meters). More appropriate to small-scale hydro are reaction turbines of which there are two basic types: Francis and propeller. For most Francis turbines (Figure 15.3), water enters a scroll case and moves to the runner through a series of guide vanes with contracting passages which convert water pressure head.into higher water velocity to drive the runner and shaft. The scroll case is a steel pressure vessel which distributes water from the penstock around tthe turbine runner. Guide vanes are steel wings placed in the: scroll case to direct the flow of water to the turbine runner, which is connected to a shaft which in turn is connected to the generator. The vanes in a Francis ‘turbine are usually adjustable, controlling the quantity and direction of the flowing water. The Francis turbine is normally mounted vertically, although horizontal axis machines are sometimes used. This type of turbine is normally. used in head ranges of 20 to. 100 meters. Common propeller turbines are designed to have an axis of rotation that is either horizontal, slightly inclined, or vertical (Figure 15.4). The usual runner has from four to eight blades mounted to a drive shaft bulb that is coupled to the generator. The blades and hub normally look very similar to a marine propeller. Several specific adaptations are common to the propeller turbine. A Deriaz type turbine is an adjustable-blade diagonal flow machine where flowing water is directed inwards as it passes through the blades. A Kaplan turbine is a propeller turbine with movable blades whose pitch can be adjusted. An axial-flow turbine-generator unit is set in line with flowing water. This type of unit may have the generator outside the 353 a SSSSSNV x C 4)) WW Ss SY Pelton Wheel KW Francis Turbine Source: Reference 4. FIGURE 15.3 TURBINES 354 \ <a VERTICAL PROPELLER — | if mio ~ i ° OPEN FLUME FRANCIS OR PROPELLER HORIZONTAL PELTON ~ I a Te | TUBE CROSSFLOW ({OSSBERGER) Source: Reference 34. FIGURE 15.4 VARIOUS TYPES OF HYDRO UNITS 355 Efficiency — Percent Curve A — Fixed-Blade Propeller Turbine Curve B — Kaplan Turbine, Biade Angle Adjusted Automatically in Coordination with Gate Opening and Load Curve C — Pelton Turbine Curve D — Francis Turbine 100 90 80 70 60 50 40 30 20 10 10 = 20 30 40 50 60 70 80 90 Output — Percent of Maximum Source: Reference 33. FIGURE 15.5 EFFICIENCY CHARACTERISTIC OF HYDRAULIC TURBINES 356 . 100 water conduit or inside the water conduit. The bulb turbine is one type where the generator is set inside the water conduit, while the inclined-axis machine is of the type where the generator is mounted outside the water conduit. Propeller turbines are normally used in head ranges of 3 to 50 meters. (32) 15.2.2 Technical Characteristics The quantity of water power in a stream is related to the flow measured in cubic meters per second, and the height of the fall of water, or head, from the intake to the generator. The selection of a turbine type is primarily dependent upon the design head and required output at that head. Manufacturers prepare sizing charts to assist in the determination, but overall economics are the final determinant, since the spread is quite wide between peak and base load needs of most villages (and households). All general types of turbines canbe effectively operated at low heads and small outputs, but their cost per kW of output will vary considerably. The efficiency characteristics of different types of hydraulic turbines are compared in Figure 15.5. The energy conversion efficiency of a hydroelectric power generating plant depends on several controlling factors but relates mainly to the efficiency of the turbine, generator, water conduit system, main electric transformer, and station service needs. Overall efficiency conversions ranging from 80% to 85% are typical of hydroelectric power installations. The annual power generation of a system is a function of the seasonal flow. In most areas of Alaska, seasonal variations of water flow reflect the melting of the snow pack. The season of high flow usually begins in March or April and peaks in August. Lowest flows are recorded in the cold months of December to February, although individual streams may vary depending on their origin. A small-scale hydro system should be ideally designed to meet the energy needs during periods of low flow. Hydro provides a continuous, yet limited source of power. Where it is the only source of power, a load management is especially important, since the total energy demand cannot exceed its availability at any given time. Battery storage may be utilized to cover peak demands with energy from off-peak periods. More often, diesel power is likely to be used for backup because batteries are costly and diesels are already in place in most villages. The turbine and generator components are usually preassembled in the manufacturer's shop to the extent possible and shipped to the project site where the components are fully assembled into operating units. Small packaged units can be shipped entirely shop-assembled ready for. operation (see availability section). 357 Geared type speed increasers are often provided for small size units to permit the use of higher speed and less costly generators. They can be used with either vertical, horizontal, or sloping shaft units. Large size water wheel driven generators and generators operating in isolated electrical systems must be alternating current synchronous type. Units over 5 MW are now usually provided with static excitation equipment, while the smaller units have shaft driven brushless exciters. Small hydro plants (under 5 MW capacity) operating in large electrical systems can generally use induction type generators, since voltage and frequency control are not necessary. Significant cost savings may be achieved due to the elimination of a standard governing system and the customary excitation equipment, plus a simplification of the electrical control system. In general an induction generator allows the use of an established utility as the storage system. ‘These units are fired by the current in the utility lines and therefore shut down during a power outage. Federal regulations that require utilities to buy power from alternate energy systems apply to Alaska only to Chugach Electric Co. because of a. required minimum size. No guidelines have been set for smaller utilities in Alaska. Therefore, it is important to approach the local utility and establish a payback rate before investing in a micro hydro system. Alaska's cold winter temperatures dictate extra precautions in the design and maintenance of a hydro system. Icing can be a problem at the intake pipe. Regular raking of the intake area is recommended. Icing in a buried penstock itself is usually not a problem as long as velocity is kept at the desired 1.5 meters per second. Turbines and generators should be housed in an insulated shed kept above freezing. It is recommended that the penstock be buried along its entire length to provide additional insulation during extreme cold air temperatures. In permafrost areas, pipes must be designed for the thermal expansion and contraction of surrounding soils. A qualified engineer should be consulted on the design of proper supports and anchors for the penstock. The sizing of pipes to match a particular flow and velocity can be done by a qualified dealer. The operating and maintenance costs of a hydro unit are usually small compared with the fuel fired alternatives. Depending upon plant design, size, and location, operating personnel may only be required for a few hours per month for inspection of control settings, equipment lubrication, and other minor tasks. Smaller machines can be maintained satisfactorily by a mechanic and an electronic technician: on an on-call basis in much the same way as, but less frequently than, for village diesel generators. 358 ween : : i 15.2.3 Environmental Issues Aside from placing the preassembled unit in the stream and running a power line from the unit to the user or village distribution system, mini and micro systems will have minimal environmental impact. Larger systems which require dams may be subject to licensing by the Federal Energy Regulatory. Commission (FERC), which requires an impact assessment, The major environmental considerations that apply to hydro units with dams that are around 1 MW in size are listed below. (22) . : "Fish and Marine Invertebrates The lakes, rivers, and streams support populations of anadromous and resident fish. The free passage and spawning of the fish must be preserved to maintain this resource. A Habitat Protection Permit would have to be obtained from the Alaska Department of Fish and Game in Ketchikan, Petersburg, Sitka, Juneau, Anchorage, or Fairbanks. , Wildlife The areas surrounding the lakes, rivers, and streams support a wide variety of mammals and birds. If the water level were raised, this would affect the banks which provide a source of food and habitat. Vegetation Vegetation is also inundated when the water level is raised. The effect can be twofold: root. rot. and physical inundation. Root rot is caused by raising the water table to the root zone, but in much of Alaska the water table is rather high and this impact would be minimal. Land Some acreage of land will be lost due to water inundation from the reservoir. This land area must be identified in the’ environmental impact statement required during the FERC licensing process." 15.2.4 Commercial Status and Availability The evolution of standard hydro packages has been rapid in recent years. There now exit a number of manufacturers who offer dependable packaged units of 0.5 to 1,000 kW capacity and with heads between 2.5 and 50 meters. Unit packages can include the turbine, generator, speed increaser, flywheel, governor or blade positioner, 359 intake water control valve, electronic equipment, and also battery storage as a power reservoir. Summarized information for selected manufacturers is presented below. AB Bofors-Nohab The Hydro Power Division of AB Bofors-Nohab, Trollhatten, Sweden, has developed standard tubular turbine designs which can be used ‘for either vertical or horizontal arrangements. These standard units have capacities ranging from 100-2,000 kW for heads ranging between 3-14 meters and runner diameter ranging between 700 and 2,000 millimeters. Units are manufactured in the U.S. by J. Leffel (see following sections). Allis-Chalmers Allis-Chalmers has developed axial flow "tubular" turbines and offers standard designs for three different basic equipment configurations. These comprise horizontal shaft TUBE units, vertical shaft axial flow turbines in the same size range as TUBE units, and low head inclined shaft axial flow turbines. All units are a propeller type and provide the highest feasible operating speed and maximum power output for a given runner size. These units are available in ten different turbine runner sizes, with outputs ranging from 50 kW to 5,000 kW. Barber Barber Hydraulic Turbine, Ltd., of Port Colborne Ontario, Canada has developed (under the sponsorship of the Canadian Government and Ontario Hydro) two small hydro generating station standard packages, designed to be transported by airplane to remote sites in Canada and installed with the minimum of labor. These units come from the factory complete in a steel housing, ready to be bolted in place on a field prepared foundation. Power output ranges between 100 and 600 kW, depending on the head. The unit has a synchronous generator driven through a gear box a approximately 1200 rpm. All pumps and governors are belt driven. Maximum turbine efficiency ranges from 85 to 95%, depending on the head at the particular site. The cost of this unit is approximately $250,000. Expected ‘turbine performance varies with the head. Escher Wyss Escher Wyss Limited of Zurich, Switzerland, famous manufacturer of large bulb units (capacities greater than 1 MW), also manufacturers right-angle bevel gear tubular turbines, complete with generator, voltage regulator, hydraulic governor, and control panel. These package units are available in four runner sizes and are suitable for heads up to 7 meters (23 ft.). Outputs range from 75 kW to 560 kW, depending upon the runner diameter and operating head. The 360 turbines are equipped with adjustable blade runners and wicket gates, thereby providing full flow control and a maximum efficiency of about 88 percent. : Gilkes The Gilbert Gilkes and Gordon Co., Kendall, England, is offering the HYDEC packaged unit consisting of a turbine and a belt- driven generator, which unit is categorized by its simplicity of construction and its low installation and maintenance costs. The unit provides outputs of 5 kW under heads. as low as 13 m up to 150 kW under a head to 100 meters, assuming the required flow is available. Hydro Energy Systems Hydro Energy System, Inc., of New York City and Societe Neyrpic of Grenoble, France, have recently joined forces to manufacture, sell and service standardized, packaged, low head hydroelectric units. These units consist of right-angle drive tubular turbines with the generators and auxiliaries located outside the water passage. They come in five standard sizes and are suitable for heads 1.8 to 18 meters. . Unit. outputs range from 100 to 1,500 kW, depending upon the runner size and operating head. The generator may be mounted directly above the turbine in a vertical position, or to one side in a horizontal position. The right angle is designed to provide a 3:1 speed ratio, thereby permitting the use of a higher speed and less costly generator. The overall unit efficiency ranges from 83. to 85 percent. IPD System The Independent Power Developers, Noxon, Montana, claim to have overcome the problems of excess cost and high environmental impact by generating DC electricity first. The DC power is stored in a battery bank which acts as a "power reservoir". When power demands fall below generator output the excess power is used to charge the batteries. As power is required it is passed through an inverter which translates the DC current into steady state 60 cycle 115 Volt AC current needed by most electrical appliances and ‘motors. This type of system does not require a dam and is capable of using all power produced. And, because electricity is stored, the system does not need to be oversized to meet peak power demands. All IPD hydroelectric power systems are designed to be owner installed and maintained. Since no dam is required, site. preparation is reduced to preparing a simple inlet arrangement, installing the inlet and outlet pipe and constructing a small enclosure to house the power plant. Hydraulic and electrical connections are standard and have been kept to a minimum to insure ease of installation. The systems come in high head. Small flow and low head large flow 361 models.. Both the high and low head models are grouped according to the following maximum output classes: peaking output 3 kW, 9 kW, and 12 kW continuous output 300 W, 700 W, 1,700 W and 8,500 W monthly power output. The systems include turbine, generator, batteries, inverter and all inner system electrical hookups. Leffel/Tampella James Leffel and Co., Springfield, Ohio, and Tampella of Sweden offer a number of packaged hydroelectric units (Hoppes units). providing outputs between 0.5 and 10 kW, for heads of 8 ft and 25 ft and discharges of 1.2 to 16.3 cubic feet/sec. For each power output, several turbine designs are offered, accommodating variations in head requirements and stream flows. In general, the packaged Leffel units employ propeller-type turbines with fixed blades and are connected directly. to the electric generator, doing away with the inconvenience, expense, and reduced efficiencies of using belts or gears. Leffel has recently undertaken a modernization program for these packaged units and has transistorized the generator and the electrical part of the packages. In addition to these units, Leffel offers its traditional line of Francis turbines known as Samson turbines, manufactured for more than 15 years. Improved vertical Samson turbines. can provide output between 13 and 1,500 kW depending upon the unit size and the available water quantities and head. Moteurs Leroy-Somer Moteurs Leroy-Somer of Angouleme, France, has developed designs for a mini hydraulic plant which is suitable for heads for 1 to 14 meters and outputs from 2 kW to 40 kW. The mini hydraulic plant consists essentially of two concentric cones, with the outer cone serving as a draft tube (diffuser) and the inner cone containing the turbine runner hub, planetary gear speed iincreaser, and asynchronous generator. Ossberger ‘The Ossberger-Turbinenfabrik Weissenbery, Bavaria, West Germany (represented in Canada) manufactures a hydroelectric package called the Hydro-light. It utilizes a turbine described as a radial impulse type, and is made in two sizes: the 3 kVa units for heads ranging from 4 to 6 meters and the 5 kVa unit for 6 to 9 meter heads of water. Their rating capacity is in the range of 100 kW. These units have been installed in many developing countries, but the only unit installed in the United States is located in Lancaster, New Hampshire, and has a rated output of 14 kW at 98 ft head. The overall efficiency of the Hydro-light system is about 83% over a broad-range of heads and flow rates. 362 ™. Schneider Dr. Daniel J. Schneider has recently designed and patented a unique hydraulic turbine termed the Schneider Lift Translator Engine. This turbine operates on a hydrofoil principle and presumably can be installed in canals, pipelines, rivers, and tidal estuaries to develop power from heads as low as 1.5 meters. A project under construction is located on an irrigation canal near the town of Richvale, California. This systems is rated at 75 kW at an effective head of 3 meters and consists of a Schneider engine, speed increaser, generator, accessory equipment, and a short 12 kV transmission line. Tampella AB/Leffel The Engineering Division of Tampella AB, which is located in Tampere, Finland, manufactures both vertical shaft and horizontal shaft tubular units. Although Tampella has ‘standardized main components for each type of.turbine, no fixed standard turbine sizes have been established to date. The horizontal shaft unit is designed for heads up to 20 meters and discharges up to 60 m3/s. Power outputs vary from 100 kW to 10,000 kW. The vertical shaft unit is designed for heads up to 25 meters and discharges up to 15 m3/s. Power outputs vary from 100 kW to 3,000 kW. The turbine may be equipped with either adjustable blades or fixed blades. Fixed guide vanes are normally supplied, although adjustable wicket gates can be provided when sites conditions warrant the additional cost involved. 15.3 Economic. Implications 15.3.1 Costs Itemized capital cost estimates account for hydro equipment (used and new), power plant, civil works, and other costs. . Used hydro equipment costs have been estimated to be about $300 kW in 1979,(25) but the availability of used turbines is limited. For new machinery, Montreal Engineering has derived a simple formula to express the costs of equipment (turbine, generator, and controls) .(15) C,,(1978) = 9 ,000(P)?° "aH Hy? 38 where C = cost in 1978 dollars, P = capacity (kW), and H = heat (m). This formula has been favorably reviewed by New York State and has been applied to a number of small-scale units. Based upon discussions with turbine manufacturers, the above figure should be increased by 30% for 1982. The Applied Physics Laboratory derived a similar formula for factory costs of units of 100-600 kW. (25) (1978) = 80,000(P)9*8(H)~0-88 363 Table 15.1 CAPITAL COST SUMMARY Technology: Small Scale Hydro { Basis: 7 Location: Bush Village Year: 1982 Constant . oa Capacity: a Input: Not Applicable . ; Output: 1,000 kW . t Estimated Useful Life: 30 years : os Construction Period: 3 years f Capital Cost (000): Equipment and Materials $1,400 Direct Labor 600 Indirect Costs 400 Home Office Costs 200 Contingency . | 400 TOTAL CAPITAL INVESTMENT (000) $3,000 364 Manufacturers are generally reluctant to quote costs unless a specific site is being investigated by an investor. It is expected, however, that small units (micro packages) of capacities 0.5-150 kW may cost between $30,000 and $60,000 each. Mini packages, or units of 200-700 kW, may cost between $100,000 and $300,000 each. (24) Power plant costs are very site specific, and there is little data for generating powerhouse and related equipment costs. Micro and mini units may not require a power plant. Small-scale units will require a powerhouse and associated equipment. The ratio of machine costs to power plant costs have been estimated from 1:2.17 to 1:1.46.(25) Civil works include dam, water intake, canal or penstock, road, powerline, etc. and can add:substantially to the above costs. There is little specific information in the literature on this cost category because it is so site dependent, but it is expected that civil work costs can be in the order of $5,000-20,000 for micro and mini units. For remote Alaskan locations, no general cost rules for civil works can be given; however, certain manufacturers produce packaged units that require less costly civil works than the other type of units. Engineering costs can vary from $10,000 for a micro hydro feasibility to $20,000 for a feasibility study of a 1 MW plant. A detailed engineering study and the construction management for the same 1 MW plant cost approximately $150,000 and $100,000, respectively. (24) Table 15.1 presents a capital investment estimate for a typical small-scale hydro plant (1 MW) which might be sited at a Bush location in Alaska. The total investment is estimated to be $3 million, or $3,000/kW. Based on the data collected annually by DOE/EIA for various hydroelectric facilities, generalized formulas, and 1978 Niagara-Mohawk operating statistics, a study by New York State concluded that operating costs for newly developed or newly renovated manually operated facilities will be about 3 mills/kWh, and 0.80 mills/kWh for automatic facilities.(23) However, costs may be as high as 4 mills for manual, and 1.2 mills for automatic small-scale facilities. Flat rate costs for small units are between $2,000 to $5,000 per year.(24) For larger units (less than 1 MW) O&M costs can vary between $5,000 to $20,000 per year depending on the turbine type and the local environmental conditions.(8,13) A figure of $40,000 was used in Table 15.2 for a remote Alaskan location. The resultant cost of power is very inexpensive by rural Alaskan standards. This suggests that hydro can compete favorably in most any location where it is available and accessible. The major constraint would be the large investment requirement of a few million dollars for a typical village. 365 Table 15.2. OPERATING COST SUMMARY Technology: Small Scale Hydro Basis: Location: Bush Village Year: 1982 Capacity (kW): Input: ‘ Not Applicable Output: Not Applicable OPERATING FACTOR: OPERATING cosTS UNIT COST FIXED COSTS: Operating & Maintenance: Taxes & Insurance: Capital Charges: TOTAL FIXED COSTS: TOTAL ANNUAL OPERATING COSTS: TOTAL ANNUAL OUTPUT: TOTAL LIFE CYCLE COST: 366 ANNUAL CONSUMPTION ANNUAL COST $ 40,000 30,000 . 285 ,000 335,000 5,256,000 kWh $0.068/kWh 15.3.2 Socioeconomic Factors Hydroelectric facilities with capacities of less than 1 MW will have a very minor impact on socioeconomic factors in Alaska. It is very unlikely that many new jobs or increased manufacturing activity will occur with the installation of more hydro facilities in Alaska, since many turbine manufacturers are already well established worldwide. If the level of activity in Alaska warranted, some manufacturers might establish service and distribution centers in the state which. could employ a few trained servicemen and other support personnel. 15.4 Impact 15.4.1 Effect on Overall Energy Supply and Use As of January 1, 1980, the national hydroelectric power capacity, developed and potentially developable amounted to about 173.9 million . kW, of which a large fraction, 33.5 million, is in Alaska. Only 0.3% of the total Alaskan hydro is currently developed. A large fraction of the remaining potential. would be large scale.(12) Alaska's small-scale hydro (less than 1 MW) potential is not exactly known, but it is not expected to supply a significant proportion of Alaska's energy needs. It may, however, contribute to the energy independence, decentralization, and reserve security of remote areas. A 100 kW micro hydro unit would be capable of producing enough power over the course of the year (at a 50% capacity factor) to displace about 50,000 gallons of diesel fuel from a typical village installation. If the village were fairly large and its base load power demand were near 100 kW, all of the power generated can be credited to fuel saving. In a smaller village, where off-peak demand were lower, some of the potential saving would be lost. The actual sizing of a hydro unit, therefore, would be importantly determined ‘by the projected demand profile of the village (or household). 15.4.2 Future Trends Small-scale hydro technology is well known and demonstrated. Improvements in recent years have been from better shop fabrication techniques, which have minimized the need for field installation and, hence, saved installation cost. Competition among the numerous suppliers worldwide is expected to keep costs down. The high delivered cost of fuel in most of Bush Alaska is the primary driving force for hydro deployment there. If petroleum prices resume their upward spiral, the trend toward hydro would continue. Further, small-scale hydro offers to play a role in the continued decentralization of rural Alaska, while maintaining a relatively secure source of energy. . 367 10. 11. 12. 18. 14, REFERENCES American Society of Civil Engineers, "Small Scale Hydropower," Lectures Series, Boston, Massachusetts, 1980. Alaska Power Authority, "Susitna Hydroelectric Project," Summary Report, Anchorage, Alaska, 1982. Alaska, State of, "Hydroelectric Commercialization Kit," Division of Energy & Power Development, Anchorage, Alaska, 1981. Alward, R., et al., "Micro-Hydro Power: Reviewing an Old Concept National Center for Appropriate Technology," Butte, Montana, 1980. . Personal communication, Baginski and Graham, Central Vermont Public Services Co., Rutland, Vermont, 1982. Brown, N.L., "Renewable Energy Resources and Rural Applications in the Developing World," AAAS __ Selected Symposium, Westview Press, Boulder, Colorado, T978. : BSCE,"Small Scale Hydro," notes of Lecture Series presented by the Boston Society of Civil Engineers," Boston, Massachusetts, 1981. Personal communication,- Dean Richard, James Leffel Company, Springfield, Ohio, 1982. DOE, "Fundamental Economic Issues in the Development of Small Scale Hydro," DOE/RA-~23-216.00.0-02, 1979. Engebretsen A.R., "Economic . Comparison of Five Electric Projects in Idaho," in Low-Head Hydro, publication by WRI, 1978, p. 72. "Hydro Turbine Manufacturers," Engineering News Record, p. 8 dune 2, 1977. . Federal Energy Reg. Commission, Washington, "Hydroelectric Power Resources of the United’ States," FERC-00700, 1980. Personal communication, James Fisher, Allis-Chalmers, York, Pennsylvania, 1982. Gillette R.W., "Small and Low-Head Hydro Plants," CH2M-Hill, Bellevue, Washington. 368 15. 16. 17. 18. 19. 20. 21. 22, 23. 24, 25. 26. a7. 28. 29. 30. 31. Gordon, J.L., "Small Hydro Sets Can Yield Competitive Energy," Energy International, 1978, p. 28. Personal communication, Ray Gorge, USGS-Alaska, Anchorage, ' Alaska, 1982. Personal communication, Scott Grundy, State of Alaska, Fairbanks, Alaska, 1982. Haroldsen R.O. & F.B. Simpson, “Micro Hydropower in the USA," ‘ EG & G Idaho, Inc., Idaho Falls, IA. Personal communication, Tony Meyer, Sheldon Jackson Junior College, Fitka, Alaska, 1978. : . Personal communication, Alfred Mohino, Barber Hydraulic Turbine, Ltd., Port Colborne, Ontario, Canada, 1982. Personal communication, Robert Mohn, State of Alaska, Power Authority, Anchorage, Alaska, 1982. National Center for Appropriate Technology, "Micro-Hydro Power, Reviewing and Old Concept," 1979. NYSERDA, "Estimates of the Costs of Renewable Energy Technologies for the NY State," Project # 114/ES-SOA/79, 1979. Personal communication, Constantine Papadakis, STS Consultants, Ann Arbor, Michigan, 1982. Pleatsikas, Christopher, "Estimates of the New York States," NYSERDA 79-7, 1979. Personal communication, Robert Reed, Great Northern Paper Co., Millinochet, Maine, 1982. Schneider, D.J., "Final Report Schneider Lift Translator System Low Head Hydro Feasibility Study," MIS-DOE/ET/01693, 1979. Susitna Hydro Studies (Jan 82/Apr 82/Jun 82). Strohmer F. & Walsh E., “Appropriate Technology for Small Turbines," Voest Alpine AG, Austria. USCOE, "Hydropower A National Energy Resource," Proceeding of March 11-16, 1979 National Conference, Fort Belvoir, VI. Verplank, W.K. Wayne, W.W., "Report on Turbogenerating Equipment for Low Head Hydro," MIS-IDO-1962-1, 1978. 369 32. 33. 34. Stone & Webster Eng. Corporation, W.W. Wayne, "Available Low-Head Hydro Technology," 1979. , Wayne, W., "Hydroelectric Equipment Selection," BSCES-ASCE, Boston, Massachusetts, 1980. Noyes, R., "Small and Micro Hydroelectric Power Plants, Noyes Data Corp., Park Ridge, N.J., 1980. 370 16.0 WIND 16.1. Introduction and Summary 16.1.1 Technical Overview Wind turbines (WT's) convert the kinetic energy of the wind into mechanical (shaft) energy which in turn can drive a generator to produce electricity. Four basic components of the system include a rotor/tower, a drive train including a gearbox and generator, a control device for overall machine control'and the control of electrical output, and a storage device if the system is operating independently from a utility network. In some applications a rectifier and/or inverter may be added depending on the type of generator or application. Blades on the rotor are connected to a hub which, in turn, rotates: a shaft to drive the electrical generator through a speed-up gear box. When the wind flows over the blades, a differential pressure and lift is created on the blades which result in rotation of the shaft. Thus, the kinetic energy of the wind is converted to mechanical energy. The major advantage of WT applications, especially in areas of Alaska where wind resources are available, is that the technology can be used to supply electricity in very remote areas where the cost of electricity is either extremely high or not even available. Large WT systems, of capacities greater than 100 kW, have seen limited applications in the Lower 48 to displace electricity loads on utility systems. Smaller scale systems can also be integrated with small diesel networks to reduce fuel usage and, in some applications, reduce the need for diesel capital equipment. Another advantage of WT's is that the energy required to run the system (the wind) is free. 16.1.2 Alaskan Perspective The most appropriate application of WT's in Alaska is in remote areas, especially in the Bush. Many populated areas such as Anchorage and Fairbanks have very high winds, but the cost of electricity in these areas are relatively low, thus making WT's comparatively uneconomical there. . In remote areas where there are no easily accessable electricity networks or where small diesel networks are used to supply power, WT's can be and are cost effective and practical. Table 16.1 provides a summary of the current factors affecting the present and future use of wind systems in Alaska. The favorable economic environment as well as government and industry support and a growing knowledge of the. operation of the WT systems will probably overshadow many of the unfavorable factors which might hamper WT development in Alaska. 371 ole TABLE 16.1 MAJOR FAVORABLE AND UNFAVORAB LE FACTORS IN THE FUTURE USE OF WIND SYSTEMS IN ALASKA Favorable Factors e Availability of loans e e Federal and State tax credits e e PURPA* to facilitate negotiating utility agreements in commercial sell-back applications e e 5-year tax depreciation period for commercial applications : e @ Government/industry support with demonstration programs e e Increasing fossil fuel and diesel costs . e e Availability of operational experience and data e@ Many remote applications mitigate concerns and reduce barriers to ready acceptance e@ Demand for cost-effective elec- tricity in remote areas *Public Utilities Regulatory Policies Act (1978) Unfavorable Factors Unfavorable weather conditions reduce performance of systems (i.e., ice, snow) High labor costs for installation and O&M on systems in remote areas Consumer lack of acceptance of the feasibility of the technology Reliability of power source and impact on local utility network systems Recent flat or depressed oil and gas prices High cost of capital (i.e., interest rates) Small-scale WT's (less than 20 kW) have seen application’ in Alaska ad ‘ the following reasons: e the diesel systems in remote areas are very small and can generally only accept up to 40 kW of power (approximately 40 to 50% of load) without. risk of damaging the diesel system; e@ there are many isolated residences or clusters of dwellings that do not have access to any type of network. Small WT's meet a portion of their energy needs very well; e larger systems often require the use of cranes for installation which are not readily available in remote areas of Alaska. Smaller-scale WT's can be _ installed’ efficiently and cost-effectively with a gin pole (an aluminum tower which uses a pulling system and winch or truck to lift sections of the tower); e small, dispersed energy sources. create a sense of energy self-sufficiency to isolate individuals and communities; and e the technical feasibility and operational experience of small-scale WT's in Alaska is better known than large-scale systems. The different climatic environment relative to the Lower 48, impacts the performance of the WT's. Aluminum blades which are typically used can ice up and become brittle during winter months. Blade materials such as sitka spruce or fiberglass must therefore be used to -add flexibility to the blades in order to prevent icing and reduced performance. Special attention must also be given to salt spray from coastal or island applications which can corrode the metal and reduce the lifetime of certain systems. In addition, the large amount of snowfall in Alaska may create cost problems related to maintenance and service and damage to WT parts. The installation of WT's in Alaska raises concerns related to safety, noise, aesthetics, environment and electromagnetic interference. Table 16.2 illustrates the reasons and level of concern for each of these issues, none of which has been serious enough to hamper the application of WT's in Alaska. Mitigation measures can be taken when necessary to minimize problem impacts posed by the WT installation. In spite of some of these apparent limitations for WT applications in Alaska, various estimates have been made that there are approximately 150 WT installations now in Alaska. It is expected that more systems above 50 kW will be installed in Alaska in the near future as the technology and operational experience with large-scale systems becomes more available. 373 PLE Concern Safety Noise Aesthetics Environment Electromagnetic Interference (EMI) TABLE 16.2 SUMMARY OF KEY CONCERNS WITH WT INSTALLATIONS IN ALASKA Reason for Concern Blades breaking off Damage to utility network and personnel Distracting to nearby populated areas Displeasing visual effect if in populated urban/ suburban areas or scenic/recreation areas Hazard to low flying birds Preclude certain land uses Interference with local television and micro- wave reception (only large WT's) Level of Concern Minor Moderate Minor for small WT's Moderate for large downwind HAWT rotors Minor to moderate Minor Minor for small machines Moderate for large machines Less severe for vertical axis WT's Mitigating Measures Employ fail-safe mechanical design or redundant safety systems Incorporation of proper relaying devices Site WT's in remote areas Nacelle upwind of tower for large WT's Careful attention to details of tower design Locate WT's to minimize visibility Locate away from sanc— tuaries and bird migra- tion routes Use of non-metallic blade materials Install machines in remote areas 16.1.3 Significance of the Technology Small-scale WT's (less than 20 kW) are technically feasible and very practical for many electrical applications and locations in Alaska. Larger-scale systems are not readily available through manufacturers, at the present time, but interest is growing throughout various communities. In many areas of the Bush where electricity prices are extremely high and there is great wind energy resource potential, WT's are very cost effective. As shown in Table 16.3, a major added cost for WT's in Alaska relative to the Lower 48 is for installation and operation and maintenance costs. Although these costs are higher due to climatic conditions and the remoteness of many of the WT sites, WT's are still a very practical and needed source of energy because electricity costs in many of these areas are extremely expensive. i In many applications, wind energy is the most promising renewable energy source in the near term. There are still many technical, legal, and institutional barriers to the widespread use of wind turbines of all sizes. However, there are also some strong incentives in terms of Federal policies reflected in energy tax credits and the Public Utilities Regulatory Policies Act (PURPA). The Alaska Division of Energy and Power Development (DEPD), the Federal Aviation Administration (F.A.A.) and the U.S. Department of Transportation are a few agencies which have also taken an active role in wind demonstration projects. These incentives, along with the growing knowledge being gained from demonstration projects and_ the widespread application of WT's will continue to promote and advance the development of this promising alternative energy technology in the state of Alaska. 16.2 Description of Technology 16.2.1 Operating Principles The sun is the primary energy source which produces the wind through the periodic and non-uniform heating of water and land masses which transfer heat to the atmosphere. Although the power in the wind is proportional to the cube of the wind velocity, only a fraction of the energy in the area swept by the blades of a wind turbine is extracted as mechanical energy. The theoretical limit of extractable energy, known as the Betz limit (14), is 59.3% of the wind energy in the swept area of the wind turbine rotor, and actual wind turbines extract even less.(2) Although there are a number of types of wind turbine generators (WT's) which have been employed or are in the concept or development stage, there are only two types of WT rotors which are commercially available and which have a sufficient amount of recent operating history to justify consideration for application in Alaska. 375 TABLE 16.3 SUMMARY OF AVERAGE SMALL-SCALE WT COSTS AND ENERGY PRODUCTION IN ALASKA Typical Range of “9 kW Installed Cost/kW Annual Energy Production (kWh*) . 1.8 $7,500 . 3,154 - 4,730 , 4 $5,900 . 7,008 - 10,512 oa 10 $3,100 17,520 - 26,280 a 25 $1,400 43,800 - 65,700 ° * Based on estimated annual average capacity factors of between 0.2 and 0,3. , ‘ 376 - These are the horizontal axis wind turbine (HAWT) and the Darrieus- type vertical axis wind turbine (VAWT), both of which are shown in Figure 16.1. Both types of machines rely on aerodynamic lift from the WT blades. Many of the older multi-bladed farm WT's, used for pumping water (Savonius rotor WT's) rely on drag forces acting on the blades. As a result, these types of machines are generally less efficient and, therefore, are not as widely used at the present time. Each type can be used to generate electricity, heat,-or merely pump water. Horizontal Axis Wind Turbines. The HAWT may have an upwind configuration, where the wind encounters the blades before it reaches the tower, or a downwind configuration where the wind first encounters the tower. Residential-scale WT's. (rated power output up to 8 kW) are usually of the upwind type, and may either use a tail vane to keep the blades pointed into the wind, or. an active yaw control to turn the blades out of the wind stream to protect the WT from extremely high winds. The upwind design minimizes cyclic blade loads and energy losses due ‘to the blocking of. the wind by the tower when the blades pass the tower, an effect which occurs with downwind machines. These loads . have been found to be greatly reduced by the use of thin cylindrical towers rather than rigid truss towers. The blades on both an upwind and downwind machine still, however, encounter gravity bending loads which can be high. The primary feature of HAWT's that has led to machine unreliability and shortened machine life has been excessive mechanical vibrations of key components. The vibrations are primarily the result of the resonant response of the drive train, tower, and blades to forced oscillations from the rotor and high speed shaft (via generator). WT designers have tried to insure that the machine does not operate for long at any critical resonance condition. However, as machines speed up and slow down, the rotor speed passes through some key. resonances. To minimize damage in these cases, the vibration modes are damped as much as possible, and speed is changed sufficiently fast to minimize the time that the machine resides in these critical conditions. Vertical Axis Wind Turbines. The VAWT responds equally well to a wind from any direction. The Darrieus. VAWT machine usually has two or three blades. It has low torque at low wind speed and is usually not self-starting. Present designs include either a separate electric motor to start the rotor when wind speed is above cut-in speed, or an induction generator which acts as a motor to start the rotor turning. Once up to speed, the motor begins to generate power. 377 Wind—— 8Lé (HAWT) a Generator FIGURE 16.1 Vertical Axis Wind Turbine (VAWT) TWO TYPES OF WIND TURBINE GENERATORS Rotor <a Blades Generator The comparative advantages and disadvantages between the HAWT and the VAWT have been discussed in detail.(3) Compared to a VAWT of the same power rating, the HAWT has the advantages of: smaller blade area; higher rotational speed leading to lighter weight gearing; self-starting capability; no shaft torque pulsations in a steady wind. In contrast, the VAWT has low tower costs because the generator and gear box are mounted at the base; lower foundation costs because guy wires, attached to the top of the VAWT (Figure 16.1) carry overturning moments; higher reliability due to absence of requirements for yaw alignment with wind and blade pitch control; absence of periodic gravitational loads which can fatigue blades. In machines that generate electricity, the HAWT or VAWT rotor is connected by a low speed shaft to a gearbox (speed increaser) which drives a generator, which may be one of the following three types: e The direct current generator which may be used for battery charging, resistive heating or feeding electricity directly into the utility network through a synchronous inverter. The DC generator is used with small WT's (of a few kW rating) and is not very commonplace. e The induction generator, generally used for WT's of less than 100 kW rated power, produces good quality 60 cycle power for utility network interconnection. However, they draw reactive power from the network which must be accommodated through the use of capacitors, or the network must supply the reactive power. e The synchronous generator, usually found in WT's with higher rated capacities, can operate in a stand alone mode and produce good quality 60 cycle power. They do not draw reactive power from the network. 16.2.2 Technical Characteristics and Design Considerations Wind resource potential is a primary factor that affects the performance of a WT. A resource study conducted by Battelle Pacific Northwest Laboratory indicates high winds in the following areas of Alaska (Figure 16.2): @ mountain summits and ridge crests of the Alaska Range, Brooks Range, Chugach Mountains, Wrangell Mountains, St. Elias Mountains, and Coast Mountains; e coastal areas of Beaufort Sea, Cape Lisburne, Wales, Nome, Chukchi; e islands in areas such as Cape Romanzof, Cape Newenham; 379 Classes of Wind Power Density at 10 m and 50 m(a) 10 m (33 ft) 50 m (164 ft) rr Wind Wind Power Wind Power Power" Density, Speed, (b) Density, Speed, (b) Class _watts/m2__m/s (mph) watts/m2__m/s (mph) Qo 0. 0. 0. 1 100-—— 4.4 ( 9.8) 200-——S.6 (12.5) 2 150—S.1 (11.5) 300 —— 6.4 (14.3) 3 200 ——5.6 (12.5) 400——7.0 (15.7) 4 250-——6.0 (13.4) 500-——7.5 (16.8) 5 300 ——6.4 (14.3) 600 ——8.0 (17.9) 6 400-———7.0 (15.7. 800 ——8.8 (19.7) 7 1000-——9.4 (21.1) 2000 ——11.9 (26.6) (4)Vertical extrapolation of wind speed based on she 1/7 power law. * (>)tean wind specd is based on Rayleigh speed distribution of equiva- lent mean wind power density. Wind spced is for standard sea-level Conditions, To maintain the same power density, speed increases 5%/5000 It (3%/1000 m) af elevation, (—) RIDGE CREST ESTIMATES 08€ 23010 _ x0 mitts O_50 100 150 KILOMETERS —— PNt-3195 WERA-10 20 160 1800 16° Source: Reference 27. FIGURE 16.2 ANNUAL AVERAGE WIND POWER IN ALASKA e Middleton Island, Kodiak Island, Bering Island, the Aleutian Islands; and e Lower Cook Inlet. Although the study indicates that generally favorable wind conditions for many WT applications in Alaska, more detailed analysis is still needed of wind resource data for specific community applications. The instantaneous energy production of small WT's can be estimated from a typical power curve such as Figure 16.3. As illustrated, most small HAWT's typically reach their rated capacities at wind speeds of approximately 25 mph. Annual energy production is determined by the. WT selected and the wind region at the site chosen. The wind resource will typically have an year-to-year variability of approximately 210% about the long term mean value. Most properly sited small WT's will have an annual average capacity factor of approximately 0.2 to 0.3, (25) assuming a 90% machine availability when the wind is available. The capacity factor, based on rated power output is about 0.2 in: areas where average annual wind speeds are approximately 12 mph and about 0.3 for areas with annual average wind speeds approximately 14 mph. The power output from WT's can be very unreliable due to the intermittent nature of the wind. For remote applications, where WT's are most appropriate in Alaska, batteries are often considered to store electricity until it is needed. Rechargeable lead acid batteries can become expensive, costing approximately $80 per kWh of storage capacity. Wind/diesel hybrid systems are, therefore, often more practical and economical. Most diesel networks in Alaska are very small, however, and can not accept more than 30 to 40 kW without significantly reducing the reliability of the diesel system. Studies by the DAF-Indal Company in Canada have shown that "the value of the wind energy is greatly improved if the diesel generator system itself is made to perform more efficiently at part load, with or without power assistance from the wind turbine. (8) As a general rule of thumb, the installed capacity from WT's that are integrated to a utility system should not exceed 10% of the network capacity to maintain control of voltage and frequency within an acceptable range. This figure may vary slightly depending on the size and type of utility network and the other sources of generation. This limitation is not a major concern to most utilities at the present time because the installed capacities of WT's integrated with the utility network are very small. The very wind resource that WT's ‘harness can also pose hazards to system integrity and life. These and other natural occurrences described below must be factored into a reliable WT design. 381 Z8é Non-Dimensional WT Output rated ala! 1.0 0.5 10 Rated Wind Speed ie 20 Hub Height Wind Speed (mph) Source: Arthur D. Little, Inc. FIGURE 16.3 TYPICAL POWER CURVE FOR A SMALL HAWT Wind Turbulence - Air turbulence consists of rapid changes in speed and/or direction of the wind. The turbulence most harmful to small WT's is the small-scale, rapid fluctuations often caused by the wind flowing over a rough surface of a barrier. The primary disadvantage of turbulence is that it induces cyclic vibrations and unequal loadings on the machinery that may eventually weaken and damage it. Wind Shear - Strong wind shear may also pose a hazard to WT's in some locations, The predominant effect is in the vertical direction and can lead to cyclic blade bending and cyclic power fluctuations in horizontal axis machines. But this is not a serious concern for small WT's. High Winds - Blades and supporting towers are both susceptible to damage from high winds. The blades become vulnerable if the protection systems designed into many machines (e.g., blade feathering system) fail.in extreme winds. Towers must be capable of supporting the small WT's in all wind speeds which normally occur in the local area. Most machines are designed to withstand a 120 mph peak wind velocity. : Ice - Ice accumulated on blades, towers, and transmission lines can cause hazards or reduce the efficiency of wind machines. There are two types of icing: rime ice and glaze ice. Rime ice is drier, less dense, and therefore less hazardous than glaze ice; however, it can, over .a period of time, build up large accumulations. Glaze icing, under certain conditions, can quickly accumulate on cold surfaces to thicknesses of several inches. This may not only reduce blade aerodynamic efficiency, but it also constitutes a severe safety hazard if suddenly released from a rotating blade. On large machines, such a problem may necessitate safety fences around all ice-susceptible WT installations or an ice detector that limits machine operation when ice builds up on the blades. Snow - Snow creates three types of difficulties for wind machines: @ maintenance and service may be difficult due to excessive snow depths; e heavy snowfalls may damage parts of the turbine or make restart difficult; and e blowing snow may infiltrate the machine and cause breakage from freezing and thawing. This can occur as easily with ice. Periods of inoperability due to large snowfalls or any of the above factors may lead to a lower utilization factor. Salt Spray - From’ the point of view of wind resource availability, many of the best sites for WT's in Alaska are in coastal or island locations. Salt spray at such sites may, however, reduces the life- 383 time of certain WT's, thereby sharply increasing the cost of wind energy to the user. Machine designs that fail to account for this hostile environment are unlikely to enjoy user acceptance in coastal regions. : 16.2.3 Environmental Issues The literature on wind generation generally supports the conclusion that small WT's have minimal environmental effects. Large WT's are, ‘however, felt to pose some potentially significant problems, but in most cases these problems can be alleviated by proper machine design and siting procedures. Although these questions have to be considered in the environmental impact and safety evaluation process, they are likely to be less significant than for those associated with conventional energy power plants. Safety - If WT's of any size are. located in populated areas, they could pose a threat to persons and property from (1) blades breaking off, (2) the support tower of an horizontal axis wind turbine falling down, or (3) the whole structure of a vertical axis wind turbine toppling over. Also, the height of larger WT's could pose a hazard to low flying aircraft. Design and operating procedures generally cover this concern very well. Another major concern is the possibility of the machine not tripping off-line when the line goes dead or in the presence of a fault. An induction generator of any size, without capacitors for reactive power compensation, will trip off if the power line is dead. A synchronous generator or compensated induction generator, however, could continue to provide power (back to a fault or even a substation) unless adequate protective relaying measures are followed.(9,10) Design and relaying (i.e., circuit breaker) procedures should provide a level of safety equal to that of other large power equipment. Noise - Small WT's do not pose a noise problem in general. If large WI's are located close to private homes the noise may be disturbing. However, the noise should be below objectionable levels for large horizontal axis wind turbines (HAWT's) that have their nacelle upwind of the tower. Thus far, no noise problems have been identified with _ vertical axis wind turbines (VAWT's) of the Darrieus design. Aesthetics - Small WT's are not felt to pose a serious or widespread aesthetics problem unless installed in rather densely populated urban/suburban areas or especially in scenic or recreation areas. However large wind turbines can create a generally more displeasing visual effect. Although a handful of wind turbines dotting the landscape may. be acceptable because of their novelty, public acceptance may wane as their numbers increase, especially in scenic or recreation areas. Careful attention to the detail of tower design can mitigate these adverse aesthetic effects. For example, the use of 384 the cylindrical tower on a current DOE funded 2.5 MW wind turbine design is felt to be vastly more aesthetically pleasing than that of the rigid truss towers used in earlier machine designs. (7) Wildlife - It has been said that WT's may be a hazard to insects. and . Tow flying birds. There have also been reports of their low frequency sound disturbing both wildlife and domestic animals. Test data thus far have not verified the presence of any problem of this nature associated with either large or small WT's. Electromagnetic _Interference (EMI) ~ Wind turbine blades create electromagnetic reflections that interfere with local television and microwave reception. This is generally not a problem for small, residential-scale WT's unless the rotor is in the direct path of the radio waves and the signals are relatively weak. ‘Use of non-metallic materials such as fiberglass and wood are generally felt to reduce the effect greatly. However, lightning protection is presently installed in the blades of large WT's, and requires some conductive materials to carry the lightning path to ground. Tests to date indicate that this protection system contributes to a continuance of the reflection problems. The geographic extent of the problem has been found to be a function. of the size of the wind turbine. For example, small WT's give rise to only very localized problems. Further, there are indications that vertical axis wind turbines do not produce microwave interference to the same degree as horizontal axis wind turbines. (1) Land Use - The installation of large WT's in certain areas preclude certain land uses. Generally, such wind machines would be sited in remote areas while small WT's would be installed in suburban, farm, and remote areas. Height and site approval restrictions may impede wind turbine use in suburban communities in the short term (1981-1985), but are expected to be changed in the distant future. If the remote land is in wilderness or restricted development areas, the land use effects would be small, although permit problems might be ‘great. In developed areas, however, WT's would preclude the use of the land for housing or commercial development, but might not preclude cattle ranching, logging, agriculture or similar activities. There are some potential legal problems associated with the use of wind. turbines in special cases, e.g., wind rights on Federal and state lands. These problems do not appear to be insurmountable and are expected to change over the next decade. 16.2.4 Commercial Status Summary Status of Small Wind Turbines Department of Energy (DOE) support for the development, testing and commercialization of small wind systems (machines with an output less than 100 kW) has been provided by the Rocky Flats Wind System Program Office for both HAWT and .VAWT's. This program is 385 intended to promote the earliest possible development of WT's designed primarily for farm, home and industrial use. In addition, approximately 60 ‘private manufacturers of small WT's have been selling, installing, and servicing their machines for several years. Based on a small WI-market study by Arthur D. Little, Inc., in 1981, it was estimated that annual sales for small WT's total in the vicinity of 1,000 to 1,500 wunits.(6,7) The industry was apparently flourishing, taking full advantage of Federal and state tax credits and other incentives. At about the same time it was publicly estimated by the president of Enertech Corp., a manufacturer of 1:8 kW machines, that, in 1981, the small WT industry would be a 10 to 20 million dollar a year business with 10 to 12 MW of capacity installed that year. (13) But in the past couple of years the industry has not experienced the expected growth in sales. Sales have dropped due to the recent oil glut, reduced Federal support for alternative energy systems, and high interest rates. For example, before August 1982, there was a Federal tax credit available to the commercial sector for the purchase of WT's. But the Tax Equity and Fiscal Responsibility Act of 1982 has since reduced the combined total of the Business Energy Tax Credit and the Investment Tax Credit has been reduced from 25% to 22%. These unexpected factors have had a significant negative impact on several small wind companies in the past year. Summary Status of Large Wind Turbines There has been a great deal of research activity with large-scale systems in the Lower 48. However, the potential application of these larger-scale systems in Alaska is not often practical due to environ- mental constraints and the fact that many potential users are dis- persed or isolated. The National Aeronautics and Space Administration, Lewis Research Center, Cleveland, Ohio, on behalf of DOE provides overall management for the large horizontal axis wind turbine development and test program. At present, their programs involve several HAWT's including a 100 kW, four 200 kW, one 1.5 WT, and a cluster of three 2.5 MW WT's. Programs of Water and Power Resources Services of the Department of Interior involve three units with power - ratings from 4 - 7.3 MW. The design and development of four 100 kW vertical axis wind turbines is also being conducted at Sandia National Laboratories in Albuquerque, New Mexico. Most of the tests will be conducted with the WT's interconnected to local electric utilities. Privately funded WT development and test programs of significant interest to electric utilities are being conducted by two major companies. WTG Energy Systems, Inc., of Buffalo, New York has four operational 200-kW HAWT's and a fifth unit has been sold to the Central Electricity Generating Board in Great Britain. FloWind, Inc., of Kent, Washington, is pursuing the design and development of a VAWT of approximately 100-kW rated power, which is very much like 386 the DOE/SANDIA 100-kKW VAWT. Independent wind turbine - developers who are not usually a part of a manufacturer's organization or an electric utility, are currently becoming a large segment of the wind energy industry in the United States. Investors are increasingly funding the development of wind farms which feed electricity directly into utility networks. Favorable economic incentives such as the business energy tax credits and a favorable . legal environment aided by the Public Utilities Regulatory Policy Act, have helped spawn the growth of this industry group. Most of the developers are installing, or plan to install, machines of rated power in the 25- to 100-kKW range. ‘At the present time, there are approximately 30 to 40 different developers in this category, while less than 10 have operating machine clusters. Most projects are located in California. : Wind Turbine Reliability and Test Experiences Most of what little data is presently available on the performance of. large, privately funded WT's has been presented in a series of technical assessment studies performed for the Electric Power Research Institute (EPRI).(15) This series points out that WT's of rated power in excess of 100. kW are still in a technical development stage. Nearly all machines installed and most data currently available are from engineering prototype machines which have been tested in conjunction with electric utilities, but not operated as normal utility apparatus. Therefore, the test data base is of limited value in evaluating the reliability of commercial units without a more in-depth analysis. Documented reliability on small WT's has been developed at the U.S. Rocky Flats Test Center in Golden, Colorado.(16, 17) These tests indicate that most early, small WT's aimed at the generation of electricity have had the following types of early development problems (16, 17): e inadequate mechanical design for minimizing WT dynamic response ‘and loads; @ inadequate ability to withstand short duration high wind speeds in a non-operating condition; e improper control system design or adjustment; and e high fatigue stresses in critical components or subsystems that have led to premature failures. These conclusions were developed from a limited amount of operating history (less than 200 hours of operating time in most cases). As a result, there is no broad test data base of long-term reliable operation of small WT's that the public can obtain. There is no 387 TABLE 16.4 WIND ENERGY CONVERSION SYSTEMS IN ALASKA Description of Small WT Systems Location of Small WT Systems (Installed and Planned) 44 installed with WT rating less than 5 kW 5 installed with WT rating between 10-20 kW 15 used for utility intertie 34 used as battery charger 34 planned for installation by end of 1982 3 Northern (Artic) Region 14 Northwestern Region 12 Southcentral Region 10 Western Region 16 Southwestern Region 4 Southeastern Region 8 Interior Region Source: Wind Systems Engineering, Inc., July 1981. 388 documented evidence that these problems have existed in privately owned small WT's, but limited informal interviews tend to support the documented test results at the Rocky Flats Test Center. Except for a very limited number of small WT's that have a good record of reliability thus far, most machines of this size have been problematic for their owners. Denmark has been a pioneering nation in the use of windpower for electricity applications for the past 45 years. As a result, they have an excellent design base from which to undertake developments. In contrast to the United States experience, Denmark has a very good record of small WT reliability that’ is documented in the public domain. The major vehicles for disseminating this information have ‘been: e reports from the Riso Test Laboratory at Roskilde, Denmark (a national test lab like the U.S. Rocky Flats Test Center), and e the Danish Association of Windmill Owners. Available test data indicate that the major ‘small WT developers. have sold a few hundred machines in sizes up to 55 kW that have operated reliably for the past two to three years. Reported uninstalled machine costs are. very reasonable, but there is very little activity in the use of large WT's in Denmark at this time. The Status of WT's in Alaska. Alaska has seen an upsurge of wind energy. development activities in _the past few years. As illustrated in Table 16.4, much of the wind development activity has been with small-scale systems with power outputs less than 5 kW. An increasing number of machines are being installed in the 10 to 20 kW range. The majority of WT installations to date have been in the Bush where electricity costs are extremely high and where there is very good wind resource potential. Anchorage and Fairbanks are populated areas in Alaska which would normally warrant WT applications, but electricity prices from conventional fuel sources in these areas are relatively lower, thus making electricity from WT's uneconomical. Research is still needed to develop a stronger information base related to the wind resource potential, as well as performance and reliability data of larger scale systems which can contribute to community electrical requirements. The lack of knowledge and commercial readiness of wind systems in the 50-1,000 kW range is impediment to WT development in Alaska. WT's in this size range are more appropriate to community utility applications, yet most of the installations in Alaska to date have been with wind systems less than 5 kW. The Division of Energy and Power Development (DEPD) has therefore begun to manage demonstration projects in Alaska that incorporate WT's with rated capacities of 10 389 068 TABLE 16.5 DIVISION OF ENERGY & POWER DEVELOPMENT (DEPD) DEMONSTRATION Location Unalakleet Skagway Nelson Lagoon Newhalen Bering Straits Skagway Kotzebue Holy Cross and Port Alexander | Size of WT 10 kW 10 kW 18 kW 8 kW PROJECTS IN ALASKA # of WI's Being Tested 3 Source: Alaska Power Authority, January 1982. Status Awaiting funding to monitor operational intertie Intertied to utility, opera— tional data available, blade broken Intermittent operation Damaged blade Appropriated $400,000 Encouraging cost sharing for $100,000. budget Appropriated $200,000 Appropriated $70,000 kW or greater. Instrumented WT's with rated capacities of 10 kW are being tested at demonstration sites in Skagway and Unalakleet. Operational data (technical and economic) from these projects is expected to be available by mid-1983.(5) Table 16.5 depicts other DEPD projects. The major objective of the projects listed in Table 16.5 is to gain operational experience and data related to WT interconnection with utilities to aid utility planners. Other major projects in Alaska include: e Sheldon Point, where a small WT prototype design has the tower of the system anchored to a foundation of a house. Five additional systems are being installed in the same manner to see if indeed cheaper installation costs can be achieved. e Gambell Village, where the village council has installed three 10 kW WT's in June 1982.. The machines power the village laundromat and will be. intertied to the Alaska Village Electrical Cooperative (AVEC). Power will be bought by the utility for 15¢/kWH. (11) : e@ Lower Kuskokwim School District, where a 4 kW Enertech system has been in operation since June 1982 to power the high school's utility puilding and excess power fed to Bethel's utility grid. The potential for continued WT activity in Alaska is significant. A study recently conducted by Polarconsult Energy Group in Alaska assessed the market penetration for WT's in Alaska.(4) Table 16.6 is a summary of the results, illustrating a significant installed capacity potential for WT's in Alaska. 16.3 Economic Implications 16.3.1 Costs The costs of a wind turbine (WT) system may be divided into first costs and recurring costs. First costs include the costs of: site selection; obtaining required permits and approvals from regulatory agencies; WT equipment; shipping; land requisition; site preparation; erection; electrical interconnection; and testing. Recurring costs include: taxes; insurance; and the labor, materials and parts generally referred to as operations and maintenance costs. No fuel is required. . The installed costs of WT's are very site specific and will be relatively high in the Bush where conventional electric power costs are also high. Installed costs in the Bush are estimated to be approximately $6,000 per kW of rated power for machines of under 100 kW rated power, and 20 to 40% less for larger machines.(19-22) O&M 391 THE MARKET PENETRATION OF Size of Number of ‘Turbine Turbines 0-50 kW 368 51-200 kW 65 201-1000 kW 14 Source: Reference 4. TABLE 16.6 25 kW 170 kW 540 kW Average 38 Standard Deviation 20 kW 250 kW 270 kW WI's IN ALASKA kW of Installed Capacity 10,160 kW 11,660 kW 7,500 kW costs generally fall in the range of 1% to 5% of installed cost per year for most applications. The same factors that lead to increased installation costs in the Bush also can result in increased 0&M costs, and this lends some justification to utilizing a similar percentage of installed cost to estimate annual O&M cost. Annual taxes and insurance are taken at 2% of installed cost.(19,21,23,24) Capital and operating cost summary data are presented in Tables 16.7 and 16.8 for a 4 kW wind turbine with no electric storage. The equivalent power cost is about $0.45/kWh which may be competitive in some communities. Figure 16.3 illustrates typical installed costs for small WT's less than 10 kW in Alaska showing the significant economies of scale achievable in- this size range. This implies that systems, when commercially ready and integrable into community systems, are a promising renewable alternative. 16.3.2 Socioeconomic Factors The incorporation and development of WT's in Alaska will not have a significant impact on socioeconomic factors such as employment and the purchase of goods in services in-state. As the market for small-scale WT's develop in Alaska, jobs will be created to sell, install and service the machines. However, it is expected that less than 100 new jobs in sales and services will be created by this growing market by 1990. Manufacturing activity will probably increase with increased demand for WT's in Alaska. But again, it is not expected that a significant number of manufacturers/companies would locate their businesses and build their machines in Alaska. 16.4 Impact 16.4.1 Effect in Overall Energy Supply and Use In small remote and isolated networks, utilization of wind energy will usually directly displace diesel fuel. Unless batteries are used, no capacity credits are taken for this type of application where the conventional energy source must be capable of supplying the entire load demand when the wind is not blowing. Fuel savings will be limited by the WT rating which will be sized to interface with the diesel generators which are to be assisted by the WT and cause not deleterious effects on diesel performance. The WT-diesel hybrid system is most effective when the diesels are designed to operate more efficiently at part load and the proper WT rotor operating speed can be chosen to minimize WT power fluctuations and the effect on voltage and frequency fluctuations on the remote network. Thus, for Bush applications, the most cost effective WT-diesel hybrid system will specify maximum WT capacity and resulting fuel savings. For example, with an annual load demand of 40,000 kWh (46-kW annual average demand), a 100-kW capacity 393 TABLE 16.7 CAPITAL COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD CAPITAL COST Equipment and Materials Direct Labor TOTAL CAPITAL INVESTMENT Wind Turbine Generators (WT) Bush Village with Water Access 1982 constant dollars Not applicable 4 kW 15 years 1 month $12,000 $12,000 $24,000 3G4 TABLE 16.8 OPERATING COST SUMMARY TECHNOLOGY — Wind Turbine Generators (WT's) BASIS Location Bush Village with Water Access Year 1982 CAPACITY Input Not applicable Output — . 4 kW OPERATING FACTOR _ 0.25 Annual Average Capacity Factor ‘Annual Operating Costs Unit Cost Consumption Annual Cost FIXED COSTS Maintenance Labor and Maintenance - . $1200 Taxes and Insurance 480 Capital Charges at 9.5% 2280 TOTAL FIXED COSTS $3960 TOTAL ANNUAL OPERATING COSTS / $3960 TOTAL ANNUAL OUTPUT 8,760 kWh TOTAL LIFE CYCLE COST $ 0.45/kWh 395 96€ 12,000 10,000 8,000 Maximum f 6,000 J Minimum 4,000 Wind Turbine Installed Costs/kW (1982 $) 2,000 1 2 3 4 5 6 7 8 9 10 Nominal kW Rating Source: Polarconsult Energy Group,inc. and Arthur D. Little Inc. estimates. FIGURE 16.3 ESTIMATED RANGE OF INSTALLED COSTS PER KW RESIDENTIAL-SCALE WIND TURBINES 7 system might eventually be optimized by utilizing two 50-kW diesels and one 50-kW WT where the average annual wind speed is 22 feet per second. In such a case, the WT would supply approximately 24% of the energy demand and reduce fuel consumption by about 22%. (26) In normal utility network applications where energy is provided by steam units, for example, WT capacity penetration is generally assumed to be limited to the range of 5 to 15% of network capacity in order to avoid adverse impacts of WT power fluctuations in network power quality. For typical capacity factors in good wind regions, WT's could characteristically supply a maximum of 5% of the network energy demand, yielding a 5% fuel savings in contrast to a 20% or greater fuel savings in Bush applications which generally employ diesel units. 16.4.2 Future Trends Many factors will influence the impact WT's will have in Alaska. The market needs for WT's will be directly impacted by the cost and availability of competing energy sources. In remote areas, electricity from WT's is often much more cost effective than. distributing the electricity from conventional centralized fuel sources. In the Bush, for example, where the price of electricity is approximately 27.5¢/kWh, WT's would be:a very attractive alternative for electrical power. Even in areas of the railbelt, outside of Anchorage, the cost of electricity is 9.2¢/kWh. As the prices of fossil fuels inevitably increase, the price of electricity from utility networks will escalate. WT will then become more cost-effective in the railbelt region. The cost of electricity in Anchorage is currently very inexpensive at 3.9¢/kWh. It is therefore unlikely that WT's will be widely implemented or cost-effective in this area in the near future. The Alaskan government is beginning to encourage the development of WT's. The Divsion of Energy and Power Development (DEPD), as mentioned in the Commercialization section, is currently promoting small-scale WT demonstration projects. Data and _ operational experience gained from these projects will give insight into areas which might necessitate technological improvement or adjustments in order to improve the performance of WT's in Alaska. In addition, the 1982 Legislature amended the alternative energy loan program from a $10,000 limit per customer to $30,000. The first $15,000 will be at 5% interest and the next $15,000 will be at 15%. Many of the current loans and grant programs are expected to continue. The 35% commercial state income tax credit for wind systems, property tax incentives and other state/local government incentive programs for the development of WT's in Alaska are also expected to continue. Government assistance will aid technological improvements for WT applications in Alaska and help make systems more economically viable for individuals wishing to install systems. The technology is 397 available, and with added government support and market demand in remote locations, the technology become a viable, practical and common source of energy for many parts of Alaska in the near future. There are, however, many constraints which could hamper the widespread use of WT's. The economics in many locations is still not favorable compared to conventional fuel sources especially if potential users are concerned with first costs. Other concerns and possible constraints are related - to technical, aesthetic, legal, and environmental issues. Potential WT users may be reluctant to purchase an energy system for which reliability of operation has not been proven. The commercialization of WT's can be slowed by the multitude of diverse, divergent and often contradictory regulatory actions emanat- ing from Federal, state and local authorities. For example, existing. building codes in some jurisdictions might not permit the installation of small WT's, or may impose burdens on WT purchasers; or the height, space, appearance, and required location of WT's may conflict with local zoning/land use ordinances; or the absence of an established body of law. or precedent for preventing obstruction of or access to wind could discourage user acceptance. Other issues could be Occupational Safety ‘and Health Administration regulations, which may be relevant in commercial and industrial applications of small WT's, and the Federal Communications Commission regulations for potential WT interference with television signals. : For large systems, utility-related issues are probably among the most significant for the commercialization of wind power. The value to the owner of a small WT interconnected with a utility is crucially influenced by the utility's rates. One-time charges for interconnection or a charge based on maximum kW demand during the billing period, an increased cost per kWh, or an increase in monthly charges have all been identified as possible reasons for consumer resistance to small WT's. Utilities may also be concerned over the question of buyback rates -- the owner of a WT may expect that surplus energy from his unit will be bought back at least at the retail rate less same reasonable profit, while the utility wishes to pay, at most, the value of the fuel saved. Utilities' concerns about assuring safe interconnections and the maintenance of the accustomed quality of power may also be a constraint to the commercialization of small WT's.(18) With very small diesel grids in remote areas, for example, most systems can only accept up to 40 to 50% of the normal load from WT's without risking damage to the diesel system. For other types of conventional generators (e.g., steam-electric units), the penetration limits would be lower because the other units may not be able to follow load changes as rapidly as diesel units. 398 10. 11. REFERENCES Senior, T. and D. Sengnupta, Wind Turbine Generator Sitin and TV_ Reception Handbook, The University of Michigan, January . Park, J. and D. Schwind, Wind Power for Farms, Homes and Small Industry. U.S. Department of Energy, REP=TOAU TOTO TENE September 1978. W. Vachon et al., Wind Turbine Performance Assessment Technology Status Report No. 4, Arthur D. Little, Inc. for the Electric Power Research Institute, AP-2456, Research Project 1996-1, June 1982. Polarconsult Energy Group, Inc., State of Alaska Market Penetration Study for Wind Generators, prepared for the Department of Commerce and Economics Development and the Division of Energy and Power, March 1982. Alaska Power Authority, Assessment of Potential Participation of Wind Turbine Generators in Alaskan Utilities, January T3Bz. . Osborn W. and W. Downey, Near-Term High Potential Counties for SWECS, Arthur D. Little, Inc. for the Solar Energy Research Institute, SERI/TR-98282-11, February 1981. Osborn, W. and W. Downey, Comparative Market Potential of Small Wind Energy Conversion Systems in Rural Counties In the nite tates, Arthur D. Little, Inc. for the Solar Energy Research Institute, September 1980. Wind Turbine Assisted Diesel Generator Systems, DAF-Indal, Ltd. Mississauga, Ontario, December 1981. Curtice, D. et al., Study. of Dispersed Small Wind Systems Interconnected with a CHT Distribution System, Systems Cc Inc. for Rockwell T C ontrol, nternational orporation. RFP-3093/9445/3533/80/7. March 1980. Curtice, D. and J. Patton, O eration of Small Wind Turbines on a_ Distribution System, Systems Control, Inc. for Rockwell International C i orporation, RFP-3177-2 UC-60, March 1981. Appropriate Energections, Vol. V., Alaska Division of Energy and Power Development, July, August, September 1982. / 399 12. 13. 14. 15. 16. 17, 18. 19, 20. 21. 22, 23. Ebasco Services Incorporated, Wind Energy: Alternative for the Railbelt Region of Alaska, prepared for Battelle Pacific Northwest Laboratories and the fice of the Governor, State of Alaska, March 1982. Testimony of the President of Enertech Corporation to the House Science and Technology Committee, March 1981. Betz, A., "Dus Maximum des Theoretisch Moglichen Ausnutzung des Windes durch Windmotoren," Zietschrift fur gas gesamte Turbinen Wesen, 17, September 20, 1920. Vachon, W. et. al., Wind Turbine Generator Performance Assessment, Technology Status Reports No. 1-5, Arthur D. Little, Inc. for EPRI Project RP1996-1, 1981-1982. Higaski, K. Dunlite Model 81/002550 Wind Turbine Generator Final Test Report, Rocky Flats Wind Systems Program, RFP-2992/3533/79-38, February 1980. , Bollmeier, W. et_al., Small Wind Systems Technology Assessment, State of the Art and Near-Term Goals, Rocky Flats Wind Systems Program, RFP-3136/5550/80/18, February 1980. Hendrick, P. and W. Downey, Wind Ener Conversion System Analysis Model (WECSAM) ‘Computer Program Documentation, rthur D. Little, Inc., for the Solar Energy Research Institute. SERI/SP-19136-4, July 1982. Osborn, W. et al., Near Term High Potential Markets for Small Wind Energy Conversion ystems, rthur D. Little, Inc. for the Solar Energy Research Institute, November 1980. Dickinson, W. and K. Brown, Economic Analysis of ‘Solar Industrial Process Heat Systems, Lawrence Livermore Laboratories, UCRL-52614, August 17, 1979. State of Alaska Long Term Energy Plan, Department of Commerce an conomic evelopment ivision of Energy and Power Development, 1982. — Alaskan Construction Cost Index from HMS, Inc., Anchorage, Alaska, March 1981, Patel, R., Preliminary Generic Assessment of the Operation and Maintenance Costs of Residential Wind Turbine Systems, Arthur D. Little, Inc. for the Solar Energy Research Institute, December 1980. 400 24, 25. - 26. 27, Downey, W., The Effect of Government Economic Incentives on the Industrial Market Penetration of rapital Intensive Solar hermal an ind Energy Systems, Arthur ittle, Ine. for Lawrence Livermore Laboratories, September 2, “1981. : Vachon, W. et al., Large Wind Turbine Generator Performance Assessment Technology Status Report No. 7, Arthur D. Little, Inc. for the Electric Power Research Institute Project RP1996-1, Spring, 1983. Schienbein, L. and D. Malcolm, Wind Turbine Assisted Diesel Generator Systems, DAF-Indal, Ltd., ATAA second Terrestrial nergy System Conference, December Le 3, 1981. Wise, J. et al, Wind Energy Resource Atlas: The Alaska Region, Vol. 10, University of Alaska for Battelle Pacific Northwest Laboratory, PNL~3195 WERA-10, December 1980. 401 je ie — - 17,0 SMALL-SCALE GASIFICATION 17.1 Introduction and Summary 17.1.1 Technical Overview Solid fuel gasification technology was first developed over 150 years ago, and by 1850 the production of gas from coal and biomass was fully commercial. Product gases were used in street lights, domestically, and somewhat later, in manufacturing processes. By about 1900 gas produced from wood or charcoal was. being used to power stationary engines and vehicles. These applications expanded rapidly through the end of World War II. By 1946 over a million vehicles were operating with low-Btu gas in Europe, and over 1000 plants sold "producer gas" in the United States. With the postwar advent of natural gas pipelines and abundant petroleum, gasification systems became obsolete and uneconomic, and virtually all were abandoned. Thirty years later, a renewed emphasis on energy from coal and renewable resources revived these technologies. Gasification, the conversion of a solid (or liquid) fuel to gaseous fuel(s), has most often been ‘practiced with coal and wood but is equally applicable, with properly designed equipment, to any other fuel. The focus of this chapter is on wood, since it is perhaps of most interest for small-scale applications in Alaska. Key differences in the gasification of other fuels are briefly discussed as appropriate. Small-scale applications of solid fuel gasification may be generally defined as those involving the production of about 20 million Btu/hour or less of product gas. For a wood feedstock containing 50% moisture, this corresponds to a feed rate of about 3.5 tons/hour or less, if the gasifier can achieve a thermal efficiency of 65% or better. (Thermal efficiency here equals energy value of the product gas divided by energy value of the feedstock.) To cover the costs of gasification, it is.only carried out when the product gases have a higher intrinsic value than the solid fuel feedstock. In small-scale applications,. this usually means that solid fuels are available and relatively cheap and that gaseous fuels are required for use in onsite combustion equipment. Two cases are most common: (1) An existing oil- or gas-fired boiler or furnace, which cannot be easily or economically modified to accept solid fuels, can be "retrofitted" or used in conjunction with a gasifier. Product gases from the gasifier may be cleaned before being burned in the boiler or furnace. 403 (2) Gaseous or liquid fuels are needed to power stationary or mobile internal combustion engines. Gases made from: gasification of solid fuels can be used, but in this case they must be cleaned and cooled. In both cases, the basic process involves: (1) preparing the solid feedstock for gasification, if necessary, by cleaning, sizing, and/or drying it; (2) conveying it into the gasifier; (3) gasifying; (4) removing ash from the gasifier; (5) collecting the product gases from the gasifier; (6) cleaning and cooling them, if necessary; and (7) transporting the gases. to downstream combustion equipment. Large-scale gasifiers can also be used to produce fuels to be burned onsite, but the major emphasis on their development has been the production of fuels which can be transported by pipeline or used as feedstocks for the production of chemicals or liquid fuels. 17.1.2 Alaskan Perspective In Alaska the potential applications for small-scale gasification are likely to be limited to: e retrofits to existing oil- or gas-fired boilers in industrial facilities, institutions, large office or residential buildings, etc.; e lumber mills in Southeast Alaska; and: e electricity generation in Bush communities. , For retrofitting existing boilers, the factors to consider are analogous to those for similar applications in the Lower 48. Fuel availability. and cost, modifications required, reliability needs, and other issues are always highly site-specific. Such applications would lie mostly in the Railbelt. , For lumber mills and other industries where waste wood is available, gasification can provide fuel for boilers, engines, or engine generator sets. However, the use of these materials as fuels in waste wood boilers is already widespread in Alaska and elsewhere. Since combustion technology is cheaper and more technically proven, it is likely to be chosen over gasification in new applications. Electricity generation in Bush Alaska presently requires the costly import of fossil fuels to the community. If solid fuels are available locally , gasification offers an alternative which may _ be cost-competitive. However, the availability of fuels and the ecological effects of their utilization require careful evaluation. ‘The issues of concern include: (1) the impact of resource depletion; (2) the ability to dispose of ash; and (3) the ability to treat and/or dispose of wastewaters. Further, an operating process requires a large amount 404 eo . “= 1, of attention and a degree of technical knowhow which may not be readily available in the Bush. For an Alaskan lumber mill application of small-scale wood waste gasification followed by electricity generation, a life cycle cost of $0.313 per kWh is predicted. Other Alaskan applications may have significantly different economics because of wide variations in equipment requirements, fuel costs, operating factor, and other items. . Excepting a demonstration project of wood gasification sponsored by the Alaska Division of Energy and Power Development, no past or present applications of small-scale gasifiers are known to exist in Alaska. This most likely results from this technology's reputation for frequent operability problems. 17.1.3 Significance of the Technology Small-scale solid fuel gasification was so widely commercialized that its technical feasibility cannot be questioned. However, significant disadvantages remain which may limit the number of cases where its present use makes practical or economic sense. The major drawback is that, in spite of a decade of renewed development work, the demonstrated operability of these processes remains poor. New units have almost always experienced recurrent problems with one or more operations, the most troublesome being feedstock or ash handling. Even after the inevitable initial problems are resolyed, the units continue to require constant operator attention and high maintenance. A large part of the problem is the inherent variability of most feedstocks. Wood experiences wide seasonal variations in moisture content, while coal can have a very different composition, with ash being the most critical component, from one seam to the next, or even within a seam. In addition to these variations in properties, physical size is also important. The feedstock particles to a gasification system must be properly sized or they will plug feeding systems and upset gasifier operation. These factors lead to a labor intensiveness which obviously has adverse effects on economics. Besides labor, other important operating costs are feedstock, capital recovery, and sometimes, maintenance materials. The feedstock cost--whether itis wood, peat, coal, or another fuel--is not likely to be small except in very limited circumstances, such as when waste wood is available in a lumber mill. Even then, the feedstock may have value for alternative uses. In ‘eases where the feedstock itself is free, e.g., agricultural residues or municipal solid wastes, the cost of collecting and transporting it may be high. 405 Capital cost for the gasifier itself is not usually exorbitant, but the required auxiliary systems may result in a high total cost. In some cases, such as lumber mills, the feedstock and its associated storage and handling equipment are already onsite. If this is not the case, the cost of feedstock receiving, storage, preparation, and conveying systems may well exceed that of the gasifier. Other required auxiliaries are an ash handling system, gas cleaning equipment, and environmental control systems. In summary, small-scale gasification has particular advantages which make it the technology of choice for a few applications; specifically, it is most often used to convert readily available solid fuels or wastes to a gaseous fuel which can be used in internal combustion engines or existing fossil-fuel-fired boilers. There are many: such applications, so the slowly increasing use of small-scale gasifiers may continue, and any upward pressure on fossil fuel prices would doubtless speed the rate of implementation. On the other hand, the difficulties with reliability and the frequently high costs are likely to limit the applications to those mentioned above. In such cases, special circumstances prevail where there are few, if any, alternatives. 17.2 Description of Technology 17.2.1 Operating Principles The term "gasification" usually implies thermal decomposition of solid fuels such as wood, peat, or coal in the presence of a limited amount of air or oxygen, producing gaseous fuels which can be burned later to supply energy (or used as feedstocks for synthesis of liquid fuels or chemicals). Gasification is one of a general group of thermochemical processes, the range of which is defined by two extremes: (1) pyrolysis--also Known as destructive distillation--or thermal decomposition in the absence of air or oxygen; and (2) combustion, or thermal decomposition in the presence of sufficient air or oxygen to convert the solid fuel. (comprised chiefly of carbon, hydrogen, and oxygen) to its combustion products (mostly carbon dioxide and water).* For these small-scale applications, in contrast to those where the product gas is to be transported by pipeline or used to synthesize liquid fuels or chemicals, it will rarely, if ever, make economic sense to use pure oxygen as the gasifying medium because of the high capital and operating costs of oxygen plants. Accordingly, the discussion of this chapter is limited to air gasification. *Gasification can also be carried out in the presence of steam or hydrogen alone, processes which do not fit the narrow definition above. While steam is. often added to air gasifiers, as discussed below, gasification in a pure steam or pure hydrogen environment is not important for small-scale applications. 406 For the case where wood is the solid fuel, air gasification can be represented . simplistically by the following overall chemical reaction: (1) CH + 0.2 O, + 0.8 N, +>CO+0.7 H, + 0.8 N 1.4°0.6 2 2 2 2 The gasification of other fuels (e.g.; coal) can be represented in a very similar fashion, with a different chemical formula for the fuel. A more complete discussion of the chemistry of gasification is included in Chapter 8. The major combustible products are carbon monoxide and hydrogen, diluted by nitrogen. In practice, since the above reaction is endothermic, more air is added to provide some (exothermic) combustion to sustain the process; thus, some carbon dioxide and water are also formed. : In addition to these major products and diluents, other products and impurities leaving the gasifier can include methane and other gaseous hydrocarbons, sulfur and nitrogen oxides, tars, ash, and unconverted carbon-bearing material (char). The relative amounts of products obtained are importantly dependent on gasifier configuration and operating conditions. A typical gas contains approximately 24% carbon monoxide, 16% hydrogen, and 4-8% hydrocarbons (mostly methane and ethylene) and has a heating value of 150 Btu/scf. Gasifiers are commonly classified by the method of air/solid contact, with four types being predominant: updraft, downdraft, fluidized bed, and entrained bed. These are shown schematically in Figure 17.1. Both updraft and downdraft gasifiers are termed "fixed-bed" or, paradoxically, "moving-bed" units because the feedstock moves downward through the gasifier by gravity and is not lifted or suspended by the gas. Updraft Gasifiers In an updraft, or "countercurrent," gasifier (Figure 17.1la) air enters through a grate at the bottom of the vessel and moves upward, while the solid fuel is introduced at the top and moves downward. The solid particles moving down are first dried and then heated to pyrolysis temperatures by hot rising gases. Pyrolysis releases volatile. products which leave the gasifier with gases coming from below. The residual char continues downward, passing next through a reduction zone where it reduces carbon dioxide and water to carbon monoxide and hydrogen, and finally through a combustion zone where combustion reactions provide process heat. 407 C + CO, = 2cO C + H,0 = CO +H, C + 0, = CO, Source: Reference 1. FIGURE 17.1a SCHEMATIC DIAGRAM OF UPDRAFT GASIFIER 2co CoO + H, Source: Reference 1. FIGURE 17.1b SCHEMATIC DIAGRAM OF DOWNDRAFT GASIFIER 408 Disengagement Section Cyclone Tar Scrubber Gas—» Ash Fluidized Bed 4 +— Blomass Feed r d <+— Alr Feed, Preheat ’ ! Source: Reference 1. Pe FIGURE 17.1¢ SCHEMATIC DIAGRAM OF FLUIDIZED BED GASIFIER Source: Reference 1. FIGURE 17.1d SCHEMATIC DIAGRAM OF SUSPENDED FUEL GASIFIER (AFTER FREDERICK/MORBACK DESIGN) 409 The product gases from an updraft gasifier are high in tars, oils, and other volatile organics produced in the pyrolysis zone. They are not usually suitable for use in internal combustion engines without further cracking or extensive cleaning. Updraft gasifiers are therefore most commonly used in close-coupled applications with boilers or furnaces. If the moisture content of the incoming fuel is low, as it is with coals, for example, steam may be added in both updraft and downdraft units to promote char/water reactions. This provides for further fuel gas production and helps to repress soot formation. For high moisture fuels such as wood (even predried wood), additional steam is usually unnecessary. Downdraft Gasifiers Downdraft (or cocurrent) gasifiers (Figure 17.1b) are commonly chosen to avoid the tars and oils characteristic of updraft. units, such as when the product gas is to be used in an internal combustion engine. Air introduced to the midsection.of the gasifier establishes a combustion zone which is continuously replenished with char from above. Conduction of heat upward permits the drying and pyrolysis of solid particles entering the gasifier. Product gases must pass through the hot char bed before leaving the gasifier, allowing for additional .reduction of carbon dioxide and water and additional cracking of tars. Because the gases and the ash both exit the bottom grate of a downdraft gasifier, there is a greater tendency for ash carryover, and the product gas from a downdraft gasifier is higher in particulates than that from an updraft unit. However, it is much lower in tar and oil content because of the additional cracking provided in the lower part of the gasifier. Fluidized Bed Gasifiers Fluidized bed gasifiers (Figure 17.1c) provide for intimate mixing of the gas and solids by maintaining an upward gas flow which is sufficient to expand and "fhiidize" the bed of particles. Unlike the fixed bed (updraft and downdraft) gasifiers, there are no distinct reaction zones, and a uniform temperature is maintained through the bed. Since the reacting particles in a fluidized bed gasifier decrease in size, much of the ash, along with some unconverted material, is carried out as fines with product gases. These fines are removed in a cyclone and may be partially recycled. Further ash removal may be accomplished by bleeding off some of the solids from the bed. 410 Inert materials, such as sand, or reactive materials, such as limestone or catalysts, may be used in the fluidized bed to improve heat transfer and, in the case of reactive materials, clean the gas or otherwise promote desired reactions. Entrained Bed Gasifiers The fourth type of gasifier, the entrained bed (Figure 17.1d), is of lesser interest to small-scale applications. It is in effect a pulverized fuel burner operated with less than sufficient air for combustion. Since a considerable amount of combustion occurs in these units, the gas is hotter than, but has a lower combustible energy value than gases from other types of gasifiers. It is important that this gas be used immediately (close-coupled with a boiler or furnace) to take advantage of its sensible heat. This gas is not suitable for use in internal combustion engines, which require cooled gases. : Some of the technical characteristics which differentiate these four types of gasifiers are further discussed in the next section. 17.2.2 Technical Characteristics As mentioned in this chapter's introduction, two types of applications are commonly considered for small-scale gasifiers: (1) retrofitting an existing oil- or gas-fired boiler with a gasifier which produces a fuel acceptable to the combustor; and (2) using the gasifier in conjunction with an internal combustion engine. An example of a process configuration for the latter application is shown in a schematic diagram in Figure 17.2. This particular diagram . is for a gasification process: demonstrated in Anchorage by the Alaska Village Electric Cooperative (AVEC).(2) The figure shows: (1) reclaiming feedstock (wood chips) from storage and conveying it to the downdraft gasifier; (2) removal of ash from the gasifier; (3) cleaning of product gases in a cyclone and scrubber; (4). preheating gasifier air by heat exchange with product gases; (5) feeding the cooled, clean gases to a spark ignition engine, originally designed to burn natural gas; and (6) an accompanying generator for electricity production. The diagram serves as an illustration of some of the important unit operations and operability aspects of gasification processes. Feedstock Storage and Preparation For the AVEC project, round wood was chipped and the chips were sereened such that only those in the size range 1/4 to, 2 inches in diameter were accepted. (In this case, both oversize and undersize materials were rejected, totaling about 3% of the input. In commercial applications, recycle of oversize material should enable losses to be kept under 2%.) The chips were next dried to less than 25% moisture 411 orp WOOD - GAS - POWER woop cHiF WOOD TO GAS TO ELECTRICITY DIAGRAM ecrupeen Source: Reference 2. FIGURE 17.2 AVEC WOOD GASIFICATION PROCESS FLOW DIAGRAM : using waste heat from the engine, and then conveyed to the storage hopper shown ‘in Figure 17.2. In general, feedstock preparation is an area of critical importance. to proper gasifier operation. Fixed and fluidized bed gasifiers can accommodate particles up to 2" or so in diameter, while entrained bed gasifiers require fine-size particles. Downdraft and entrained bed gasifiers usually require fairly dry feedstocks (less than 25% moisture), while updraft and fluidized bed types can accept higher moisture fuels (up to about 50-55%). For wood, size reduction below chip size is difficult and expensive, so that an entrained bed gasifier is not likely to be: selected unless sawdust is available. . For peat and coal, grinding to small] particles is simpler and not as costly as for wood. Cold weather usually necessitates enclosed fuel storage facilities and other provisions against freezing. Ash Removal Ash is usually removed in dry form to a lock hopper via a rotating grate. Dry ash operation requires maintaining ash temperatures below about 2000°F. "Slagging" gasifiers operate with ash temperatures above 2400°F, and ash is removed as a liquid and quenched. Dry ash operation is more common, especially for wood gasification where quantities of ash are small. Fluidized bed gasifiers are usually followed by cyclones for ash removal. Entrained bed gasifiers. may be operated in nonslagging mode with cyclones or in slagging mode with ash quench systems. Gas Cleaning Gas cleaning is of critical importance when internal combustion engines are downstream, since tars will rapidly ruin an engine. Reportedly, engines run on wood and coal gas in Europe during World War II required complete rebuilding once a year.(3) Such frequent and extensive maintenance would likely render a current project uneconomic. For close-coupled applications with boilers and furnaces, it may be possible to burn dirty gases directly. The extent of cleaning required is dependent on the design of the combustor. Gasifier Selection and Design Neither fluidized bed nor entrained bed gasifiers have been much employed in small-scale applications. Fluidized bed gasifiers have more complex control requirements than fixed bed units, lesser 413 turndown capabilities, and are relatively new. Entrained bed gasifiers require dry, fine-size feed materials, also have poor turndown capabilities, and have received very little development attention for small-scale applications. Fixed bed gasifiers generally have good turndown ratios (about 5:1), but response to load changes is somewhat slow. They may be best applied for base load generation with incremental requirements (peaking) supplied by another storable fuel. The high temperatures at which gasifiers operate impose constraints on materials of construction. Most gasifiers are refractory lined, although provisions for heat recovery, such as water jackets, are used in some cases. In most downdraft gasifiers the product gases flow upward through an annulus around the shaft so that some heat transfer to incoming feed material is effected. Lower temperatures are required to volatilize wood as compared to coal, and wood also gasifies faster than coal under given operating. conditions. However, wood is also less dense than coal, so that it requires more volume per unit heating value. Because of these factors, the energy output of a gasifier of certain physical Size is similar whether operated with wood or coal feedstock. Reported thermal efficiencies of gasifiers have varied from less than 50% to over 90%, with most falling between 70% and 80%. Gas cooling and cleaning prior to combustion introduce additional thermal penalties. Minor efficiency losses are also incurred when operating the gasifier at less than design capacity. Most small air gasifiers operate at or close to atmospheric pressure to avoid the increased costs of pressurized construction and to minimize sealing problems. Some gasifiers (suction type) operate at less than atmospheric pressure to further minimize sealing problems and avoid potentially dangerous gas leaks. The vacuum is usually created with the use of induced draft stack blowers on the boiler exhaust or by the use of a blower on the wood gas exiting the gasifier. In contrast, large gasification processes designed to produce pipeline gas or chemical feedstocks usually find it economical to operate at high pressure in order to reduce downstream compression costs. Some fixed bed gasifiers have devices such as stirrers to agitate the fuel bed and maintain more uniformity. Process Operability Virtually all present systems for which information is available report varying degrees of startup and operating difficulties. The usual culprits are feed and ash handling systems and product gas cleaning systems. 414 Bee _ Cl ay Feedstock handling is probably the largest single source of problems. This is particularly true when there is too much size variation among the pieces of feedstock. Bridging and jamming are the common result. Plugging of dry ash. handling systems is another problem frequently encountered. These problems quite often have resulted in the redesign of solid conveying systems. The’ gasifiers themselves can give problems if the feedstock is high in moisture or ash or if it is not properly sized. High ash contents have caused flows to be blocked by ash accumulation or, if temperatures are high, have also caused slag to form in the gasifier. The continual duties of feedstock receiving and handling, gasifier adjustments, etc., will usually require a full-time operator. Additional maintenance labor on a part-time basis is also required. In short, reliable operation of a gasification system is only achievable by proper initial design, careful attention to feedstock quality and size, and continual monitoring of the operation. Safety Other precautions are. necessary to ensure safety. The product gases, consisting of carbon monoxide, hydrogen,. and other components, are explosive at a wide range of concentrations in air (approximately 4-70%) and must be kept from sparks or fires. Furthermore, carbon monoxide, which is odorless and colorless, . constitutes a toxicity hazard. Detection devices should always be installed near the gasifier or worn by operating personnel. Finally, proper training of operators and adequate safety clothing should. be used to minimize the dangers of burns, eye injuries, and other accidents. 17.2.3 Environmental Issues The important environmental issues relative to the application of small-scale gasification in Alaska are resource depletion and the handling of liquid effluent ‘streams. The availability of fuel is in serious question in remote communities in tne Bush. For example, the generation of 50 kW of electrical energy in a gasifier/diesel engine/generator set would require about 200 Ib/hour of wood at 50% moisture--about 2.5 tons per day year round. If driftwood is used for such a _ purpose, its gathering becomes a full-time occupation. In contrast, a typical lumber mill would experience no difficulty in supplying a much larger amount of wood waste to an onsite gasifier. 415 TABLE 17.1 EXAMPLES OF COMMERCIAL SMALL-SCALE GASIFIERS Gasifier Type* Fuel Scale Demonstrated *. Applied Engineering (USA) U Wood 82 tons/day . Biotherm (USA) E Wood 16 tons/day i Biomass Corporation (USA) D . Wood 3-20 tons/day . Coal . Lo: Duvant (France) D Wood 7-25 tons/day — Entropie (France) D Wood 6-26 tons/day i Forest Fuels Inc. (USA) U Wood 3+20 tons/day ms Halcyon Associates (USA) U Wood 25 tons/day ae Imbert (Germany) D Wood 3.5 tons/day if Pillard (France) D Wood — 0.5-42 tons/day ~ E Wood 0.5-42 tons/day if *y = updraft, D = downdraft, E = entrained bed : { Source: Arthur D. Little, Inc. 7 | i } oh 416 The liquid condensate collected in the gas cleaning system is likely to ‘be a significant disposal problem. Contaminated with phenols, organic acids, and many other organic species, it has a very high biological and chemical oxygen demand.(4) If discharged directly, it would be highly toxic to aquatic life. The amount of this wastewater is dependent on the manner in which the gas is scrubbed, moisture content of the fuel, and the type of gasifier. Its quantity can- be reduced by drying the fuel. The tars can be removed by skimming the wastewaters and further evaporating with exhaust gases. In any event, the contaminants themselves must be treated. Various methods: have been proposed,(5) including separation and/or evaporation followed by reinjection into the gasifier, but more work is needed to demonstrate their effectiveness. The combination of gasification, gas cleaning, and combustion is . relatively clean in regard to air emissions. Particulate emissions are well controlled. Sulfur and nitrogen oxides emitted will be comparable to those from direct combustion facilities and are likely to be relatively minor. Ash disposal may be an issue for some fuels and/or in some very fragile ecologies. 17.2.4 Commercial Status As mentioned earlier, gasification is an old technology. which was widely used up through the end of World War II. This knowhow was largely lost, although some current gasifier manufacturers are survivors from former days. . Some’ examples of small-scale gasifier manufacturers ‘are shown in Table 17.1. A much larger list of commercial developers and researchers is given by Reed. (1) Users of these and other systems are worldwide. For example: e The U.S. manufacturers shown in Table 17.1 have installed about 20 commercial units in the United States.(1,6) Many other pilot and demonstration scale facilities have been built. e The Alaska Village Electric Cooperative project included a 3 million Btu/hour Biomass Corporation downdraft gasifier. Numerous operational problems were encountered, including wood and ash handling, gas cleaning and cooling, and gasifier control. Long-term operability was not demonstrated. e The French manufacturers have built many commercial units in Europe. Duvant and Pillard have also reportedly sold many units in developing countries. (6) 417 e Imbert built over half a million gasifiers up to 1945 and has recently revived the old technology. Two systems have been sold in Canada, one (60 kW) to Saskatchewan Power, and one (120 kW) to a private owner. (6) The implementation of gasification within the past decade has been rapid relative to that of less developed technologies such as solid fuel liquefaction but slow relative to the better understood combustion process. Small-scale gasifiers are usually chosen for specific applications where their advantages over direct combustion are apparent, e.g., for retrofitting an existing gas-fired boiler or furnace, or for feeding to internal combustion engines. The only constraints to more rapid implementation in those applications is the remaining lack of confidence about the reliability and operability of these units. Added deterrents in Alaska are geographic remoteness, weather,. number of cases where fuel is available, a lack of construction or operating labor in some situations, and the fragile environment. There are a large number of research and development programs aimed at solving the remaining problems with gasification processes. Of particular note is the continuing program at the Solar Energy Research Institute,(1) which includes laboratory and _ pilot-scale research, contract funding, and technology transfer. 17.3 Economic Implications 17.3.1 Costs The task of choosing a representative small-scale gasification system, in order to examine its costs, is difficult, especially for Alaska. Fuel type, system size, and load characteristics vary widely between, say, industrial applications and Bush communities. The system chosen here, a waste wood gasifier used to produce electricity in a lumber mill, may thus be regarded as typical but not representative; that is, its costs cannot be easily generalized to other applications. In the final report relating the Alaska Village Electric Cooperative's experiences with wood gasification, Marenco, Inc. reported that the Mitkof Lumber Company of Petersburg, Alaska planned to install a gasification/power generation system in early 1983.(2) The mill was reported to produce 6,870 tons/year of wood waste, 2,245 tons of which are sawdust and the remainder, bark and trim ends. The annual electric power consumption was estimated to be 600,000 kWh. The cost estimates presented here are for a system which would satisfy these electricity needs. The system evaluated does not necessarily correspond to that which may be chosen by Mitkof, nor -are any of the costs’ necessarily relevant to that particular application. 418 The following assumptions were made: e The gasification/power generation system will operate 12 hours/day, 250 days/year. e Sawdust, with a moisture content of 50% (wet basis) and a heating value of 8,600 Btu/dry Ib, is used in the gasifier.* This wood waste was assumed to have a value of $30/oven dry ton, at the recommendation of another Alaska lumber mill which operates a waste wood boiler. (8) ‘e@ The gasifier and gas cleaning system achieve an overall thermal efficiency of 65%. (3) e The electric power produced has dedicated uses; thus, no provisions for hookup to an electric utility are included in the capital costs. e Equipment and materials costs are 20% higher than in the Lower 48, corresponding to Cost Zone 3, but direct labor costs for installation are 1.82 times those in the Lower 48, corresponding to Cost Zone 2. The latter factor was chosen to reflect the relatively larger infrastructure and labor base in the Southeast relative to more remote Bush communities. A 3.5 million Btu/hour (of product gas) gasifier and 300 kW diesel engine generator are chosen. It is assumed that the engine generator is derated 20%, to 240 kW, when used with low Btu gas.(7) It is further assumed that the engine requires a wood gas to diesel fuel ratio of 6:1, on a Btu basis. (Supplemental diesel fuel is required for ignition.) The equipment costs for the downdraft gasifier and the diesel engine generator set were updated from Levelton(6) for a Duvant system. Auxiliary systems added included wood handling and preparation (including drying), gas cleaning and cooling, and diesel fuel storage. A contingency of 10% of total capital investment was added.: The results are shown in Tables 17.2 and 17.3. Total capital investment is estimated at $659,000, and total life cycle cost for this -~9ll system is calculated to be $0.313/kWh. ~ Since the reported electricity cost for the mill is $0.1156/kWh, the system would appear to be uneconomic. *The successful use of sawdust in downdraft gasifiers has been demonstrated,(7) but in general, larger particle sizes are required. 419 TABLE 17.2 . CAPITAL COST SUMMARY TECHNOLOGY Small-Scale Gasification ! BASIS Location Southeast Alaska : Year 1982 Constant Dollars / to CAPACITY ' Input 1630 tons/year Wood Waste (@ 50% moisture) t Out put 720,000 kWh/year Electricity [ ESTIMATED USEFUL LIFE 15 years CONSTRUCTION PERIOD 3 months ‘ CAPITAL COST Equipment and Materials $515 thousand wil Direct Labor 78 Indirect Costs -- Home Office Costs - as Contingency 66 TOTAL CAPITAL INVESTMENT $659 thousand 420 TABLE 17.3 OPERATING COST SUMMARY : TECHNOLOGY Small-Scale Gasification : BASIS “ Location Southeast Alaska Year . 1982 Constant Dollars ‘ CAPACITY : Input 1,630 tons/yr Wood Waste (@ 50% moisture) , Output 720,000 kWh/yr Electricity j OPERATING FACTOR : a Annual Annual . Operating Costs Unit Cost Consumption Cost VARIABLE COSTS Fuel (wood waste, 50% $15/ton 1.630 $24,500 moisture) Electricity $1.80/gal. 10,500 18,900 (diescl fuel) Catalysts, Chemicals, i etc. IL. TOTAL VARIABLE COSTS $43,400 ' FIXED COSTS j Labor Operating, $58,240/man~year ol 58,200 ‘ Maintenance - $58,240/man-year 0.25 14,600 : Maintenance Materials : 33,000 ‘ Taxes and Insurance 13,200 ‘- Capital Charges 62 ,600 t TOTAL FIXED COSTS 181,600 : TOTAL ANNUAL OPERATING COSTS 225 ,000 TOTAL ANNUAL OUTPUT / 720,000 kWh TOTAL LIFE CYCLE COST 0.313 kWh 421 Several points must be noted: e The assumed operating factor is low, since the mill is assumed to be open 12 hours/day, 250 days/year. If the system operated 24 hours/day, 330 days/year, the annual electricity production would be 1.9 million kWh, and life cycle cost would be about $0.25/kWh. e@ The wood waste, charged at $30/oven dry ton, accounted for only $0.034/kWh of the total life cycle cost. e Operating and maintenance labor costs charged to this system, accounting for $0.101/kWh, may be partially or totally avoided if present personnel are used. However, special training is required. . e Capital recovery accounts for $0.087/kWh, the largest single contributor to life cycle cost. The economies of scale are significant for these small systems because: e labor costs are the same for a 1 MW system as for a 200 kW ' system; e there are practical limits on the minimum size of most : components of the wood feeding system, so the equipment purchased for the smallest gasifiers serve equally well for those several times larger; and @ good economies of scale can be obtained for the gasifier, gas cleaning system, wood dryer, and diesel fuel storage. The effect of increasing system size from 200 kW to 1 MW on total life eycle cost is shown in Figure 17.3. 17.3.2 Socioeconomic Factors The implementation of small-scale gasification in Alaska would have small overall impact on the socioeconomic structure of the state, but local impacts might be significant. This is especially true in the Bush. Small-scale gasifiers require constant operator attention and frequent maintenance. Their implementation would therefore create new jobs unless prior employed labor was available. The number of jobs would depend on the size and complexity of the installation and its operating rate. Typically, a small system would require one operator per shift (i.e., up to four full-time employees) and one part-time or full-time maintenance person. Operator training is required, but such training would likely be of relatively short duration. It is likely that, with 422 €eP Life Cycle Cost, dollars/kWh 0.40 0.30 0.20 0.10 1000 400 600 Electric Generating Capacity, kW 2000 3000 4000 5000 6000 Wood Consumption, tons/year (50% moisture) FIGURE 17.3 COST OF WOOD GASIFICATION 7000 training, local personnel could satisfy the labor needs for operation. Maintenance requires more technical knowhow, the availability of which may be severely limited in the Bush. In forested areas, the additional consumption of wood or wood wastes which would result from implementation of gasification may provide additional jobs and other local benefits. These benefits are associated with gathering and transporting wood and wood wastes to new gasification facilities. The equipment required to implement small-scale gasification would be mostly imported. No other significant purchases of goods or services within AlasKa are envisioned. 17.4 Impact 17.4.1 Effect_on Overall Energy Supply and Use Since solid fuel gasification followed by fuel gas combustion is not as efficient as liquid fuel or natural gas combustion, the implementation of small-scale gasification will result in increased overall fuel usage (on the basis of total Btu's consumed). It is, of course, the substitution of wastes or readily available solid fuels for premium gaseous and liquid fuels which provides the impetus to implementing gasification. This substitution of solid fuels can provide savings not only in a reduced unit cost for the fuel itself but also by the elimination of distribution and storage requirements associated with the formerly used fuel. In lumber mills, for example, facilities to handle wood wastes are required anyway, so the use of these wastes as fuel instead of oil, for example, is generally advantageous. Likewise, the elimination of transportation costs for fuels imported to Bush communities may be sufficient incentive to use alternative fuels. These effects also apply to combustion and other solid fuel conversion technologies in addition to gasification. The advantages peculiar to _ gasification lie in its ability to provide gaseous fuels for onsite combustion. 17.4.2° Future Trends Small-scale solid fuel gasification will continue to find limited use in special circumstances such as retrofits. to existing boilers or with small-scale electricity generators. In Alaska, Bush applications such as lumber mills or isolated communities may be most relevant. There may also be industrial or other applications in the Railbelt (retrofitting boilers), as there are in the Lower 48. 424 In these favored situations, gasification offers the same advantages as other solid fuel conversion technologies, i.e., those associated with substitution of cheaper or more available fuels and possible elimination of distribution/storage requirements. It also offers the benefits of some positive socioeconomic effects such as the creation of new jobs. These advantages must be weighed against high capital and operating costs, a poor history of operability and reliability, and the potentially harmful ecological impacts of resource depletion and waste disposal. The aim of ongoing research and development activities is to solve the remaining problems with gasification processes, thereby improving their economics and increasing their use. 425 roe 3. REFERENCES Reed, 1T.B., ed., Biomass Gasification Principles and Technology, Noyes Data Corp., Park Ridge, N.J., 1981. Marenco, Inc. and Alaska Village Electric Cooperative, Inc. Wood Gasification - Power Generation Development Project, Final Report to U.S. DOE, Alaska Division oF Energy and Power Development, March 1982. Levelton, B.H. & Associates, Ltd., An Evaluation of Wood-Waste Energy Conversion Systems report to British Columbia Wood Waste Energy Co-Ordinating Committee, March 31, 1978. Institute of Gas Technology, Qualitative Assessment of Marenco Wood Gasification System, report to Alaska DEPD, March 1982 (included in Reference 2). Wilkinson, L.E., Results of Preliminary Tests on Condensate from AVEC Gasifier and Proposed Methods for Treatment and Utilization, report to Marenco, Inc., February 1982 (included in Reference 2). Levelton, B.H. & Associates, Ltd., An Evaluation of Wood Waste Energy Conversion Systems, ENFORK Project C-IiI, Canadian Forestry Service, Ottawa, Ontario, March 1981. Mahin, D.B., Thermochemical Conversion of Biomass for Energy, Bioenergy Systems Report, U.S. Agency for International Development, Washington, D.C., June 1982. Personal communication, Mr. Bob Allensworth, Alaska Lumber & Pulp Co. (907)-742-2242, with C.G. Greenwald, Arthur D. Little, Inc., November 16, 1982. 426 18.0 FUEL CELLS 18.1. Introduction and Summary 18.1.1 Technical Overview A fuel cell is an energy conversion device which can continuously convert the chemical energy of a fuel and oxidant to electricity electrochemically.. It generates electricity by galvanic action, as does as storage battery. Like a fuel-fired power generator, and unlike a storage battery, a fuel cell generates power from a continuous feed of fuel and oxidant, and exhausts a continuous stream of . oxidation products. ‘ : All fuel cells consist of the same very basic components: an anode where a fuel containing hydrogen releases electrons; an electrolyte, which separates the cathode from the anode and conducts the cell current, aS ions, between the two; and a cathode, at which the oxidant accepts electrons. The electrons released at the anode are externally conducted to the cathode when: the circuit is completed. Numerous basic fuel cell types have been investigated. The principal identifying characteristic is generally considered to be the choice of electrolyte. In some fuel cell concepts, the fuel is oxidized at the anode by losing electrons to form positive ions which then migrate through the electrolyte to combine with, and reduce the oxidant at the cathode. In effect, the fuel. becomes the charge carrier. In others, the oxidant is reduced at the cathode to form negative ions which migrate through the electrolyte to combine with, and oxidize, the fuel at the anode. In these cells, the oxidant carries the charge. In either case, the charge then passes through an external circuit, delivering usable electrical energy. Since fuel cells only operate on hydrogen or synthesis gas, an integrated fuel processor is required for operation with commercially available, fuels such as natural gas or petroleum distillates. Fuel cells offer very. high power generation energy efficiencies. Unlike heat engines, fuel cells convert energy isothermally (i.e., at constant temperature). Therefore, their efficiency is not Carnot-cycle limited. This means that theoretically nearly all of the chemical energy* in the fuel can be converted to electricity. The maximum theoretical efficiency of a fuel cell running on hydrogen is about 95% compared to 40-50% for heat engines(1). However, for a practical fuel cell system designed to run on natural gas, the fuel-to-net-electricity efficiency is in the range of 38-40%, still an improvement over conventional systems. *The maximum amount of chemical energy that can be converted is equivalent to the Gibbs free energy. 427 Fuel cell conversion systems have several features which make them attractive candidates for dispersed power generation. Briefly, these include: e Siteability. Fuel cell power plants are environmentally "clean" compared to other generating systems. Gaseous emissions consist mainly of water vapor and carbon dioxide with neglegible amounts of hydrocarbons, NO, and SO_. They produce no liquid waste streams and are also relatively~ quiet. e Constructability. Fuel cell power plants lend themselves to factory assembly. It is envisioned that preassembled modules ean be delivered to the site for final installation. This will reduce lead time and construction costs for incremental - generating capacity. e Load Management. The efficiency versus load characteristic for a fuel cell is nearly constant between 20 and 100% of load. Therefore, a fuel cell can be modulated over a wide range without loss of fuel efficiency. e Unattended Operation. Currently, fuel cell power plants are being designed to follow load automatically without on-site operator assistance. This is of particular significance for remote power generation applications. 18.1.2 Alaska. Perspective Currently, village power is provided mainly by local (dispersed) power plants based. on diesel generators. The delivered cost of diesel fuel is equivalent to about $12.00 per million Btu; consequently, the resulting power cost is very high. Fuel cell technology has the potential for providing reliable electric power to Alaskan villages at rates that are equivalent to or lower than those now experienced. In addition, fueled with methanol, fuel cell power plants would reduce the state's diesel fuel product demand which would ease the current distillate product supply/demand imbalance. At present there are no fuel cells used in Alaska for this purpose. The technology status is such that prototype power plants in the size range appropriate for village application (40 KW) are just now being field tested in the Lower 48, Fuel cell power generation is still considered to be a high cost option for many applications, particularly where extensive energy distribution networks already exist. However, the Alaskan environment may offer a unique opportunity for fuel cell deployment. The power costs in remote villages are presently in a range where mass-produced fuel cell units might offer a cost advantage. In addition, the fuel cell generators can operate efficiently on methanol. Due to the state's large natural gas 428 and coal resources, the production of methanol for various end uses is a distinct possibility in the future. Large-scale production of methanol could reduce the delivered fuel costs to a level below diesel fuel on a Btu basis, assuming Alaskan natural gas is priced below petroleum on a Btu equivalent basis. 18.1.3 Significance of the Technology The use of fuel cells for remote, distributed power generation is technically feasible today; the major obstacles are economic. The development of .fuel cell hardware has advanced to the point where operational units could be ordered for delivery within a two to three ‘year timeframe. Two potential manufacturers, United Technologies and Westinghouse, will accept orders for phosphoric acid fuel cell power plants. . However, the owner would accept certain risks if he were to buy today. ' Complete power plants are just now being field tested in actual power generation applications. Hence, there is essentially no field operating experience regarding performance versus load, maintenance requirements, or reliability. Limited data on these aspects for judging the merits of fuel cell generation will be available by 1985. The estimated life-cycle cost for fuel cell generated energy from a methanol fuel is 19¢/kWh of electrical output. This cost is based on projected commercial fuel cell hardware cost taking into account engineering cost reductions and manufacturing cost savings. Using prototype system costs, the life-cycle cost increases to about 29¢/kWh. Both estimates are based on no recovery of waste heat for district heating. It appears that on the basis of current or near-term phosphoric acid technology, there is a potential role for fuel cell generators in Alaska. Even with the high anticipated hardware costs. for pilot production modules, the life-cycle costs appear to be comparable with the current village energy costs. With modest hardware cost reduction, the fuel cell generator would offer cheaper power than a diesel generator. 18.2 Description of Technology 18.2.1 Operating Principles Three basic types of fuel cells are actively being developed for terrestrial applications. Identified by electrolyte, they are referred to as phosphoric acid, molten carbonate, and solid oxide fuel cells. Although their operating conditions are somewhat different, the basic principles of operation are quite similar. These operating principles will be explained using a single phosphoric acid cell as an example. Figure 18.1 is a schematic diagrdm of one repeating element in a stack of fuel cells. A functional power plant will contain cell "stacks" which 429 FUEL- HYDROGEN CONTAINING ANODE- OXIDATION REACTION ;—ELECTROLYTE-ION CARRIER CURRENT COLLECTOR FUEL IN memes ep FUEL OUT LOAD emp = OXIDANT OUT L_ CATHODE-REDUCTION REACTION OXIDANT- OXYGEN CONTAINING CURRENT. COLLECTOR A7906130! Source: Reference 1. FIGURE 18.1 CONCEPTUAL DIAGRAM OF A PHOSPHORIC ACID FUEL CELL 430 consist of many cells stacked one above the other with electrical interconnections. .Each cell element is made up of an anode, where oxidation of the fuel occurs; an electrolyte, to separate the anode and cathode and to conduct the ions between them; and a cathode, where reduction of the oxidant occurs. The operation of the cell involves many complex mechanisms, which are conceptually simplified in the following discussions. (1) Phosphoric Acid Fuel Cells These cells utilize hydrogen as fuel and charge carrier. At the anode, gaseous hydrogen is electrochemically oxidized (i.e., loses electrons): H, + 2H* + 2e7 (1) The electrons are transported through the external circuit, and the hydrogen ions are conducted through the electrolyte to the cathode reaction sites. Oxygen, which has diffused through the cathode, reacts with the hydrogen ions and electrons (the oxygen is reduced) , and the product water diffuses back out of the cathode: 1/2 O, + 2H" + 2e + H,O(g) (2) The net reaction is H, + 1/2 0, + H,O(g) (3) Operated at relatively low temperatures (around 375°F), phosphoric acid fuel cells require the use of platinum on both electrodes to catalyze hydrogen dissolution and oxidation and oxidant reductions. Phosphoric acid fuel cells are tolerant of all significant atmospheric constituents; hence, they use air as the source of oxygen. They are less tolerant of sulfur and carbon monoxide. Thus, sulfur compounds must be removed from the fuel stream to a limit of 100 ppm by volume. Carbon monoxide must be less than 1% in the fuel to the stack and is normally shifted to hydrogen and carbon dioxide prior to fuel use. Furthermore, since hydrogen is the only fuel which the cell will actually utilize for power generation, a source of hydrogen (usually produced on-site from other fuels) must be available. Molten Carbonate Fuel Cell This fuel cell typically uses a molten mixture of two alkali metal carbonates as the electrolyte. Operated at temperatures around 1200°F, they do not require expensive noble metal catalysts on the electrodes. Electrodes are fabricated of nickel, with the anode forming nickel oxide in service. In this cell, the oxidant carries the charge through the electrolyte. Oxygen and carbon dioxide diffuse through the cathode, accepting electrons to form carbonate ions at the cathode/electrolyte interface: 431 ot 2e + co, (4) These carbonate ions migrate through the electrolyte, combining with hydrogen to yield water, carbon dioxide, and electrons at the anode: CO, + 1/20 H, + CO, + H,O + CO, + 2e (5) 2 3 2 Molten carbonate fuel cells are tolerant of major atmospheric species, so they use air as the oxidant. They require continuous partial recycle of carbon dioxide from the exhaust to the air feed in order to have sufficient carbon dioxide in that stream to allow carbonate ion formation with the required amount of oxygen. Molten carbonate fuel cells have the capability to utilize mixtures of hydrogen and carbon monoxide directly, which makes them particularly suited for integration with coal gasifiers. The potential for formation of nickel sulfide on the electrodes makes them very intolerant of sulfur. Limits are estimated to be on the order of 1 ppm by volume sulfur in the fuel gas. Solid Oxide Fuel Cell These cells use a ceramic oxide, such as a mix of zirconia and yttria as the electrolyte. Operated at temperatures around 1800°F, electrodes contain no high cost catalysts, but must have excellent high temperature strength and corrosion resistance. Accordingly, a variety of electrically conductive ceramics are being considered for electrode ‘construction. At these temperatures, oxygen is conducted through the electrolyte as an O ion. Molecular oxygen diffuses through the cathode, and accepts four electrons per molecule, forming two ions, at the cathode/electrolyte interface: O,+4e > 20- ; (6) Passing through the electrolyte, the ion arrives at the anode where it yields its excess electrons to that component and combines with. fuel gases available there: 200 +20. + 2CO, + 4e- (7) Air is expected to be used as the source of oxygen. In electrochemical theory, a broad range of fuels could be used; in engineering practice, simple, stable’ fuel molecules which will not deposit coke on the system components can be used. Hydrogen, carbon monoxide, and methane are © examples. Due to the primitive state of system development, tolerance of impurities in the fuel cannot be assessed accurately at this time. The remainder of the technology description focuses primarily on the characteristics of the phosphoric acid fuel cell, since they are similar for all types of fuel cells. Furthermore, the phosphoric acid technology is the most technically advanced. 432 Generator System Description A viable plant for power generation based on any of the fuel cell concepts described above will include four basic engineered systems: e a fuel processor, which produces a fuel, free of damaging concentrations of contaminants, which the cell can utilize; e a cell stack, which securely mounts a usable number of cells in a configuration which allows continuous fuel and oxidant feed and exhaust and waste heat withdrawal; e a thermal management system, which utilizes and/or disposes of -waste heat; and @ a power conditioning system, which processes the power generated into a form appropriate for end use or utility transmission, A simplified block diagram is shown in Figure 18.2. These systems vary among fuel cell concepts, available fuels, system scale, waste heat utilization opportunities, and power end uses. Each basic system is discussed in turn, with the range of likely variations noted. Fuel Processor. As previously noted, the phosphoric acid fuel cell utilizes only hydrogen for power generation. Other major fuel gas stream constituents will serve only as diluents or, in the case of carbon monoxide, will inhibit reaction. Sulfur-bearing species will damage the cell. Accordingly, the fuel processor must produce a fuel stream which is rich in hydrogen and contains very little carbon monoxide and sulfur. To achieve this, a variety of fuel processing options are appropriate to use with different raw fuels. When a source of essentially pure hydrogen is available (as is typically the case on spacecraft), it can be used directly. Next to hydrogen, the simplest fuel to utilize is methanol. It can be reformed at low temperatures, using fuel cell waste heat to provide much of the energy required by the endothermic reforming process depicted by the overall reaction equation: CH,OH +H,O ¢ CQ + 2H (8) Moreover, methanol is a very low sulfur product; thus, no sulfur control step is required. : Other light fuels, such as natural gas, LPG, ethanol, and naphtha, typically require a multiple stage fuel processor. The sequence of steps involved includes feed desulfurization, catalytic steam reforming to form hydrogen and carbon monoxide, and finally, shift conversion of 433 Fuel Pre-Processor Source: Reference 6. . Fuel Recycle Fuel Exhaust Low Grade Heat Recovery Processor Power Section and Thermal Management : Steam ; Subsystem Air in FIGURE 18.2. A FUEL CELL GENERATION SYSTEM 434 High Grade Heat Recovery AC Power Out carbon monoxide and steam to carbon dioxide and hydrogen. The overall chemical reactions involved include: Steam Reforming CnHm + nHoO + nt r H + nco (9) Shift Conversion co + H,0 t Hy + CO, (10) 2 High speed diesel oil and heavier liquid fuels cannot be reformed in conventional steam reformers, due to the tendency to produce carbon deposits in the reformer and difficulties in controlling sulfur sufficiently to protect the conventional reformer's catalyst. Two methods are being developed to process heavy liquids: autothermal reforming and high temperature steam reforming. In autothermal reforming, the raw fuel is partially burned with a substoichiometric quantity of oxygen, and in the presence of steam. ‘The resulting gas is reformed and shift-converted using a partitioned bed of sulfur-tolerant catalyst. High temperature steam reforming typically involves passing the fuel-steam mixture through two successive catalyst beds, at higher temperatures than used for conventional steam reforming. The first bed utilizes a catalyst which promotes partial: reforming but inhibits earbon deposition. The second bed, which promotes more complete reforming than the first, typically contains a nickel catalyst, similar to that used in a-conventional steam reformer. — Cell Stack. The geometry of an individual fuel cell is planar. Electrically it consists of a cathode plate, an electrolyte, and an anode plate as previously shown in Figure 18.1. Each cell will develop a small voltage, about 0.55-0.7 volt, operate at current densities ranging approximately 0.1-0.4 amps/cm". To minimize internal losses. and maximize power output it is desirable to minimize the thickness of the electrolyte, which separates the two electrodes, In order to maintain cell physical integrity and generate usable quantities of power in the smallest possible volume, demonstration and commercial system designs involve arrangement of the planar cells into stacks. Electrodes are structured in order to allow manifolding of fuel and oxidant in and exhaust out along the cell edges, which constitute the sides of the cell stack. Electrolyte is dispersed through a thin, porous plate of structural material,’ which serves to separate cathode from anode in each cell. Cells are electrically connected in series to produce stack power outputs of tens to thousands of volts. A large power plant will contain a multiplicity of stacks. - Stacks are also structured to allow cooling. Phosphoric acid fuel cell stack designs may incorporate air, purified water, or organic liquid coolants. Thermal Management System. Two classes of waste heat are produced: "high grade" heat at stack temperature (about 375°F), absorbed into 435 Source: Reference 3. FIGURE 18.3 40 kW FUEL CELL POWER PLANT PROTOTYPE ~ 436 the stack cooling fluid; and "low grade" heat from fuel processor and fuel cell exhausts, available at temperatures up to approximately 190°F. This heat can be used for space conditioning, for low temperature process heat, or it can be rejected to the atmosphere, depending on economics. Waste heat management is coupled to water management. The reformer in the fuel processor requires a continuous supply of steam, which can be generated by the stack cooling system if it is water cooled. The required water supply can be recovered from the stack and fuel processor exhausts, and requires no independent water source under most weather conditions. This is the concept applied in the 40 kW fuel cell on-site power generation program sponsored by the Gas Research Institute (GRI). Power Conditioning. The fuel cells themselves produce direct current. For most uses, this power is inverted to produce alternating current required by most standard household appliances (60 Hz in the United States). This is accomplished with a solid state inverter. 18.2.2 Technical Characteristics Phosphoric acid fuel cell power plants currently under development. span a range from less than 1 kW to 10 MW. The U.S. gas utility industry through GRI is sponsoring a field demonstration test of a 40 kW on-site fuel cell power plant for supplying total energy requirements of commercial buildings.(4) This power plant was built by .United Technology Corporation and uses a cell stack which operates very near atmospheric pressure (Figure 18.3). Units larger than approximately 100 kW can economically operate at 3-5 atmospheres of pressure. A unit in this size range would be appropriate for application in villages in the Bush. The most significant performance feature of the fuel cell power plant is its energy conversion efficiency. The intrinsic electrical efficiency of a fuel cell is typically in the 80-90% range. However, in operation when current is drawn from the cell, other sources of inefficiency* come into effect. With these effects minimized to the fullest extent, cell stack efficiencies of 55-60% can be achieved; if the inefficiencies of the. fuel processor and system integration are also taken into account, the electrical efficiency falls to about 40%. However, quite often a portion of the wasted heat can be utilized for space heating. In such instances, the overall energy efficiency can approach 80% depending on thermal heat demand. * These inefficiencies are caused by activation and concentration polarizations, and internal electrical resistance losses. (1) 437 60 - HIGH EFFICIENCY AT ALL RATINGS 50 FUEL CELL SYSTEMS 40 EFFICIENCY * 30 PERCENT & GAS TURBINE 10K GASOLINE SYSTEMS . ELECTRIC 1 10 100 1,000 10,000 100,000 POWER OUTPUT ~ KILOWATTS Source: Reference 3. FIGURE 18.4 COMPARATIVE EFFICIENCY OF ELECTRIC POWER SYSTEMS RELATIVE PART LOAD . EFFICIENCY 0.25 20 30 40 50 60 70 80 9:0 100 PERCENT RATED OUTPUT Source: Reference 3. FIGURE 18.5 COMPARATIVE EFFICIENCY OF POWER SYSTEMS AT PART LOAD 438 abet A comparison of fuel cell system efficiency with other generating systems at full-load is shown in Figure 18.4.(3) Diesel electric and gas turbine systems efficiencies approach fuel cells only at the extreme end of their size range. Of equal importance is the part-load efficiency of _ the fuel cell (Figure 18.5).(3) The part-load efficiency of a fuel cell is actually higher than the full-load value down to about 20% of load. This aspect makes a fuel cell attractive in applications where the load variations are large as is the case in Alaska. The performance of the fuel cell power plant will not be adversely affected by the low ambient temperatures in Alaska. The primary cell reaction is exothermic and releases sufficient heat to maintain the stack temperature. Close control of stack temperature is accomplished by adjusting cooling fluid flow designed to remove internally generated heat not used in the cell reaction. The fuel cell efficiency is relatively insensitive to changes in ambient temperature during steady operation. A minor loss in efficiency on the order of 3.5% will occur with reduced ambient temperature due to increased reformer heat losses which are offset by increasing the fuel processor burner rate at lower outdoor temperature. Offsetting this, the more dense air would reduce the parasitic power load associated with air compression. In addition, there could be hardware cost savings due to reduced requirements for air cooled heat exchangers. Fuel cell power plants for on-site power generation will be designed to operate (including normal start-up and shutdown). with minimum operator presence. The control system on the power plant provides fully automatic operation over the complete power range. During the cold start-up sequence, the control function is semi-automatic and the presence of a service-person is required. Shutdown is manual or automatic. Automatic shutdown is based on monitoring power plant out-of-limits conditions. Externally supplied electric power from the electric bus and a separate de power supply unit is required during power plant start to provide power for electric heaters, pumps, and control components. Special procedures must be followed during the initial start-up. (4) The power plant responds automatically to changes in load demand by sensing the power being delivered and metering the fuel flow as required. Output voltage recovery after a step change will occur within two cycles. Overloads above rated power will be maintained for up to five seconds. Shutdown of the fuel cell power plant requires special considerations, especially in the Alaskan environment. Some external source of power is required to provide for an even cooldown and to maintain the cell stacks at 100°F. Below a temperature of 100°F, phosphoric acid electrolyte could solidify in the event of a system power failure. This could result in a slight loss in subsequent stack performance. . About 439 > > o Demineralizer External Condenser Surfaces and Water Cooler Air Filter for Process Air Blower Acid Tank Inverter Air Filter Safety Valves Steam Separator High Grade Heat Exchanger Low Grade Heat Exchanger Water Tank Coolant Filter Charcoal Filter Reformer Source: Reference (4). TABLE 18.1 SCHEDULED MAINTENANCE ITEMS Every Every Every 2000 Hours 4000 Hours 8000 Hours Change Clean Change Change Requires Power 8 Hour Plant Shut Down Every 8000 Hours Change Actuate (local codes apply) Internal Inspection (local codes apply) Clean customer side if Clean necessary . Clean Clean Change Check Gaskets, Thermocouples 20 such events can be tolerated before the cell stack would require replacements. Current fuel cell demonstration units are water cooled. Due to the tight integration of the coolant and process steam systems, it is not practical to introduce antifreeze into the water. A practical design for subarctic conditions would require either an air or hydrocarbon liquid coolant to avoid damage from freezing water. Power plant designs based on air cooling are presently under development by Energy Research Corporation and Westinghouse.(7) Since moisture also exists in the processed fuel, means for automatically purging the system of fuel after shutdown must be provided. This is normally done as part of the safe shutdown procedures. , Maintenance requirements for fuel cells have not yet been verified. The design life goal for mature, commercial power plants with periodic overhaul and maintenance is twenty years with an operational goal of no more than one unscheduled shutdown per year. However, it is anticipated that the GRI field test power plants will be operated for 8000 hours each and these preprototype power plants may experience unscheduled shutdowns in excess of the above goal. (4) Two types of scheduled maintenance are required by the power plant: (1) minor component changes or cleaning which can be accomplished with the power plant operating, and (2) annual procedures which require power plant shutdown. The equipment arrangement inside the power plant is designed to permit efficient and safe maintenance, adjustment, and repair. After every 2000 hours of power plant operation .the demineralizer requires changing, while for every 4000 hours of operation, the air filter for the process air blower must be changed and the external condenser surfaces and water cooler must be cleaned. These three operations can be safely conducted during power plant operation and require minimal training to accomplish. A number of components require checking, cleaning, or changing on an annual basis. The annual maintenance procedure would require an eight hour shutdown of the power plant.(4) Scheduled maintenance items are summarized in Table 18.1. The operating reliability of fuel cells generators is judged to be slightly better than conventional systems based on projected component reliability. The estimated mean time between failures (MTBF) for the major components such as the cell stack and reformer are quite long (Table 18.2). Like other conversion systems, the fuel cell has ‘auxillary pumps, blowers and controls, which generally have shorter MTBF's. However, the fuel cell power plant has fewer pieces of rotating equipment and no reciprocating devices. Presently, there is insufficient field experience to verify the presumed reliability of phosphoric acid fuel cells. : Of more concern is the potential for deterioration from contaminants which could be experienced in remote applications. Exposure to sulfur 441 TABLE 18.2 FUEL COMPONENT MTBF (hours) Cell Stack Reformer Pumps, Blower, and Controls Power Conditioner Source: Reference (5). 442 250,000 55,000 2,630 6,670 } se pee memento” or chloride contaminants introduced from accidental mixing of fuels, incompatible with the fuel processor, during storage will deactivate the fuel processor catalyst fairly rapidly. The poisoned catalyst will not generate sufficient hydrogen to operate the -cell stack. Continued operation with contaminated fuel would eventually ruin the cell stack as well, Another aspect of fuel cells which relates to reliability is that it takes several hours for start-up from a cold condition (100°F). The practical operational solution to this problem is to design a system so that at least two fuel cells are always operating at part load. If one cuts out, the other one can pick up the load. 18.2.3 Environmental Issues A fuel cell generator is quite benign for a fossil-fuel-based conversion system, Results of emissions tests conducted by UTC on a preprototype 40 kW power plant are shown in Table 18.3. These tests were made at steady-state operation on natural gas fuel. The emissions reach a maximum at 50% of load. For all conditions, the emissions are significantly below established state or federal standards for stationary power plants as depicted in Table 18.4. The emission rates of the 40 kW fuel cell power plant also compare very favorably with the emission rates of a domestic gas furnace. At full rated power, the power plant has lower, or equivalent, emission rates than a gas furnace for the emission constituents listed. A phosphoric acid fuel cell generating system produces no contaminated water discharges. Also, it produces no significant quantities of solid wastes, A fuel cell power plant has various fans, blowers, pumps, and other mechanical components that produce noise. The primary noise producers are the process air blower and the condenser fan.(5) The power plant free field noise level at the 4.6m (15 ft) perimeter was measured at 61 dBA at full power operation. The noise level was found to vary little over the output range of the power plant. Based on the 61 dBA reading for one power plant, an installation of two power plants will produce a free field noise level 64 dBA at 4.6m. The power plant noise levels are below the present OSHA exposure limit of 90 dBA for an eight-hour period and the proposed modified limit of 85 dBA. They are substantially quieter than diesel generators of the same capacity. This would allow fuel cells to be located closer to residential and commercial buildings than is possible with diesel generators. In some situations, the fuel cells could be located in large building basements. These factors could potentially reduce the cost of district heating installations based on the use of waste heat from fuel cells as compared with diesels. 443 TABLE 18.3 40 kW FUEL CELL POWER PLANT EXHAUST EMISSIONS IN kg/GJ HEAT INPUT FOR VARIOUS NET OUTPUT LEVELS (ib per million Btu) 0 kW 20 kW ' 38.5 kW Net Power Net Power Net Power NO, . -0027(.0056) -00062( .0013) -00029( .00060) SO, -000017(.000035) -000015( .000032) - 000016 ( .000034) Particulates .000072(.0015) -0010(.0021)° -0000(.0000)* Smoke None None None Total Hydrocarbons -021(.043) -0031( .0065) .0012(.0025) Source: Reference (4). *Possible sampling problem. 444 8 7 Source; Reference (4), 445 TABLE 18.5 PARTICIPANTS IN FUEL CELL DEVELOPMENT : Sponsors Hardware Suppliers U.S. Department of Defense Energy Research Corp. U.S. Department of Energy Engelhard Industries Electric Power Research General Electric Institute Westinghouse Ministry of International Trade and Industry (Japan) United Technology Corp. National Aeronautics and Space Administration 446 Users Electric Utilities Gas Utilities Military Services oa Fuel cell power plant commercialization requires a fuel system for producing and delivering liquid and gaseous fuels. Fuel extraction and processing typically have major environmental consequences. The national fuel production scenario and its resulting impacts should not be measurably expanded or altered by fuel cell deployment, however, since fuel cells should be able to utilize nearly all fuel types and fuel cell deployment will replace the fuel demand of displaced power plants rather than’ creating additional demand. Since fuel cells are more energy efficient than most competing technologies, overall fuel production impacts may lessen as a_é result of fuel _ cell commercialization. (4) . Most of the fuels that are compatible with fuel cells such as natural gas, natural gas liquids, and naphtha, have a= relatively high form-value established by other end-uses not involving electricity generation. Using these fuels for power generation may not constitute the optimum allocation of resources. In fact, the Power Plant and Industrial Fuel Use Act (PL-95-620) discourages the use of natural gas for electricity generation. However, these restrictions apply only to very large plants (larger than 100 million Btu/hr). In general, fuel cells are exempt from these restrictions. (4) 18.2.4 Commercial Status Fuel cell technology has not yet: reached the stage of commercial production. The phosphoric acid fuel is the closest to reaching this milestone. One of the major remaining hurdles is underwriting the cost of a full-scale manufacturing facility in the absence of firm user commitments. , Development of the technology is fairly active involving a number of sponsor/manufacturer/user organizations. A summary of the participants in each category is shown in Table 18.5. The major portion of U.S. funding comes from DOE, EPRI and GRI. Among the hardware suppliers, UTC is the acknowledged leader in terms of ability to demonstrate system performance and capability to deliver assembled power plants. They have built two 4.8 MW demonstrator units for field testing and are supplying forty-eight 40 kW units for the GRI field test program. The commercial development of phosphoric and fuel cell technology has reached the stage where prototype power plants are being field tested at two size ranges. At the large-end are two 4.8 MW demonstrator units being tested at Consolidated Edison Company and Tokyo Electric Company. These units are designed for operation with natural gas or petroleum naphtha. The ConEd unit was undergoing initial start-up in mid-1982 when problems developed in the fuel processor. Modifications are underway and. the unit is expected to be operational in 1983. A second 4.8 MW unit installed at Tokyo Electric was expected to be running by the end of 1982. At the smaller end of the spectrum is the GRI 40 kW program. The goal of this effort is to install 48 units at 447 approximately 24 sites with different applications. The test sites will include residential, commercial, and light industrial applications. The primary GRI interest in the field test is the assessment of the commercial potential of on-site phosphoric acid fuel cell power plants. (4) The major constraint to implementation of the technology is the power plant system cost. Preprototype costs range from $3,000/kW for large scale systems to $12,000/kW for the 40 kW test units. Pilot production units are projected to cost anywhere from $2,000 to $4,000/kW in dollars. The commercialization cost goal is $700-1000/kW. The real crux of the problem is that, in order to achieve this cost goal, the application of assembly-line production methods and the learning associated with such methods is required. However, none of the potential users of fuel cells are sufficiently committed to their deployment to justify private-industry investment in a manufacturing facility. : 18.3 Economic Implications The most plausible application for: fuel cells in Alaska is as power generators for villages in the Bush which are located on navigable water. Presently many of these villages obtain power from the Alaskan Village Electricity Cooperative (AVEC) which operates distributed generation facilities in the village. The generation facility usually consists of three diesel/electric generators. The delivered price of diesel fuel in 1982 dollars is about $1.80/gal or $12.00/million Btu, and the resulting electricity rate is in the vicinity of 27¢/kWh. A fuel cell generator using methanol produced from Alaskan gas might be attractive provided the life cycle costs are competitive with diesels. The potential benefits of the fuel cell systems are: e utilization of domestic resources and decreased dependence on refined petroleum products produced out of state; e efficient resource utilization; @ compatibility with district heating concept; and e compatibility with methanol production for use in vehicles. Other fuel alternatives for fuel cells include natural gas, naphtha (natural gasoline fraction of crude oil) and diesel fuel. Natural gas is not currently available in the Bush and is not expected to be in the near future. Naphtha is not commercially traded or distributed in Alaska and would have to be purchased from the refineries or imported. Diesel or No. 2 fuel oil would. be the preferred fuel cell fuel, however, the availability of demonstrated fuel processor technology is much farther away than for methanol. Given the availability of natural gas in Alaska and the fuel processor development status, methanol appears to 448 be the fuel of choice for initial deployment. For a modest increase in capital cost, the fuel cell could be designed to run on the entire range of fuels from natural gas to naphtha including LPG's and alcohols. r 18.3.1 Costs : The first step in estimating fuel cell system costs is the development of a a conceptual system configuration. This was done by first establishing _ the energy requirements for a representative village. Based on a 1 review of energy demand profiles for a number of Bush villages (2), ~ the energy demands for Kiana were selected and are shown in Table = 18.6. Both electricity and potential district heating thermal demand are presented. . r . During most of the year, two 50 kW fuel cell units would be able to carry the load. The operating units would be running between 70% and 90% of full load. During the winter demand period, the operation of - three such units would be required at certain times to cover the peak _ demand. The overall annual electric load factor based on installed capacity would be about 48%. . Based on the available data on thermal demand, it appears that nearly all of the available waste heat from the power plant could be utilized, except during the mid-summer. months. For the purpose. of the life-cycle cost analysis, waste heat recovery is ~ not included. However, the seasonal electrical and thermal demand i. changes coincide reasonably well, such that most of the available waste ‘ heat could be utilized for building space conditioning. Conceptually, the fuel cell generating system configuration for a typical village would consist of three 50 kW fuel cells fueled with methanol. The methanol fuel would be stored in an atmospheric tank and delivered by pump to the fuel cells at 30-75 psig. The fuel cell generators would ™ be air or liquid (except water) cooled. Provisions for service water . and drains would be required. The generator units would be located in . a building enclosure with a minimum floor space of 1000 square feet. This configuration is the basis for the cost analysis which is presented ‘later. An alternative configuration involving a combination of fuel cell and diesel generators is interesting from the standpoint of operating flexibility. A diesel generator has better fast-start capability than a fuel cell. Hence, it could quickly pick up load if one of the two operating fuel cells were to shut down. It could also provide the power -~ needed to start and shut down the fuel cells. However, it would © : require handling of two fuels at the site which would increase the probability of contaminating the methanol. i The estimated capital cost for fuel cell generator systems is shown in i Table 18.7. Cost estimates are presented for prototype and commercial units. The costs shown are for a three unit (50 kW each) facility. The largest cost element is equipment and materials. The reason is that the factory assembled generator modules will require a minimum of .. = 449 TABLE 18.6 ALASKAN VILLAGE ENERGY DEMAND PROFILE (KIANA) Electricity, kW(electric) District Heat, kW(thermal) Monthly Ratio, kW(t)/kW(e) Annual Power Consumption, kWh Source: Reference (2) -¥Potential peak demand. Peak(D) Minimum(N) Peak(D) Minimun(N) 140 600* Winter 2.0 450 90 N.A. 90 630,000 Summer 0.9 70 48 TABLE 18.7 CAPITAL COST SUMMARY TECHNOLOGY Methanol Fueled, Phosphoric Acid Fuel Cell Pressurized Stack BASIS Location Coastal Village Year 1982 constant dollars CAPACITY Input 132 kW(t) Output Electricity - 50 kW(e) Thermal - 50 kW(t) ESTIMATED USEFUL LIFE 20 years CONSTRUCTION PERIOD 6 month Factory/3 month Field CAPITAL COST, $(000) Prototype Commercial Equipment and Materials* 653 178 _Direct Labor 41 41 Indirect Costs 7 7 Home Office Costs 7 7 Contingency 8 6 TOTAL CAPITAL INVESTMENT 716 239 *Includes delivered modules and site bulk materials. 451 OPERATING COST SUMMARY TECHNOLOGY Pressurized Stack BASIS Location Coastal Village, the Bush Year 1982 CAPACITY Per Unit Input, kW(t) 132 Output, kW(e) 50 kW(t) 50 OPERATING FACTOR 0.48 Unit Cost ($) VARIABLE COSTS Fuel - Methanol Catalysts and Chemicals TOTAL VARIABLE COSTS FIXED COSTS Labor Operating Maintenance Maintenance Materials Taxes and Insurance, 2% Capital Charges, 9.5% TOTAL FIXED COSTS TOTAL ANNUAL OPERATING COSTS TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST TABLE 18.8 0.68/gal 0.1/kg 452 Methanol Fueled, Phosphoric Acid Fuel Cell, Total System Consumption 97,480 3,000 395 150 150 Annual Cost 66,300 300 66,600 12,000 7,200 3,700 4,800 22,700 50,400 117,000 630,000 kWh $0.186/kWh i field installation once they arrive at the site. Note there is a factor of three difference between the prototype and commercial cost projections. The operating and life-cycle cost for a village energy system is shown in Table 18.8. The major variable cost item is-methanol fuel. The unit cost of methanol includes the plant gate price ($0.28/gal) as developed in Chapter 7 and transportation to the Bush ($0.40/gal). The. major fixed costs include labor and capital charges. - Operating - and maintenance labor accounts for 38% of the fixed costs, and the capital charges are 45%. The computed life-cycle cost. is 18.6¢/kWh of electrical energy. The cost to the customer would be somewhat higher allowing for power distribution. This could:be reduced somewhat by installation of a heat recovery system for. building heating. Even -using prototype fuel cell system cost, the estimated life-cycle cost of energy is only 29¢/kWh. Consequently, commercial fuel cell technology appears to have potential for reducing the cost of village power generation and reducing the State's dependence on imported petroleum products. 18.3.2 Socioeconomic Factors The deployment of fuel cell power plants within the state 'would have a marginal direct impact on the current business activity. The modular units would be engineered and manufactured in the Lower 48. Site preparation and hook-up require a relatively small amount of construction work, which could be handled by existing local civil and HVAC contractors. Consequently, fuel cell construction would not - create any additional new jobs when compard to the baseline employment for diesel generators. Once in operation, fuel cells probably would not provide additional net employment but would require upgrading of skills. In terms of direct operating labor, the requirements would be similar to diesels. However, opportunity may exist for an in-state service business to perform routine and annual maintenance on the power plants. At present, only the potential manufacturers have sufficient knowledge of the system to perform maintenance on the power plants. Considering the distance involved, the manufacturer would probably . establish authorized service representative in the state at some level of deployment. The number of persons involved in such activities would probably be relatively small; on the order of 10-20. These jobs would be created at the expense of similar, less skilled jobs for servicing diesel-generators. Indirect benefits attributable to fuel cell deployment would include a pro rata share of the direct benefits identified (in Chapter 7). for production of methanol from Alaskan natural gas. Fuel cell fuel demand would account for about 5-10% of a 2500 tpd methanol plant, assuming deployment in all the villages. Again, these would be displacing indirect benefits related to the production and _ distribution of petroleum-derived fuels. 453 TABLE 18.9 ENERGY UTILIZATION IN VILLAGE POWER GENERATION Fuel Cell vs. Diesel Generator Diesel Generator Excluding Including WHR* WHR_ Annual Electricity Output, kWh/yr 630,000 630,000 Fuel Type Diesel Diesel Fuel Required, mWh/yr 2840 2840 Electrical Efficiency, $(LHV) 22 22 Space Heating Demand, mWh/yr 1020 1020 Waste Heat Available, mWh/yr 2140 2140. Waste Heat Utilized mWh/yr -- 1020 % 0 48 Overall Energy Efficiency, % 22 58 Source: Reference (2), Arthur D, Little Estimates. *Waste heat recovery. 454 Fuel Cell Excluding Including WHR WHR 630 ,000 630 ,000 Methanol Methanol 1620 1620 39 39 1020 1020 480 480 -- 480 0 100 39 69 m 18.4 Impact 18.4.1 Effect on Overall Energy Use and Supply Fuel cell power generation in the Bush would result in measurable energy saving to the villages. A comparison of energy utilization by diesel and fuel. cell generators in village power generation is presented in Table 18.9. The analysis is developed for systems without and with waste heat recovery. For the case without waste heat recovery, the fuel cell system energy efficiency is 39% compared to 22% for the diesel generator. This efficiency difference is equivalent to a 44% reduction in fuel consumption for power generation. In the case with waste heat recovery, the resulting reduction in fuel usage would be about 16%. Perhaps, equally important as fuel. savings is the potential for elimination of liquid: petroleum fuel. The fuel cell generator. can be designed to run on methanol using available technology. The methanol can be economically produced in Alaska from natural gas. The delivered price of methanol in the Bush should be equal to or lower than diesel fuel, if large scale production of methanol for motor vehicles becomes a reality. This will obviously be strongly influenced by _natural gas pricing policy in Alaska relative to petroleum. Initially a dedicated fuel distribution and storage system would be required. This is necessary to avoid contamination of the methanol, for which the consequences would be distruction of the fuel cell. 18.4.2 Future Trends _ The major emphasis in phosphoric acid fuel cell development now and in . the immediate future is to demonstrate the performance and reliability of the technology, and to implement hardware cost reductions through engineering design and application of factory manufacturing methods. Development of other, fuel cell technologies will also continue, with primary emphasis on molten carbonate technology in combination with coal gasification. Currently, the economics of fuel cell power generation in industrialized areas does not look favorable. However, if world petroleum prices resume an upward trend, then the improved energy conversion efficiency available with fuel cells would eventually make them economically attractive. Consequently, the timing for wide-scale deployment of fuel cells in the Lower 48 is believed to be beyond 1990. Yet, it appears from this assessment, that opportunities exist in Alaska for nearer term consideration of this technology for remote power generation. A general need in Alaska is to reduce the cost of electricity generation in rural villages. Fuel cell power plants could play a role in meeting 455 the objective. attractive in this application are: Consequently, high fuel efficency; good part-load efficiency; unattended operation; availability of waste heat; fuel flexibility; and modular construction. basically refine these features. The major features of a fuel cell which make them fuel cell technology as currently configured is quite responsive to the rural Alaskan environment. 456 Further developments will REFERENCES Institute of Gas Technology, Handbook of Fuel Cell Performance, DOE Report Contract No. EC-77-C-03-1545, Chicago, [linois, 1980. Raj Bhargava Associates, Rural Waste Heat Capture and_ District Heating Project, Alaska Power Authority Report, Anchorage, Alaska, . United Technologies Corporation, National Benefits Associated with Commercial Application of Fuel Cell Power Plants, ERDA Report Purchase Order No. WA-76-3405, South Windsor, Connecticut, 1976. Aerospace Corporation, Environmental Assessment of the 40-Kilowatt Fuel Cell System Field Test Operation, DOE Report DOE/NASA/270i-1, El Segundo, California, 1982. Arthur D. Little, Inc., Thermoelectric Generator and Fuel Cell Technolo Growth Potential, U.S. Army MERAD-COM Report, Contract No: DAAK70-79-D-0036, Cambridge, Massachusetts, 1981. United Technologies Corporation, On-site 40-Kilowett Fuel Cell Power Plant Preliminary Model Specification, DOE/GRI Report, Contract No. ET-77-C-03-1471, South Windsor, Connecticut, 1978. J.J. Buggy, Fuel Cell Technology, National Fuel Cell Seminar, Westinghouse Electric Corporation, an 23-25, 1981. 457 19.0 _. OVERVIEW OF RESIDENTIAL AND COMMERCIAL END-USE SYSTEMS : 19.1 Introduction The primary needs requiring: energy in the home and in commercial and institutional buildings include: e@ space heating for health and comfort, including ventilation, humidification, and dehumidification ; e hot water heating for comfort, sanitation, and laundry processing; e cooking; e appliances for convenience, safety, and entertainment; and e lighting for convenience and safety. The development of technology and mass produced equipment to. satisfy these needs has been a major contributor to the modern standard of living. The most rapid advancements have taken place in the quarter century after World War II and have been guided by the availability of plentiful, low-cost, convenient liquid and gaseous fuels. Accordingly, end-use technologies and buildings were designed with relatively little consideration given to energy conservation. Reliability, degree of automation and ability to meet peak needs on demand were of primary interest. . Then, sharpened by the 1973 OPEC oil embargo, depletion of domestic oil and gas reserves, growing dependence on oil imports and increasing energy costs focused national interest on the search for ways to reduce the growth in energy demand. Modifications to older buildings have been made, such as increasing the quality of construction materials and adding insulation. New buildings have been constructed to much more stringent energy-use standards. Consumers of energy have become willing to sacrifice convenience by dressing more warmly and by switching to fully-automated space heating systems to solid~fuel burning radiant stoves. Management. of energy use has increased through better controls (e.g. setback thermostats). In some instances, rate differentials have been implemented to encourage off-peak electricity use. Solar energy systems has begun to play a role through both active and passive systems. This chapter presents an overview of energy use technologies in. the residential and commercial end-use sector by category. Major attention is focused on the major categories of space heating. and 459 water heating. Some additional discussions are provided on other use categories such as energy storage, controls, energy management, lighting, and appliances. 19.2 Comfort and Space Heating Human Comfort Perhaps the most important consideration in determining space heating requirements is human comfort. Human comfort is influenced by the interaction of a variety of physiological as well as physical factors. The primary factor is ambient air temperature (room temperature). In a 70°F room, most normally attired persons feel comfortable because the rate of heat loss from their bodies is in equilibrium (balances) the heat loss produced by their own metabolism. Some persons, particularly older persons, may require higher temperatures to feel comfortable because of differences in their metabolic and circulatory conditions. Numerous other factors also affect the human comfort and the ambient temperature at which this comfort is achieved. .Humidity level affects the rate of evaporative heat loss from skin. Since about one-quarter of a resting person's heat loss is by evaporation, raising the humidity can slow evaporation and permit a drop in ambient temperature with no change in one's perception of comfort. Similarly, wall temperature plays a role in the comfort process. About half of one's heat loss is by radiation to the surrounding walls. If wall temperature is low, radiation hastens and comfort deteriorates unless ambient temperature is increased. - Therefore, well insulated walls not only retain the warmth of room air but also retard radiation cooling. Well selected wall coverings, e.g. light colors with some texture can have a beneficial effect on comfort. Other factors which impact human comfort and ultimately heating costs are: e presence of air currents (drafts) which cause cooling by convection; . e the amount and effectiveness of clothing which acts to insulate the body from heat loss to the room; and e activity level which is a prime determinant of’ metabolic rate and hence body heat production. System Considerations Space heating is an integrated system of factors which affect the building heat loss and gain: infiltration, transmission, solar gain, occupants, and equipment. Infiltration refers to the passage of 460 outside air into a building through cracks and clearances, particularly around doors and _ windows. Air infiltration through poorly constructed walls, both wood frame and masonry may be significant, particularly during high winds. Transmission, or conduction, refers to the amount of heat transmitted from a building. to cold outside air. The amount of heat flow caused by transmission depends on the difference between indoor and outdoor . temperatures. The rate of transmission depends on the composition of building materials and how much insulation is used. Ventilation may be introduced deliberately to remove smoke, odors, and products of combustion or metabolism and to supply fresh oxygen where natural ventilation. (opening and closing of doors and windows) and infiltration are not sufficient. The effect of deliberate ventilation on energy consumption is greater than that of infiltration since the loss due to the temperature change is augmented by the blower work necessary to move air into and out of the conditioned space. In tight buildings in cold climates, air-to-air heat exchangers offer the potential to save energy by preheating incoming outside air as it cools warm outgoing stale air. Solar heat gain through: building walls depends on the geographical location, intensity and direction of rays, the construction material, and the color and texture of walls and the area of glazed openings in the walls. In Alaska, the solar contribution is smaller in winter than in temperate climates, but in late fall and early spring may be _ significant. Lighting contributes directly to a building's heat gain in direct proportion to the wattage of lamps used. “However, lights are not generally placed to provide heat efficiently and electrical heating is usually more expensive than other methods. People and equipment also contribute heat. People also affect the moisture content in the air through the process of perspiration’ and respiration. Similarly, many human activities, bathing; cooking, laundry, ete., add. humidity to the air. - Conservation measures which can reduce energy demands can be implemented in the following areas: . e Setback controls - setback thermostats reduce energy demand during periods when buildings or zones are normally unoccupied or during sleeping hours when lower temperatures can be more easily tolerated. e Ventilation - in air systems, filters and chemical absorbers ean be used to reduce ventilation requirements. Nighttime setbacks and shutdowns can also result in substantial _ Savings. : 461 Heat Exchanger Gas Burner Air Blower FIGURE 19.1 CONVENTIONAL GAS FURNACE 462 e Infiltration - infiltration can be reduced by improving the fit of doors, caulking and weatherstripping, the use of vestibules, plastering or otherwise sealing porous walls, and external windbreakers. Windbreakers can be_ structural additions on or near the building, carefully designed ground contours or selective plantings. A closely spaced line of coniferous trees makes an excellent screen. e Insulation - the use of drapes, storm windows, and wall, floor, and ceiling insulation is a very effective way to reduce . demand. e Clothing - wearing warm clothing and accepting a lower temperature will reduce demand. e Radiant heat - for example, small, low-wattage quartz heaters are available to provide radiant heat to a stationary person who as a-result feels more comfortable in lower ambient air temperatures. Various residential and commercial heating technologies are briefly described below: conventional and condensing boilers and furnaces, heat pumps, solar systems, and solid fuel stoves. Oil and Gas Systems Natural gas, propane, and oil burning devices are the most common residential space heating fuels in Alaska. Natural gas-fired systems are the predominant heating systems employed in the Anchorage area. Oil-fired boilers with hot water baseboard distribution are the predominant systems used in other parts of the state. Oil- and gas-fired units can be furnaces or boilers. A furnace (Figure 19.1) is a unit delivering warm air which is distributed throughout the building while a boiler delivers hot water (or steam) which is used as the distribution medium. They both convert fuel to useful space heat at about 65-80% efficiency (steady-state efficiency). Hot water systems are more readily zoned to heat different areas of the residence independently. This is because changes in distribution and the water flow rate have little effect on the thermal performance of the boiler while redistribution and reduced -air flow in a hot air zone system will adversely affect the furnace efficiency. Warm air systems are typically less expensive to install and they offer the distinct advantage that, if a power outage or component. failure causes the central unit to stop operating, they are not subject to bursting pipes as a result of boiler water freeze up. 463 ” / OUTDOOR TEMPERATURE} CHARGE Cd SENSOR TIME CONTROLLER BRICK CORE DAMPER Source: Arthur D. Little, Inc. FIGURE 19.2 ROOM STORAGE HEATER : 464 ELECTRICAL DISTRIBUTION PANEL INSULATION HEATING ELEMENTS ROOM THERMOSTAT Warm air systems also tend to have lower standby losses than boilers. During the standby condition, boiler water is normally maintained at the design condition (140-160°F), and heat is lost up the flue by natural draft, as warm. indoor air rises through the chimney because of buoyancy to be above the colder, more dense outdoor air. This can cause seasonal efficiency to drop 10-15% below steady state efficiency. Flue dampers.can reduce the flue loss by closing during standby and preventing this unwanted natural draft. Boilers and furnaces can be designed to recover as much as 95% (seasonally) of the fossil fuel energy for useful space heating. At recovery efficiencies above approximately. 85%, the water vapor in the products of combustion condense and special heat exchangers are required to manage the condensate for the high efficiency units. While these units are more expensive than conventional gas or oil furnaces/boilers, the energy cost savings are substantial. Also, these condensing units do not require chimneys to provide proper draft and can save on this cost in new installations. Electrical Resistance Heating Electrical resistance heating uses electrical energy flow through a resistive circuit to release heat to the room. Usually, these devices are mounted in. baseboard. units and are operated by thermostatic controls in’ each room. Resistance heating is inexpensive to install because there is no central unit and need for circulating the heating medium; the distribution is electrical (wires). Also resistance heating is 100% efficient in terms of purchased power converted to useful heat. However, since power production from fuel is only 30-40% efficient at the electrical utility's central station, resistance heating tend to be more expensive overall than fuel-fired heating. In some states, electric utilities have instituted time-of-day rates to discourage use of electricity during peak periods. With such a structure, peak rates are higher than average and off-peak rates lower than average. Where the rate spread is wide enough it can encourage the user to invest in a thermal storage system’ (Figure 19.2) which is charged during off-peak periods and discharged as needed. , . Heat Pumps Electric heat pumps offer the potential of electrically heating at greater than 100% efficiency. Heat pumps are refrigeration systems in which the useful output is the warm air produced by the condenser rather than the cool air generated by the evaporator. In a heat pump (Figure 19.3), the ‘refrigerant compressor increases pressure and therefore the temperature at which. the refrigerant transfers its heat to the condenser cooling fluid (room air). The condensed refrigerant is then passed through an expansion valve which lowers 465 Fluid Pressure (psia) Expansion Device Compressor (Input Energy W) ~Refrigerant Path Heat Evaporator 1000 800 600 <— Saturated Liquid Line Two-Phase - id 400 Gas & Liqu 300 260 200 Gas Condensation . @ T = 117°F Expansion 100 80 60 Evaporation 1 SS 40 Compression 20 20 42.5 0 4t Enthalpy (h) Btu Source: Arthur D. Little, Inc., estimates based on a 45°F outdoor temperature and a 20°F air-to-heat exchanger temperature difference, standard fluid properties of refrigerant 22, and an ideal gas compression process. Derived from General Electric sales literature entitled ‘General Performance Curves for Packaged Units,” July 1973. FIGURE 19.3 TYPICAL FLUID STATES DURING HEAT PUMP HEATING MODE 466 the pressure and, therefore, the temperature at which heat flows from the cold region (e.g. outdoor air) into the refrigerant. The electrical energy consumption is thus determined by the temperature range over which the system operates. The wider the range, the lower the coefficient of performance (COP), which is the ratio of heat supplied to electrical energy input. Heat pumps most frequently use outside air (air source heat pumps), because they tend to be least expensive to install. However, below freezing temperatures, the COP drops rapidly to a point where the heat pump -hardly performs better than electrical resistance heating. Therefore, in cold climates where the high COP's cannot be expected with air source heat pumps for much of the heating season, alternate heat sources may be groundwater or the earth itself; these sources vary in temperature over a much narrower range over the year than outside air. However, they are more expensive to install, requiring buried coils or groundwater pumping. In some parts of Alaska, permafrost may prohibit consideration of such options. Heat-actuated heat pumps are being developed. for use in residential and commercial applications. These heat pumps can be gas, oil, or solid fuel-fired and also offer the potential of space heating at an efficiency of greater than 100%. They operate like standard electric heat pumps, but the electric motor is replaced by an _ internal combustion gas-fired engine with the refrigeration equipment approximately the same as the heat pump cycle described above. More advanced heat pump systems, such as the absorption heat pump (Figure 19.4) in which two fluids are used in an arrangement in which heat can be applied to separate the fluids causing one fluid (refrigerant) to circulate through the refrigeration cycle, are under development and appear to offer efficiencies greater than 100% though substantial development work still lies ahead. Other heat-actuated heat pumps, including a reverse Brayton cycle, are under development, moving close to field demonstration. No heat-actuated heat pump is widely available for commercial distribution. Several European natural gas-fired internal combustion heat pumps have been developed, but these are not readily available in the marketplace. Solid Fuel Equipment Wood and other solid fuel stoves are frequently used as a secondary space heating unit where the user obtains the tangible benefit of a lower fuel cost and/or intangible benefits: the aesthetics of an active fire in a decorative heating device, the sense of energy security, or a hands-on opportunity to size and handle solid fuel. Solid fuel burning stoves do not offer the convenience of automatic operation since they require stoking and periodic flue cleanings to remove creosote buildup. However, they are simple in design and operation. Fuel is burned in a firebox and the combustion is controlled by limiting the flow of the inlet air. The key is to limit the rate of 467 Heat of evaporation + heat of solution Heat of evaporation (refrigerating effect) Heat of condensation + heat of solution Symbo! Typical System Ammonia R — refers to refrigerant Lithium Bromide C — refers to absorbent FIGURE 19.4 A SIMPLE ABSORPTION REFRIGERATION SYSTEM 468 combustion so that not too much hot combustion gas can escape up the flue wastefully. In addition, there are central units which offer multifuel capability as well as automatic fuel firing and thermostatic control. Solar Energy Solar energy may be collected to make a contribution to the annual heating requirement, and reduce fuel costs. An active solar system involves circulating a fluid, usually water or an antifreeze mixture, through the coils of a collector located on or near the building. The heat is stored in a water tank or rockbed and discharged to the living space using a combination of pumps and/or fans and _ heat exchangers. Passive solar heating uses the building itself as the collector by appropriate placement of glazing. Storage in the form of thermal mass, usually masonry materials or concrete incorporated into the architectural design of the building, prevents overheating during sunny days and assures continued heat availability at night. Movable night insulation, which can be shutters, thermal draperies, or other materials, are required to conserve energy at night, on cloudy days, or on especially cold days when solar gain is smaller than heat loss. Solar energy is less available in the coldest winter months in Alaska than in more southerly locations. Therefore, the contribution of solar energy as a fraction of the total annual heating requirement cannot practically be as large. However, in absolute energy terms (i.e., Btu per year), the saving can be even larger because the total heating load in most of Alaska is greater than that in most of the Lower 48. Commercial Buildings Commercial buildings often require some space cooling in the internal or interior spaces which are surrounded by heated spaces. The multi-zone system must deliver air for heating as well as cooling in these types of buildings. This can be accomplished through the use of multiple duct air supplies in which warm and cool air are provided and a separate zone-controlled mixing system determines’ the appropriate amount of each stream to meet the space conditioning demand for that zone. All-air systems distribute hot and cold air in ducts throughout the building. The local (zone) heating and cooling requirements established by local thermostats are used to control dampers or electric heaters to adjust the mixed air temperature to meet the local space conditioning demand. In a _ reheat system, cool (air conditioned) air is distributed throughout the building and is locally reheated to the desired temperature. The reheat system is not an 469 Source: A.O, Smith Corp. Inlet, Outlet Fittings Insulation Glass-Lined Steel Tank Dip Tube Anode Flue Baffle Thermostat Burner Pilot Figure 19.5 Gas-Fired Water Heater 470 wee ; i energy efficient system and would not be a good choice for Alaskan buildings with their severe winters. The variable air volume (VAV) system controls individual zone temperatures by throttling the amount of heated air supplied to each zone. The dual duct system mixes the desired amounts of hot and cold air locally to achieve the desired air temperature to be delivered to the space. While these two systems are more complicated and expensive than reheat systems, they are more suited to the high heating demand in Alaska. Water systems (either air-water or all water) distribute hot and cold water in pipes. Fan coil units, which have both hot and cold water supplies, cycle the fan to achieve the desired room temperature by blowing air across coils and transferring heat accordingly. A system employed principally in hotels and motels where occupancy is highly varied is the local package system in which space heating is provided by an electrical resistance heater controlled locally to the appropriate temperature and master-controlled from a central system depending on occupancy. These types of system can have automatic setback or other energy-conserving functions operated from a central control to reduce the heating requirements when there is _ not occupancy. 19.3 Water Heating As in the case of space heating, water heating can be accomplished through the use of propane, electricity, and solid fuels. Oil is a more prevalent domestic water heating fuel in the remote Alaskan locations while electricity and gas water heating are more prevalent in urban areas. Flame-fired water heating involves the combustion of the fuel beneath the reservoir of water as shown in Figure 19.5 with the hot gases passing through a flue either in the center or around the tank. These systems have substantial standby losses when the combustor is off. Air continues to rise through the flue during the standby condition, drawing valuable heat from the stored water and possibly the heated space nearby, exhausting it into the atmosphere. Flue dampers for water heaters have begun to be employed, and a large number of gas water heaters use flue dampers today in Alaska. Electric water heaters have the advantage of being flueless and can be heavily insulated to reduce the standby loss to 5-15%, depending on the level of water demand. Where time-of-day rates are employed, electric water heating is accomplished preferentially during lower cost off-peak periods. Commercially available electric heat pump water heaters have water heating efficiencies above 100% as in the case of space heating. These systems generally draw heat from the surrounding air (which 471 GLP integral Collector Storage Forced Circulation Thermosyphon FIGURE 19.6 MAJOR TYPES OF COLLECTOR-STORAGE INTEGRATION so ai. during the heating season is first heated by the space heating system) to heat the water supply to the desired temperature. The benefit of this technology in Alaska is relatively less than in warmer climates where they can be operated effectively drawing heat from the outdoor air, rather than from air which must have been first heated by a central space heater. Solar water heating offers a means of saving fuel by heating water with renewable solar energy. A properly oriented collector (Figure 19.6) captures incident radiation which heats a fluid passing through its coils. The fluid may be the domestic water itself, as shown in the figure, or an intermediate medium such as an antifreeze solution which transmits heat to the domestic water through a heat exchanger. Since solar energy is available for limited periods, the heated water must be stored until it is needed later. In Alaska, solar intensity and duration during the weeks about the winter solstice is very limited while heat losses from the collector are large. With the strong possibility of freeze-ups, this implies that the most cost-effective solar water heating system for Alaska may be one which is only intended to operate during the warm weather months when solar availability and intensity are greatest and freeze protection is not required. : 19.4 Refrigeration and Cooking Residential and commercial refrigeration is normally accomplished with electric-driven refrigeration units based on _ standard vapor compression principals already discussed under heat pumps (Figure 19.3). These systems normally use a reciprocating or vane type rotary compressor to compress the refrigerant gas to a high pressure. It is then passed through a heat exchanger (condenser) which removes heat, thereby condensing the high pressure gases to high pressure liquid. This high pressure liquid is expanded through a simple valve arrangement which causes it to drop precipitously in temperature, so that it can be circulated to the zone which requires cooling. Next the refrigerant returns to the compressor to repeat the cycle. Propane and natural gas driven refrigeration cycles have been employed in remote locations and are based on absorption refrigeration technology. Figure 19.4 has shown a typical absorption refrigeration cycle based.on ammonia-water. The mixture of ammonia and water in the boiler section is heated to a point where the ammonia is released from the water at relatively high pressure and circulates through a condenser valve and refrigeration heat exchanger exactly as in a vapor compression cycle, whereupon it is returned to a vessel where it is reabsorbed by the water. The generator section shown in Figure 19.4, in combination with the absorber section, acts like the compressor of a vapor compression cycle creating the pressure differential which drives the refrigerant through the condenser valve and evaporator. 473 Cooking is accomplished with electricity, natural gas, or propane. Propane units are typically modified natural gas units where the gas valve and the burner are adjusted for complete propane combustion. 19.5 Energy Storage Several types of energy storage are available, some being employed already in Alaska, and others not yet introduced offering substantial potential for energy cost savings. The conventional storage water heater is common in Alaska and offers a means of storing substantial amounts of energy for meeting peak demands while requiring relatively low power levels to maintain. Diurnal space heating storage has found a little penetration in Alaska and the Lower 48, though it is commonplace in Europe. These systems are used principally with electric resistance heat and allow the owner to draw electric power during the off-peak hours of the electric utility and store the energy locally for use to meet the space heating demand. A variety of materials are used for the off-peak electric energy storage,. principally ceramics and other high temperature solids. Seasonal storage of energy has had some prototype testing in the Lower 48 and offers potential savings in locations where heating and cooling energy demands are approximately equal, annually. The seasonal storage concept simply causes waste heat from the air conditioning process to be stored for use in the winter season. Clearly, this type of storage makes little, if any, sense in Alaska where there is no cooling requirement. 19.6 Controls and Energy Management Control systems and energy management are .probably the least understood and most overlooked technology for end use energy systems. Proper thermostatic control for space heating in typical residences is rarely achieved. Proper adjustment .and location of modern thermostats is essential, including anticipator adjustments. The thermostatic anticipator is designed to compensate for the thermal capacitance of the heating system in the building; it terminates the heating cycle prior to the thermostat achieving the set point since the thermal capacitance of the space heating and distribution system will be sufficient to carry the room temperature up to the desired set point as a result of the stored energy in the system. Control systems to segregate the heating heating requirements of different zones in residential application are infrequent and largely overlooked. Most controls companies offer zone -control systems for both hydronic and all-air systems; however, most regional and local 474 distributors are unfamiliar with the design, installation, and application of these energy-saving technologies, with the result that they are not widely applied. Automatic setbacks and shutdowns of equipment ventilation and heating systems are a very effective means for reducing demand. Additionally, gap and floating thermal envelopes (versus fixed point control) have produced major savings in large complex structures containing areas with different heating needs. Self-heating can be accomplished by the flow of energy from warm to cool areas without specific hardware to effect the transfer. In addition to energy conservation, controls are also used to move electric demands from peak times in order to improve the central utility load factors. Simple time of day water heater controls have been employed throughout the Lower 48 states and some specific applications in Alaska. These systems terminate the electric heater eycle during specific peak electric hours while allowing the electric storage water heater to charge itself with hot water during offpeak hours. Little, if any, performance reduction is noticed by the user while substantial improvements in load management for the utilities are achieved. More complex load shedding equipment is becoming available for residential applications, though they are not in wide distribution in the United States. The user can often share in the utility's benefit of wiring such systems if a time-of-day rate structure is employed to encourage such participation. Load shedding equipment for commercial and industrial systems are rather widespread as a result of the substantial cost benefits to the commercial and industrial users. These systems shed electric loads during peak hours on a prioritized basis. 19.7 Lighting and Appliances Lighting levels are frequently high and can be reduced with no loss of comfort or efficiency. Unnecessarily high light intensities have been sold on the basis that the heat as well as light is provided. Lights are usually not placed to provide heat efficiently and electrical resistance heating is one of the most expensive methods of providing heat. Replacement of general lighting with task lighting, accompanied by reduction of light intensity in noncritical areas, can amount to substantial energy savings. Energy demands for lighting, television, and laundry are highly varied in Alaska. In remote locations, laundry facilities are generally performed at a central community location, while television and other electronic and lighting functions are, of course, demands within the residence. 475 New solid state electronics has dramatically reduced the electric requirements for televisions and other entertainment electronics, while also reducing the first cost. Most new receivers are inherently energy-efficient as compared to the nonsolid state predecessors of the early 1970's. 476 20.0 HEAT PUMPS 20.1 Introduction and Summary 20.1.1 Technical Overview A heat pump is a thermodynamic refrigeration cycle machine which moves heat from a low-temperature source to a higher-temperature sink by the addition of work. When low temperatures are desired, as in air conditioning or refrigeration, the heat pump removes unwanted heat and expels it to a higher temperature sink (such as warm outside air or water). When high temperatures are desired, the heat pump supplies this energy by drawing it from a low-temperature source (usually cold outside air or water). Thus, there is little difference between the operation of a refrigerator and a heat pump; the difference is entirely in the purpose of the process. ‘ In many applications, including most residential and commercial uses, the same machine may at one time operate as an air conditioner, and at an another time as a heater. In some industrial applications, the heat pump may be used for cooling at one temperature level and simultaneously for heating at another temperature. For instance, in a brewery in Florida,(1) a heat pump is being considered to remove heat (i.e., refrigerate) from the fermentation vats and "pump" this heat up to heat water to be used elsewhere in the facility. In large commercial buildings, heat pumps have been used’ to "pump" heat from the warm side of the building (sun load or internal heat generation) to the colder side of the building. The predominant use of heat pumps sold for residential and commercial purposes is for space heating in the winter and space cooling in the summer. During the winter, heat is extracted from outdoor air and “pumped up" to the desired temperature levels for. heating indoor air; during the summer, where the climate requires it, the heat pump cycle can be reversed and operated as a conventional air conditioner wherein heat is removed from the indoor space and rejected to higher temperature outdoor air. Figure 20.1 shows a typical residential heat pump installation. An outdoor section (it may be round or rectangular) draws heat from the ambient air during the heating season and supplies the heat in the form of compressed refrigerated fluid through 3/8" to 3/4" diameter copper tubing to the indoor section which contains another heat exchanger, indoor circulating fan and controls. Electric heat pumps for a heating only function have not made inroads in the U.S. and have met only limited applications Europe where the economics for electric heating are somewhat more favorable. Electric heat pumps are generally chosen for applications with a heating and cooling seasons to take advantage of the dual mode capability of the heat pump. Otherwise,. the heat pump rarely offers an economic advantage over conventional heating systems. 477 INDOOR UNIT “ELECTRIC HEAT <1 B z.\ § LIQUID LINE re | (| 6 lay _/ : \ \__/ \_/ \_/ \/. ELECTRIC HEATER TAT VAPOR LINE THERMOS: CONDENSATE PAN t AUXILIARY DISCONN FUSED 478 e 5 1 irr} x 3 TYPICAL RESIDENTIAL HEAT PUMP SYSTEM FIGURE 20.1 Electric energy used to run the heat pump also is provided to the indoor environment, supplementing the amount of heat taken from the outside. As such, the heat pump offers the potential of heating a space with electricity at an efficiency (due to the heat drawn from the outside air) greater than 100%. The ability to draw heat from the ambient air decreases as the ambient temperature decreases. Both the efficiency and the heating rate of the refrigeration cycle decline with outdoor temperature. For most practical systems, a _ point (balance point) is reached when auxiliary electric power is required to supplement the heating system. During heating in temperate conditions (outdoor temperature above 40°F) when the heat pump cycle is most efficient, frost will tend to accumulate on the outside heat exchanger as it is colder than the ambient air. Means for defrosting of the outside heat exchanger are required. Normally this is accomplished by reversing’ the refrigeration cycle and drawing heat from the building onto the outdoor heat exchanger, thus reducing the effective heating. Because of its ability to move heat "uphill" from a low-temperature source to a higher temperature, the heat pump can also be used as a heat recovery unit. In a 1977 study for ERDA,(2) a heat pump was incorporated into a residential drain water heat-recovery system to supply heat to a hot water tank. Large (5 million Btu/hr) heat pumps are employed in industrial processes to recover waste heat (at 90°F) from liquids for water heating up to 200°F). Another use of heat pumps has been in the annual cycle energy storage (ACES) concept(3) in which a heat pump is used to draw heat from a large container of water during the winter for use in heating the house, and this large chilled container is then used for space cooling in the summer. There is a variety of electric heat pump configurations, designed to draw heat from other sources, such as ground heat or groundwater heat. The use of these is highly site specific, clearly depending on the availability of groundwater. 20.1.2 Alaskan Perspective As indicated throughout this discussion, the use of air source heat pumps is very dependent on the ambient operating temperature and the size of the heat pump compared to the thermal load. The average heating season temperature in Alaska is sufficiently low that normal sizing practices employed in the Lower 48 which would establish a balance point at approximately 35° would not probably be applicable in Alaska, and lower balance ‘points (larger ratio of heat pump to building size) would be employed. As a consequence, heat pump installations in Alaska will tend to require larger and more expensive heat pumps than those used in comparable size buildings in the Lower 48. 479 Heat Transfer Heat Transfer FIGURE 20.2 Heat Sink (heat pump) Heat Source A THERMODYNAMIC SYSTEM ACTING AS A HEAT PUMP 480 Another relative disadvantage of heat pumps in Alaska versus much of the Lower 48 is that, in the contiguous states, the heat pump can also serve as an air conditioner in summer. Therefore, the value of the air conditioner would be effectively taken as a credit. A heat pump is more expensive to install and operate in Alaska than a gas or oil furnace. In certain situations where gas or oil are either unavailable or not suitable for the space heating application, there are electric air source heat pumps which can be competitive with electric baseboard heating, though the payback periods (operating time required to offset higher initial cost of a heat pump) may be quite long. Under these circumstances a heating only heat pump in Alaska would be quite different than a conventional air source heat pump in that the compressor should be located in the indoor section to recover jacket heat for it. If gas and oil fuel supplies are available at average Alaskan costs and the gas and oil heating equipment is available through normal distribution channels, then electric heat pump installations probably are not economical. 20.1.3 Significance of the Technology The electric heat pump is quite practical and fairly well developed for use in temperate climates (average outdoor temperature during the heating season approximately 4 °F). There have been a number of prototype developments of northern climate heat pumps designed to operate in. colder environments, and these would enhance the attractiveness of such a heat pump in Alaska. However, the heat pumps would have a higher first cost since multiple stage compressors and larger heat exchangers would be required. A gas-fired heat pump may be considered for Anchorage where the ratio of natural gas costs to electricity costs appear appropriate. These technologies are in the development stage and are not likely to produce either a residential or commercial device for two to five years. 20.2 Description of .Technology 20.2.1 Operating Principles Figure 20.2 shows the thermodynamic system. The absolute temperature of the source is T, degrees, and the heat extracted from the source is Q,. The energy rejected to the sink at temperature T degrees is Q,,.. Both effects are accompanied by the input of work, W. From the first law of thermodynamics: Qy = Qy + Ws 481 Expansion Device Compressor : \. (Input Energy W) Condenser ~Refrigerant Path Heat Evaporator 1000 800 rm { - 600 Two-Phase | “—— Saturated Liquid ts! - Liquid Line , 400 Gas & Liqu: L : rm. as hoe 300 Condensation 2 ~ 260 7 g @ T = 117°F eo ‘a 200 Dk a Q a 7 Expansion \ % 100 € 80 Evaporation . 3 60 - & —S i! 40 Compression 20 20 42.5 0 41 Enthalpy (nh) Se Source: Arthur D. Little, Inc., estimates based on a 45°F outdoor temperature and a 20°F air-to-heat exchanger a temperature difference, standard fluid properties of refrigerant 22, and an ideal gas compression process. Derived from General Electric sales literature entitled ‘General Performance Curves for Packaged Units,” = July 1973. FIGURE 20.3 TYPICAL FLUID STATES DURING HEAT PUMP HEATING MODE f | / 482 that is, the heat output, Q,,, is the sum of that heat extracted from the cold region plus the energy added to the heat pump. Therefore, the total heat output will always be greater than the energy required to run the device, and thus the heating coefficient of performance (COP) will always be greater than unity: COP p = Qy/W >1. It is important to note that sometimes a portion of the heat, Q,,, is not delivered to the desired space. For example, when outdoor compressor units are used, some of the energy available for space heating is lost before it can enter the house. Under these conditions, the actual COP based on delivered Q,, can be less than 1. If a heat only heat pump were developed for use in Alaska, the compressor should be located in the indoor section to minimize heat loss. Conversely, heat removed in air conditioning or refrigeration is less than the heat rejected, and therefore, the COP of a heat pump in the cooling mode, COP, Ga» is less than when it is used for heating. It can be shown that COP AG + Q,/W = COPip 1. Vapor cycles utilizing mechanical compression are most common for heat pumps. The majority of heat pumps used in the United States are refrigeration cycle machines which typically use liquid-vapor refrigerant as the working fluid in a vapor-compression cycle. The primary components of a vapor-compression cycle and a typical set of fluid states during cycle operation are shown in Figure 20.3. Compressed vapor is liquefied in the condenser and subcooled to a temperature 5° to 10°F less than the condensing temperature. The subcooled liquid is expanded through a flow restriction (a valve or capillary tube), creating a two-phase, low-pressure, cold fluid which passes through the evaporator. There heat is drawn from the environment to be cooled, evaporating the cold liquid so that gas may be returned for compression. Compressors used in_ residential applications are typically positive displacement devices and are capable of providing compressed gas over a wide range of pressures. Whether the heat pump application is conventional air-to-air or water-to-air or a heat-recovery unit, the heat pump performance is highly dependent on the two temperatures between which it operates. For a typical heat pump during the heating season, the following performance variations with outdoor and indoor temperature variations were found by Kelley: (9) e +1.1% increase in COP per 1°F increase in outdoor temperature; e +1.3% increase in COP per 1°F decrease in return indoor air temperature. 483 cop 12 Ideal Heat Pump = ° Th 100°F = - ° T. T) 10°F 10 L Commercial Heat Pump 2 Ty CF) COP denotes the coefficient of performance (COP) for a heat pump (HP) during the heating mode (H). Source: Reference 9 for commercial heat pump. FIGURE 20.4 COP VERSUS Toyrgip¢(To) 484 This dependence on temperature is common for all thermodynamic refrigeration cycles, for which an "ideal" thermodynamic efficiency ean be defined. The French physicist Carnot hypothesized a perfect thermodynamic cycle using perfect machines for pumping heat between two temperature sources. -For this cycle the upper limit of the coefficient of performance for the heating function is found to be: Ty THT, COP = Absolute temperatures are used ss or Kelvin) for the high temperature, T,,, and low temperature, The maximum Carnot (ideal) efficiency, assuming ideal heat Srchslizers, is shown in Figure 20.4, along with Kelley's data representing the. actual measured efficiency of a particular heat pump presently sold for residential air-to-air application. The deviation between the ideal and the actual is due to three real machine constraints: '@ finite size of heat exchangers, which causes. the heat pump to operate between greater temperature differences than the . differential between room temperature and - outdoor temperature, in order to transfer the heat pumped; e fan power requirements which do not add to the heating capacity but add to the power requirement; and @ nonideal compressor efficiencies (typically around 60-70% isentropic efficiency). The nonideal heat exchangers with limited size cause the refrigeration unit to reach elevated temperatures on the condensing side in order to reject the heat and depressed temperatures on the evaporator side to absorb the heat. Figure 20.5 shows the effect of the real heat exchanger capacity on the system performance when compared to an ideal heat exchanger.* If the nonideal features of a compressor are added, the final COP drops to the level shown in Figure 20.6. Aliso shown in this figure is the range of reported COPs for air-to-air heat pumps presently sold in the United States. *The ideal heat exchanger used here is one in which a constant temperature difference is maintained between the refrigerant and the air temperature. Maintaining constant temperature would in actuality require a variable-flow-rate fan that just matches the heating capacity with the airflow volume. 485 coP to 9 8 Ideal Heat Exchanger and No Fan Power; Ideal Compressor 7 6 5 Real Heat Exchangers and 4 Real Fan; Ideal Compressor 3 2 po -20 0 20 40 60 * Outside Air Temp °F Source: Arthur D. Little, inc., calculations based on a typical unit performance. FIGURE 20.5 EFFECT OF REAL HEAT EXCHANGERS ON HEAT PUMP COP (FOR 70°F INDOOR TEMPERATURE) 486 Real Heat Exchanger and Fan, Idesl Compressor a Screening Range of a° Published Data “~w,, | i 2 8 * £ { Real Heat Exchanger [y - and Fan - Real Compressor bi ARI ARI Temperature gltigh-Tenperature 1 Rating Point Rating Point = : 0 10 20 30 40 50 60 Outdoor Temperature (°F) KEY TO FIGURE: HEAT PUMP CAPACITY (Btu/hr) , - © WESTINGHOUSE HLO 30C 24 TONSPLIT 30,000 A © SOUTHWEST SHP 251 2'%2 TON SPLIT 34,000 -- & TRANE RPHA 40 4 TON SPLIT 45,000 ; O TRANE SPHC 506 5'/2 TON SPLIT 65,000 *® Air Conditioning and Refrigeration Institute. - Source: Arthur D. Little, Inc., calculations and manufacturers’ product literature as follows: : Westinghouse data from “An Investigation. of Methods.to Improve Heat Pump Performance and Reliability in a Northern Climate,” Final Report, Vol. 11, App. A, prepared by Electric Power Research Institute, Report No. EPRI EM-319. Southwest data from newspaper advertisement (date unknown). Trane data from untitled 1977 product literature on models SPHC 506 and RPHA 40. FIGURE 20.6 EFFECT OF COMPRESSOR EFFICIENCY ON HEAT PUMP PERFORMANCE 487 TABLE 20.1 CHARACTERISTICS OF A STANDARD ELECTRIC AIR SOURCE HEAT PUMP HEATING CAP. (BTUH/1000) AT TOTAL POWER IN K.W. AT ; CORRECTION FACTORS — OTHER AIRFLOWS AQDICATED INDOOR D.B. TEMP. INDICATED INDOOR 0.8. TEMP. | (Value at 1350 CFM Times Corr. Factor - Value at New Airflow) 92 84 8.0 75 10.3 94 89 8.5 AIRFLOW 1200 1350 1500 11.5 10.5 10.0 9.5 HEATING CAP. x0.980 x1.000 *1.020 13.0 11.9 11.4 10.8 COMPR. KW *1.025 x1.000 x0.975 47° 135 129° «123 VALUES AT ARI RATING CONDITIONS OF: 165 . 152 146 14.0 47/43-70 (High Temp. Point) 17/18-70 (Low Temp. Point) 185 | (17.1 16.4 15.8 206 192. 185177 AIRFLOW = 1275 CFM 231 216 20.8 20.0 HEATING CAP. (High Temp.) = 34000 BTUH 257 2a 23.3 22.4 HEATING CAP. (Low Temp.) = 19000 BTUH 285 26.7 25.9 25.0 COMPR. POWER (High Temp.) = 3040 WATTS 31.2 29.4 28.5 27.6 COMPR. POWER (Low Temp.) = 2290 WATTS aa 7 COEFF. OF PERF. (High Temp.) = 2.5 382343 334 325 COEFF. OF PERF. (Low Temp.) = 1.7 ws (364 355 348 OUTDOOR FAN POWER = 410 WATTS ‘ " " INDOOR FAN POWER = 500 WATTS 399 38.1 37.2 36.3 40.9 39.1 38.3 37.4 NOTE: RATED WITH 25 FEET OF 7/8 41.2 39.6 38.8 38.0 SUCT. AND 5/16 LIQUID LINE 413 398 39.1 38.4 Inenniy lanauw 5 6 7 8 9 1 2 | > oO oe oN=| COND) Dawn CWON| anEw 2 2. 2. 2. 2: 3. 3. 3 3. 3 3. 3. 4 4. 4. 4. 4. 4, 4. aaa] poon!/ wows] ounn! yom aaa] Paoo/ovwos| Gov: yunr Mal O=$O0)/NUNEWINOOD. NAM Paa|pRaw!l OWE) COGN Onn ONOo Bane N=Oo Now| aBwW20| enone From Dwg. 218125539 Rev. 2 Source: General Electric product data for BWB936A with BWV942G Split System Weathertron heat pumps. 20.2.2 Technical Characteristics The characteristics of a standard electric air source heat pump are given in Table 20.1. The heating capacity and total electric power required are given as a function of indoor and outdoor temperatures. The characteristics of the heat pump efficiency discussed in the above are based on steady-state efficiencies of the units. Three non- steady-state factors affect the actual annual seasonal performance factor (SPF), defined as the total seasonal heating divided by the total energy consumption: e@ supplemental electric resistance heating is required when the heat pump capacity is unable to keep up with the heating load; e during mild, humid, outdoor temperature conditions (below 45°F), the outdoor coil will accumulate frost which limits heat exchange and must be removed, this requires additional energy that does not contribute to the useful space heating; and @ cycling of the heat pump will accumulate energy losses because of the cooldown of the heating coil when it is not contributing to the space heating. As the outdoor temperature falls, the capacity of the heat pump also falls, while the heating requirements of the building rise. This is shown in Figure 20.7. At the cross-over point or the "balance point," the heat pump must be operated continuously in order to keep up with the heating requirement. At temperatures below this balance point, auxiliary electric resistance heating is required.* At these lower temperatures, the COP of the heat pump begins to approach that of an electrical resistance system (i.e., COP = 1). As the outdoor temperature falls below approximately 45°F, the outdoor heat exchanger temperature is generally below freezing, and frost accumulates on the outdoor coil. Most heat pumps defrost the outdoor coil by reversing the cycle temporarily (2 to 10 minutes) to provide warm refrigerant to the outdoor coil to melt the frost. Depending on the outdoor temperature, reversing-cycle frost may add between 5% and 10% to the annual energy cost of the heat pump. (13) The reverse cycling of the heat pump not only causes the annual energy consumption to rise, but is thought to reduce the life of the unit. A sudden thermal shock to the system components occurs during the reversing operation, and after many thousands of these reversals, the performance of the compressor may deteriorate. (14) *Growing attention has been given to supplementing the electric heat pump with a gas-fired backup because of the lower cost of gas per end-point Btu. Lennox sells a Fuelmaster Plus based on _ this principle; see Reference 12. 489 Heating Load 1000 Btu/hr 50 a 45 h 40 Supplemental Heating Requirements: Direct - element Electric or Fossil - fuel Furnace 35 30 25 20 15 10 Major Frosting Region deg F 10 20 30 40 50 60 70 degC —12.2 -6.7 -1.1 4.4 10.0 15.6 21.1 Outdoor Temperature FIGURE 20.7 COMPARISON OF HEAT PUMP OUTPUT TO THE HEATING REQUIREMENTS OF A TYPICAL RESIDENCE (600 BTU/HR°F OR ABOUT 1600 SQ. FT: FLOOR AREA) VS. OUTDOOR TEMPERATURE 490 7 Each start and stop of the heat pump has a loss or inefficiency associated with it. There are two causes of this transient loss: e heat stored in the indoor coil is allowed to dissipate into a nonuseful space; and : @ each compressor start must separate the refrigerant from the lubricating oil within the system; this requires energy, causing a small but noticeable loss in system performance. Kelley's data(9) on the defrost and the start-up and- stop requirements of a heat pump indicate the distinct drop in COP of about 20% at 38°F. The energy cost of defrost, start and stop, and the auxiliary electric resistance heating lowers the annual SPF from an average value of 2-2.5 to 1.5-2.0. Most heat pumps used in the United States locate the compressor unit outdoors because of the noise associated with compressor operation. Some of the heat rejected by the heat pump is lost from the hot compressor to the cold outside air. Electrical resistance heaters are sometimes employed to keep the compressor warm to offset some of the adverse start-up problems with a cold compressor; this means that additional energy is lost if the compressor is located outside. Present Air Conditioning and Refrigeration Institute (ARI) test procedures permit the inclusion of the compressor heat as useful space heating if indeed this heat will be usefully provided to the indoor conditions. Some manufacturers have recognized this advantage and have begun to package compressors on the indoor units and provide sound insulation to make this application acceptable, although compressor heat rejected indoors in summer may decrease the heat-pump cooling capacity. The winter benefit of this approach would be worthwhile in Alaska, while the summer penalty would not be a factor. Ground and water source heat pumps have been developed and draw heat from ground or water which are at higher temperatures than the surrounding air. These heat pumps’ generally offer higher efficiencies but require special outside heat exchangers matched to the heat source. Advanced Heat Pumps An improvement under development is the two-stage or two-speed compressor. In this case, a single stage of compressor capacity is used when the outdoor temperature is above the balance point, and a second stage of heat pump capacity is brought on line below the outdoor balance point. The details of this concept are discussed in References (1) and (9). : 491 Heat of evaporation + heat of solution evaporation (refrigerating effect) Y 4 Heat of condensation + heat of solution Symbol : . Typical System Ammonia R — refers to refrigerant C — refers to absorbent Lithium Bromide Source: Reference 17. FIGURE 20.8 A SIMPLE ABSORPTION REFRIGERATION SYSTEM 492 Advanced heat pumps using higher motor-compressor efficiencies, variable speed compressors, rotary compressors, increased heat-exchangers surface areas, and improved fans have been under study by the Department of Energy. (15) The greatest disadvantage of the electric vapor compression refrigeration system is the expenditure of electricity and the associated fuel consumption at a power plant. The elimination of this work (except for a relatively small liquid pumping: requirement) is largely responsible for the development work on various. chemical absorption and absorption-vapor refrigeration systems. These systems operate on the principle that certain substances attract and hold large quantities of vapor of other substances at a relatively low temperature which are driven off when the temperature is raised. When the vapor is retained on the surface, or in the capillary pores of the solid, the process is adsorption. When the two substances react, or when the vapor goes into solution, the process is absorption; and for practical combinations of refrigerant and absor- bent (exothermic reactions), the chemical reaction heat must he removed along with the latent heat of the vapor. Since the refrigerant-absorbent solution is liquid, it can be pumped to condenser pressure with very little shaft work, and the vapor refrigerant is then driven from the absorbent by heating. Thus the absorption system is attractive where electricity is costly and natural gas or low-grade heat energy (such as waste steam) is plentiful. The Department of Energy, the Gas Research Institute, and Allied Chemical Corporation. are undertaking research to develop an absorption heat pump. It is expected that the machine will be similar in principle to the standard absorption refrigeration cycle shown in - Figure 20.8, although: specific components and the organic working fluids (absorbent and refrigerant) will be designed for a heat pump characterized by higher temperature differentials than those experienced in a refrigeration unit. (16) In addition to the gas-fired absorption heat pump, DOE is sponsoring the development of the Stirling/Rankine gas heat pump which has a gas-fired Stirling engine that produces power for driving a conventional Rankine cycle heat pump. An alternative absorption system is based on batch chemical processes as opposed to flow processes as in the Allied Chemical absorption heat pump. This concept is presently under investigation by the Department of Energy and Sandia Laboratories. In this system, shown in Figure 20.9, heat from a solar collector or fossil fuel is used to evolve a gas, B, over to the condensing unit to form a chemical compound, BC, releasing heat. After this has been accomplished, the chemical, B, is once again driven off the compound BC, but at a substantially lower temperature, which provides the refrigeration effect at the low-temperature. unit. During the heating 493 Ce ey \ Big) transport | \ ' ’ . —_— _ Heat in/out C—>] AB = A + Bia) BC = C+ Bly) Heat our/in Ph High temperature Low temperature unit Source: Reference 18. FIGURE 20.9 PROPOSED CHEMICAL ABSORPTION HEAT PUMP 494 mode, low-grade energy is used to evolve the gas at _ the low-temperature source, BC, from which it is transported to the high-temperature unit. Here the gas, B, condenses to form chemical compound AB, thereby liberating thermal energy for space heating. The difficulty of using this system is that it is inherently a batch or nonflow system. This means that duplicate systems must run on alternate cycles in order to maintain a constant heat pump capacity. Four chemical systems are presently under consideration: salt-methanol, an ammoniated salt, a hydrated salt, and sulfuric acid and water. 20.2.3 Environmental Issues There are no significant direct environmental issues related to the operation of residential or commercial air source heat pumps. The potential release of R-22 refrigerant (amounting to about 6 lbs) is not considered hazardous due to the low toxicity of the refrigerant. Some depletion of the ozone layer in the upper atmosphere is implicated in the release of flurocarbon refrigerants. However, the extent of the depletion and its effect in the quality of life has not been established. Air source heat pumps do generate outside noise that furnaces do not. Typical sound ratings of 20 (ARI Standard 270) can be expected from the outdoor fan. Water source and ground source heat pumps may have substantial effects on the environment depending on the location of the outdoor heat exchangers. Specific applications in Alaska would have to be analyzed. Electric heat pumps have a secondary environmental impact in that fuel is burned at a power plant to provide the electricity. The environmental impact of electric power generation should be considered. 20.2.4 Commercial Status Air source heat pumps are available from numerous manufacturers and range in size from room size units of 10,000 Btu/hr to large commercial building units in excess of 1,000,000 Btu/hr. The Air Conditioning and Refrigeration Institute published a Directory of residential air source heat pumps which provides a _ convenient overview of the products available. Table 20.2 is an excerpt from the July 1982 ratings. The heating performance is given for 47°F and 17°F conditions. (The Department of Energy and the heat pump industry have established the rating point for heat pumps at these two temperatures.) The HSPF represents the heating seasonal 495 TABLE 20.2 January 1- June 30, 1982 ARI STANDARD RATINGS HP-9 : Eee ' ting ating i High Temperature Low Temperature Model — Capacity____Btuh/Watt = Capacity Capacity Footnote(s) Designation (Btuh) EER SEER (Btuh) HSPF COP (47) (Btuh) COP (17) 641CB042 508B060 41,000 —- 7.95 42,600 6.20 2.60 24,200 1.85 H 641CB042 607D/518A048 41,000 - 7.95 43,000 6.30 2.65 24,000 1.85 645AB042 518A/607D060 41,000 - 9800 43,600 7.10 2.90 26,200 2.26 —- : 645AB042 619B/CO60 : 41,000 - 880 43,000 6.85 2.80 26,000 220 1 641CB042 607D/518A060 42,000 - 805 43,600 6.36 2.70 24,200 1.86 yy 641CB042 519B/C080 42,000 - 68.10 43,000 6.36 2.70 24,200 1.85 s 641CB048 §19B/C048 45,000 —- 7.85 48,500 6.25 2.60 30,000 2.00 641CB048 606B048 45,500 = 770 48,000 6.20 2.60 30,100 2.00 641CB04B 607D/518A048 46,000 - 7.76 49,000 6.35 2.65 30,600 2.00 S45AB048 606B048 46,000 - 826 47,000 6.20 2.65 27,800 2.05 645AB048 +. 519B/C048 46,000 - &20 47,000 6.20 2.60 27,600 2.05 1 ' 641CB048 508B080 46,500 - 7.76 48,600 6.25 2.60 30,300 2.00 S45AB048 518A/507D048 48,500 - 830 47,500 635 2.70 28,200 © 2.10 a 641CB048 619B/CO60 47,500 - 795 49,500 645 2.60 30,600 2.05 ? 641CB048 % 607D/518A080 47,500 — 7.85 60,000 6.55 2.65 30,800 2.00 ) 645AB048 506B060 47,500 - 830 47,000 6.26 2.66 28,000 2.05 tos 645AB048 518A/507D060 48,000 - 835 48,500 6.45 2.76 28,600 2.16 645AB048 519B/CO60 48,500 - 846 48,000 6.35 2.70 28,400 2.10 " 645AB060 $ §19B/CO60 53,000 - 7.85 69,500 6.85 2.76 38,500 2.25 ‘ 645AB060 S06B080 53,600 - 7.76 67,500 6.75 2.60 35,200 2.10 645AB060 618A/507D060 53,600 - 7.76 69,600 7.00 2.76 36,000 2.16 TYPE: HRCU-A-CB Outdoor Unit tndoor Unit 641CJ018 $ =SI7ENO18 17,000 - 7.70 18,800 6.00 2.60 11,600 1.80 641/018 6130018 . 17,200 =- 7.70 18,400 6.00 2.60 12,200 1.85 641CJ018 517EN024 17,500 - 7.70 18,800 6.00 “2.60 11,800 1.80 641CJ018 6130024 18,200 - 626 18,800 6.16 2.60 11,500 1.65, ‘ 6410018 517EN030 18,300 - 7.95 19,300 6.10 2.65 11,800 1.80 641/018 619B/C036+620B042 18,500 — 7.85 19,600 6.25 2.65 12,300 1.80 6410018 617EN036 18,600 - 800 19,400 6.30 2.60 12,200 1.85 6410018 518A036+520B042 18,700 — 7.85 19,700 6.25 2.60 12,100 1.80 . 541C0J018 : 6130030 18,900 - 805 19,600 6.26 2.60 12,000 1.85 645AJ024 617EN024 20,000 - . 7.76 21,600 6.10 2.76 11,800 1.60 645AJ024 6130024 20,400 - 6.10 21,400 6.30 2.80 11,700 1.85 545AJ024 517ENO30 20,800 - 7.90 21,800 6.20 2.80 12,000 1.85 645AJ024 * §18A036+520B042 21,000 - 7.66 22,800 6.35 2.85 12,700 1.90 645AJ024 §19B/C036+620B042 21,000 - 765 22,400 6.35 2.80 12,500 1,85 S45AJ024 6130030 21,200 - 7.85 22,400 6.35 2.85 12,400 1.90 we 645AJ024 617EN036, 21,200 - 7.90 22,200 6.30 2.80 12,300 1.85 §410J024 $ =SI7ENO24 22,000 - 7.60 22,400 6.60 2.60 12,600 1.76 - 64104024 5130024 22,800 - 7.95 21,800 6.85 2.50 12,000 1.65 . 541CJ024 517EN030 23,000 - 7.80 22,200 6.70 2.60 12,400 1.76 641CJ024 517EN038 23,600 —- 7.86 22,400 5.60 2.50 12,600 1.80 — 641CJ024 6130030 23,800 - 7.80 22,600 6.85 2.60 12,600 1.76 64105024 619B/C038+620B042 23,800 - 7.80 22,400 5.80 2.50 12,600 1.76 541CJ024 - 618A036+6520B042 23,800 - 7.90 22,800 6.80 2.65 12,700 1.80 §45AJ030 517EN030 25,800 - 7.56 28,600 6.25 2.65 16,800 1.95 645AJ030 §18A036+520B042 26,400 - 760 29,800 6.40 2.76 17,600 2.00 ue 645AJ030 §18A042+5208042 26,600. - 7.56 29,600 6.35 2.70 17,600 1.95 : S45AJ030 5130030 . 26,600 - 7.70 29,400 6.45 2.80 17,200 2.00 B45AJ030 © 519B/C036+6208042 26,600 ~ 7.60 29,400 6.30 2.70 17,500 1.05 545AJ030 517EN038 26,800 - 7.76 29,000 8.40 2.76 17,100 2.00 §45AJ030 §19B/C042+6520B042 27,200 - 7.55 29,800 6.45 2.75 17,700 2.00 é 6410030 ¢ =§17ENO30 28,000 - 7.665 30,400 6.35 2.70 17,600 1.95 641CJ030 5130030 28,200 - 7.60 31,000 6.60 2.70 18,000 1.85 641CJ030 618A036+520B042 28,600 - 7.70 31,000 6.55 2.76 18,000 2.00 , 64105030 §19B/C036+520B042 28,600 - 7.70 30,600 6.50 2.70 17,900 2.00 ‘ 4105030 617EN036 28,800 - 7.85 30,400 6.65 2.70 17,800 | 2.00 2 410/030 §18A042+620B042 28,800 — 7.80 30,800 6.60 2.70 © 18,000 2.00 641CJ030 S17EN042 29,200 - 7.80 31,000 6.55 2.80 17,800 2.00 on 641CJ030 §19B/C042+620B042 29,200 - 7.80 31,000 6.55 2.76 18,000 2.00 ’ S45AJ036 : 618A036+520B042 33,200 - 7.30 38,600 6.60 2.65 22,600 2.00 { B45AJ036 518A042+520B042 33,400 - 7.36 38,200 6.40 2.60 22,400 2.00 . i 64SAJ036 519B/C036+520B042 33,400 - 7.35 386,400 6.45 2.65 22,400 2.00 645AJ038 + 517ENO38 33,600 - 750 36,200 6.45 2.65 22,200 . 2.00 641CJ036 . 518A038+520B042 34,200 - 760 37,000 6.00 2.76 21,600 2.00 8410/0386 619B/C036+520B042 34,200 - 7.60 37,200 6.00 2.70 21,500 1.95 45AJ036 519B/C042+520B042 34,200 - 745 36,800 6.50 2.66 22,600 2.00 wa 641CJ036 + =517EN036 - 34,400 - 7.60 37,000 6.20 2.70 21,400 2.10 6410036 §18A048+620B042 34,600 - 7.60 37,400 6.10 2.76 21,800 2.00 = S45AJ036 618A048+620B042 34,600 - 7.60 37,200 6.65 2.76 23,000 2.05 1 4 ne 496 my TABLE 20.2 (Continued) "-. ARI STANDARD RATINGS January 1- June 30, 1: er LON enn rence i ——High Temperature _____Low Temperature _ i Model : Capacity___Btun/Watt___ Capacity Capacity y Footnote(s) » Designation (Btuh) = EER SEER (Btun) HSPF COP (47) (Btuh))=— COP (1 . BRYANT AIR CONDITIONING (CONTINUED) i Outdoor Unit indoor Unit 6410038 616A042+6208042 38,000 = 780 $7,200 606 270 21,500 2.00 a 6410036 617EN042 35,200 - 7.88 37,200 630 «(2.78 21,000 2.00 4 6410038 6108/C042+6208042 35,200 - 7.65 37,400 6.08 2.76 21,700 1.98 é 64104038 617ENG4B 36,800 - 7.70 37,400 6.38 2.80 22,100 2.08 . 648AB042 818A042+6208042 38,500 - 830 42,000. 8.85 2.60 26,800 1.65 64108042 618A042+6208042 39,000 = 7.85 42,500 6.00. 2.65 24,000 1.80 645AB042 6108/C042+6208042 39,000 - 860 43,000 «6.76 2.80 28,200 2.16 64108042 $ S17ENO42 39,500 - 715 44,000 «6.16 =—-2.85 24,500 1.86 645AB042 618A048+6208042 39,600 - 660 43,600 6.86 2.85 26,800 2.168 64108042 618A048/6208042 40,000 = 7.76 43,600 620 265 24,200 1.88 . 64108042 619B/C042+6208042 40,000 - 7.10 42,600 6.08 2.66 24,200 1.80 ac 645AB042 618A060+5208042 40,600 = 860 44,000 6.95 2.90 26,800 2.20 645AB042 610B8/C060+6208042 40,500 = 880 43,600 6.85 2.85 26,600 2.16 : , 641CBO42 617EN048 41,000 - 7.80 44,500 6.26 2.70 24,700 1.88 - 641CB042 §16A060+5208042 41,000 - 7.85 44,000 6.26 270 24,400 1.88 64108042 6198/C080+620B042 ° 41,600 - 7.80 44,000 6.20 2.65 24,500 1.85 ~ 64108042 617EN060 42,000 - 7.80 _ 48,000 8.38 2.70 28,000 1.88 64108048 619B/C048+6208048 48,000 - 7.60 60,000 635 = 2.66 30,900 2.00 64108048 518A048+5208048 48,500 - 7.85 49,500 630 268 30,800 2.05 . 6A6ABO48 6198/C048+6208048 48,800 - 7.88 47,600 610 © 2.60 28,600 2.00 645AB048 616A048+6208048, 48,000 ~. 7.96 48,000. 630 268 28,800 2.06 - 64108048 $ BI7ENO4S 48,500, - 7.88 40,600 630 265 31,300 2.05 64108048 518A080+6208048 48,600 - 7.85 60,600 620 260 31,200 2.00 64108048 §19B/C060+520B048 48,500 - 7.70 49,600 6.30 2.85 30,800 2.08 641CB048 617ENO60 47,000 - 7.65 60,500 6.45 2.60 32,300 2.10 646AB048 618A060+6208048 47,000 - 8.00 49,000 6.35 2.70 29,200 2.05 ~~ 645AB048 §198/C060+5208048 47,000 - 6.05 48,500 6.20 2.66 29,000 2.05 { 45AB060 . 618A060 + 5208048 62,000 - 7.60 68,600 6.80 2.66 36,600 2.10 . 645AB060 519B/C060+5208048 52,000 - 7.60 68,600 6.80 2.60 36,200 2.05 S45AB060 §18A060+ 5208080 62,600 - 7.30 60,600 6.86 2.65 37,000 2.10 645AB060 6198/C060+520B080 63,500 - 740 69,600 670 . 270 38,600 2.08 PRODUCTS NOT COVERED BY DOE TYPE: HSP-A t 642EP036 35,400 = 7.60 - 35,800 - 2.70 20,800 2.00 4 642EP042 40,000 7.80 - 41,500 - 2.70 21,400 1.80 G 642EE048 47,000 7.70 - 48,500 - 2.80 27,400 2.00 642EP048 : 47,000 7.70 - 48,500 - 2.80 27,400 2.00 Me 642DE,EE060 $7,500 7.60 - 58,000 - 2.70 932,400 1.90 Y 6420P,EP060 57,500 7.60 - * 68,000 - 2.70 32,400 | 1.90 . 642D0S0HPD 87,000 8.60 - 89,000 - 2.05" 49,000 1.06 642D120HPD 115,000 7.50 - 128,000 - 2.65" 69,000 1.88" — TYPE: _HRCU-A-C Outdoor Unit Indoor Unit 641CKO36 §198/C036,5108036 94,800 7.60 7 38,800 7 2.80 21,100 2.00 641CK036 6068036 34,800 7.70 - 37,000 - 2.80 21,400 2.00 641CKOI6 507D/518A036 34800 7.60 - 937,000 - 2.80 21,100 2.00 \ 641CK038 $07D/518A042 35,000 7.70 - 37,000 - 2.80 21,300 2.00 : 641CK036 068042 38,400 7.70 ~ 36,800 - 2.80 21,400 2.00 \ 641CK038 6198/0042 35,400 7.80 - 37,000 - 2.00 21,400 2.00 ‘- 641CK038 5068048 35,600 7.70 - 37,000 = 2.80 21,500 2.00 641CKO36 607D/518A048 36,000 7.80 - 37400, 2.80 21,400 2.00 6410K042 6068042 40,000 7.60 - 42,000 - 2.80 23,900 1.80 641CK042 6198/C042 40,000 7.60 7 = 42,000 - 2.80 23,900 1.90 641CKO042 6070/518A042 40,000 7.60 - 42,500 - 2.60 23,800 1.90 ~ 641CK042 6068048 40,600 7.60 «- 42,500 - 2.60 24,000 1.90 641CK042 6088060 41,000 7.60 ~ 42,600 a 2.60 24,200 1.00 : 641CK042 5070/6 18A048 41,000 7.60 - 43,000 = 2.70 24,000 1.90 i 641CKO42 619B/C060 42,000 760° = 43,000 - 2.70 24,200 1.90 wo 641CK042 $07D/618A060 42,000 750° ~ 43,600 - 2.70 24,200 1.90 641CK048 619B/Cosa 45,000 7.60 - 48,500 ~ 2.60 30,000 2.00 641CK048 5088048 45,500 7.50 - 48,000 - 2.60 30,100 2.00 B41CK048 607D/518A048 48,000 7.50 - 49,000 - 2.80 30,600 2.00 497 SOURCE: TABLE 20.3 PARTIAL LIST OF MANUFACTURERS OFFERING AIR SOURCE HEAT PUMPS . Trade or Brand Name Adaptomatic Air-Ease, Armstrong, Gaffers & Sattler, Johnson, Williams Airtemp All Seasons Amana Bard Bryant Century Champion Climatrol Climizer _ Comfort-Aire Comfortmaker Commercial Heat Pump Weathermaker Compact Heat Pump : Day & Night ENVI-RO-TEMP Enmod Flexhermetic " Flexhermetic i Fraser & Johnston Friedrich Fuelmaster + Goett! Heatwave Heil Legend Lennox Luxaire Marair Moncrief Montgomery Ward Patco Payne Rheem Rheem RPCA Series Rheem RPNA Series Ruud " TRADE OR BRAND NAME INDEX 498 Company FEDDERS CORPORATION MAGIC CHEF AIR CONDITIONING & HEATING DIVISION AIRTEMP CORPORATION AIRTEMP CORPORATION AMANA REFRIGERATION, INC. BARD MANUFACTURING CO. BRYANT AIR CONDITIONING CENTURY BY HEAT CONTROLLER YORK HEATING & AIR CONDITIONING CLIMATROL SALES COMPANY CLIMATROL SALES COMPANY HEAT CONTROLLER, INC. SINGER COMPANY, THE Climate Control Division (St. Louis) CARRIER AIR CONDITIONING CARRIER AIR CONDITIONING DAY & NIGHT AIR CONDITIONING LOWE’S COMPANIES, INC. YORK HEATING & AIR CONDITIONING FEDDERS CORPORATION FEDDERS CORPORATION FRASER & JOHNSTON HEATING & AIR CONDITIONING NORTHRUP, INC. LENNOX INDUSTRIES, INC. GOETTL AIR-CONDITIONING, INC. SOUTHWEST MANUFACTURING Division of McNeil Corporation HEIL-QUAKER CORPORATION LENNOX INDUSTRIES, INC. LENNOX INDUSTRIES, INC. LUXAIRE HEATING & AIR CONDITIONING MARVAIR COMPANY MONCRIEF HEATING & AIR CONDITIONING MONTGOMERY WARD & CO., INC. PATCO, INC. PAYNE AIR CONDITIONING RHEEM AIR CONDITIONING DIVISION City Investing Company | RHEEM AIR CONDITIONING DIVISION City Investing Company : RHEEM AIR CONDITIONING DIVISION City Investing Company RUUD AIR CONDITIONING DIVISION * City Investing Company Air Conditioning and Refrigeration Institute i TABLE 20.3 (Continued) ry PARTIAL LIST OF MANUFACTURERS OFFERING AIR SOURCE HEAT PUMPS TRADE OR BRAND NAME INDEX | “ Trade or Brand Name Company ~ Ruud UPCA Series RUUD AIR CONDITIONING DIVISION oe City Investing Company ‘ i Ruud UPNA Series RUUD AIR CONDITIONING DIVISION oo City Investing Company - Sears SEARS, ROEBUCK & CO., INC. Single Package Heat Pump Sun Dial Sunpath Super Energy Saver T.H.E. Heat Pump Thermizer Trane WeatherKing Weatherite Weathermaster Ili Weathertron Heat Pump Westinghouse Whirlpool Williamson 499 CARRIER AIR CONDITIONING SQUARE D COMPANY, THE YORK HEATING & AIR CONDITIONING COLEMAN COMPANY, INC., THE COLEMAN COMPANY, INC., THE FEDDERS CORPORATION TRANE COMPANY, THE ADDISON PRODUCTS COMPANY INTERTHERM, INC, CARRIER AIR CONDITIONING GENERAL ELECTRIC COMPANY : WESTINGHOUSE ELECTRIC CORPORATION Commercial Industrial Air Conditioning Div. " WHIRLPOOL HEATING & COOLING PRODUCTS WILLIAMSON COMPANY, THE TABLE 20.4 CASE STUDY-HEAT PUMP PERFORMANCE IN ANCHORAGE (Heat Pump Nominal Capacity — 36,000 Btu/hr at 47°F) Outside Hours Heat Pump Heating Load, Temperature per Year Power Output 1000 Btu/hr °F at Temp. COP* kW 1000 Btu/hr House "A" House "B" House "C" -30 1 0.77. 2.30 6 20 30 60 -20 9 1.02 2.30 8 18 27 54 -10 68 1.28 3.40 10 16 24 48 0 271 1.38 2.55 12 14 21 42 10 556 1.60 2.75 15 12 ™ ar 36 20 923 1.89 2.95 19 . 10 15 30 30 : 1362 2.13 3.30 24 8 12 24%* 40 1531 2.55 3.45 30 6 9 18 50 1435 2.90 3.70 35 4 _ 6 12 60 1933 2.90 3.95 39 2 3 6 Effective Seasonal COP -1.87 1.83 1.66 *Not including startup and reversing cycle losses. **Denotes approximate balance point SOURCE: Arthur D. Little, Inc. Nd 500 ~s ey . performance factor (heat delivered: to.electric power required) for the heat pump operated in a climate approximately equivalent to Memphis, Tennessee. Table 20.3 is a partial list of the manufacturers offering air source heat pumps along with the unit brand name. 20.3 Economic Implications 20.3.1 Costs To understand the economic implications of the use of a heat pump in Alaska, a case study of a residential heat pump unit operating in Anchorage was undertaken. As outlined in Séction 20.2.1, the performance of a residential unit will vary by climatic zone as well as . by the size of house (due to the change in balance point). As a 36,000 Btu/hr is used in larger houses (Table 20.4), the effective seasonal efficiency goes down from 1.87 to 1.66 as a result of the increased fraction of electrical resistance space heating (at an efficiency of 1.0) that must be performed in Anchorage to meet the demand. : : Table 20.5 summarizes the energy cost of a 36,000 Btu/hr heat pump, standard resistance and natural gas furnace in Anchorage for a 3.9¢/KWh electric rate and $1.7 per million Btu natural gas cost with 70% efficient conventional furnace. At these fuel prices the heat pump has significantly lower operating cost than a straight resistance unit and more than a gas furnace. Typically the cost of installing an air source heat pump (Table 20.6) is higher than that for a conventional oil or gas system. Since the fuel cost has been: shown to be higher in Anchorage than for a natural gas system, it does not appear that a heat pump would be economical in most circumstances. In other Alaskan locations, where the cost of residential heating fuel is more expensive than gas in Anchorage, the cost of electricity is likewise higher (since power and space heat usually depends predominantly on the same fuels at any given location). Therefore, the-heat pump will suffer a similar economic disadvantage in most of Alaska. Table 20.7 summarizes the cost of delivering space heat to a typical residence in Anchorage (House "B"). 20.3.2 Socioeconomic Factors There are no significant socioeconomic impacts associated with the installation or use of residential or commercial air source heat pumps in Alaska. We would expect that the normal distribution channels would be used and some new field service training would be required, but it would be wholly within the operation of the existing manufacturers' distribution systems. 501 COMPARATIVE ENERGY COST OF RESIDENTIAL TABLE 20.5 SPACE HEAT IN ANCHORAGE House "A" Heat Load, Btu/hr - °F 200 Annual Heat Load, million Btu 51 Heat Pump Seasonal COP 1.87 House "B" 300 76 1.83 House "'C" 600 152 1.66 Annual Fuel Cost, $/Year Heat Pump Power at COP, 3.9¢/kWH 310 Electrical Resistance at 100%, 3.9¢/kWh 579 Conventional Gas at 70%, $1.70/million Btu 123 502 476 867 185 1046 1738 369 TECHNOLOGY: BASIS: Location: - Years CAPACITY: Input: Output: ESTIMATED USEFUL LIFE: CONSTRUCTION PERIOD: CAPITAL COST: TABLE 20.6 CAPITAL COST SUMMARY Electric Heat Pumps Anchorage 1982 constant dollars 12,195 kWh/year 76 million Btu/year 10 to 15 years 2 man days installation Equipment and Materials: 1650 Direct Labor: Indirect Costs: 360 290 TOTAL CAPITAL INVESTMENT: $2300 503 TABLE 20.7 OPERATING COST SUMMARY TECHNOLOGY: Electric Heat Pump BASIS: Location: Anchorage Year: 1982 CAPACITY: Input: 12,195 kWh/year Output: 76 million Btu/year OPERATING FACTOR: Annual Operating Costs Unit Cost Consumption VARIABLE COSTS: Energy Electricity: 3.9¢/kWh 12195 kWh TOTAL VARIABLE COSTS: 3.9¢/kWh 12195 kWh FIXED COSTS: Maintenance: Capital Charge, 9.5% of Capital Investment TOTAL FIXED COSTS: TOTAL ANNUAL OPERATING COSTS: TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST: 504 Annual Cost $475.60 475.60 62.00 218.50 280.50 $756.10 76 million Btu $9.95/million Btu 20.4 Impact 20.4.1 Effect on Overall Energy Supply and Use The primary attraction of a heat pump for space heating is that it is capable of supplying more heat to a load than the electrical energy it takes to run it. However, since the efficiency of generating electricity from fossil fuel at a central station is about 30-35 percent, this benefit is largely lost in terms of the overall energy supply. This is especially true if, as in the cold Alaskan climate, considerable resistance heating is required over the year. If the source of electricity is one which is locally available (e.g., hydro or coal), heat pumps could become a means of supplying these secure, stably priced resources to the residential and commercial sectors. However, this- would also exacerbate the already high electricity demand peaks in winter in Alaska. One of the advantages of heat pumps in temperate climates, which is irrelevant in Alaska, is that the heat pump can be reversed and utilized as an air conditioner. Thus, the overall cost of heating and cooling can be reduced by using a single system. And since the seasonal performance factor in these locations is quite good, the overall economic attractiveness of heat pumps is substantially better than in Alaska. , : 20.4.2 Future Trends Air source heat pumps have found widespread use in the Lower 48 and have been designed to operate under a wide variety of conditions: Installation procedures and design codes for the heat ‘pump are well established, and there are no major changes likely. California and several other states have considered efficiency standards for heat pumps, but the. impact of these standards has already been absorbed by the industry so that the products presently available are likely to remain available. In the future, as sources of electric power become less closely tied to the fuels used for home heating, electricity may become more economical. Examples would be Susitna in the Railbelt and a variety of renewables in the Bush. Heat pumps could help extend the resource because of their ability to deliver more energy than they consume. , 505 10. Il. REFERENCES Looney, Q., Electric Power Research Institute, Palo Alto, California. Personal communications, April 1978. Arthur D. Little, Inc., Design, Development and Demonstration of a Promising Integrated Appliance, Phase I - Design, prepared under Contract No. EY-76-C-03-1209 for the Energy Research and Development Administration, September 1977. Moyers, J.C., et al., "The Annual Cycle Energy System Concept and Application," Proceedings of the International Conference on Energy Use Management, Tucson, Arizona, October 24-28, 1977, Vol. II, Pergamon Press, New York, 1978, pp. (231 ff. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., ASHRAE Handbook and Product Directory: 1973 Systems, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., New York, 1973. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., ASHRAE Handbook and Product Directory: 1975 Equipment, American Society of Heating, Refrigerating, and Air Conditioning Engineers, Inc., New York, 1975. "Manufacturers' Share of Market or Order of Ranking," Appliance Manufacturer, January 1976, pp. 81 ff. Duffy, G. "Statistical Panorama XIV," Air Conditioning, Heating, and Refrigeration News 146(5):3, January 29, 1979. Duffy, G. "Statistical Panorama XIII," Air Conditioning, Heating, and Refrigeration News 143(5):3, January 30, 1978. Kelley, G.E., et al., Dynamic Performance of a_Residential Air-to-Air Heat Pump, National Bureau. of Standards, Washington, D.C., 1977. . U.S. Department of Commerce, "Air Conditioning and Refrigeration Equipment, 1974," Current Industrial Reports, Series MA-35M, Table 7. Hiller, C.C., et al., “Energy Savings in Vapor Compression Refrigeration and Heating Devices," Proceedings of the Technical Opportunities for ENergy Conservation in Appliances, Boston, Massachusetts, May ll, 1976. Conference conducted by Arthur D. Little, Inc., for Energy Research and Development Administration, 1976. NTIS Report No. CONF-7605139. 506 12. 13. 14, b. 16. 17. 18. 19. 20. Consdorf, A.P., "Stage Research for Heat Pump Boom," Appliance Manufacturer, November 1975, pp. 43 ff. . Comley, J., et. al., Heat Pumps: Limitations and Potential, . Corporate Research, and Development, General Electric Company, Schenectady, New York, 1975. Groff, G., et al., "Investigation of Heat Pump Performance in the Northern Climate through Field Monitoring and Computer Simulation," preprint, American Society of Heating, Refrigerating and Air-Conditioning Engineers 1978 Semiannual Meeting, Atlanta, Georgia, January 29-February 2, 1978. Published in ASHRAE Transactions 84, (Part 1), 1978. Hirst, E., Energy and Economic Benefits of Residential ENergy Conservation RD and «» prepared by Oak Ridge National Laboratory, February 1978. Murphy, K.P., Allied. Chemical Corporation, Morristown, New Jersey. Personal communications, December 1, 1978. Wood, B.D., Applications of Thermodynamics, Addison-Wesley, Reading, Massachusetts, 1969. (Addison-Wesley Series in Mechanics and Thermdynamics.), pp. 205 ff. . Miller, M., "Chemical Heat Pumps Now Seem _ Technically Feasible," Air Conditioning, Heating, and Refrigeration News 143(12):14, March 20, 1978. , : Gordian Associates, Evaluation of the Air-to-Air Heat Pump for Residential Space Conditioning, prepared for the Federal Energy Administration, April 23, 1976. NTIS Report No. PB 255 652. American Air Conditioning Company, Boston, Massachusetts. Personal communications, March 1978. 507 21.0 CONDENSING FURNACES AND BOILERS 21.1 Introduction and Summary 21.1.1 Technical Overview Condensing furnaces and boilers are space heating systems that achieve a high efficiency in heat transfer between the products of combustion and the working fluid (room air or hot water). The systems are designed so that the temperature of the exhaust combustion products is reduced below the dew point, thereby causing _ the water vapor contained in the products to condense and release the latent heat of vaporization into the working fluid. As a result, these systems operate at a steady-state efficiency in the 88% to 94% range. Because off-cycle losses are also reduced, a seasonal effi- ciency in the 82% to 92% range can be achieved. This is 15% to 30% better than conventional systems. Some condensing systems are designed to also provide water heating at the same high efficiency. The installation of a condensing system is a little more complex, in ‘general, than the installation of a conventional heating system. Condensing systems require no chimney, though, because the exhaust products can be vented through the wall through two-inch plastic pipes. This offsets part of the higher cost of these systems and makes them cost effective in new buildings. Condensing furnaces and boilers appeared in the U.S. and in Europe in 1979. Since then, several manufacturers have been developing systems, and the market is growing and becoming competitive. About seven different condensing systems are expected to be available on the European market by the end of 1983; there were only two in 1980. All systems presently on the United States market and most systems on the European market operate with natural gas and/or pro- . pane. One developer in Europe has developed oil-fired units that have been on the market there since 1979. 21.1.2 Alaskan Perspective Use of condensing systems is advantageous, especially in cold climates and/or in applications where the cost of fuel is high.. In either case, the energy cost savings resulting from the high efficiency of the systems can pay back the added cost of installation of these systems in a relatively short period. Alaska's colder and longer winters than the Lower 48 make the use of condensing heating systems more advantageous here. In the Bush where fuel costs are relatively high, the benefits are even better. 509 FIGURE 21.1 Heat Exchanger Gas Burner CONVENTIONAL GAS FURNACE 510 21.1.3 Significance The reliability of the condensing furnaces and boilers, although only on the market of the Lower 48 states for three years, is claimed to be high. Potential problems with condensate disposal and heat exchanger corrosion have been addressed and solved. Some manufacturers provide a 20-year warranty on the heat exchanger. A regular ' maintenance service once a year (similar to the service done on conventional oil furnaces) is recommended to change filters and check | the ignitor, flame sensor and general combustion operation. |The spark plug and the air or fuel inlet valves.in the pulse combustion systems are expected to be replaced every five years. ~ The payback period for a new house installation in Anchorage is two to three years for an oil unit and ten to fifteen years for a natural gas-fired unit. This estimate is based on an estimated annual heating . load of 150 million Btu/year (equivalent to slightly more than 1,000 gallons/year of fuel oil), the cost of fuel oil of $1.31/gallon and the cost of natural gas of $1.70/million Btu. The added possibility of some of these systems (boilers) to also provide water heating at the same. very high efficiency, makes the estimated payback periods. shorter. : 21.2 Description of Technology 21.2.1 Operating Principles Conventional Systems Space heating systems, natural gas or oil fired, are characterized as "furnaces" of "boilers," depending on the working fluid that is being heated by the products of combustion of the fuel. In furnaces, the . working fluid is air, whereas in boilers it is water. Figure 21.1 shows the major components of a conventional gas-fired furnace. The gas burners are long, slotted pipes, usually ignited by a small pilot burner when the furnace calls for heat. The air for combustion of the gas is drawn "naturally" to the flame of the burning gas by the buoyancy of..the flame (natural draft systems). The products of combustion release their heat to the working fluid (air in a furnace) through metal heat exchangers. The air, driven by a blower (shown in the lower section of the furnace of Figure 21.1), is heated to a temperature of 100°F to 120°F and is distributed to the house through air ducts; it returns to the furnace through return ducts at a temperature of 60°F to 70°F (house temperature). In a boiler, the water is heated to a temperature of 160°F to 180°F, is distributed to the house through pipes, releases its energy through radiators in the living space, and returns to the boiler at a temperature of 120°F to 160°F. The products of combustion, having released their energy to the working fluid, exit the heat exchanger at - 511 = > 13 12 11 10 20% EA 9 ‘ a 8 // S 7 ~—— 20% EA 5 6 70°F x a) 4 3 2 Fuel: Natural Gas (CH3.95) 15 Relative Humidity: 50% 70 90 110 130 150 170 190 210 Bulk Flue Temperature °F FIGURE 21.2 HEAT LOSS (PERCENT OF HIGHER HEATING VALUE OF FUEL) IN EXHAUST COMBUSTION PRODUCTS (FLUE GAS) AS A FUNCTION OF TEMPERATURE; EXCESS AIR: EA = 0%, 20%, COMBUSTION AIR TEMPERATURE: 40°F, 70°F 512 ' ‘ a temperature ranging between 400°F to 600°F in conventional systems. The operation of an oil-fired furnace or boiler is essentially similar to that described for the gas furnace of Figure 21.1, except that the air for combustion of oil is not "naturally" drawn by the flame, but forced by a small blower. A portion of the combustion air is also used to atomize the oil into small more readily combustible droplets as they pass through an orifice (nozzle) and into the flame zone. The efficiency of a furnace or boiler, both natural gas or oil fired, is expressed as a fraction of the higher heating value of the fuel. The efficiency, which is the useful heat to the working fluid (air or water) divided by the heat input of the fuel, can also be determined by difference from the losses of the exiting combustion products during system operation (steady-state losses) and the losses of the system during stand-by (stand-by or off-cycle losses). A fraction of the heat loss in the combustion products: is latent heat loss, i.e., the heat required to retain the water of combustion as a vapor. This amounts to 9.5% of the higher heating value of the fuel for natural gas, 6.5% for distillate oil. The rest of the heat loss in the | combustion products, usually "20% to 30% of the higher heating value . of the fuel in conventional systems, is sensible heat loss. Condensing Systems To improve the efficiency of a furnace or boiler beyond 90% and recover part of the latent heat. loss, the temperature of the exhaust combustion products needs to drop below 130°F; at this temperature the water vapor in the combustion products starts to condense. Figure 21.2 shows the exhaust products heat loss, expressed as a percent of the higher heating value of the fuel, as a. function of the temperature of the exhaust combustion products (flue gas). As can be seen from this figure, substantial heat can be recovered, due to water vapor condensation, if the exhaust gas temperature drops below 110°F. Low exhaust products temperatures can, in principle, be reached if there is a stream sufficiently cool and of sufficient volume to extract the heat for a useful purpose. This can be in a furnace with return air temperature 60°F to 70°F; or in a boiler only during | start-up with cold water temperatures, but not during steady-state operation with return water temperature ranging between 120°F and 160°F. In order to achieve flue. gas condensation it is required that the heat exchanger be made larger and/or more. effective than that for non-condensing systems. . In practice, this precludes the use of natural. convection. Therefore, the two most common ways of achieving flue'.gas condensation are by the use of either a power burner or a pulse combustion burner. 513 BACK PRESSURE | IGNTION Source: Lennox Furnace Literature. FIGURE 21.3 1 - Gas and air enter and mix in combus- tion chamber. . 2-To start the cycle a spark is used to ignite the gas and air mixture. (This is one ‘pulse’). 3- Positive pressure from combustion closes flapper valves and forces exhaust gases down a tailpipe. 4- Exhaust gases leaving the chamber create a negative pressure. This opens the flapper valves drawing in gas and air. 5 - At the same instant part of the pres- sure pulse is reflected back from the tailpipe causing the new gas and air mixture to ignite. No spark is needed. (This is another ‘pulse’). "6-Steps 4 and 5 repeat 60 to 70 times per second forming consecutive « ‘pulses’ of 1/4 to 1/2 Btu each. PRINCIPLE OF PULSE COMBUSTION OPERATION 514 eee at r A power burner, conventionally used in oil-fired systems, may be used in gas-fired systems as well. It involves using a power blower on the inlet (forced draft) or on the exit side (induced draft) of the unit to promote the flow of combustion products through the heat exchanger. The power blower can overcome a higher pressure drop through the heat exchanger than do conventional natural draft systems and can, thus, achieve higher efficiency by forcing the e combustion products to flow either at higher velocity and/or through a larger area heat exchanger. These effect the necessary heat transfer to achieve condensing flue gas temperatures. Further, 7 during the off-cycle, standby losses are also lower because the increased resistance to flow through the system impedes natural [ convection up the flue. Therefore, the systems can achieve a seasonal efficiency which may range between 82% and 92%. The principle of operation of a pulse combustion burner is shown in Figure 21.3. A spark plug ignites the fuel-air mixture in the ' combustion chamber during the first cycle. The pressure generated during combustion forces the hot gases through the heat exchanger where they cool below the dew point, thereby condensing and releasing the latent heat of vaporization into the heating medium (water or air). The exiting combustion gases create a vacuum in the ‘ combustion chamber, thus causing fresh air/gas mixture to be drawn in. The mixture is ignited by the residual heat from the previous combustion cycle, and repeats the cycle at a rate of 60 to 70 times i per second. 21.2.2 Technical Characteristics Scale of Operation High efficiency condensing systems emerged during the last decade as a response to a need for energy conservation. Presently, there is a fairly large number of systems in various stages of development, both { in the United States and abroad. The systems are designed for residential use with an input heating capacity ranging from 40,000 to 100,000 Btu/hr. Hydrotherm, Inc., is also introducing two Hydro-Pulse models at 125,000 and 150,000 Btu/hr, and Lennox Industries is introducing a model at 130,000 Btu/hr. The Hydro-Pulse boiler with a 100,000 Btu/hr input capacity, has a diameter of 14 inches and a height of 48 inches. The boiler is small enough to fit into a closet or a utility room. The Lennox furnaces are 49 inches high, 21 inches wide, and 26 inches deep. Pulse combustion condensing furnaces or boilers are smaller than conventional furnaces or boilers because the heat transfer rates of the pulsating products of combustion are much higher (by a factor of two to five) than they are in conventional furnaces or boilers, thus requiring a considerably smaller heat exchanger. 515 Increased Distribution Capacity (UA = 1000 Btu/hr°F) Increases Condensation Standard Distribution Capacity ‘| (UA = 440-730 Btu/hr°F) Qin "Tgs Projected Symbol Site. CFH % Seasonal 1H Space Heating Efficiency, % 50 ° SC-W4 95 91.0 86.3% 9 BG-H6 93 90.2 84.7% 4 BU-H2 96 91.0 82.6% o BG-H2 93 89.2 81.7% 10 Vv BG-H1 90 91.5 AVG. Lab Tests 0 10 20 30 40 50 60 70 Load Factor, % Source: Reference 2. FIGURE 21.4 FIELD TEST SPACE HEATING DATA FOR CONDENSING BOILERS 516 Performance Space heating systems cycle on and off to satisfy the variable heating load. On the average, the fraction of burner on-time (also called load factor) is 20% to 25%. The space heating efficiency during part-load operation is less than the efficiency during full load (or steady-state) operation because of heating losses during stand-by. The space heating efficiency of condensing space heating systems, as a function of load factor is shown in Figure 21.4, based on the results of field test measurements by Caron, et al(1)with five different condensing boilers. One important characteristic of condensing systems is that. they can operate at avery high efficiency, even at very low load factors. At a load factor of 10%, the efficiency is still over 80%. This compares with conventional space heating systems where the efficiency at a 10% load factor is usually less than half the steady-state efficiency. Another characteristic of the condensing boilers is that their efficiency: also depends on the capacity of the distribution system. An increased distribution capacity (large radiators) results in a higher space heating efficiency. This is because the temperature of the water returning from the radiator to the boiler is lower in an installation with large radiators. Colder return water temperatures in the boiler cause the products of combustion to cool further, thus allowing a larger fraction of the water vapor to condense and release a larger fraction of the latent heat of vaporization. Although the same basic relationships hold for conventional boilers, their space heating efficiency is not as sensitive to the return water temperature because the exhaust combustion products are always much hotter than the boiler water and they never condense. Simplicity of Operation and _ Reliability Condensing systems are, in general, as simple to operate as are conventional space heating systems. The reliability of the systems, although only on the market in the United States for three years, is claimed to be high. Potential problems with condensate disposal and heat exchanger corrosion have been addressed(2,3)and appear to be under control; new materials, corrosion resistant, have been developed for heat exchangers. Some manufacturers provide a 20-year warranty on the heat exchanger. A regular maintenance service once a year (as for conventional oil furnaces) is recommended to change filters and check the ignitor, flame sensor and general combustion operation. The spark plug and the air or fuel inlet valves in the pulse combustion systems are expected to be replaced every five years. One disadvantage of the pulse combustion condensing systems is the higher level of noise (68-70 dB in the room where the boiler is installed) compared to conventional systems. This level, although not 517 as high as the noise level from clothes washers, is not acceptable for relaxation and sleep. If the boiler can be installed in a basement, the high noise level is not a particular problem. However, if the boiler has to be installed close to the living area, as would be the case in Bush homes and many others in Alaska, additional noise reduction mufflers or sound attenuating walls or doors must be used to reduce the noise. The installation of a condensing system is a little more complex, in general, than the installation of a conventional heating system. Pulse combustion systems need a vibration eliminator and noise control mufflers; they also need to be connected to the drain system of the house to allow for the disposal of the condensing water vapor in the flue products (three to 10 gallons/day). The control system is also a little more complex than it is in conventional systems. Condensing systems require no chimney, though, because the products of combustion are directly exhausted outside through a (usually two inch) PVC pipe. The inlet air for combustion is also drawn from outside through a similar PVC pipe. The fact that condensing systems need no chimney offsets part of the higher cost of these systems and can make them cost effective in new buildings. Materials of Construction The combustion chamber and the first portion of the heat exchanger where the temperatures are highest, is generally made of cast iron. In some systems, steel fins are brazed onto the combustion chamber and/or the first part of the heat exchanger tail pipe to increase the rate of heat transfer. . The latter part of the heat exchanger tail pipe, where the exhaust gas temperature drops below 400°F, is made of copper or aluminized steel. One manufacturer uses a copper heat exchanger covered with a metallic coating for the latter part of the heat exchanger. Other manufacturers are considering using alternate materials such as teflon coated tubes, nickel alloys, or stainless steel to withstand the corrosive effects of acidic condensate. The exhaust pipe in most systems where the temperatures are less than 150°F, is usually a two-inch PVC plastic pipe. 21.2.3 Environmental Issues The American National Standards Institute sets a limit of 400 ppm (parts per million) for the concentration of carbon monoxide in the exhaust products of gas-fired furnaces or boilers. All condensing furnaces or boilers that are presently available in the U.S. market conform with this limit other. Other emissions are generally the same as for conventional systems. 518 When condensing furnaces first appeared there was an issue of disposal of the acidic condensate (pH 3.4 to 4.0) through the drain system of the house. The issue was addressed and solved.(2,3) In situ. corrosion-coupon exposure tests of metals, mortar and PVC plastic did not reveal accelerated corrosion or deterioration as a result of condensate addition to the drain system. It was concluded that the addition of the condensate to the municipal sewage treatment should not upset the sewage treatment system process, even under the condition of 100 percent market penetration of condensing furnaces. 21.2.4 Commercial Status Presently, there is a fairly large number of systems in various stages of development, both in the United States and abroad.(4) Several of the more important systems and their development status are summarized in Table 21.1. Of these, three manufacturers are in the United States: Hydrotherm, Lennox, and Smith-Jones. In Europe, seven different condensing systems will be on the market by year end 1983;(5)there were only two in 1980. ‘Extensive field tests with condensing systems over a period of three years pointed out areas where attention must be given but have also pointed out the feasibility of commercializing the condensing furnaces or boilers. The two significant areas of attention are: e additional sound attenuation; and e need for a_ service personnel training program. Pulse combustion condensing systems employ a completely different, yet simple, new design. Plumbers and service personnel need to be trained to install condensing systems and do a simple, once-a~year regular maintenance service. 21.3 Economic Implications 21.3.1 Costs Table 21.2 summarizes the estimated costs of installing new condensing and conventional gas-fired furnaces in Anchorage. The difference between them is about $1,100. Tables 21.3 and 21.4 show the life cycle costs of delivering the requires space heat to a residence. Based on the relatively low value of natural gas in Anchorage, it does not appear that this incremental investment can be economically justified on the basis of the value of the fuel cost saving. If the same analysis were done for an oil-fired system (not shown), assuming that the value of oil were $1.31/gallon ($9.35/million Btu), the delivered cost of space heat would be $14.48/million Btu for the 519 02S System Hydro-Pulse Boiler Lennox Furnace Turbopuls Boiler nie boiler Nefit Turbo Boiler Tappan Furnaces Raytheon-Amana Boiler Yankee Boiler Shock Hydrodynamics Furnace TABLE 21.1 STATUS OF CONDENSING FURNACE/BOILER DEVELOPMENT Developer Hydrotherm, Inc. Northvale, N.J. AGA Laboratories Cleveland, Ohio and Lennox Industries Carroll, Texas Turbopuls Geneva, Switzerland Nederlandse Netherlands Netherlands and KRUPP Warmetechnik Germany Smith-Jones Corporation Elyria, Ohio Raytheon-Amana Waltham, Massachusetts Yankee Engineering West Haven, Connecticut Shock Hydrodynamics Division Whittaker Corporatian North Hollywood California *Intermittent ignition device. Source: Compiled by Arthur D. Little, Inc. Burner Type Pulse (natural gas) Pulse (natural gas or propane) Pulse (natural gas or oil) Power (natural gas) Power (natural gas) ° Power with TID* (natural gas) Power with LID*™ (natural gas) Pulse (propane) Pulse (natural gas) Heating Capacity Btu/Hr 40,000 to 150,000 40,000 to 130,000 80,000 to 230,000 55,000 61,000 40,000 to 120,000 120,000 35,000 and 60,000 Up to 80,000 Status of Development On the market since 1979, approximately 20,000 units sold to date. On the market since 1981. Approximately 10,000 units sold to date. Oil-fired units have been on the market in Europe since 1979. Field prototype gas-fired units are ready and intended for the U.S. market. Free licenses granted to ten Nether- land manufacturers, five of them ready to commercialize the product. On the Netherlands market since 1981. KRUPP Warmetechnik will pro- duce Turbo in Germany. . On the market since 1979 under the Tappan trade name; 2,000 units sold to date. Two prototype units field-tested by ADL during 1981-1982. Propane-fired prototypes. Single prototype developed and laboratory tested. TABLE 21.2 CAPITAL COST SUMMARY TECHNOLOGY: Condensing and Conventional Furnace for Residential Use BASIS: Location: Anchorage Year: 1982 constant dollars CAPACITY: Condensing Input: 87 million Btu/yr Output: 76 million Btu/yr ESTIMATED USEFUL LIFE: 20 Years CONSTRUCTION PERIOD: Commensurate with building CAPITAL COST: Condensing Equipment and Materials 2,000 Direct Labor 400 Indirect Costs: 100 TOTAL CAPITAL INVESTMENT: $2,500 521 Conventional 109 million Btu/yr 76 million Btu/yr Conventional 500 250 650 1,400 TABLE 21.3 OPERATING COST SUMMARY TECHNOLOGY: Condensing Furnace for Residential Use 522 BASIS: Location: Anchorage Year: 1982 constant dollars CAPACITY: Input: 87 million Btu/yr Output: 76 million Btu/yr OPERATING FACTOR: ANNUAL ANNUAL OPERATING COSTS UNIT COST CONSUMPTION COST VARIABLE COSTS: Natural gas: $1.70/million Btu 87 million Btu $147.90 TOTAL VARIABLE COSTS $147.90 FIXED COSTS: Maintenance: . / 50.00 Capital Charges: 9.5% of Capital Investment 237.50 TOTAL FIXED COSTS 287.50 TOTAL ANNUAL OPERATING COSTS: 435.40 TOTAL ANNUAL OUTPUT: 76 million Btu TOTAL LIFE CYCLE COST: - $5.72/million Btu TABLE 21.4 OPERATING COST SUMMARY TECHNOLOGY: Conventional Furnace for Residential Use BASIS: 7 Location: Anchorage ~ Year: 1982 constant dollars be CAPACITY: ~ Input: . 109 million Btu/yr iy Output: 76 million Btu/yr OPERATING FACTOR: “ANNUAL ANNUAL OPERATING COSTS UNIT-COST . CONSUMPTION COST | VARIABLE COSTS: - Natural Gas: $1.70/million Btu 109 million. Btu $185.30 | TOTAL VARIABLE COSTS: 185.30 : FIXED COSTS: Maintenance . . 50.00 r Capital Charges: 9.5% of Capital Investment 133.00 I TOTAL FIXED costs : , , $183.00 i! TOTAL ANNUAL OPERATING COSTS: . $368.30 \ . TOTAL ANNUAL OUTPUT: 76 million Btu TOTAL LIFE CYCLE COSTS: - $4.85/million Btu 523 condensing system, and $15.82/million Btu for the conventional system. This implies that in other parts of Alaska, where oil costs are frequently higher than $1.31/gallon, condensing systems may be economical for new installations. 21.3.2 Socioeconomic Issues There are no significant socioeconomic impacts associated with the installation or use of residential condensing furnaces or boilers. We would expect that the normal distribution channels would be used for the distribution and installation of condensing systems. Some additional fluid service training required for plumbers and/or service personnel could be easily undertaken within the operation of the existing manufacturers distribution systems. 21.4’ Impact 21.4.1 Effect on Overall Energy Supply and Use Condensing furnaces and boilers have the potential to save 15% to 30% of the energy use for residential space heating compared to conventionally fired systems. Although this saving is potentially applicable for both oi] and natural gas systems, equipment only for natural gas systems has been offered in the U.S. to date. European companies have offered systems for both fuels. An analysis of the market potential of condensing systems showed that condensing systems could capture approximately 15% to 35% of the new construction and 12% of retrofit installations (upgrading of boilers for higher efficiency).(7) This analysis was limited to gas-fired systems, since all condensing systems presently in the United States are fired with natural gas or propane. But similar penetrations could also apply for oil-fired systems if they were available. In Alaska, the penetration of these systems would be quite variable, depending on the value of the fuel saved. For example, in Anchorage the relatively low cost of natural gas would diminish the attractiveness of a condensing system. At the opposite, in the Bush where delivered oil cost is quite high, the economics of conservation would be very favorable. 21.4.2 Future Trends New condensing systems have become available fairly rapidly in the United States over the past three years with several manufacturers offering residential systems. However, offerings of oil-fired systems have been limited thus far to Europe, and no plans are known for making them available in the U.S. in the near term. If they were available in Alaska, the major benefit of oil-fired condensing systems would be reduction in the cost of heating residences outside the Anchorage area where fuel costs are comparatively higher. 524 REFERENCES Caron, R.N., et al., "Field Measurements of the Performance of a Condensing Residential - Boiler," Volume -II Proceedings, Symposium on Condensing Heat Exchangers, Atlanta, Georgia, March 3-4, 1982, pp. 19-1 to 19-20. Stickford, G.H., et al., Batelle-Columbus Laboratories, "Flue Gas Condensate Disposal from High-Efficiency Gas Furnaces - A Field Investigation," Volume II Proceedings on Condensing Heat Exchangers, Atlanta, Georgia, March 3-4, 1982, pp. 21-1 to. 21-22 . / . DeWerth, D.W., and Waters, A.M.,; American Gas Association Laboratories, "Effect of Flue-Gas Condensate on Heat Exchanger’ Materials of Constructions," Volume II Proceedings on Condensing Heat Exchangers, pp. 2-1 to 2-17. New York State Plan to Encourage Commercialization of More Efficient Gas Technologies and Appliances, prepared for New York State Energy Research and Development Authority, Agency Building No. 2, Rockefeller Plaza, Albany, New York, .12223 and New York Gas Group NYSEARCH Committee, 500 Fifth Avenue, Suite 4120, New York, New York, 10110, by Arthur D. Little, Inc., August 1981. Noir, D., and Houlman, N., Batelle Geneva Research Centres, Switzerland, "European Technology in Condensing Flue-Gas Systems," Volume II Proceedings on Condensin Heat Exchangers, Atlanta, Georgia, March 3-4, 1982, pp. 10-1 to 10-15. : Personal communication with distributors of the Hydropuls boiler and Lennox furnace in the Lower 48. . Demonstration of Integrated Systems for Space Heating and Domestic Hot Water, Arthur D. Little Report #83390, prepared for Brookhaven National Laboratory, under Contract 473275-S, BNL Account No. 05178. 525 22.0 DISTRICT HEATING 22.1 Introduction and Summary 22.1.1 Technical Overview District heating is a method of supplying thermal energy from a centralized source to meet the needs of several consumers within a small geographic area. It was first developed in the United States in the late 19th Century and is now used widely throughout northern and eastern Europe. District heating systems include a heat source and a means of distributing the heat to the consumers. The heat source is most often a fossil fuel-fired boiler or an electric cogeneration plant, but other sources that have been used include municipal solid waste incinerators, local geothermal resources, nuclear power stations, and local industries with surplus process heat. The distribution system consists of a network of pipes, usually underground, which carries hot water or steam to consumers. Several benefits are derived from this concept. In European countries where fossil fuel is largely imported, it makes better use of this resource. District heating can reduce the cost of space and water heating for residents of a population center by using less expensive fuel and using it more efficiently than a number of separate residential units. Environmental benefits that can be derived from district heating include improved air quality and a means of disposal of municipal solid waste. In addition, these systems create jobs and stimulate the local economy. 22.1.2 Alaskan Perspective The suitability of district heating to any locality depends largely on local conditions and requires site-specific analysis. These systems have not received general acceptance in this country where the climate is relatively mild and energy has been both inexpensive and abundant. In contrast, European countries such as Sweden, Denmark, Germany, and the Soviet Union all have extensive systems in many population centers. In most cases this acceptance is a result of the colder climate and limited energy resources. Alaska is unique in that it is generally colder than most of the Lower 48. It has large energy resources, some of which are low-cost, particularly natural gas in the Anchorage area. In addition, there are only two population centers of sufficient size (population greater than 40,000) to support conventional district heating systems. In the larger, Anchorage, the low cost of energy makes the system difficult to justify economically, although the climate and population are favorable. In a case like this, considerations such 527 as the future cost and availability of fuel, as well as the construction period for a large municipal system (10 to 20 years) may influence system plans. One important consideration is the availability of geothermal energy. This is a non-polluting domestic resource with stable and predictable annual costs. In 1982, some 40 localities in the U.S. were studying or building geothermal district heating systems. In the Bush or other rural districts, village populations are small but the fuel costs and availability can be very high. In these cases, small systems may be justified, although care must be given to the additional costs of construction in that environment. A 1980 study of eleven rural villages in Alaska recommended cogeneration-based district heating for seven of them. The villages are depicted in Figure 22.1 and the recommendations are summarized in Table 22.1. In summary, the unique characteristics of Alaskan cities and villages may negate some of the generalizations concerning district heating and an in-depth analysis of site-specific conditions is needed to make a decision on a specific project. 22.1.3 Significance The technology of district heating is well developed. The practice is widespread in Europe and it is. receiving renewed interest in this country. There are no technical barriers to the development of these systems, although care must be given to unique environmental issues such as designing distribution systems for permafrost regions. The economics of district heating are deeply affected by local conditions, including construction costs as well as present and future. fuel costs. In a hypothetical DH project in Anchorage, heat would be supplied to the user for about 5.25/million Btu, while gas customers in that locality pay about $2.25/million Btu (assuming 75% conversion efficiency). In the Bush, however, the cost of heat can be more than $16/million Btu. Given this wide range of economic conditions, it can be concluded that district heating systems can only be evaluated on a case-by-case basis where the costs and impacts of each case are identified. 22.2.1 Operating Principles The basic principle of district heating is the distribution of heat from a single central source to a number of individual consumers. This permits a community to capitalize on the economies of scale while providing an efficient and reliable source of thermal energy for heating and domestic hot water. 528 ae KIANA e, AMBLER SAVOONGA © SHUNGAK ELIM @ KALTAG | | - GOODNEWS KODIAK br © Ap® - . *°AKHIOK i , iu i _- | | FIGURE 22.1 LOCATIONS OF 11 COGENERATION WASTE HEAT RECOVERY SITES i i 529 TABLE 22.1 te SUMMARY OF RECOMMENDATIONS FOR ELEVEN ALASKAN VILLAGES Site Recommended Plan ey Akhiok ” Plan 0 ve Ambler Plan 0 Angoon Plan 1 Elin Plan 2 Goodnews Bay - Plan 2 “ Grayling Plan 0 Kaltag : Plan 2 Kiana Plan 2 Savoonga “ Plan 2 o Shungnak Plan 1 1 White Mountain Plan 0 Key Plan 0 - No change Plan 1 ~ Diesel cogeneration, district heat with engine jacket ~ heat only . i | Plan 2 - Diesel cogeneration, district heat with engine jacket and exhaust heat recovery ut Source: Reference 1. . Wo 530 The heat sources for district heating (DH) are varied. In European systems, the most common sources are cogenerating electric plants and fossil-fueled boilers. Experience with these systems in Europe has been extensive. For example, Denmark supplies 35% of its total ‘heating requirement through 450 DH systems, and one third of this energy is from cogeneration plants.(2) The Soviet Union supplies 70% . the space heating and hot water requirements for urban areas and 4% of those for the entire country with DH (2) and 85% of that is copenerated. (2) . Other heat sources for DH include industrial process waste heat, municipal solid waste incineration, geothermal and nuclear energy. The thermal energy generated is distributed to the consumers in the form of either hot water or steam. Water, usually 150°F to 250°F, or steam (15 to 1500 psig) is distributed through a network of insulated, underground pipes located either in tunnels or buried in direct contact with the earth. DH systems can be either the two or the three pipe variety including supply and return lines for the. thermal transport fluid and in. the three pipe systems, a separate pipe delivers potable domestic hot water. In the two-pipe systems, domestic hot water, as well as space heating requirements, are met through heat exchange with the carrier fluid. Heat can be supplied for both warm air and hydronic space heating systems. Hydronic systems use a water-to-water or steam-to-water heat exchanger to heat the hydronic fluid which circulates through the building. Warm air systems tap the heat supply by including a water-to-air or, steam-to-air heat exchanger in the primary air supply duct. District systems can also meet cooling loads if necessary. A central chiller plant can be included in the system and chilled water can be distributed in separate pipes parallel to the hot water lines. An alternative is to use the existing high temperature water or steam supply, coupled with a local heat activated air conditioner such as an absorption chiller. As for much of northern Europe, the need for air conditioning in any Alaska community is virtually nonexistent and therefore this analysis will not consider chilled water distribution or thermal requirements resulting from heat activated chillers. 22.2.2 Technical Characteristics Scale of Operation ’ Although the economics of a particular district heating system are site dependent, some generalizations can be made concerning the economies of scale as they relate to DH systems. Some of these factors are: e reduced capital cost per unit of capacity of equipment with increased size; 531 e@ reduced operating costs from: (a) lower fuel costs resulting from bulk purchases; (b) lower fuel costs resulting from use of lower grade fuel; (c) lower fuel costs resulting from more efficient operation near capacity limits (sometimes including cogeneration), and operation with more effective control systems. (d) reduced maintenance resulting from the use of more rugged equipment and centralized, better trained maintenance staff. In addition to savings related to size alone, the practice of aggregating many loads tends to level the load curve which improves overall efficiency and reduces the system peak load below the total of the aggregated peaks. Safety is improved because equipment is. better maintained and removed from the sites where the heat is used. The delivery system is more reliable than overland transport by truck. Finally, the removal of most HVAC equipment produces additional living space in residential buildings and rental space in commercial buildings. The size of existing DH systems varies considerably. Many small systems serve the local needs of university campuses, military bases and industrial facilities both in the U.S. and abroad. Larger systems, serving the needs of entire towns or cities, are now commonly found in Europe as well as about 50 cities in this country. Because of the relatively large capital investment involved in the installation of DH and the variation of efficiency with size, a population threshold exists below which. district heat is not economical. West German studies in 197 indicated: that towns with populations of over 40,000 were suitable for these systems. (4) Maximum system size is only limited by population density. The Moscow district heating system contains more than 1,100 miles of major distribution lines alone,(5) and the West German government is studying a national: district heating grid supplied by large power stations. (4) Typically, the radius of service around an individual power plant extends 8-24 miles. Studies have shown, however, that transmission up to 50 miles can be practical and recent experimental work has opened the prospect of transmission distances of 100-200 miles. (6) 532 Performance The overall performance of a district heating system is a function of both the efficiency of the heat source and the losses in the distribution system. Distribution losses include heat leaks, fluid leaks and frictional loads. The efficiency of the heat source varies depending on the type of source and operating conditions. Design distribution losses are often assumed fixed at 10% of the system peak capacity because heat loss is a function of temperature which is relatively constant in the distribution system. A 1978 study of one operating system showed that thermal losses can be three to four times the design value in cases where insulation was wet or deteriorated.(7) Because the losses are fixed at a percentage of the peak load and the average annual load factor in a temperate climate is between 15% and 25%,(8) the annual integrated losses can be as high as 50% of the energy delivered. Improved materials, such as high temperature plastic pipe and foam insulations, and design practices based on experience with modern systems, will help to reduce some of the losses in distribution systems. The move toward hot water vs. steam distribution systems in: Europe is related to the following issues: e thermal losses per unit length of pipe are less for water systems due.to lower distribution temperatures; e steam systems without condensate returns lose sensible heat in the condensate; e steam systems with condensate returns have water quality problems related to pipe corrosion; e water ‘distribution systems cost less to install than steam systems; : e lower temperature systems favor electric production in steam turbine cogenerations systems; e pump energy for water systems is a small fraction of the total energy generated. For these same reasons, most DH systems now being considered for the U.S. are hot water based. : The performance of thermal plants varies widely by type. Efficiencies of some dedicated boilers and cogeneration systems are listed in Table 22.2. . Cogeneration systems include. steam extraction, gas turbines, and reciprocating internal combustion engines. Steam extraction systems can be used to supply steam or hot water. Gas 533 TABLE 22.2 AVAILABLE PERCENTAGE OF INPUT ENERGY FOR VARIOUS DISTRICT HEAT SOURCES Boilers Electric Fossil Fired Cogenerators Steam Extraction Reciprocating Engine Gas Turbine * Total Thermal Electric Usable 98 0) 98 85 0 85 75 10 85 38 27 65 43 18 61 Source: References 9, 21, and Arthur D. Little. 534 Can range as high as 30% depending on system configuration and temperature requirement of the waste heat. ™ turbines also produce high quality (900°F to 1100°F). heat which can produce either steam or hot water. Reciprocating engines, on the other hand, produce approximately 60% of their available waste heat in the form of hot water (180°F), or in some cases low pressure steam (15 psig). The balance of the recoverable energy from the reciprocating engine is in the engine exhaust at temperatures in excess of 350°F. The case of cogeneration of electric power and thermal energy has received a great deal of attention in the literature. Typical condensing steam turbine plants operate with conversion efficiencies in the range of 33% to 42%. The losses include 10% from the boiler and exhaust stack, 5% for auxiliary equipment, and 43% to 52% is condenser waste heat.(10) In the standard plant, steam is expanded through a power turbine to an exhaust pressure of about 0.6 psia (85°F). This exhaust temperature, however, is too low to provide usable district heat. The turbines can be modified to provide higher temperature energy either by reducing the overall expansion of the steam such as in a back pressure turbine or by bleeding off some of the steam in the upstream turbine stages. The latter case is known as.an extraction turbine. Both of these modifications reduce the overall electric. output of the plant, but for each kilowatt of electricity lost, five to ten kilowatts of usable thermal energy are made available for district heating use.(12) These turbines are illustrated in Figure 22.2. ‘ Simplicity of Operation District heating systems are simple from the consumer's standpoint. In most cases, the only system component to which the consumer is exposed is a heat exchanger and a circulating pump or blower. The nature of the concept is to centralize the complex components of the heating system. This approach justifies the full-time commitment of skilled personnel to system maintenance and repair. Additionally, large industrial sized equipment tends to be built more ruggedly than mass produced residential units. This results in less overall maintenance per unit of output than residential units. Reliability Reliability of district heating systems is closely related to the simplicity of operation. The distribution system is not susceptible to disruption due to adverse weather, as is the case with oil delivered by truck or barge. In most cases, the heat generating equipment is less prone to failure because it is more rugged and better maintained than residential equipment. Large units typically have a multi-fuel capability which makes them less susceptible to outages resulting from shortages of a particular type of fuel. Finally, because most systems operate at a load factor less than 40%, they can be designed with enough reserve heating capacity to meet system loads if one unit fails or is down for maintenance. 535 High pressure steam Generator District heater Medium Pressure steam Medium Condensate Pressure steam Condensate Condensate District heating To and from district heating system Feed pump a b. Conventional condensing turbine; without district heat a. b. Back pressure turbine; steam is not fully expanded in the turbine and residual energy is used for district heating. Heat to electric ratio is fixed. c Extraction turbine; steam for district heat is removed before the final turbine stages; system can be operated as full condensing or full back pressure. Heating and electricity production are decoupled. Source: New Scientist, April 8, 1982. FIGURE 22.2 COGENERATION SCHEMES FOR DISTRICT HEATING 536 ‘ oo The reliability of an underground distribution system is determined by the quality of corrosion protection, network water temperature, network age, pipe thicknesses, type of installation (tunnel or direct contact), local hydrological conditions, and adjoining utility systems. Failures are usually detected through visual inspection, pressure testing, infrared scanning and monitoring make-up water. Based on information from nine utilities in Europe and Japan, daily make-up water requirements range between 0.12% and 2.0% of the circulation rate. Time to repair failures has ranged between two hours and two months with a mean of about three or four days. The repair cost per failure averaged $8,400 (1978 dollars) and system maintenance costs varied between 1% and 6% of the network installation cost or between $1,800 and $18,000/mile per year (1978 dollars).(5) , 22.2.3 Environmental Issues The environmental issues concerning district heat center around reduced air pollution. Additional benefits are reduced thermal pollution: and the possible effective use of municipal solid waste. Air pollution is affected by cogeneration-based district heating in two ways: reduced total fuel consumption and cleaner fuel consumption. In the case where a centralized boiler is used to supply heat to a number of buildings and the boiler uses the same fuel as the systems it replaces, pollutants will be reduced because the central plant will operate more efficiently and emissions from such plants are regulated. In cogeneration case, much of the waste heat from the power plant directly replaces the use of fuel. As a result, less overall fuel is consumed and lower emissions are generated. These generalizations do not hold if the fuel for district heat differs from that which it replaces. For example, in the case of a coal-fired plant replacing gas space heating, increases of 200% or more are predicted for sulfur dioxide, nitrous oxides and particulates, along with decreases in.carbon monoxide and hydrocarbon emissions. (3) In addition to the impacts on air quality, several other possible environmental effects are noteworthy. Municipal solid waste can be used to augment boiler fuel, thereby reducing fuel requirements and the need for waste disposal. Noise and air pollution associated with fuel deliveries in residential and commercial areas are eliminated. Finally, thermal pollution at power plant sites is eliminated because waste heat is used effectively. . 22.2.4 Technology Status District heating. is a well-developed technology, particularly in Europe. “As noted earlier, large portions of countries such: as the Soviet Union, West Germany, Denmark and Sweden are supplied by district heat and, in most cases, these portions are growing. 537 ' Although the technology is not as broadly applied in this country, some 50 cities have had DH for as much as 100 years, and several modern systems are either under construction or being planned for both conventional and nonconventional heat sources. A 1982 paper identifies some 40 DH systems in the U.S. which are planned for using geothermal energy.(12) Some cities are examining the feasibility of supplying or augmenting DH systems with municipal solid waste incineration. Cooperative arrangements between industry and municipalities to make use of waste process heat are under consideration in several localities. Table 22.3 lists some of the unconventional DH projects that have been reported in ‘the literature. The technology for efficient extraction steam turbines is available, and in use both in this country and abroad in both DH and a wide variety of other applications. The technology currently used for distribution systems, however, is dated and is now beginning to change. New pipe materials such as fiber-reinforced plastics, pre-stressed concrete, and cross-linked polyethylene are under test. New insulations such as polyurethane foam have improved thermal characteristics and are impervious to ground water. These materials advances hold the potential of seriously reducing both installation and maintenance costs. Cross-linked polyethylene pipe ean be obtained in continuous rolls of up to 820 ft. With this pipe, modern trenching equipment and prefabricated polvurethane foam insulation blocks, a twin pipe system can be laid in a small fraction of the time needed to weld together short sections of carbon steel pipes. Maintenance costs are reduced with these materials because pipe corrosion is eliminated. Presently, cross-linked polyethylene pipes are only available in small diameters (4.3 in) and, therefore, not suitable for large transmission lines. Larger non-corroding pipes using materials such as fiberglass and prestressed concrete are under test at higher temperatures and pressures (850 psi at 200°F). Preliminary reports of these tests indicate that the results are promising. (13) 22.3 Economic Implications 22.3.1 Costs The economic characteristics were examined for a district heating system in a portion of the Municipality of Anchorage. This area, about one square mile, was suggested by the Anchorage Municipal Energy Coordinator because of the relatively high levels of population and commercial activities. The site (Figure 22.3) is part of an area previously studied for district heat, and it contains approximately 20 percent of the city's commercial fuel consumers and 13% of the city population. The city wide annual gas consumption for the two sectors 538 TABLE 22.3 / DISTRICT HEATING PROJECTS USING UNCONVENTIONAL HEAT SOURCES ; : : Geothermal ; &. m California - Susanville be : oo Colorado - Pagosa Springs o ~ / Idaho - Boise , : Madison County - : Preston i St - . Nevada - Elko [ . Hawthorn a _ Reno (- |. Oregon - Klamath Falls Lakeview i Washington - Ephrata | North Bonneville Thermopolis - Municipal Solid Waste | gs Akron, OH r i i Industrial. Process Heat q° Bellingham, WA 539 GLENN HIGHWAY Wi dda WLLL ULLLLLLLLLLLLLLM LLL LLL } | z ° oO enact <P is 7.97 trillion Btu and 7.94 trillion Btu, respectively, and the total population is 174,000.(14) Based on these data, the DH system would serve a population of approximately 23,200 and provide 1.99 trillion Btu annually (assuming an average annual heating plant efficiency of 0.75). The system peak capacity was estimated to be 394 million Btu/hr, based on experience in Finland which has shown peak per capita loads of about 17,000 Btu/hr.(15) If waste heat could be recovered at a rate of about 5,880 Btu/KWh from the Municipality's gas turbines at Plant #1, Units #1-4 would be able to supply a sufficient quantity of space heat (losses at 10%). This would require the installation of waste heat boilers, but no additional fuel would have to be burned. Based on the size of this system and the construction times reported in the literature,(17) the construction period is estimated to be 10 years. The system will require continuous maintenance and periodic replacement of all components. The average lifecycle is expected to be about 20 years.(18) The total: system capital cost was estimated using an average cost of -$225/KW of thermal heat delivery capacity (1978 dollars), based on the Minneapolis/St. Paul system.(16) _ This figure was inflated at 10% per year and increased by a factor of 1.71 to represent costs in Anchorage. Adding $10.5 million for waste heat boilers on this basis, the capital cost was estimated to be $75 million (Table 22.4). Of this figure, 80% is labor and 20% represents materials costs. Pump power was estimated to be 1.4% of the annual thermal load or 8.16 13 10° KWh at power plant efficiency;(19) the cost in gas is 6.9 x 10° Btu. Operating and maintenance was estimated to be 3% of the capital cost.(5) Taxes and insurance were assumed to be 1% of the initial investment and capital charges were estimated based on a 9.5% rate. Based on these figures, the total annual operating cost is $10,443,000 (Table 22.5). This translates to a delivered thermal energy cost of $5.25/million Btu. 22.3.2 Socioeconomic Factors In addition to saving energy and reducing pollution, district heating offers several other potential benefits to a community. These benefits include the creation of additional jobs, the retention of large sums of money, and the attraction of industry to the community. Construction and maintenance of a district heating system create jobs in the immediate population centers where thev are to be utilized. Built over a. period of 10 to 20 years, the conceptual Anchorage DH project would provide approximately 500 man-years of construction 541 TABLE 22.4 CAPITAL COST SUMMARY Technology: District Heat Basis: Location - Anchorage Year: 1982 Capacity: Input = Waste heat at no value Output = 394 million Btu/hr net peak Estimated Life: 20 years Construction Period: 10 years Capital Cost: Equipment and Materials: $15,000,000 Direct Labor 60,000,000 Total Capital Investment: $75,000,000 542 TABLE 22.5 OPERATING COST SUMMARY Technology: District Heat Basis Location: Anchorage Year 1982 Capacity: Input = Waste heat at no value Output = 394 million Btu/hr net peak Operating Cost Unit Cost Fuel: Natural gas Electricity (pumping) $3.9¢/KWh Fixed Costs: Operating and Maintenance Taxes and Insurance Capital Charges 9.5% of Capital Investment Total Fixed Costs Total Annual Operating Cost Total Annual Output Total Lifecycle Cost a 543 Annual Consumption 8,160,000 KWh Annual Cost $ 318,000 2,250,000 750,000 7,125,000 10,125,000 10,443,000 1.99 x 10! Btu $5:25/million Btu employment (based on Reference 11). In -addition, the 20-year lifecycle ensures that system will require continuous maintenance. The labor for both construction and maintenance can be _ supplied almost entirely by the community and the long-term development helps to alleviate the boom-bust cycle of most construction projects. 22.4 Impact 22.4.1 Effect on Overall Energy Supply and Use The major impact of district heating is a net reduction in fuel consumption. In addition, a district heating system can create jobs and improve local air quality. DH can help exploit local resources such as lunicipal Solid Waste (MSW) and geothermal energy, while decreasing dependence on premium value fossil fuels. At present, however, the economics of these systems may only be acceptable in isolated cases, and there are some regulatory issues that need to be resolved before the systems are widely used. District heating reduces overall energy consumption in a community. In the case where a heat-only boiler is used, the savings are. not large and result from the improved efficiency of central systems when compared to a number of individual units. In the cogeneration case, primary fuel consumption may be reduced by 50% to 85%.(3) In the Anchorage case discussed above, gas consumption of 2.63 trillion Btu/yr is eliminated by using power plant waste heat. This net reduction in fuel consumption can be an important factor for isolated rural villages which are dependent on fuel distribution over long distances. The availability of alternate fuel or heat sources is an important benefit. Alaska has a large geothermal potential which would be difficult to exploit for individual residences, but could be used to supply or augment a district system, especially in rural villages located near this resource. Municipal solid waste is another local resource that can be exploited by DH. MSW has a heat value 3,000 to 7,500 Btu/Ib (29) or about one-half that of coal. The preferred method of MSW _ disposal (sanitary landfill) is becoming more expensive and incineration now appears to be a viable alternative. Incinerators designed to extract heat. from the combustion products not only provide energy for the district heat system, but seriously reduce waste disposal problems. 22.4.2 Future Trends In the future, interest in these systems will remain high and applications will expand. Technical innovations such as_ high temperature . plastic pipe, improved insulations, and improved installation techniques will reduce the capital cost of new systems. 544 Finally, the continued increase in the cost of energy will make the systems more economically attractive for smaller population centers. There are several regulatory issues that need to be addressed with respect to district heating. The issue of public vs. private ownership of all or part of the system will impact the economic success of any system. In particular, if utility regulations limit yearly profits, they may need to be modified to insure that initial construction costs can be recovered. For city-owned systems, there is an issue concerning taxes subsidizing a service that serves only part of the community. This issue must be equitably resolved. In general, all of the regulations concerning utility operations need to be reviewed. in order to identify the operating constraints under which district heating must work. 545 10. 11. 12.- 13. 14, REFERENCES Raj Bargava Associates, "Rural Waste Heat Capture and District Heating Project," Alaska Power Authority, 1982. Strauss, S. D., "District Heating Linkes with Cogeneration," Power, August 1979, pp. 72-75. Sessler, G., "Environmental Effects of District Heating," Proceedings of Environmental Aspects of Non-Conventional Energy Resources, Denver, September 1978. ‘ Oliker, I., "Cogeneration Power Plants Serve District Heating Systems," Mechanical Engineering, July 1978. Oliker, I., "District Heating Survey: Piping Corrosion is Key to System Reliability," Power, October 1980. "Chemical Heat Pipelines - New Energy Delivery Concept," Pipeline and Gas Journal, May 1979. Vanderweil, G. and J. Donahue, "In-Place Testing of Thermal Distribution Systems," Heating, Piping, Air Conditioning, April 1978, Coad, W. J., “A Renewed Interest In District Heat," Heating, Piping, Air Conditioning, April 1981. "Electric Boiler for District Heating System," Power Engineering, November 1977. OO Denesdi, L., "Fuel Savings with Turbines Modified for District Heating," Power Engineering, February 1980. Leighton, G. R., "District Heating, Job Creation, and Economic Revitalization," Proceedings of the 8th Energy Technology Conference, Washington, D.C., March 1981. Fornes, A. O., "Direct Use District Heating Projects in the U.S. - A Summary," Gas Thermal Energy, March 1982. Margen, P., "New Types of Hot Water Distribution Systems for Low Density Heat Areas," Energy Engineering, January 1979. Estimates supplied by Peter Poray, Municipal Energy Coordinator, Anchorage via telephone conversations, December 20, 1982. 546 15. 16. = 17. L 18. - 19. bo 20. i 21. Tarjanne, R., "Prospects for Using Low Temperature Nuclear Heat," Nuclear Engineering, August 1977. Meyer, C. F., Potential Benefits of Thermal Energy Storage in the Proposed Twin Cities District Heating - Cogeneration System, U.S. Department of Energy No. ORNL/Sub-7604/2, October 1979. Schwarzenbach, B., "Economical Design of District Heating Power Plants," Brown Boveri Review, September 1977. ASHRAE Handbook and _ Product Directory, 1976 Systems, American Society of Heating, Refrigeration, and Air Conditioning Engineers, New York, 1976. Huxtable, D.D., and M. Olszewski, "Industrial. Waste Heat Utilization for District Heating," Proceedings of the 15th Intersociety Energy Conversion Engineering Conference, Seattle, Washington, August 1980. Ohlsson, O. O, "The Role of Refuse Derived Fuel (RDF) as an Alternative Energy Source for District Heating and Power Generation," Proceedings of. the 15th Intersociety Energy Conversion Engineering Conference, Seattle, Washington, August 1980. Cummings, .C.R., Cogeneration Energy Systems Assessment, Science Applications, Inc., Ladolla, CA. March 1982. 547 23.0 ENERGY STORAGE 23.1 Introduction and Summary 23.1.1 Technical Overview Energy storage applies to a number of technologies designed to increase system efficiency or reduce operating costs by storing energy. Three basic types of storage frequently used are mechanical, electrical, and thermal. Mechanical systems typically employ flywheels for short term storage although DOE is investigating the use of flywheels for daily storage of wind energy.(1) Electrical storage is done. by batteries for applications such as electric vehicles and wind turbine systems.(2) Thermal energy storage is used mainly for solar applications and for utility load leveling. DOE has sponsored a variety of projects aimed at developing new or improved storage systems for these applications.(3) This chapter deals with applications for energy storage which can benefit the electric utilities' generation mix, particularly those devices which could have immediate applicability in Alaska. : The use of thermal storage to provide space heating using off-peak electricity is intended to reduce utility operating costs by improving their load factors, e.g., shifting load from: the peak periods during the day to the off-peak periods at night. This has the effect of reducing operating costs by displacing more expensive peaking. fuel, such as gas and oil with less expensive fuel, such as nuclear, hydro or coal, and reducing capital costs by eliminating a need for a large reserve margin. capacity. The reduced cost is passed on to consumers in the form of an off-peak rate structure which provides an incentive to invest in thermal storage devices. The technique has been in use in Europe for many years and has successfully improved load factors for utilities there as shown in Figure 23.1. Although it can in some circumstances, in general the technology does not save energy as the operation of the utility is not necessarily more efficient due to the improved load factor. Also, there are increased losses at the point of use due to standby losses from the storage systems, which do not contribute to useful space heating. However, the shifting of load from peak to off-peak periods results in a reduction in the use of premium fuels and a substitution of less expensive, renewable, or more easily replaceable fuels. A number of companies presently offer various configurations of heat storage systems for residential and commercial applications. For residential applications, several different configurations are available. These are: 549 oss BRITAIN WEST GERMANY 1960 1,600 1972 1,200 % OF AVERAGE ‘DAILY LOAD 800 Mw 400 oO 0 12 24. 0 12 HOUR HOUR SOUTHWEST ELECTRICITY HAMBURG ELECTRIC WORKS BOARD (WINTER PEAK DAY) (JANUARY DAY) Sources: References 5 and 6. FIGURE 23.1 EFFECT OF HEAT STORAGE ON UTILITY LOAD PROFILES 1973 1971 1968 24 4 @ room heaters, which consist of blocks of ceramic materials heated electrically to temperatures as high as 1200°F enclosed in insulation in a cabinet. These units usually are equipped with a fan and thermostat to control the discharge of heat to the room; e central space heating furnaces for warm air systems in which a larger number of ceramic blocks are heated electrically. In this case, a conventional air handling system is used to heat the house; and e water storage in which water is heated in large tanks at night and circulated through a hydronic heating system during the day to provide heat as needed using conventional circulating pumps ‘and thermostats. For commercial installations, two main types of systems are utilized. In one type, used in buildings constructed on a slab, electric resistance coils are imbedded below the slab above a layer of sand.” The heaters are activated at night, heating up the concrete slab and the surrounding scil. During the day, heat leaks from the slab into the building being heated. No means of controlling the amount of heat release are available. However, the amount of energy put into the slab at night is controlled by sensing the outdoor temperature to match the amount of heat in storage to the space heating load likely to be encountered in the following day. In Canada, slab storage systems have also been used in multistory buildings with poured eoncrete floors. The second type of system uses large tanks of pressurized water to store energy. These tanks are constructed to take water at temperatures and corresponding pressures up to 280°F. Users of residential and commercial thermal storage systems are given a rate structure which contains a differential cost between peak and off-peak rates. The off-peak rate is substantially lower than the rate offered to customers without storage and the savings from this differential justifies the investment in the storage equipment. Some rate structures make electric space heating competitive with alternative fuels. In commercial applications, the benefit to the user comes, not only from the rate differential between day time and night time use, but also from the reduction in peak demand for electricity, and which results in smaller monthly demand charges. 23.1.2 Alaskan Perspective Thermal storage systems should be quite suitable for use in Alaska. They are most appropriate to areas where utilities are winter peaking and where space heating comprises a significant part of the utility load. The technology is not sophisticated and the fabrication and installation of the units can be done using labor which is relatively unskilled. Environmental impact would be negligible except for slab 551 incremental Costs( 10°3 dollars / kWh ) | ht Lt ; “7 ty Nuclear . ~ 0 1 2 3 4 5 6 7 > Power Demand (10© Kilowatts) ry Source: Reference 8. ; ot FIGURE 23.2 INCREMENTAL COSTS AS A FUNCTION OF DEMAND FOR THE REFERENCE POWER POOL . 552 storage systems which could not be used in permafrost regions. The switch to off-peak use should enable utilities to substitute the use of the electricity generated by the hydro power which is readily available for higher priced, non-renewable fuels. In Bush region communities where there are smaller population bases and little industrial or commercial loads to even out demands, peaks are more exaggerated, thermal storage systems could be used to provide a more even demand. This could be particularly attractive with diesel generators which could operate more efficiently. In addition, loads might be met with fewer or smaller diesel units. 23.1.3 Significance of the Technology As mentioned above, the technology employs commercially available units which are manufactured by a number of firms. The economic benefits to Alaskan utilities can really only be evaluated by carrying out a comprehensive analysis of the utility-generating mix and the benefits from shifting load off-peak. This must be done before the appropriate rate structure can be developed. However, because of the large space heating load. and the magnitude of Alaskan winter peaks, it seems probable that a rate differential which would make storage systems attractive to residential customers could be justified. Based on the analysis below (Section 23.3.1) a differential of 4¢/KWh or greater would produce a reasonable customer payback. 23.2 Description of Technology 23.2.1 Operating Principles Basis for Use of Energy Storage Storage of electricity generated during off-peak times is designed to reduce utility operating costs. The cost savings come from the use of less expensive or more efficiently used fuel for generating electricity and from reduced capital costs due to a reduction in the reserve margin required. Typically, utilities use hydro, nuclear or coal-fired plants to generate the electricity required to meet their base load. Peak demands are met using oil or gas-fired generators at a substantially higher cost/kilowatt hour generated as shown in Figure 23.2 which illustrates the incremental costs to meet different power demands for a typical utility.(7) In Anchorage where natural gas is the predominant fuel for both baseload and peaking, eliminating some peak load demand means producing a larger proportion of electricity in more efficient combined cycles. Thermal storage is not principally a conservation technology. Energy use (Btu/kilowatt hour) may be reduced slightly at the utility depending on the mix of generating capacity as would be the case in Anchorage. However, in some cases more energy may be required to 553 meet the load with storage than without. Referring to Figure 23.2, several representative heat rates (the energy input/kilowatt hour output) for equipment which would be dispatched to meet different load levels are shown. For each type of input (e.g., coal, oil, gas) there is a range of efficiencies. Because of the difference in the cost of fuel, in many cases it costs less to operate less efficient baseload units than to operate more efficient equipment. For example, switching load which would have been met by an oil unit with a heat rate of 9,300 Btu/KWh (A) to a coal-fired unit with a heat rate of 12,000 Btu/KWh (B) would result in less efficient but more cost effective operation. Thus in this case, an energy savings would not result from an improved load factor, even though utility operating costs were reduced. However, there should always be a reduction in the amount of fuel used to meet peaks which is usually oil or gas. In addition to the possibility for a less efficient operation at the utility, storage systems suffer from standby losses at the point of use. Depending on the type of thermal storage system used, (e.g., room heaters), some of the standby losses will be useful. For others, these losses will not be recovered and an additional loss in energy efficiency will result. Storage of off-peak electricity as domestic hot water has also been used by some utility systems. Again, since a larger water tank with greater standby losses is used, more energy is required per gallon of hot water. Thus, although a cost benefit may result for the user of the system due to reduced electric rates resulting from lower utility operating costs, the actual energy use for space and hot water heating will increase. The amount of increase will depend on whether the storage units are located in a normally occupied place. The actual benefits to the utility from the use of off-peak storage depend to a large extent on the load curve for the utility and the type of generating equipment which is available. Thermal storage has been used in Europe for a number of years and in some cases, the reduction in the peak demand coupled with an increase in off-peak demand, has resulted in an overloading of the distribution system. This has occurred because the cost effectiveness of electric space heating and water heating has increased due to the reduced generating costs. These problems can be prevented by adequate planning. Description of Thermal Storage Systems In general, a heat storage medium is heated electrically at night using lower cost off-peak electricity and heat is withdrawn from the storage system during the day to provide space heating as required. These systems may be utilized in residential or commercial applications. The basic types of systems currently in commercial use are ceramic bricks (room and central), pressurized water and slab heaters. 554 The first type, which has been in use in Europe for the last 20 to 30 years, utilizes ceramic bricks or cast iron as the heat storage medium. Electric resistance heaters are positioned between blocks of heat storage medium. Three different configurations are used. In the first, the unit is positioned in a room, heated at night, and discharged during the day passively. Minimal manual controls are provided to adjust the amount of heat given off by the unit. In the second, shown schematically in Figure 23.3, a fan and thermostat are added to the unit to provide controlled heat withdrawal. In this case, heat still leaks into the room through the insulation, but higher rates of heat exchange can be obtained by turning on the fan. The third configuration is a central storage unit. The system operates, in principle, the same as the controlled heat withdrawal unit. In this case, however, a much larger heat storage capacity is available. In addition, the units may be designed to be adaptable to the use of a central air conditioner. An illustration of a central system is shown in Figure 23.4. These systems are all tailored to residential installations. They may be equipped with a more sophisticated control system which senses the outdoor temperature and adjusts the overnight charge accordingly. This adjusts the system charge to the load and prevents room overheating due to heat leaks through the insulation during the day. In addition, the standby losses are reduced by maintaining lower storage temperatures when less space heating is required. The second type of system which may be emploved in residential* or commercial buildings uses water as the heat storage medium. In this case, water in a large storage tank is heated during the night by resistance heaters and circulated through the building's hydronic heating system or through a fan coil unit in a hot air duct during the day to provide space heating see Figure 23.5. The storage temperature varies depending on the construction of the heat storage tank. Megatherm Corporation manufactures a system which uses tanks designed to withstand relatively high pressures (e.g., 45 - 50 psig) and temperatures (as high as 280°F.) In this case, a heat exchanger in the tank is used to heat the water which is circulated through the building so that the hydronic heating system itself does not experience high temperatures. These systems have been most successful in commercial applications such as hospitals, schools, etc. Steibel-Eltron (West Germany) offers a system which uses more conven tional storage tanks and operates at pressures. typically encountered in hydronic systems. They suggest the use of the system with tubing buried in concrete floors for commercial buildings and with conventional forced water convectors for _ residential buildings. * Residential systems were offered in a 240 gallon size by Megatherm in the late 1970's but are now not produced. 555 ELECTRICAL DISTRIBUTION PANEL OUTDOOR TEMPERATURE} CHARGE SENSOR TIME CONTROLLER © INSULATION BRICK CORE |; A> HEATING ELEMENTS ROOM THERMOSTAT DAMPER Source: Arthur D. Little, Inc. FIGURE 23.3 ROOM STORAGE HEATER 556 BRICK CORE INSULATION HEATING ELEMENTS MOTORIZED DAMPER (DAY POSITION) NIGHT HEATING SECTION SUPPLY AIR Source: Arthur D. Little, Inc. FIGURE 23.4 CENTRAL STORAGE HEATING SYSTEM 557 Supply Air Supplemental Resistance Heat House Thermostat |(|- Heat Storage | Off-Peak Resistance Heat lO " Return Air Source: Arthur D. Little, Inc. FIGURE 23.5 PRESSURIZED WATER STORAGE HEATING SYSTEM . 558 ” ’ The third type of storage system, used primarily in commercial buildings, utilizes the storage capacity of the concrete floor of a slab type building and the sand or earth under the floor as the storage medium (see Figure 23.6). Electric resistance heating ‘is provided either through conventional heating elements or by electric heating of the steel reinforcing bars. Again, the system is charged by heating the floor area electrically at night using off-peak electricity. In this case, no means of controlling the heat discharged during the day is provided. However, the temperature swing is small, and the amount of charge can be adjusted depending on the outdoor temperature. 23.2.2 Technical Characteristics | The input to the storage systems and the amount of storage used depends on the building thermal load and the length of time over which the system is to provide heating. Since it is necessary to meet night time loads plus charge the system, the connected electric load will be 100% to 150% greater than the design load. Table 23.1 summarizes characteristics of typical systems. TABLE 23.1 CHARACTERISTICS OF THERMAL STORAGE SYSTEMS Input (KW) Capacity (KWh) Central Storage Brick 12-30 80-220 Room Heater Bricks 2-6 15-40 Water Storage (Commercial) 40-800 250-3000 Slab (Commercial) 100-800 . 1200-9600 Thermal storage units utilizing ceramic bricks for residential space heating are offered in a variety of sizes which cover the range of applications likely to be encountered in Alaska.. In general, a unit would be selected based on the house heating load and the desired storage period. Installation of the system is relatively straight- forward requiring only the installation of wiring carrying 220 Volts, the wiring of thermostat, and the installation of the device itself which, due to the weight of bricks used for storage, is typically shipped to the site in boxes and assembled on location. Installation requires the use of an electrician but the hook-up is no more complicated than for conventional baseboard system heaters or an electric furnace. The operation of the units is quite simple. Residential ceramic brick units require little servicing and are analogous to baseboard resistance heaters for the room units, or a central electric resistance heating furnace for the central units. Systems employing water for hydronic systems are analogous to an electric boiler or water heater. 559 09s Supplementary Resistance Heat slab f" Slab Heating Reinforcing Rods Heat Reservoir Source: Arthur D. Little, Inc. FIGURE 23.6 GROUND STORAGE HEATING SYSTEMS BLELOZ CUR ZEISS Insulation Heating Mat Below Slab ee ee Commercial installations using slab heaters involve a technology which has been in use for many years to heat buildings. Should a heater fault develop, repair is somewhat more complicated as the heating element is buried under the slab. Typically multiple heaters are used which are wired together external to the slab so that a defective heating unit can be bypassed. In general, the reliability and satety of all the storage systems should be high. An important consideration in the use of off-peak storage is the method which will be used to control charging the storage systems. At present, time clocks ,at the user's location are typically used. These have the disadvantage that power outages interrupt the clocks, causing the storage system to get out of the desired operating cycle. Batteries operated backups would help alleviate the problem but they are not currently used. Several systems have been developed which _ permit the utility to control the charging of the system from a central location. Possibilities include various methods for sending a signal over the power lines to a receiver at the house.and the use of radio controlled devices. These: types of systems will permit the utility to turn off devices depending on the peak load they are experiencing and thus to have a greater control in the event that the peak demand exceeds the generating capacity. Various operating strategies have been evaluated. At this time, no large scale implementation of these systems has been undertaken by American utilities. Detroit Edison has experimented with ratio controlled water heaters on several thousand -homes.* 23.2.3 Environmental Issues Storage heaters for residences should involve no environmental issues in the Alaskan environment. These units are so similar to the present technologies in operating concept that no risk should be involved. Systems employing the heated slab method could obviously not be used where permafrost protection was a consideration. At the central utility, more baseload fuel and less peaking fuel would be burned. Depending on the generating mix, this could increase or decrease air pollution. The use of storage plus hydro baseload would minimize this impact. 23.2.4 Technology Status The systems described above are available from a number of suppliers (Table 23.2.) In some cases, they are prototypes rather than production units. The only area where development work is occurring at this time, involves a search for better heat storage * Reference 9 summarizes these tests plus tests by a number of other U.S. utilities. . 561 TABLE 23.2 MANUFACTURERS OF HEAT STORAGE EQUIPMENT Residential Systems Ceramic Bricks (Room Units) Control Electric Corportion Burlington, VT. 05401 (802) 658-3026 Thermal Acoustics Corporation Rutland, VI 05701 (802) 775-2020 Siemens Corporation Iselin, NJ 08830 (201) 494-1000 Stiebel Eltron Boston, MA (617) 367-1669 Emerson Electric St. Louis MO 63042 (314) 595-2500 Ceramic Brick (Central Units) TPI Corporation Johnson City, TN 37601 (615) 929-1761 Pressurized Water Methatherm Corporation East Providence, RI 02914 (401) 438-3800 Patterson-Kelly Co. East Stroudsburg, PA 18301 (717) 421-7500 Ground Heat (Slab) Smith Gates Farmington, CT 06032 (203) 677-2657 Easy Heat Corporation New Carlisle, IN 46552 (219) 654-3144 Peak Supervision Controls, LTD.Laval, P. Quebec H7L 353 (514) 687-2710 Modular Comport Systems, Inc.Camillus, NY 13031 (315) 672-3168 Gen Tech, Inc. Quebec, P. Quebec G1P 359 (418) 651-8000 MacDonald-Wadman Co. In Needham, Heights, MA 02194 (617) 449-2500 . 562 materials. The ceramic blocks. used are cycled from room temperature to approximately 1200°F. A continuing effort is being made to identify storage materials which would have an equivalent volumetric storage capacity with a lower operating temperature range or a reduced cost.. However, the present materials have no technological problems. 23.3 Economic Implications 23.3.1 Costs “( ‘In residential applications, the benefit to the user results from the price differential between peak and off-peak electricity. In general, the economics of storage heater use is very sensitive to the differential. For the commercial user, most utilities have a demand charge which is based on the 15-minute peak demand from each site during the billing eycle. The use of storage systems, coupled with load control devices can limit these peaks and result in significant savings due to the reduced demand charges in addition to savings which result from the differential between peak and off-peak electric costs. For most commercial applications, the demand charge reduction is the more important of the cost savings available at this time. In order to evaluate the economics of a heat storage application in Alaska, we have selected a system which would meet the requirements of a house with a design heat loss at the coldest temperature (-30°F) of approximately 30,000 Btu/hr in the Anchorage area (House "B" - Table 23.3). Over the coldest day, the heat loss would be less due to variations in temperature. Based on historical temperature data, the load for the worst day would be 180 KWh or an average rate over the day of about 25,600 Btu/hr. The system is designed to provide 16 hours of peak heating with an 8 hour charge period. The estimated installed cost of this system is $6000, including the room heating units, -wiring, thermostats, and the system which modifies the charge level based on _ outside temperature. Approximately $2,200 of this cost would be for installation labor. As the economics of the use of these systems is highly dependent on the rate incentive, the results are presented on a parametric basis relating the years to payback for the system to the rate differential between off-peak electricity and the average conventional residential rate. It is assumed that all space heating requirements are met using off-peak rate electricity. The relationship of years to payback and rate differential are presented in Figure 23.7. The baseline economic calculations for the figure are summarized in Tables 23.4 and 23.5. 563 Annual Load (kWh) ‘House "A" 14,850 House "B" 22,280 House "C" 44,560 TABLE 23.3 HEAT LOAD AND STORAGE REQUIREMENTS FOR THREE HOUSES IN ANCHORAGE Part of Load During Utility Peak (kWh) 9,900 14,850 29,700 Assumptions: e 16 hours of storage for peak Worst Day Load (kWh/day) 120 180 360 Required Storage (kWh) 80 120 240 e Load uniform throughout 24 day (nighttime set-back) e Standby losses uniform throughout season due to use of outdoor temperature sensor. 564 Years to Pay Back Incremental Cost of Storage System Installed Cost of Conventional Electric — $2700 Installed Cost Room Storage System — $6000 22,000 kWh/yr Space Heating Load 90% Storage Efficiency All Loads at Off-peak Rate 15 10 Typical Differential Used in Utility Demonstrations 1 2 3 4 5 6 7 8 9 10 Rate Differential (Average Rate to Nonstorage User — Rate to Off-peak Storage User) (¢ /kWh) Source: Arthur D. Little, Inc. FIGURE 23.7. IMPACT OF RATE DIFFERENTIAL ON CUSTOMER PAYBACK FOR TYPICAL SPACE HEATING APPLICATIONS 565 TABLE 23.4 CAPITAL COST SUMMARY TECHNOLOGY : Thermal storage for electric space heating BASIS: Location: Anchorage Year: 1982 constant dollars: CAPACITY: Input: 20 kW Rate/160 KWh Storage capacity * Output: 90% of input is useful ESTIMATED USEFUL LIFE: 20 Years CONSTRUCTION PERIOD: Concurrent with construction of buildings. CAPITAL COST: Equipment and materials $3,800 Direct Labor 2,200 TOTAL CAPITAL INVESTMENT: $6 ,000 * Based on 8 room units with 20 kWh capacity to present reasonable distribution in house. 566 TABLE 23.5 OPERATING COST SUMMARY TECHNOLOGY : Thermal storage for electric space heating BASIS: Location: Anchorage Year: 1982 Constant dollar CAPACITY: Input: 20 kW Rate/160 kWh storage capacity Output: 90% of input is useful ANNUAL ANNUAL OPERATING COSTS UNIT COST CONSUMPTION cost. VARIABLE COSTS: Energy: * : kk Electricity: $2.5¢/kW 24756 kWh $619 TOTAL VARIABLE COSTS $619 FIXED COSTS: Maintenance: Assumed negligible Taxes and Insurance: Assumed 1% of total capital cost $60 Capital Charges: Assumed 9 1/2% of total capital cost $570 TOTAL FIXED COSTS: $630 TOTAL ANNUAL OPERATING COSTS $1249 TOTAL ANNUAL OUTPUT: 22,280 kWh 76 million Btu TOTAL LIFE CYCLE COST: $16.43/million Btu : . Assumed off-peak rate lower than average (3.9¢/kWh) ke 22,280/0.90 567 23.3.2 Socioeconomic Issues Some of the heat storage units described above are currently manufactured in Europe or in the United States under a license from European manufacturers. The technology is straightforward and the units could easily be manufactured in Alaska using conventional fabrication techniques. Due to the weight of the brick storage units, the shipping costs would be high if the units were manufactured elsewhere. As noted above, installation of the units is straightforward and could be done using electricians who are familiar with installing electric heating systems. Manufacturer's installation instructions are straight- forward and difficulties with building codes should not be encoun- tered. 23.4 Impact 23.4.1 Effect on Overall Energy Supply and Use Heat storage units can be used in new homes or retrofit to existing homes. The room units could replace existing baseboard units in electrically heated homes. A central hot air furnace could be used in place of an existing electric furnace or fuel fired furnace. Depending on the price of various fuels, low off-peak rates could encourage customers to switch from one type of fuel (e.g., gas or oil) to electric. For example, electric heat at an efficiency of 90% (to allow for storage losses) and off-peak cost of 3¢/kilowatt hour would cost (energy only) $9.77/million Btu of delivered space heating. An oil furnace with a seasonal efficiency of 65% and an oil cost of $1.40/gallon would provide useful space heating at a cost of $15.38/million Btu (energy only). . 23.4.2 Future Trends The primary factor relating to the economics of storage heaters is the rate differential offered by the electric utilities. This is typically controlled by a public service commission. Should the _ utilities determine that off-peak storage would offer significant advantages based on their load curves, the implementation of the rate structure changes would follow normal procedures. However, the trend by some rate setting agencies and consumer activist groups to promote flat rates for all users removes the incentive as it does not recognize the utility cost savings from storage and pass it on to users. No other barriers to the use of the heat storage systems are likely to be encountered. Manufacturers are currently offering units for sale on a commercial basis and would be eager to promote the use of the technology in Alaska’ should an appropriate rate structure be developed. - 568 In the Railbelt, if Susitna is implemented, the low fuel cost of this large hydro source may make differential rates even more attractive for thermal storage in electrically heated homes. Similarly, small scale hydro or other locally available sources for baseload power in the Bush could encourage the use of thermal storage, especially if it enables optimal loading - or even elimination of diesel generators. 569 REFERENCES Barlow,T.M., et al., "Mechanical Energy Storage Technology Project: Annual Report for Calendar Year 1980," Lawrence Livermore National Laboratory UCRI-50056-80 May 1, 1981. O'Connell, L.G., et al., "Energy-Storage Systems for Automobile Propulsion: Final Report Volume 1: Overview and Findings," Lawrence Livermore National Laboratory UCRI-53053-80, Vol. 1 and Vol. 2, December 15, 1980. MCC Associates, Inc., "Proceedings of the Sixth Annual Thermal and Chemical Storage Contractors’ Review Meeting," for U.S. Department of Energy Assistant Secretary, Conservation and Renewable Energy, Division of Energy Storage Under Contract No. W-7405-ENG-26 - Conf-810940, September 14-16, 1981. Asbury, J.G. and A. Kouvalis, "Electric Storage Heating: The Experience in England and Wales and in the Federal Republic of Germany," ANL/ES-50 Argonne National Laboratory, prepared by U.S. Energy Research and Development Administration under Contract W-31-109-ENG-38, May 1976. Peddie, R.A. "Peak Lopping and Load Shaping of the CEGB's Demand, Institution of Electrical Engineers Seventeenth Hunter Memorial Lecture, Engineering Services Department, CEGB (1975). Die Oeffentliche Elektrizitaetsversorgung 1973, Vereinigung Deutscher Elektrizitats-Werk (VDEW), Frankfurt am Main, West Germany. , Lawrence, W.T., et al. "Energy Conservation Opportunities in Appliances using Energy Storage," Final Report, prepared by Arthur D. Little, Inc. for Oak Ridge National Laboratory/Union Carbide Corporation for the U. S. Department of Energy ORNL/Sub-7303/1 Dist. Category UC-95d, November 30, 1978. "An Assessment of Energy Storage Systems Suitable for Use. by Electric Utilities," EPRI Research Project #22 5, ERDA #E(11-1) 2501, Report #2-64 July, 1976, Public Service Electric and Gas Company, Newark, New Jersey. "Survey of Utility Energy Storage, Load Management and Energy Conservation Projects," Phase II Preliminary Report (draft), Energy Utilization Systems, Inc., Pittsburgh, Pennsylvania, prepared for ERDA Power Distribution Branch under Contract ORNL/SUB-7713509/1 and for the Electric Power Research Institute August 5, 1977. , 570 24,0 SOLAR SPACE HEATING - PASSIVE 24.1 Introduction & Summary 24.1.1 Technical Overview Passive solar heating is a series of building practices and techniques for arranging architectural components which work together to gather and store energy from the sun. Many of the concepts of passive solar heating are not new and have been applied for many years, but some, such as phase change storage involve new materials or new design ideas. Passive solar heating can potentially cut fuel bills on new residences by 20-80%. The techniques are primarily applicable to new construction as retrofitting would usually need to be extensive. The reduction in fuel use is given as a broad range because it depends so heavily on individual designs. In general, designs which heavily cut fuel use are also expensive. In Alaska, attempting to cut fuel use by more than 30-40% is usually not economically justified. Implicit in sound passive solar design is energy conservation to minimize heat loads and, hence, the need to increase heat gain. However, conservation is equally well applied when other heat sources are used. Therefore, this chapter focuses on the principles of heat gain and heat storage. 24.1.2 Alaskan Perspective Winters in Alaska are colder and less sunny than in the Lower 48. The most economical scale system in Alaska is therefore probably not going to save as large a percentage of the annual energy use as an optimal system in the Lower 48. However, with heating loads being very high in Alaska, even a smaller percentage can be a considerable fuel savings. The primary cause of this reduced percentage of energy savings is the very limited sunlight available for the 45 days before and after the winter solstice in Alaska. During those three months, throughout most of Alaska, the amount of energy captured from sunlight entering the south-facing windows will not compensate for the energy lost from having the windows exposed to the sun. During that period, windows should be covered with proper insulation devices in order to reduce heat loss substantially. Other factors affect the use of passive solar heating in Alaska: @ permafrost, which causes many Alaskan buildings to be built above direct contact with the ground, to avoid melting the tundra beneath them. This makes the common practice of storing solar heat gain in a masonry floor infeasible; 571 e materials availability, particularly in the Bush where, for example, concrete is an expensive commodity. The common practice of using masonry as a heat storage medium is relatively expensive in many areas of Alaska; in fact, most construction materials are expensive to ship and, hence, the tradeoffs within Alaska vary substantially and are very different from the Lower 48; : e limited Alaskan research on passive heating, requiring most of the design information available to be based on Lower 48 experience. In many cases, charts for design use do not cover above latitude above 52° north, and thus Alaska is left without the required design assistance; e low humidity in the winter which causes high static charges on insulating materials. This makes some movable insulation systems hard to operate, as the insulation clings to the window; and e extreme cold and heavy winds, which affects -movable insulation. Ice can form between movable insulation panels and glazing, firmly attaching the insulation to the glazing; external movable insulation is ravaged by high winds and icing outside, making this application impractical. Nevertheless, Alaskans have an important inducement to using passive solar; the cost of heating a home can be extreme. Liquid fuel in the Bush is expensive and the heating season is long. Alaskans have, therefore, begun to incorporate many passive solar techniques into building designs. The Anchorage Health Center, for example, is using passive solar heating and movable insulation. Several passive solar’ homes have also been built in Anchorage, Homer, Juneau, Delta Junction, Copper Center, Ambler, and Fairbanks. Homeowners in Alaska are beginning to design their own passive systems or incorporate specific passive techniques. In comparison to the Lower 48, however, Alaska has relatively few passive solar homes. 24.1.3 Significance of the Technology Passive solar heating techniques are technically feasible and widely applied throughout the United States. It is difficult to measure the cost of adding passive solar features, as they are an integral part of the building design, although an additional cost of 10-15% of the total house cost is considered a good rule of thumb. Most passive heating techniques are not easily retrofitted to existing buildings. Hence, passive solar heating is not likely in the near term to alter radically residential energy use in Alaska. But, as the housing stock is replaced during the next few decades, many passive 572 design features will probably become standard. Heavy insulation and reduced north-facing windows, for example, are already widely followed building practices. 24.2 Description of Technology 24.2.1 Operating Principles The purpose of designing a passive solar heating system is to choose materials and locate them, using the principles of heat transfer, to take advantage of some predictable solar facts and to accommodate human. living styles. Therefore, starting with the needs of people; defining the sun's properties, describing heat transfer, discussing material properties and, finally, presenting passive designs is a logical approach to explaining the operating principles of passive solar heating systems. ‘ , The Needs of People In general, the more rapidly a person loses his or her body heat to the surroundings, the cooler and less comfortable the person will’ feel. Typically, in a 74°F (air temperature) room with 50% humidity, a person loses 25% of his or her. heat loss through the evaporation of perspiration, 25% through convection to the surrounding air, and 50% through radiation to the walls.. All other factors equal, if the wall temperature increases, the amount of heat lost by radiation will decrease and cause the person to feel warmer, even if the air ‘temperature is unchanged. These circumstances promote the use of passive solar designs with the thermal mass on the inside of the building and insulation. on the outside of walls. This keeps wall temperatures higher and improves human thermal comfort, while it increases thermal storage. Similarly, the need to minimize convection (drafts) implies that the rate of air circulation by natural or forced convection should be minimized in passive solar designs. Solar Properties There are two facts about solar radiation that affect passive solar design: e@ the average total amount of radiation ‘per day; and e@ the angle at which the radiation strikes the collection surface. The average amount of radiation striking a particular collection surface influences the amount of energy that can be usefully derived from the sun over a period of time. The sun's radiation is partially blocked by the atmosphere, and that portion which reaches the earth's surface is the "Total Global Radiation." . Some portion of the radiation is then blocked. on cloudy days; this portion is usually 573 expressed as a "percent cloudiness" or "global cloudiness index." Hence, one minus the index is the percent of the energy that reaches the earth. . The other fact regarding solar radiation that is important to designers is the angle at which solar radiation arrives. This angle changes both with the time of day and the time of year. The effects of the sun's angle are two: first, the angle at which radiation strikes glass influences the amount of radiation which penetrates the glass--the sharper the angle, the less penetration; second, the angle- influences shading. An overhang can shade the summer sun from penetrating, and probably overheating, a room with southern exposure. As a rule of thumb, for Alaska a _ horizontal projection of 50% to 70% of the window height will give nearly complete summer shading at noon. Since the sun is positioned lower in the sky in winter than in summer, the overhang will not interfere with . the solar radiation reaching the window in winter when the energy is needed. Heat Transfer There are three fundamental processes of heat transfer: convection, conduction, and radiation. In a typical passive solar heating system, all three play a role. Shortwave solar radiation enters the living space through a glass window. The radiation strikes the rear wall or collection surface and heats it. This heat spreads through the wall by a conduction process, warming the wall which then reradiates the ‘heat ‘into the space as longwave _ radiation. Glass, which is transparent to shortwave solar radiation, retains the longwave solar heat in the living space. (Trapping solar radiation in this way is known as the greenhouse effect.) Energy leaves the space by convection from infiltration of outside air through cracks and from ventilation. It also leaves by conduction through the insulation to the outside, where wind convects it away. . The objective of passive solar design is to maximize the heat gains. In the next two sections, the materials and the designs for maximizing solar gain will be discussed. Materials Properties Two materials are important in passive solar heating systems: e glass which selectively allows radiation energy to pass through; and e heat storage materials which reduce temperature fluctuations within the space. 574 - | i J ee Glass, and many other transparent materials, are not, in fact, transparent to all wavelengths of radiation. As mentioned before, high energy (short wavelength) radiation penetrates through glazing materials but low energy (long wavelength) radiation does not. (Low energy radiation is also not visible but can be felt as heat.) This is the principle on which south-facing glass windows work to trap solar energy. Glass is not the only material with this property. Table 24.1 contains a list of other. materials which can and have been used for glazings on passive solar heating systems. For Alaska, plastic materials may be preferred in transport, as they are less fragile, lighter, and generally as efficient. Passive solar buildings also make use of materials that store heat to control temperature fluctuations. As an example, if a container of water is heated from 70°F room to 80°F, it has stored heat. If that 80°F water were returned to a 70°F room, it would cool back down to 70°F and give up the heat to the room. All materials store heat, but certain materials have good thermal mass and are very appropriate for retaining heat in passive solar systems. Brick, adobe, concrete, and stone, for example, are materials commonly used in passive systems because of their excellent thermal properties and reasonable cost. An example of how thermal mass is used to control temperature fluctuations is illustrated in Figure 24.1 with water as the thermal mass. Space 1 heats and cools rapidly depending on whether the sun is shining. As a result, the temperature swings from uncomfortably | warm and uncomfortably cool. Space 2, with a container of water in © it, heats up less quickly during the day because the water absorbs the excess energy. During the night it cools off less quickly because the water releases heat to the air. Therefore, the thermal mass of the water has a stabilizing effect on the room temperature and, hence, its comfort level for habitation. As mentioned, many materials can be used as thermal mass for heat storage in. this way; concrete and water are most commonly used. Concrete can store about 2 Btu in a pound of material heated ten degrees; water can store 10 Btu under the same conditions. Therefore, water is more effective as a storage material, but cannot be used as a structural material. Concrete can be used for the walls, ceilings and floors and to store heat. oo Another class of materials that can be used for heat storage are “phase change" or "heat of fusion" materials. Typically, these eutectic salts come packed in cans, tubes, or boxes that are permanently sealed. They melt or freeze at an appropriate temperature for a specific application, typically 70-100°F. Their benefit is that-they can store more heat: (heat of fusion) per pound than water or concrete. For example, a box of phase change material 2 feet by 2. feet by.6 feet can store enough heat to supply a house 575 TABLE 24.1 TYPICAL GLAZING MATERIALS FOR PASSIVE SOLAR SYSTEMS . Thermal Thickness Expansion (inches) Transmittance (°F-! x 10)-5 Strength Remarks Water White Glass 0.125 0.90 0.47 Good (tempered) Very durable - no degradation "Solatex" (ASG) : Float Glass 0.125 0.84 0.47 Window Glass 0.090 0.91 0.47 Poor (non=empered) Fragile (ASG SS Lustroglass) Sunite Premium IL 0.040 0.88 2.00 Very good Maximum temperature - 300°F (Kalwall) Filon w/Tedlar --- 0.86 2.30 Very good Maximum temperature - 300°F (Vistrom Corp.) Flexiguard 7410 7 mil 0.89 _— Good Maximum temperature - 275°F (3M) Tedlar 4 mil 0.95 2.80 Good 4-5 year lifetime at 150°F (Dupont )* Some embrittlement Teflon FEP 100A 1 mil 0.96 5.85 Fair, not for. Maximum temperature - 300°F a (Dupont ) exterior glazing n Swedcast 300 Acrylic 0.125 0.93 4 Very good Maximum service temperature (Swedlow Inc.) - 200°F Lucite Acrylic 0.125 0.92 4 Very good Maximum temperature ~ 200°F (Dupont) . Tuffak-Twinwall --- Equiv. to 0.89 3.3 High impact strength 5% reduction in transmittance (Rhom & Haas) for 1 layer fatigue cracking over 5 years Acrylite SDP --- Equiv. to 0.93 4 Good Maximum temperature - 230°F (Cyro) for 1 layer Sunlite Insulated Panels -— Equiv. to 0.88 0.47 Good Maximum temperature - 300°F (Kalwal1) for 1 layer Solar Glass Panels --- Equiv. to 0.90 0.47 Good Very durable (ASG) for 1 layer Regular Polyethylene 1,5,2,4,8,10 90% - Must be stretched tight Has a tendency to split on the (untreated plain) mil Diffused or inflated for strength fold. In shaded areas this mate- rial will last 2 years or more. Vinyl Film 4.8 mil 90-95% . Clear - Diffused Source: Division of Energy & Power Development, Alaska Energy Extension Service. Air Temperature “— ma ) 20° 7 eZ 80° 70° 60° 50° Sunrise Sunset Sunrise Space 1 Space 1 3 fl) Air Temperature — ( ) * 0 s : =—4 80 Water Temperature Water Temperature |" . 70° 60° 50° Sunrise Sunset Sunrise Space 2 Space 2 FIGURE 24.1 CONTROL OF TEMPERATURE FLUCTUATIONS USING WATER AS A THERMAL MASS Living Space Direct Gain Living Space \ Collector Wall Attached Sun Space FIGURE 24.2 CONCEPTUAL PASSIVE SOLAR HEATING DESIGNS 578 for up to 5 hours. However, they are more expensive than other thermal mass materials per unit of storage capacity and tend to find less application. Conceptual Passive Solar Heating System Designs Probably the most basic choice a designer faces is between a direct gain and a separated collection wall or sun space design. Figure 24.2 illustrates the differences. In the direct gain system, solar radiation enters the living space; in a collection wall system, the sun warms a volume located between the inner wall and the outer glazing; in a sun-space system, the solar radiation first warms a separate sun- space and then the heat is transferred, as needed, from the sun-space to the living space. Direct gain systems are usually just south-facing windows, although alternative insulation and storage schemes cause variation in designs. Collector wall designs have many variations -- the wall can be used as a heat storage device or, as an insulator between the collection volume and the living space. Heat can be transferred between the collector space and the living space by conduction, or by convection, either by using natural ventilation or by using fans. Attached sun-space systems are quite similar to collector wall systems, except that the collector volume is large enough to be useful for other purposes -- a greenhouse, or a porch. Naturally, an individual home can mix these basic design types. Another basic choice, applicable to all three system types, is the use of movable insulation to block the loss of energy through the glazing at night. Movable insulation can be quilt-like curtains that draw down, shutters that are rigid and latch closed, batts of urethane foam which can be inserted, or even a patented system that blows in polystyrene beads. In Alaska, insulating glazing areas is a recommended practice because throughout much of the winter, more heat is lost overnight through a window than can be gained from the sun during the day. Storage of heat is the next major choice a designer faces. In Alaska, winter heat needs are relatively large and continuous so that storage is often not too helpful -- the house uses all the heat that any glazing exposed to the sun can collect at the moment that it is collected. This is not, however, a firm rule and not a practice that can be quantifiably analyzed without the use of sophisticated design tools or a computer. For example, heat storage may be more effective in late fall or late winter. Figure 24.3 shows various options for storing heat. Again, any of these options can be mixed. In Figure 24.3a heat is stored in the walls and/or floors of the structure. In this scheme, insulation should be placed on the outside of the masonry wall so that the temperature within the structure will be tempered by the walls storing and discharging heat. 579 ~ Ee ABDAL DMO Tes ODDEN _ a. Store Heat in Masonry Walls or Floors d. Store Heat in a Bed of Rocks “} . or Phase Change Materials i b. Store Heat in a Masonry “Frombe” Wall a ‘e. Thermosiphon System Le c. Store Heat in Water Tanks, Used Here as a Collector Wall : . FIGURE 24.3 HEAT STORAGE FOR PASSIVE SOLAR SYSTEMS “4 580 Figure 24.3b shows heat being stored in a vented masonry wall which separates the sun-heated space from the living space. These vents allow for natural convection of heat from the solar heated space to the living space. Also, heat travels by conduction through the wall to the living space. These two heat transfer mechanisms provide for a nearly constant temperature in the living space. The vents are usually manually controllable, allowing the occupants to bring in heat from the solar heated space during the day if the living space becomes uncomfortably cool. At night, the vents are closed and heat travels through the wall only by conduction. Conduction through the wall is a gradual process, however, and the heat may continue to move into the living space hours after sunset. This tends to match the need for heat in the evening with the availability of heat from the wall. In Figure 24.3c, the masonry wall has been replaced by a "water wall". The effect is the same--the heat storage capacity of the wall tempers the room temperature. Water can store much more heat, however, than concrete for a given volume and mass. Containing it can be, however, a problem. Solutions range from fiberglass tubes, 18 inches in diameter and 6 feet tall to 5 gallon oil drums. Even used wine bottles have been employed. In Figure 24.3d, storage of heat is accomplished in a bed of rocks or of containers of phase change materials. A small fan is shown, rendering this a hybrid passive/active system, but this is not essential. Figure 24.3e shows a thermosiphon system. In it, hot air from the collector rises, leaves its heat in a bed of rocks, phase change material, or water containers. The cooled air falls back to the base of the collector to repeat the cycle. In Alaska, where houses are often built above ground to protect the permafrost, clever design could perhaps produce a workable thermosiphon system but would require very careful sizing of ducts and storage in order to work properly. All of these concepts have been successfully employed and are also not exhaustive--other successful designs have been built. Detail on how to size the glazing, storage, and insulation to produce a comfortable building can be found in the references at the end of the chapter. 24.2.2 Technical Characteristics A key measure of passive solar performance is the quantity of total heat load which can be obtained from solar energy. This measure is called solar savings fraction (SSF). Each passive solar-heated house is unique. Homes have been built from 5% to 100% SSF, that have varied from the simplicity of south-facing windows to complex thermosiphon systems. The actual performance of a passive solar house, however, is not only a function of the specific design, but also of local climate. 581 TABLE 24.2 SOLAR SAVINGS FRACTION IN DIFFERENT LOCATIONS House House House House House House House cf A B c D E F G ij Load Collector Ratio, Btu/ft--degree-day Location 121 52 30 20 13 9 6 _ Duluth 0.10 0.20 0.30 0.40 0.50 0.60 0.70 . Milwaukee 0.12 0.23 0.35 0.46 0.58 0.70 0.87 St. Louis 0.17 0.32 0.48 0.62 0.74 0.82 0.90 Dallas 0.23 0.54 0.72 0.846 0.90 0.94 0.96 SOURCE: Arthur D. Little, Inc., based on Reference 1. TABLE 24.3 RELATIVE ENERGY SAVINGS : Location _ A B Cc D E F G - Duluth 1.85 3.70 5.55 7.40 9.26 11.11 12.96 7 Milwaukee 1.70 3.25 4.94 6.50 8.15 9.85 12.25 Hy St. Louis 1.53 2.88 4.32 5.55 6.67 7.39 8.11 ~ Dallas 1.00 2.35 3.13 3.65 3.91 4.09 4.17 i SOURCE: Arthur D. Little, Inc., based on References 1 and 2. ~ my i] my . my ny 582 Se To illustrate this, Table 24.2 shows how seven identical houses would perform in four different locations in the Lower 48.* Each of the seven houses is assumed to be optimally oriented, otherwise identical, and characterized by a load collector ratio (LCR). LCR is a parameter which indicates the magnitude of the heat load (Btu/degree-day) relative to the solar collector area, e.g. window area in square feet. A high value of LCR would indicate a-low expected solar savings fraction. : The seven houses selected in Table 24.2 differ in.their LCR's, the values having been selected to produce SSF's of 0.1-0.7 in equal increments in Duluth. The results indicate superior performance in the more southerly locations. However, this can be very misleading since the actual heat loads differ among the locations. For example, the heat load in Dallas is only about one-fourth that in Duluth; thus an SSF of 0.1 in Duluth results in a greater Btu saving than in Dallas by nearly a factor of two. Table 24.3 shows the relative energy savings of each house in each location, normalized to House A in Dallas. From this we can conclude that a given passive solar design, operated optimally (e.g., R-9 night insulation over. the glazing), offers greater benefit in more northerly climates. Of course, the actual design of a. solar house would vary somewhat among these climates, but the results are nevertheless indicative of a trend. This trend appears to have some strong implications for the applicability of passive solar in Alaska. While the economically achievable SSF may be low by Lower 48 standards, the very high annual heating loads provide a strong driving force. | . A preliminary indicator of daily SSF is the ratio of incident solar energy to the building heat load. The National Weather Service reports, such monthly average data as total global radiation (Btu/ft"-day) and heating degree-days, respectively. High values of this ratio indicate a potential for overheating and thus all the collectible heat would not be useful. Low values mean that the incident solar is hardly worth capturing since heat losses through the collector are offsetting. Table 24.4 lists the ratio of these parameters (which have been defined as "passive solar index") by month for several U.S. cities including five in Alaska. The climates vary from virtually no heat * To develop data of this type is a tedious calculation process which usually requires computer simulation to be reasonably accurate. Because of a common perception -- perhaps a misconception -- that passive solar is relatively unsuitable in Alaska, comparable results for Alaska. locations are not available in the literature. Nevertheless, the discussion leads to some potentially useful conclusions regarding the applicability of passive solar in Alaska. 583 TABLE 24.4 PASSIVE SOLAR INDEX* Annual Location Degree~Days Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Miami, FL 205 20.33 19,61 ~ » ° © © » - ° * 18.20 Dallas, TX 2,290 1.35 2.46 4.54 23.20 ° o « o o 23.60 3.31 1.50 St. Louis, MO 4,748 0.60 1.06 1.51 5.75 18.20 © bad ° o 4.93 1.20 0.56 ——___ Milwaukee, WI 7,443 0.34 0.62 1.05 2.37 5.10 22.00 ° 47.75 9.42 2.07 0.61 0.30 Duluth, MN 9,756 0.22 0.45 0.80 1.73 3.41 9.16 27.67 15.04 3.45 1.19 0.35 0.19 oo -_—_——_—, f Juneau, AK 9,005 0.09 0.27 0.59 1.34 2.29 | 4.01 4.44 ' 2.98 1.35 0.45 0.15 0.05 Anchorage, AK 10,911 0.08 0.42 0.50 1.42 2.84 5,23 6.91 4.48 1.69 0.33 0.10 0.06 Fairbanks, AK 14,342 0.01 0.12 0.39 1.10 2.92 | 8.30 10.42 3.70 1.15 0.24 0.04 0.00 Kotzebue, AK 16,038 0.00 0.08 0.29 0.76 1.55 2.85 4.09 2.36 0.91 0.20 0.02 0.00 Barrow, AK 20,264 0.00 0.03 0.20 0.53 0.80 1.59 1.79 1.01 0.40 0.08 0.00 0.00 a oo _ * Total Global Radiation Btu (Monthly) Passive Solar Index = Heating Degree Days = ft ~degree-day y SOURCE: Arthur D. Little, Inc., based on National Weather Service Data. load (205 degree-days) in Miami, Florida, to the very high heat load (20,264 degree-days) in Barrow, Alaska. The enclosed area encompasses all values of the index between 0.20 and 3.0. This arbitrary range has been selected as means of extropolating apparently favorable climatic conditions from the Lower 48 to Alaska. From this, it appears that conditions for passive solar heating in Alaska may be favorable for as many months of the year as in the Lower 48. The difference is that the optimum months would vary among cities and that the expected annual solar savings fraction would diminish with more northerly latitude. Nevertheless, the total energy saving potential would probably also rise with more northerly latitude. 24.2.3 Environmental Issues There is no evidence that passive solar heating is anything but environmentally benign. In fact, total air pollution is probably reduced, from the avoided fossil fuels that would have had to be burned to supply the same amount of heat that can now be met by passive solar systems. 24.2.4 Commercial Status Many hundreds of homes have been built that aim to provide 20-80% of their energy requirements from the sun, and certainly many thousands have been built that avoid north facing windows to minimize heat loss and have south facing windows to maximize heat gain. Reports on many such buildings can be found in books and pamphlets including some in the references at the end of this chapter. In Alaska, passive homes have been built in Anchorage, Homer, Juneau, Delta Junction, Copper Center, Ambler, and _ Fairbanks. Two commercial applications exist -- a school and a health center in Anchorage. 24.3 Economic Implications 24.3.1 Costs The cost of passive solar heating is very site-specific, being dependent on local climate (solar radiation and temperature), building design and orientation, and the cost of construction. To assess the solar gain for any particular concept involves simulation of climatic conditions (hourly or daily) using calculation aids and/or a computer. The result is limited to the particular building analyzed and is difficult to extrapolate to another building, even in the same geographic area. As a result, there are very few examples in the literature which quantify the economics of passive solar systems. This is particularly true of examples in Alaska where the optimum solar savings fraction is less than most of the Lower 48, and 585 TABLE 24.5 CAPITAL COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD" CAPITAL COST Equipment and Materials Direct Labor TOTAL CAPITAL INVESTMENT Passive Solar Heating ‘Anchorage 1982 constant dollars 8.3 million Btu/year solar radiation 8.3 million Btu/year useful space heat 20 years 3-6 months $2,875 4,625 $7,500 586 therefore, passive solar has not been extensively studied. As demonstrated in Section 24.2.2, however, although the solar savings fraction in Alaska is relatively low, the heat load is high, causing the total energy savings and, hence, payout to be comparable to or better than other climates. The example selected here is one based on a passive solar home with a heat loss characteristic very similar to that of the homes analyzed in Chapters 20-23. Taken from a published analysis (3) of the performance of a passive solar home in three U.S. locations, the building has a 1500 square foot plan area. With about 330 square feet of double glazing on the south wall, R-4 night insulation, and 700 square feet of 6-inch thick concrete for thermal mass, the net incremental cost of this new installation was calculated to be about $5000 in the Lower 48. In Anchorage the cost would be about $7500 (Table 24.5). Using computer programs to simulate the hourly performance of the passive solar system, it was estimated that this home in Boston, Massachusetts, would gain about 5.7 million Btu per year of usable solar energy. Based on the trends developed in Table 24.3 above, we estimate very approximately that this same home in Anchorage may gain about 8.3 million Btu per year. Table 24.6 shows that this heat gain would be very expensive ($104/million Btu) compared to providing the same heat conventionally, by burning natural gas at $1.70/million Btu or even oil at $9.35/million Btu with a 60% seasonal efficiency. Of course, the effective cost of passive solar heat may be reduced by a do-it-yourselfer who does not factor the cost of his own labor into the project, but nevertheless, this project would be rather costly per unit of energy saved. This result does not necessarily imply that passive solar is especially uneconomical in Alaska. Rather, it appears somewhat costly everywhere. The basic example cited (3) shows a heat gain of 5.7 million Btu per year resulting from a $5000 investment in Boston. This, too, would be uneconomical at prevailing energy prices. 24.3.2 Socioeconomic Factors Producing passive solar heated residences involves the use of labor and materials. In Alaska, these costs would be split approximately 60% for additional labor costs and 40% for additional materials costs. The materials required are mostly glazing, insulation, and concrete. It is anticipated, therefore, that the widespread adoption of passive solar heating would be beneficial to employment in construction and, possibly, in the light industrial business of window panel assembly. No particular new labor pool skills would be required in either business, as passive solar heating relies on familiar residential construction techniques. 587 TABLE 24.6 OPERATING COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output OPERATING FACTOR Operating Costs FIXED COSTS Taxes and Insurance, 2% of Capital Investment Capital Charges, 9.5% of Capital Investment TOTAL FIXED COSTS TOTAL ANNUAL OPERATING COSTS TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST Passive Solar Heating Anchorage 1982 constant dollars 8.3 million Btu/year solar radiation 8.3 million Btu/year useful space heat Not applicable Annual Unit Cost Consumption Annual Cost $150.00 712.50 862.50 862.50 8.3.million Btu $103.92/million Btu 588 24.4 Impact 24.4.1 Effect on Overall Energy Supply and Use Passive solar heating has some apparently paradoxical effects on energy use. The reduced total energy demand of a passive home encourages the use of electric heat for back-up heating. This is justified by the fact that the cost of electric heating equipment is less than the cost of fuel heating equipment initially, although the. operating cost is higher. The lower utilization of heating equipment in a passive home makes this operating cost less important and - encourages the use of the lower capital cost system. As a result, electric utilities can experience higher peak loads from passive homes. Electric back up for passive solar heating is a low base-load but high peak-load device. A windy, cold, cloudy winter’ day will be a day on which both passive homes and conventional homes call for power. For homes that do use oil-fired back-up heat, passive solar heating will, of course, reduce demand. As there is a fixed cost to delivering oil, reduced demand at individual sites may reduce the economic efficiency of. the distribution system slightly, but the effect is probably insignificant. Finally, the economics of homes which seek to use the sun's energy for more than a few percent of their heating needs are not so favorable that passive solar heating is likely to become common in the next decade. Therefore, its impacts are likely to be slight. 24.4.2 Future Trends Solar radiation intensity and duration during winter in Alaska is. not conducive to using passive solar energy for heat. In general,’ the trend, even in states where solar is relatively better, is not towards solar space heating. Solar is often cost effective and common for water heating, but in many locations solar is still not a popular means’ for space heating in a living unit. Therefore, it is unlikely. that passive solar will become significant in the next decade in Alaska. At the same time, heating homes in. many parts of. Alaska is expensive and passive solar does create opportunities for innovative heating designs with the use of skylights, trombe walls, greenhouses, and solar porches. In general, consumers will choose between alternative products to meet needs on the basis of cost and utility. To meet their need for heat, they will perhaps consider passive design. However, so long as insulation and other competitive products remain more effective than passive solar heating, passive solar heating will lag behind these other options. 589 10. 11. 1977 Fundamentals, 197 . REFERENCES Balcomb, J. D., et al. "Passive Solar Design Handbook - Volume Two of Two Volumes: Passive Solar Design Analysis", Los Alamos Scientific Laboratory, University of California for U.S. Department of Energy, January 1980. Knapp, C. L., et al. "“Insolation Data Manual - Long-term Monthly Averages of Solar Radiation, Temperature, Degree-days and Global K,, for 248 National Weather Service Stations", Solar Energy Information Data Bank, Solar Energy Research Institute for U.S. Department of Energy, October 1980. Kohler, J., and D. Lewis, "Passive Principles: Glass and Mass", Solar Age, February 1982, pp. 32-3 . Anderson, B. with M. Riordan, The Solar Home Book, Cheshire Books, 1976. - Mazria E., The Passive Solar Energy Book, Rodale Press. Daniels, Farrington, Direct Use of the Sun's Energy,: Yale University Press, 1964. The AIA Research Corporation, "Regional Guidelines for Building Passive. Energy Conserving Homes", for U.S. Department of Housing and Urban Development Office of Policy Development and Research in Cooperation with the U.S. Department of Energy. International Energy Agency (IEA), "Air Infiltration - in Buildings", Draft Program Plan for U.S. Department of Energy, October 1979. : American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE), ASHRAE Handbook & Product Directory, Solar Energy Research Institute, "Passive Design: It's a Natural", for U.S. Department of Energy,. August 1980. Seifert, R., A Solar Desi Manual for Alaska, Institute of Water Resources, University of Alaska, Fairbanks, 1980. 590 25.0 SOLID FUEL FIRED STOVES 25.1 Introduction and Summary 25.1.1 Technical Overview Solid fuel stoves are the oldest and simplest devices for residential heating. Fuel, usually wood or coal, is placed in a firebox and ignited. The fire intensity is controlled by limiting the air flow into the firebox. The 1973 oil embargo sparked new interest in this old method of heating. The new concern with efficiency led to the development of a number of air tight stove designs with baffle and draft control to maximize heat delivery to a residence. Steady state efficiencies of up to 81% have been measured. However, the air tight design also promotes the build-up of creosote in the chimney, with the concommitant potential of chimney fires. For the last few years, a great deal of information has been generated and disseminated to the public to reduce the fire hazards associated with solid fuel stoves. There is also concern about the emission of air pollutants (smoke, ash, soot, carbon monoxide and nitrogen oxides) from solid fuel stoves. New devices (such as catalytic burners) are now being developed to obtain cleaner burning. They have been installed in some models on the market, but their acceptance has not been established. 25.1.2 Alaskan Perspective In Alaska, wood is widely used as a fuel for residential space heating. Coal is currently used in a very limited way. We have focused our discussions on wood, though the benefits and issues related to wood and coal burning are similar. A survey conducted by Battelle in the spring of 1981 found that 23% of the homes in Anchorage used wood as a secondary heating fuel. Other areas of the Railbelt reported surprisingly high proportions of homes that use wood as a primary heating fuel (13% of the homes in the Valdez-Glenallen area, and 15% of the homes in the Matanuska Valley area).(1) The economics of wood, oil and gas for the Municipality of Anchorage are roughly as follows: 591 Mixed No. 2 Natural Alaskan Hardwoods Fuel Oil Gas Approx. Heating Value, 13.8 million/Cord 140,000/gal 1,000 ft? Btu Estimated Unit Price,$ 80 - 130/Cord 1.31/gal 0.17/ft3 Appliance Efficiency, % 50 65 . 65 Price, -$/million Btu 11.6 - 18.9 14.4 2.6 Delivered In Anchorage, for the same Btu delivered, purchased wood may be competitive with oil but is generally not competitive with gas. In rural areas, or areas where gas is unavailable, wood gains economic attractiveness. Wood can be obtained cheaply by those who cut and split their own wood. Additionally, wood stoves are simple to use and maintain, and thus are well suited to rural areas where skilled repair people or materials may not be readily available. Resource availability is generally the restraining factor in the expanded use of wood for space heating. A family in Anchorage would need to access approximately 18 acres of timberland per season. Two acres of timberland are required to support a sustained annual harvest of 1 cord and 9 cords of wood (with an equivalent heating value of approximately 1000 gallons of oil) are required to fulfill the average Anchorage household's thermal demands. In regions where the land is more sparsely forested, such as the Interior, estimates suggest that it would take about 6 acres to support a one-cord sustained annual forest.(2) Maintaining adequate bleactad supplies would be difficult in such areas. Other concerns which may be of special interest in Alaska are the possible effects low temperatures may have on creosote formation and pollutant levels. Creosote is formed when flue gases drop below approximately 210°F. Maintaining sufficiently high flue temperatures may require particular effort or design characteristics in Alaska. Pollution from wood stoves, of some concern in Anchorage primarily in the form of carbon monoxide and soot or particulates, may be of less concern in the rest of Alaska due to relatively low densities of population and heavy industry densities. 592 25.1.3 Significance of the Technology Residential heating with wood has been used for a long time in Alaska. Numerous stoves are available on the market covering a wide range of features and prices. A critical comparison of heating with wood vs. other fuels is difficult to undertake for several reasons. First, fuel prices vary significantly from area to area. Second, wood is available in many areas where other fuels are not available. Third, wood can be obtained free by collecting drift wood, forest wastes or harvesting wood on personal property. Unlike conventional fuels, wood usage also involves intangible factors such as_ the aesthetic satisfaction of a glowing fire, owning a private wood lot, chopping wood, and being independent of conventional energy sources. On the other hand, wood stoves require atterition unlike automatically operating oil and gas heating equipment. Consequently, wood will be used as a secondary fuel in many areas, and less fre- quently as a primary fuel. 25.2 Description of Technology 25.2.1 Operating Principles Stoves used in the home to burn solid fuels are varied, differing greatly in design, efficiency and price. In all devices, the operating principles are essentially the same regardless of the fuel, be it coal, wood or peat. The fundamentals of solid fuel combustion are described below. The initial stage of the combustion process involves the heating of the solid fuel and the vaporization of its moisture. Fresh fuel placed on - glowing coals loses moisture before it begins to provide useful heat. Thus, the more water in the fuel, the greater the heat loss carried by the moisture. Additionally, steam in the firebox inhibits the complete burning of other gases liberated from the fuel. Wood curing or aging significantly reduces the moisture content of wood and increases the thermal output of the space heating unit. Wood that is cured contains approximately 20% water. Fresh cut wood is usually very high in water content (45-80%). Whereas a dry cord of wood delivers approximately 20 million Btu, green wood delivers only 16.5 million Btu. Once the heat of the fire has eliminated the moisture, the fuel starts to break down chemically into volatiles (pyrolysis) and eventually charcoal. Pyrolysis is usually a net endothermic reaction. Heat is produced by the reaction of the volatiles and char with oxygen. Volatile matter burns with a gaseous flame; char glows. When combustion is incomplete, the volatiles are carried up the chimney or flue as smoke. The products of incomplete combustion include carbon monoxide, partially burned hydrocarbons and soot. If 593 the temperature of the exhaust gases drop below the dew: point of water (212°F), partially burned oxygenated hydrocarbons will be deposited on the inside. surface of the chimney as pyroligneous acid. (Water is believed to act as a catalyst in this deposition.) The acid normally dries on the chimney wall and forms creosote, a sticky black substance that is highly flammable at temperatures around 1000°F. In complete combustion, carbon is completely oxidized to carbon dioxide and hydrogen to water vapor. Nitrogen and excess oxygen go up the chimney, taking energy in the form of sensible heat with them. This heat loss is called the dry flue gas loss. Other heat losses include the hydrogen loss and the moisture loss. The hydrogen loss is the heat loss incurred by the latent heat of water vapor (formed in the combustion of hydrogen) and the heating of that vapor to flue gas temperature. The moisture loss is the heat loss incurred by heating the moisture in the fuel to flue gas temperature. The room is heated primarily by two heat transfer processes: radiation and convection. Radiative heat transfer takes place mainly from the glowing char and luminous flames to the stove, and from the stove to the surrounding bodies in the room. Convection is the heat transfer process occurring via the room air. Room air near the stove heats up, rises due to buoyancy and is replaced by colder room air. 25.2.2 Technical Characteristics Fuels, device type, performance, materials, installation, and creosote and fire hazards are described below. The terms combustion efficiency and appliance efficiency are first briefly discussed, since these concepts are'essential in the comparative evaluation of solid fuel burning devices and systems. : Efficiency Combustion efficiency (3) is the measure of the completeness of combustion. Efficient heating systems are . designed for the optimization of the combustion process. The raw fuel, as well as the gases or volatiles driven from the fuel during pyrolysis, are burned. In most oil and gas furnaces, combustion is effectively complete. Products of incomplete combustion such as carbon monoxide, hydrocarbons and soot are emitted in relatively small amounts. The combustion efficiency is nearly always greater than 99%. For wood-fired appliances, combustion is less complete and may be lower than 95%. Products of incomplete combustion (and thus potential heat) are lost to the atmosphere. Appliance efficiency is the measure of the ability of a device to extract heat from fuel and deliver it to the house. It is strongly dependent on the degree of completeness of combustion and the ability 594 of the device to transfer the heat from the combustion process to the stove's surroundings. The stove's mass and surface area are important characteristics in this heat transfer process. Appliance efficiency can be measured directly by using a calorimeter room or indirectly, by measuring all the heat losses (primarily comprised of losses up the stack). To assure a reasonable degree of combustion completeness, more air is normally introduced than _ that which is theoretically or stoichiometrically required. The greater the amount of excess air, however, the greater the dry flue gas loss since this excess air must be heated to flue gas temperature. Additionally, the residence time of hot gases in the stove is shortened, decreasing heat transfer efficiency. Turning down the damper results in a cool, slow burning fire with good heat transfer efficiency. However, because of the low temperature, some gases produced by pyrolysis leave the fire unburned and conditions conducive to creosote formation exist. Experimentation with individual stoves is the best way to determine optimum damper and air inlet positions since efficiency is dependent on stove design. Fuels The three major solid fuels are wood, coal and peat. Although these fuels differ in their chemical composition, physical properties and moisture content, their combustion behavior is essentially similar. The main differences between the combustion of coal and wood are: e The pyrolysis temperature: wood is pyrolyzed _ at approximately 100°F less than coal. Thus more heat must be retained within the fuel bed to keep a coal fire properly ignited. Coal has a lower moisture content than most hardwoods so less heat is wasted during the drying process and less heavy oils, tars, or creosote are formed in the coal burning process. e Flame characteristics: a properly maintained anthracite coal fire has a deep fuel bed of glowing coals with the volatile gases burning just above the fuel bed with a short, blue flame. Approximately 20% of the heat from the coal is released as gases and 80% is released or produced in the glowing coals. In a wood fire, 50% of its heat is in the volatiles driven off and burned as long yellow flames with the remaining 50% being heat produced by the char or fixed carbon of the wood. The practical result of these differences is that the firebox and grate in coal stoves must be made to withstand higher temperatures than wood stoves. 595 Peat is a vast resource in Alaska. Estimates of the resource range up to 700 quadrillion Btu, the equivalent of 120 billion barrels of oil.(4) Utilizing peat as a fuel as is done in Scandinavia and Ireland may be possible in certain areas of rural Alaska. However, since Alaskan peat is of a lower quality and the required extraction and distribution system have yet to be developed, its cost competitiveness remains in question. Further investigation of this potential fuel is necessary. Wood is the fuel of greatest current interest and is the most preva- lent residential solid fuel in Alaska. As a fuel, its most important characteristics is its heating value. Certain species of wood are denser, giving off more heat when burned, and are priced accordingly. Wood is generally characterized as either softwood or hardwood, although heating values vary greatly within these two groups. Given below are heating values for various species of Alaskan trees. (5) . Million Btu/cord (25% moisture) SOFTWOODS ‘Black Spruce 13.7 White Spruce : 13.7 Sitka Spruce 13.7 Western Hemlock 15.4 Alaska Cedar 15.1 Tamarack 18.2 Million Btu/cord (25% moisture) _ HARDWOODS : Balsam Poplar 11.6 Black Cottonwood . 12.0 Quaking Aspen * 13.0 Paper Burch . 18.9 ' Willow Spp 13.4 Red Alder / 14.0 As previously stated, wood curing or aging significantly increases the relative heat value of wood. More heat is delivered per cord and starting a fire is easier. Dry wood (seasoned 12 months) has a typical moisture content of 25%. Green wood with a high moisture content (e.g., 65-80%) has a heat value of about 89-93% of that of dry wood. Wood should be cured a minimum of 3 months, and 6 to 12 months if possible. The length of time required depends on how it is stored, temperature levels, relative humidity, and its exposure to sunlight or rainfall.: The low temperatures typical of the Alaskan climate will tend to lengthen the drying process. 596 Stove Types Solid fuel stoves can be divided into two major classifications, airtight and non-airtight, the latter including fireplaces. Airtights, where combustion air is closely controlled, can be further subdivided into several types depending on the flow of gases within the combustion ‘chamber of the stove. The open fireplace, either free standing (Figure 25.1) or built-in, is . the most common type of wood-fired appliance. It has a very low heating efficiency, typically 10%. Field tests show that a conventional built-in fireplace can actually increase overall fuel consumption because of the high draft generated. The conventional fireplace has no controlled combustion air delivery, the only effective combustion air control being a damper downstream of the appliance. Fireplace modifications which can increase efficiency include: (1) metal-lined heat exchangers (in the chimney or flue pipe) that allow room air to be heated and re-emitted out of the top of the unit, (2) tightly fitted glass doors and outside air vents which may double the performance of the fireplace, and (3) hollow tube grates for extraction of additional. heat. These may operate with a blower or simply with natural convection. Manufacturers claim heat gains of 5000 to 50,000 Btu/hr from these devices. Additionally, fireplace inserts, stoves which can be retrofitted to existing masonry fireplaces, can also dramatically increase efficiency. — Other non-airtight stoves, popular because of their simplicity and low initial cost relative to airtight stoves, include simple box stoves, Franklin stoves, pot belly stoves, and sheet metal stoves. Air leaks due to unsealed seams result in poor combustion control. The efficiency of these stoves is typically 20 to 30%. Like a fireplace, these stoves have a straight air path either across or through the fire allowing sufficient air for combustion of the wood (primary air) with no accommodation for additional air to support flame combustion or combustion of volatiles (secondary air). : : Airtight stoves generally made from cast iron range in efficiency from approximately 35 to’65%. The simplest unit is the airtight updraft (box) stove (Figure 25.1). It has controlled combustion air, but no effective internal baffling. As in the case of non-airtight devices, the flame and volatiles have direct access to the exit from the stove. More efficient stoves employ baffles or other design features to control air flow, preheat. intake air, and promote combustion of volatiles. Examples of these airtight types include the horizontal baffle design, the down draft design and the side draft design. (6) These are described below. The airtight horizontal baffle stove (Figure 25.1) has a horizontal baffle running from the rear of the stove toward the front to at least mid-length. . Combustion air is controlled at the front, where the 597 cr Combustion betay'a | Air et | Schematic of Free Standing Fireplace Combustion Air Schematic of Airtight Box Stove Corrs -_—— ccc Gas \ \\ Combustion Air Schematic of Horizontal Baffle Stove FIGURE 25.1 FIREPLACE AND STOVE ARRANGEMENTS 598 Je le wood tends to burn. Smoke and other products of incomplete combustion are released from the back of the stove, remote from the flame, and then are forced forward under the baffle toward the flame where they can be re-ignited and burned if brought into contact with the flame. The flue gases travel back over the baffle to the flue exit. : In the airtight down draft design (Figure 25.2), combustion air is controlled and the wood sits on a grate. When the fire is lit, a vertical damper is opened at the top, making the stove behave like an updraft, to promote rapid. combustion. When a hot bed of embers is developed, this: damper is closed; the flue gases are forced back down near or through the bed before going up and out the flue. The airtight side draft stove (Figure 25.2) has a vertical baffle with an opening (typically 1-1/2 - 2 inches high) at the bottom and an open-and-shut damper at the _ top. The latter is open when combustion is initiated, as with the down draft. When enough heat has been developed to sustain adequate draft, the damper is closed. The flame moves horizontally, parallel to the fuel bed, and out the baffle opening. Smoke and volatile matter released remote from the flame are pulled down under the vertical baffle. If the flame is adequate at that point, the incomplete products of combustion may be ignited and burned before passing up and out the flue pipe. The performance of different stoves within a specific type can vary widely and depend on how effectively the particular design utilizes the potential performance advantages of its type. Design characteristics that improve the overall performance of stoves generally provide one or more of the following functions: e intake air preheating (usually by employing a chamber); e intake air flow control; e@ .secondary air flow control; e turbulence or mixing of air and volatiles; or e improvement of heat transfer from the combustion chamber to the room (e.g. incorporation of a heat exchanger). Side draft stoves and horizontal baffle stoves are generally considered to offer the best overall performance. Seasonal efficiencies of up to 65% have been measured for these devices, a value comparable to most existing oil and gas central heating systems. Simple updraft airtight stoves range in efficiency from 35 to 50%. Table 25.1 lists typical performance data determined experimentally by Hayden and Braaten.(7) Clearly, the fireplace performs the most - poorly in efficiency. The test efficiency, determined by the indirect 599 Combustion op Schematic of Downdraft Stove Combustion Air Schematic of Sidedraft Stove FIGURE 25.2. STOVE ARRANGEMENTS 600 ie. ~ TABLE 25.1 TYPICAL PERFORMANCE DATA FOR FOUR STOVE TYPES, FUELED WITH SUGAR MAPLE * Airtight design SOURCE: Reference 6 601 Total Flue Eee’ Burn Rate a rir Gas Stove Type Efficiency kg/hr Air 3 % 8 % mn” /kg — Freestanding Fireplace 44 5.90 830 44.6 k Updraft (box) 78 0.91 298 19.4 Horizogtal Baffle 80 1.36 304 19.7 * Sidedraft 81 1.81 165 17.9 heat loss method, is higher than an average fireplace, but still the lowest of the four devices. The fireplace burn rate and excess air level far exceed values measured for the other devices. The test efficiencies determined for the updraft (box) stove, the horizontal baffle and the side draft stove are similar, ranging from 78 to 81%. However, again, the burn rates differ greatly. About twice as much fuel is burned per hour in the sidedraft stove. compared to the updraft. Materials Stoves can be made of sheet metal, steel plate, or cast iron. Sheet metal stoves made from thin steel sheets (typically less than 1/8 inch thick) are either welded or bolted together. They require a smaller — initial investment but burn out faster than cast iron and plate steel stoves. The material characteristics between plate steel (typically 1/4-1/2 inch) and cast iron are not significantly different. The thicker the metal, the longer the stove should last. Cast iron and plate steel retain roughly the same amount of heat per pound and both cost roughly the same. Cast iron stoves can be blackened or “enameled and can also be decorated with designs in the casting. In general, whether the material is steel plate or cast iron is secondary to the workmanship and design. Some stoves have fire bricks or metal plates to prevent burnout, increase the lifetime of the stove, and increase the thermal mass. (Typically, a 500 pound stove can continue to deliver usable heat four hours after the fire has been extinguished. ) Installation Stoves must be away from combustible materials to prevent fires from radiant heat. This requirement raises at least two issues: the need for careful installation and, when considering the purchase of a stove, the potential loss of living space. : Many stoves are tested by national laboratories (such as Underwriters Laboratory) that specify clearance distances. A stove which is not listed should be installed with ample clearance. Typically, a -stove should be placed at least 36 inches from a combustible wall. Noncombustible material placed one inch from the wall will reduce the distance requirement. Asbestos millboard will allow an 18 inch stove-to-wall clearance. Twenty-eight gauge sheet metal will allow a 12 inch clearance. Safe floor clearances are substantially less than those for walls because the heat radiated from the bottom of a stove is generally less. than from either the sides or the top. During a fire, ashes fall to 602 the bottom of a stove. This has an insulating effect, blocking the flow of heat downward. This process may be started in a new stove by placing a layer of sand in the bottom. Clearances for proper wall and floor protection are given in Table 25.2. Clearly, these clearance requirements add to the cost of stove installation and may modify the appearance of the living space. Creosote and Fire Hazards While a wood burning stove is relatively simple to operate, special care must be taken to avoid creosote buildup and fires. This is particularly important because the cold weather of Alaska enhances the formation of creosote. Creosote is acidic with a typical pH value of four. It is corrosive to iron, steel and even galvanized steel. Creosote also has a significant insulation effect which reduces the heat output to the room when formed on heat transfer surfaces. It is impossible to prevent creosote formation entirely, but its quantity can be greatly reduced if the following steps are taken: (9) e Maintaining a minimum stack temperature of 250°F will prevent the condensation of water. Water acts as a catalyst in the deposition of creosote. An inexpensive thermometer that will adhere to the stove pipe can help the homeowner monitor stack temperature. e Insulating all stacks inside and outside the house will help reduce creosote condensation by minimizing heat loss and keeping the flue gas the maximum temperature. e Using medium to well seasoned wood instead of wet wood gives better overall performance. Generally, the lower the moisture content of the wood, the higher the firebox and flue gas temperatures. Medium seasoned wood, however, has been shown to produce less creosote than well seasoned wood.(10) It has been suggested that the water vapor dilutes the other constituents of smoke, lowers the vapor pressure, and thus lowers the temperature at which creosote will form. (11) e When starting a fire in a cold stove, burning a hot fire the first 30 minutes will quickly warm up the stove and the ductwork. This will make a good coal base for maintaining a higher efficiency fire. It has also been suggested that the daily burning of a very hot fire for a short period of time will burn off some portion of the creosote. 603 TABLE 25.2 TYPICAL MINIMUM CLEARANCES AND PROTECTION MATERIALS MINIMUM CLEARANCES 18 12 18 12 12 9 12 9 LEG LENGTH fp Stove Stove TYPE OF PROTECTION Sides Pipe 36 18 Unprotected 36 18 i" asbestos millboard 1" asbestos millboard spaced out 1" 28 ga. sheet metal on % asbestos mill- board 28 ga. sheet metal spaced out 1" 28 ga. sheet metal on 1/8" asbestos millboard spaced out TYPE OF PROTECTION 24 gauge layer of sheet metal 24 gauge layer of sheet metal over % inch layer of asbestos millboard 4 inches of hollow masonry laid to provide air circulation through the masonry layer covered by a sheet of 24 gauge sheet metal Source: References 8. 604 e The inside of the flue should be checked periodically. If ereosote is caked more than 1/4 inch thick, the stack should be cleaned. Falling embers and sparks present an additional safety problem that is _ often ignored. . Noncombustile floor protection of masonry, ceramics, asbestos, or metal plate will provide adequate protection if extended to a reasonable length. 25.2.3 Environmental Issues The major environmental issues associated with solid fuel burning are air pollution and wood resource depletion. These issues are discussed below. Air Pollutants. - Complete combustion of wood to carbon dioxide and hydrogen to water vapor does not. produce any health-related air pollutants. Unfortunately,- even when operated at optimum conditions, wood stoves do not achieve complete combustion throughout the entire burning cycle. Generally, stoves emit 30 to 250 times more solid particles and up to 1,000 time more carbon monoxide than oil-fired furnaces (on a_ heat-equivalent basis). Table 25.3 provides a comparison of emissions from gas, oil, coal, and wood-fired spacing heating equipment. Though these emission levels do not take device efficiency into account, there is a clear indication that carbon monoxide levels and particulates are significantly higher from wood-burning stoves than from gas- and oil-fired residential sources. These compounds do have health-related effects as described below and hence are considered to be air pollutants. Smoke, ash, and creosote matter are complex substances primarily composed of polycyclic aromatic hydrocarbons. Inhalation of these particles is generally considered to be health hazardous. Low wind conditions and temperature inversions promote smoke accumulation that impair visibility and increase health risks. Valley locations are particularly susceptible to stagnant air conditions. Carbon monoxide (CO) is another product of incomplete combustion that poses health threats. CO, because it is odorless and colorless, may go undetected inside the home. Thus, proper drafts are essential for the safe operation of stoves. The potential health effects of carbon monoxide are: decrease in awareness and reaction time, a restriction in the blood's ability to carry oxygen, at high CO levels, death caused by suffocation. There are currently no federal or state regulations controlling emissions from solid fuel stoves. 605 909 TABLE 25.3 COMPARISON OF RESIDENTIAL SPACE HEATING EMISSIONS I Emissions, 1b/10° Btu (ng/J) Filterable co HC NO, (as NO) particulate Gas / 0.02 (8.6) 0.01 (4.3) 0.08 (34.4) 0.005-0.015 (2.15-6.45) Oil 0.04 (17.2) 0:01 (4.3) 0.16 (68.8) 0.014 (6.02) Coal 3.46 (1488) 0.77 (331) 0.12 (51.6) 0.77 (331) Wood (fireplace) 3.0 (1290) 2.6 (1118) 0.25 (108) 0.33 (142) (Stove) , 22 (9460) 0.28 (120) 0.07 (30.1) 0.5 (215) Note: Emissions are based on heat input and do not take equipment efficiency into account. SOURCE: Reference 12 Effective resource management including the potential to make forested lands more productive is essential to widespread wood use for home heating. While regional forestry agencies have recently opened more areas for personal-use cutting by the public, these are on lands designated for agriculture disposal. They will not be reforested and will be excluded from future forestry use. It appears evident that the state government, the federal government and industry must work together to effectively implement forest management if future demands are to be met. 25.2.4 Technology Status It is estimated that 7% of existing stoves have been installed since 1973 when the oil embargo caused increased interest in wood burning for residential heating. Regional market saturation and stabilized oil prices have contributed to a decrease in sales and dwindling diversity in product lines. It is unlikely that major new products will be introduced over the next few years. Instead, product changes will be incremental, including: e improvements to existing lines such as higher efficiency models, see-through doors, and more easily installed and serviced stoves; e@ some line extensions such as smaller stoves for use in fireplaces; and e clean burning devices fitted with catalytic reactors to burn smoke. Industry leaders also believe that marketing and distribution are the key factors for success in the future. New market areas need to be identified such as replacement markets for first stoves, second stove in same residence, and shifts from rural to suburban areas. Table 25.4 is a list of major solid fuel manufacturers in the United States. 25.3 Economic Implications 25.3.1 Costs To illustrate the economic implications of the use of wood stoves in Alaska, a case study of a 30,000 Btu/hr residential unit operating in Anchorage was undertaken. Table 25.5 summarizes estimates of the capital cost associated with this unit. These estimates were obtained from a number of sources including telephone conversations with suppliers. Although a range of values can be identified for each item, average values are given. Stove prices vary from $300 to $1,100. Installation costs depend on whether the installation is done 607 TABLE 25.4 LIST OF MAJOR SOLID FUEL STOVE MANUFACTURERS IN THE U.S.A. Atlanta Stove Works * Franco-Balje Fisher Stove International * Jotul, U.S.A. Inc. Locke Stove Co., Inc. Martin Industries, Inc. Quaker Stove Co., Inc. Shenandoah Manufacturing Co. United States Stove Co., Inc. Vermont Castings, Inc. * Local representatives of foreign manufacturers 608 Atlanta, GA New York, NY Portland, OR Toronto, Ontario Kansas City, MO Florence, AL Trumbaursville, PA Harrisonburg, VA Chattanooga, TN Randolph, VT TABLE 25.5 CAPITAL COST SUMMARY TECHNOLOGY: Solid fuel stoves BASIS: Location: Anchorage Year: 1982 constant dollars CAPACITY: Input: 60,000 Btu/hr Output: 30,000 Btu/hr ESTIMATED USEFUL LIFE: 20 years CONSTRUCTION PERIOD: Can purchase immediately * CAPITAL COST Stove: #500 Direct Labor for Stove Installation: 180 Pipes, Protection, Plates, Tools 200 TOTAL CAPITAL INVESTMENT : $880 * Y Loans are available up to 10,000 at 25% interest for purchase installation of alternating heating systems.2 609 TECHNOLOGY: BASIS: Location: Year: CAPACITY: Input : Output: Operating Costs Variable Costs Wood Fixed Costs Maintenance TABLE 25.6 OPERATING COST SUMMARY Solid Fuel Stoves Anchorage 1982 constant dollars 60,000 Btu/hr 30,000 Btu/hr Unit Cost Annual Consumption $5.00/million 152 million Btu Btu Capital Charges 9.5% of Capital Investment TOTAL FIXED COSTS TOTAL ANNUAL COSTS: TOTAL ANNUAL OUTPUT: TOTAL LIFECYCLE COST: 610 Annual Cost $760.00 100.00 83.60 183.60 943.60 76 million Btu $12.42/million Btu i by professionals or the homeowner, on the existence and condition of the chimney, and on the cost of auxiliary equipment. (Auxiliary equipment might include pipe, protective materials and installation tools.) - . ‘ | Estimates of the typical annual operating costs are summarized in . Table 25.6. Actual values depend on factors such as house size and (om construction, weather conditions, family size and life style, and most importantly, wood cost. The figure of $5/million Btu is equivalent to about $100/cord. Although this can be higher in some locations, wood can also be obtained under some circumstances at little or no i. cost. For example, an individual with access to a wood lot who does &. not charge himself for the labor of transporting, cutting, and 7 splitting wood may have no out-of-pocket costs. Similarly, in Bush : villages, individuals who gather driftwood or locally available coal or i. peat may have essentially free fuel. 25.3.2 Socioeconomic Factors The principal favorable socioeconomic impact which results from the use of solid fuel-fired stoves for Alaskan residential heating is the satisfaction and/or cost benefits received through its use. Though fossil fuel costs are relatively stable, and currently these fuels are readily available, the use of wood (or other solid fuels) as a second fuel in the home will provide some cushion to the consumer if fossil fuel prices suddenly rise in the future. The use of wood stoves for residential. heating also has benefits for local wood-related industries. It offers a value-added outlet for wood wastes from lumber operations and managed timberlands. 25.4 Impact 25.4.1 Effect on Overall Energy Supply and Use Solid fuel fired stoves currently have and will continue to have their most common application in residential heating. In Alaska, the greatest benefit of solid fuel stoves will continue to be in rural areas where oil and natural gas are not readily available and where wood, coal, or peat are in relatively plentiful supply locally. It is unlikely that wood, coal, or peat will replace fossil fuels on a broad basis because they are not cost competitive in many instances, and because current solid fuel fired stoves are not automatically fired and fed. On the other hand, large fossil fuel price increases and/or effective forest management will make solid fuels more attractive to the consumer. 611 25.4.2 Future Trends There are no major technical improvements in solid fuel stoves in the offing which are likely to alter their role as. a space heating source. Nor ‘are there regulatory actions pending which will have an impact. Solid fuel stoves re-entered the market in the 1970's on the basis of economics and perceptions of energy security. These are likely to remain the major driving forces. 612 £ ' 10. 11. 12. REFERENCES State of Alaska, "1982 Long Term Energy Plan," II-16. Ibid, II-16. Hayden, A. C. S., and Braaten, R. W., “Effect of Wood Stove Design on _ Performance," Canadian Combustion Research Laboratory, March 1980, p. 1. State of Alaska, "1982 Long Term Energy Plan," II-17. Alaska Division of Energy and Power Development," Your Wood Burning Stove," (WSUN/126 UC-61la). "Effect of Wood Stove Design on Performance," pp. 2-3. Hayden, A. C. S., and Braaten, "Efficient Wood Stove Design and Performance," January 1980, Wood Heating Seminar Proceedings 1980/1981, Wood Heating Alliance, February 21-24, 1981, New Orleans, Louisiana, pp. 35-36. New England Regional Commission, "Heating with Wood, Burning Wood Safely," p. 15. Alaska Department of Environmental Conservation and the Alaska Energy Extension Service, "Wood Stoves." Allen, J. M., "Control of Emissions from Residential Wood Burning by Combustion Modifications," Battelle Columbus Laboratory, in Wood Heating Seminar Proceedings, 1980/1981, Wood Heating Alliance, February 21-24, 1981, Row Orleans, Louisiana. Juger, S. J., Maxwell, T. T., Dyer, D. F., Maples, G., "Improving the Efficiency, Safety, and Utility of Wood Burning Units," Auburn Unviersity, Auburn, Alabama, U. S. Department of Energy, Division of Buildings and Community Systems, Contract No. EC-77-S-05-5552. Hall, R. E., "EPA Research Program for Controlling Residential Wood Combustion Emissions," Wood Heating Seminar Proceedings 1980/1981, U. S. Environmental Protection Agency, Industrial Environmental Research Laboratory, Combustion Research Branch, Research Triangle Park, North Carolina, p. 53. 613 26.0 SOLAR DOMESTIC HOT WATER 26.1 Introduction and Summary ~ 26.1.1 Technical Overview Solar water heaters have enjoyed commercial success in warm, sunny climates, such as Australia, Israel and Japan for over 30 years. An active market also existed in Florida in the 1930's. However, the start of the current solar hot water industry in the United States _ dates from 1973 and the oil embargo which sparked a number of government activities in support of renewable energy technologies. Solar domestic hot water has shown continued growth in United States in the last several years. From 1980 to 1981 gross sales of dedicated solar water heaters increased from 190 to 330 million dollars--the 1981 sales total represents 92,000 systems, of which 88,000 were installed in single family residences.(1) This chapter primarily focuses on the relatively small (and simple) residential systems with some comment on the salient characteristics of the larger, more complex, commercial systems. Solar water heating is a simple’ application of solar energy comparable with relatively low technology, components and system arrangement -- and sometimes employing some reasonably sophisticated equipment. The key element in a solar water heater is the solar collector which traps the incident energy and which must present a reasonably large, typically 40 to 80 ft”, and inexpensive area to the incident solar energy. The remainder of the system consists primarily of means of transferring the collected heat to storage and to the load. The major types of solar collector design are flat-plate, evacuated- tube, and-concentrating. The flat plate collector, consisting of one or more transparent covers over an absorbing surface is the simplest design and is generally the near term choice for solar water heaters. The evacuated tube collector represents a lower heat loss and currently more expensive design which is not normally cost effective for water heaters. Concentrators which are available in a variety of designs represent the most complex, highest temperature option--they are commonly used only in very large systems and primarily in areas of high direct solar radiation. System designs for coupling the collector to storage and load range from self-driven thermosyphon systems to a wide variety of pumped, forced circulation designs. The indirect forced circulation design using an antifreeze loop and heat exchanger is the design most commonly used to achieve freeze protection in northern areas of the Lower 48. - : , 615 The development of the solar industry, particularly in Northern areas, has not been smooth. Early applications were marred by a high degree of failures attributable largely to installer inexperience. Initially, certain key components such as controllers also had an inordinately high failure rate. However, with increasing experience, operational reliability is improving greatly. ‘ 26.1.2 Alaskan Perspective In Alaska, the reasonably low level of solar insolation combined with high labor and material costs, and (currently) low fuel costs in the more populated areas, makes the economics of solar water heaters relatively poor. Also, Alaska does not currently offer financial incentives comparable to many states in the Lower 48. However, there may be some specific opportunities where the combination of high fuel costs and reduced system costs may make the economics viable. Owner-installed systems -- and possibly owner-fabricated systems -- could effect substantial reductions in out-of-pocket system costs which could greatly improve economics. The use of seasonal systems designed without freeze protection and. deployed and operated only in the summer when performance is best also presents an opportunity for greatly reduced costs without a commensurate loss in annual solar heat collection. ‘ : If solar water heating were to gain a wide market in Alaska, it could serve as a source of new permanent employment in the manufacture and installation of systems. Since flat plate collectors might logically be manufactured within Alaska this would result in increased purchase © of in-state product. 26.1.3 Significance of the Technology The technical feasibility of solar hot water is well established. The technology is simple and although there has been a rash of mechanical problems in early installations, particularly in northern areas, this’ situation is rapidly improving. The economic example of Section 26.3.1 shows the economics for a residential water heater to be very poor in the Anchorage area. Although energy costs are higher in other locations, there are no obviously attractive targets where solar could displace a high cost energy source normally used for water heating. For example, the high electric cost in the Bush could make solar water heating economically feasible if displacing electric resistance water heating. However, electric water heating is rarely, if ever, used in this area. . . Future trends in solar water heating will include means of increasing performance per unit of installed cost. These might include low cost plastic collectors, improved integral storage and thermosyphon 616 systems, seasonal collection systems, and owner assembled/installed systems. In addition, high volume production of evacuated tube collectors could result in reduced cost equipment delivering more energy per dollar invested than flat plate collectors. 26.2 Description of Technology 26.2.1 Operating Principles The function of a solar water heater is to absorb solar radiation and convert it into useful heat which produces hot water. The main elements of a solar water heater are: e a solar collector which absorbs solar radiation when available; e a means of storage (i.e., insulated water tank) which accepts the collected heat when available and supplies it on demand when needed; e an auxiliary energy source which accommodates demand beyond that which the solar system can supply; and e The necessary piping, controls, valves, pumps, etc. required to integrate the above elements into an operating system. Figure 26.1 illustrates the components of a typical solar water heater. Solar collectors are configured to trap solar energy--both by maximizing the amount of incoming solar radiation absorbed and by minimizing the energy dissipated as heat loss. The efficiency of a solar collector is defined as the ratio of collected energy to incident solar energy. The collected heat is the difference between the solar radiation absorbed by the collector and the heat loss from the collec- tor to its surroundings. Therefore, the efficiency can be expressed as the difference between two terms: e an optical efficiency expressing the fraction of incident solar radiation which is transmitted to, and absorbed by, the collection surface; and e the heat loss per unit incident solar radiation. With inclusion of a factor to account for the difference between inlet fluid temperature (which is convenient to measure) and mean collector surface temperature, the efficiency of a solar collector can be described with the following equation: (2) UL at n= FR (at), - FR i 617 8T9 Domestic Water Absorber Inlet ' Plate >: > Outlet Absorber Surface Auxiliary Heater Collector Controls Insulation Storage Tank Frame Heat-Transfer Pump Fluid Heat Exchanger Insulated Piping FIGURE 26.1 SOLAR WATER HEATER COMPONENTS \ 4 ~ , 1 Where: n = collection efficiency (at), = effective transmittance absorptance factor (where absorbance of absorber plate and 1 is transmittan cover panes) FR = heat removal factor* U, = overall heat loss coefficient AT = temperature difference between inlet fluid and amb I = incident solar radiation The above equation is derived for a flat plate collector. Howeve is also valid for arrays of evacuated tube collectors and, with modifications, for concentrators as well. When the collection temperature is near ambient, heat loss will small, and collection efficiency will be limited by optical efficie (the transmittance-absorptance product). However, when collect temperatures exceed ambient temperatures by 20 to 30°F, the h loss from an exposed absorber plate would be significant, and collector must be designed to minimize the heat loss coefficient. Three major types of solar collectors are flat plate, evacuated tu and concentrating designs as illustrated in Figure 26.2. Flat Plate Collectors Flat plate collectors, contain an absorber plate which is covered | one or more glazings on the sunlit side and insulated on the ba side. The cover panes of glass or plastic are transparent to incide (short wave) solar radiation but impede heat loss by air convecti and infrared (long wave) radiation from the absorber surface. Bo single and double panes are used. The double pane design is mo effective in reducing heat loss--a particular advantage when ambie temperatures are low. However, in addition to having greater co and weight, a double pane collector also has reduced optic efficiency. Some types of flat plate collectors for very low *Accounts. for differences between inlet fluid temperature a! absorber surfaces average temperature due to fluid-absorber he transfer and absorber plate fin effects for an infinitely conducti absorber plate with infinite fluid heat transfer, FR = 1.0. In mo actual designs, FR is between 0.9 and 1.0. 619 TRANSPARENT COVERS FLAT PLATE WEATHER TIGHT FRAME — ABSORBER PLAT THERMAL INSULATION FLUID PASSAGES MANIFOLD REFLECTIVE BACKING EVACUATED TUBE MC COLLECTOR ARRAY FEEDER TUBE EVACUATED TUBE COVER TUBE SELECTIVE BLACK, ABSORBER HEAT TRANSFER { FLUID TUBE DETAIL (PER REFERENCE 3B) CURVED REFLECTOR. CONCENTRATOR (PARABOLIC, TROUSH Must follow sun / 620 FIGURE 26.2 SOLAR COLLECTOR OPTIONS 2 temperature needs (i.e., swimming pool heating) consist merely of unglazed, uninsulated, black metal or plastic absorber sheets through which liquid is heated... However, this design is not suitable for . achieving temperatures needed for domestic water heating. The absorber which contains the heat transfer passages is normally copper (for good heat transfer.and corrosion resistance) and coated with a flat black or selective black absorbing surface. A flat black surface (typically paint. or ceramic) is both a good absorber of solar radiationand emitter of heat. A selective black surface (typically a compositive surface like chromium. oxide electrodeposited on a bright nickel substrate) is a good absorber of shortwave solar radiation but a poor emitter of long wave infrared radiation. In the Lower 48 states, flat plate collectors are almost always the design of choice for. residential solar water heaters and many commercial installations. Single pane collectors tend to be most popu- lar--normally combined with flat black absorber surfaces in Southern regions and selective black in Northern regions. However, in Alaska, the lower ambient temperatures may favor the two pane collector. --or even the evacuated tube. Evacuated Tube Collectors An evacuated tube collector consists of a manifolded array of evacuated glass tubes, each containing an inner selective black absorber. The evacuated tubes. are backed by a reflective sheet which may be either diffuse or spectral and slightly concentrating. The combination of the selective black absorber with high vacuum is extremely effective in. reducing radiation and air conduction losses. The degree of concentration in the backing sheet and the spacing of the tubes provides some freedom in tailoring performance anc cost to application. Closely packed tubes have maximum optical efficiency (no reflector loss) but also somewhat higher heat: loss (more absorber area) and higher cost. More widely spaced tubes, with concentrating backing, can reduce costs--and in applications requiring extremely high temperature, the reduced heat loss may offset the lower optical efficiency. Evacuated tube collectors are currently more expensive than flat plate collectors. ‘ Since the material costs are roughly comparable for these two types, it has been proposed that the cost differential could be eliminated by high volume production. Also, the extremely high stagnation (or no flow) temperatures resulting from their highly effective insulation can complicate system design. At present, they find their greatest use in commercial systems and are not commonly used for residential water heaters in the Lower 48 since their performance gain is small relative to their increased cost and system complexity. However, at least one manufacturer (Sunmaster) (3) does 621 offer a residential system with evacuated tube collectors--and the colder, cloudy climate such as might occur in western coastal regions of Alaska would tend to favor this use. Solar Concentrators The. concentration of solar energy increases collection efficiency by reducing the absorber area relative to the aperture (collection) area. A variety of designs exist. Many involve focusing optics such that the sun's rays are brought to a single point or line focus. On the other hand, certain non-imaging designs act as light -funnels, concentrating the radiation in a smaller area but without the precision required for focusing and with a greater tolerance in terms of sun tracking. The methods of tracking vary widely, including designs where either the absorber, the complete reflector, or segments of the reflector move.’ Typical designs include: e Cylindrical collectors, which result in a line focus. High concentration parabolic cylinders aligned with axes in the north-south direction can achieve high temperature at the expense of continuous, although rotation only, tracking. Lower concentrating troughs aligned in. the east-west position--in approximately the direction of the apparent sun path--may require only periodic, seasonal adjustments. e The compound parabolic reflector (4) is an example of a non-imaging concentrator which for a given concentration ratio may be more tolerant of off-axis radiation than focusing designs. e The Fresnel lens concentrator, which accomplishes concentration with a Fresnel lens instead of a reflector mirror. (5) . e The Solar Linear Array Thermal Systems (SLATS) collector utilizes a series of slat-like mirror segments which are moved slightly during the course of the day to track the sun. (6) The concentrating designs, and particularly the more _ highly concentrating versions, are not effective in collecting diffuse radiation; and, therefore, the applicability of these collectors may be limited to regions with a large amount of direct insolation. Also, higher temperature performance generally entails penalties in terms of cost and system complexity. Although mentioned as_ general background with regard.to domestic hot water heating, these designs are much more likely to find application in commercial installations requiring air conditioning or industrial process heat applications rather than residential water heating applications. 622 I § <é 26.2.2 Technical Characteristics 26.2.2.1 System Design Options Figure 26.3 presents a classification of generic solar water heater systems based on the type of integration between solar collectors and storage. : The three major categories of system-types listed in Figure 26.3 and illustrated schematically in Figure 26.4 (with the storage volume shaded) are thermosyphon, integral collection-storage, and forced circulation systems. In the thermosyphon and forced circulation systems, the storage, which is invariably potable water, is located remote from the collectors and connected to them by either a direct heat-transfer loop as shown in Figure 26. , or an indirect heat- transfer loop (heat exchanger). The storage reservoir of the integral collector storage is always in direct thermal contact with the collector absorber; however, this storage may be either the potable water passing through the collector as shown in Figure 26.5, or a liquid separated from the potable water by a heat exchanger. The general features of these types of systems are discussed below. Thermosyphon Systems Thermosyphon systems have the advantage of being. self controlled--the natural circulation loop is solar heat-driven--thereby eliminating the need for a pump, controls and ancillary power. Although the thermosyphon system has the potential for low cost and high performance, it also requires that the storage.tank be physically located above the collector, which frequently poses aesthetic and structural problems, and possible vulnerability to freezing. (The potable water lines must be. routed to and from a storage container which is likely to be located in an unheated space.) Thermosyphon systems using the direct circulation of potable water have been used extensively in mild climates, such as Israel and. Australia, but in the United States, their use is confined to extreme southern regions. Even in Florida, solar purchasers apparently feel the aesthetic and structural advantages of the thermosyphon outweigh its potential performance and cost advantages. In recent years, indirect thermosyphon systems using a phase change heat-transfer fluid have arrived. on the market, and they can operate in freezing climates. Although thermosyphon systems. have considerable merit, they account for only a few percent of the United States solar water heater market. 623 - 89 System Types Forced Integral Thermosyphon Circulation Storage FIGURE 26.3 CLASSIFICATION OF SOLAR WATER HEATER GENERIC DESIGNS Re-circulation S29 Thermosyphon FIGURE 26.4 Integral Collector Storage Forced Circulation MAJOR TYPES OF COLLECTOR-STORAGE INTEGRATION 929 Drain Back FIGURE 26.5 Antifreeze-Water Heat Exchanger - 3 Antifreeze Air-Water Heat Exchanger and Blower Air INDIRECT FORCED CIRCULATION SYSTEM OPTIONS _ Integral Collector-Storage System The integral collection-storage type, like the thermosyphon, needs no active circulation equipment. Low-cost versions of this design have been used commonly in Japan and less commonly in the United States, Although these systems generally were not compatible with freezing temperatures, they could-be used. for summer operation. Improved versions, currently becoming available, are designed for freezing temperatures, Some types of integral collector-storage systems can .avoid the aesthetic problem of the thermosyphon system and minimize the structural problems, since the storage volume and weight can be distributed among several collectors and ‘the systems are similar in appearance to conventional systems. Freeze vulnerability is roughly comparable to the indirect thermosyphon systems: the storage can be adequately insulated (by insulation and/or front surface glazing); however, the water lines leading to and from the system must be protected. The integral storage system will experience. a higher night time heat loss’ than either thermosyphon or forced convection systems, counterbalanced to some extent by the elimination of heat loss from the circulating loop connecting storage and collectors. The resulting overall collection efficiency is both somewhat lower than that of the other. types and highly sensitive to use patterns. (Concentrating water usage in the late afternoon maximizes efficiency; concentrating it in the morning, minimizes efficiency. ) Relative to forced circulation systems, the integral collector-storage approach has the potential for cost reductions which may offset its lower performance. The integral storage design is even less common than the thermosyphon design in the U.S. solar water market. Forced-Circulation Systems Forced-circulation. systems are commonly available with various types of either indirect or direct heat transfer loops. Figure 26.5 illustrates the indirect (or closed loop) forced-circulation systems, where .a- secondary heat transfer fluid and heat exchanger are employed for heat transfer between collector and storage. The three major types of indirect systems and their distinguishing features are described below. : ‘ e Antifreeze--the antifreeze design, using a heat exchanger to isolate the antifreeze liquid from the potable water, has been considered to be a generally reliable option for freezing climates and is currently, by far, the most common type of indirect system. However, some major problems and unresolved issues intrinsic to this system are: the lack of a "perfect" antifreeze having optimal compatibility, stability, and heat transfer and fluid flow. properties, and an 627 uncertainty as to the need (or requirement) for double wall heat exchangers. In colder regions of Alaska, conventional ethylene glycol antifreeze may not offer sufficient protection against low temperatures. e Drainback--in the indirect drainback system, the heat transfer fluid drains back to a sump in contact with the potable water whenever the heat transfer loop pump is shut off. An important feature of this design is that the heat transfer fluid need not be exposed to either freezing temperatures or high stagnation temperatures in the collector, thereby allowing the use of a low cost, highly stable fluid with excellent heat transfer and fluid properties, such as corrosion inhibited water. Drainback systems are likely to become somewhat more common, even though the drainback concept does require that the lines leading from the collector to storage maintain a constant downward pitch. e Air--air collectors have been of interest for solar water heaters and, more particularly, for solar space heating systems, because of the very benign properties of air with respect to high and low temperatures, corrosion and toxicity. However, air is inferior to liquid with respect to heat transfer and flow characteristics (heat-transfer coefficients are low, ducts must be large, and circulating power can be high), and the resulting systems have been found to have both higher cost and lower performance* than liquid systems. It is not expected that air systems will be a major factor in the solar water heater market. In the direct systems, the heat-transfer loop contains potable water at water main pressure which is circulated directly between collector and storage without an isolation heat exchanger. The elimination of the heat exchanger serves both to improve efficiency and to reduce systems costs. Freeze protection is generally achieved by control schemes employing either recirculation or draindown. e. Recirculation--in the recirculation type, the control activates the heat-transfer loop pump to circulate water when the collector temperature approaches freezing. This system, which is well adapted to regions of infrequent freezing temperatures, is the most common type of direct system. * These remarks apply specifically to solar water heaters. Space heating systems using rock storage and air collectors can operate with relatively low air flows and large temperature gradients across collector and storage to achieve performance and costs comparable to liquid systems. 628 e Draindown--in the draindown system, the control activates a drain valve(s) (typically solenoids) which allows the water in the collector and lines to drain to an external sump when the collector temperature .approaches — freezing. Like the drainback system, routing of lines is critical. Also, the successful operation of the draindown system requires high reliability of controls, drain valves, and air vent valves. The system options discussed above are primarily concerned with methods of interconnecting collectors and storage. However, there are also different ways of integrating the solar storage with the auxiliary or back-up heating system. As shown in Figure 26.6: e the solar water heater may function as a preheater to an existing water heating system (frequently referred to as a - two-tank system); or e@ the solar water heater ‘and auxiliary heater may utilize a common integrated storage tank (i.e., a one-tank system). The preheat approach is functionally compatible with any auxiliary heating system and any collector storage integration methods, and permits large storage capacity. Since, the preheat system has dedicated solar storage, collection efficiency is maximized; however, storage losses tend to be high. The integrated tank design has the advantage of compactness and low costs in new installations--or where the existing water heater is being replaced due to age, malfunction, or low efficiency. Results of recent National Bureau of Standards tests indicate the integrated tank design has higher overall system collection efficiency because of reduced storage losses. - Until recently, only solar electric integrated tanks were commercially available. However, some solar/gas-integrated tanks are now arriving on the market and new designs are in development. In an integrated tank, the auxiliary heater typically is located in the top half of the tank and the solar connections are located at the bottom, so as to encourage thermal stratification and minimize solar collector temperature. 26.2.2.2 Performance Specification and Estimation Figure 26.7 illustrates the performance of three typical collectors--a single pane flat plate collector, a vacuum insulated tubular collector, and a concentrating collector of the SLATS variety. This plot shows collector efficiency as a function of the parameter AT/I, where AT is the temperature difference between inlet fluid and ambient, and I is the incident solar radiation. This method of presenting solar collector efficiency data is based on the equation discussed in Section 26.2.1 above. (In the case of. concentrating collectors, the incident radiation I should be interpreted as the direct component of total insolation. ) 629 0€9 Separate Solar Pre-heat and Auxiliary Integrated Soiar/Auxiliary Tank FIGURE 26.6 STORAGE INTEGRATION METHODS 7 Collector Efficiency Flat-Plate Daystar Mode! 1600 ~ Concentrator (Suntec SLATs)® Evacuated Tubular (Owens-Illinois SSR) 1.0 1.5 5 ° 2 (°F -hr-ft’ AT/I ( Btu / AT = Inlet Fluid Temperature — Ambient Air Temperature - , Incident Radiation Source: Reference 3. FIGURE 26.7 COLLECTOR PERFORMANCE DATA 631 2.0 As shown, the high temperature performance of both the evacuated tube and concentrator is superior to that of the flat plate collector. It is difficult to use Figure 26.3 as an indication of monthly or annual performance since temperatures and incident radiation are constantly shifting and experience has shown it is difficult to assign meaningful average values. However, some qualitative observations are possible. For solar water heating in a warm sunny climate, (i.e., AT = 120-80 = 40°F, I = 300 Btu/hr-ft“-°F, AT/I = .133), the evacuated tube and concentrating collector have little or no performance advantage on the flat plate collector. However, for cold, cloudy conditions, (i.e., AT = 120-30 = 90°F, I = 200 Btu/hr-ft"-°F, AT/I = .45, the advantage of the evacuated tube could be significant. A standard test for solar collector instantaneous efficiency has been established by ASHRAE which results in a plot of instantancous efficiency vs. the parameter AT/I--of the type presented in Figure 26.7. Several test laboratories are now equipped to perform the ASHRAE 93-77(2) test and manufacturers commonly use this data in their product literature. : A procedure for testing overall solar water heating system performance (ASHRAE 95-80) (7) which utilizes the ASHRAE 93-77 collector test results has recently been completed. In its current form, it is only applicable to forced circulation systems. Solar water heaters are normally designed to supply 40 to 70% of annual water heating demand. Designing for near 100% solar fraction* almost always requires uneconomically large collectors and storage. However, it is not uncommon for an owner to turn off the auxiliary during the summer, when solar collection is at its best, letting the system operate in the 100% solar mode and adapting to the available hot water. : To assist in assessing optimal design, there are several methods of predicting the monthly and annual performance of solar water heating systems: very sophisticated dynamic simulation performance useful for optimizing system design, various look-up tables provided by manufacturers, and government publications which provide rough estimation guides. (8) F-Chart, a correlative program, based on results obtained from the TRYNSIS dynamic simulation model, is the most commonly used method for predicting monthly and annual solar fraction.(9) It is the basis for many of sizing estimation guides prepared by the manufacturer. F-Chart can be applied to some 266 locations including several. in *Solar fraction is defined as that portion of the load that is supplied by solar 632 js he Alaska. Input data includes collector area and orientation, collector efficiency data (intercept and slope of plots of efficiency vs. AT/I like those in Figure 26.7), water usage, and certain system characteristics like storage quantity and collector loop heat exchanger effectiveness. Plots of solar fraction as a function of solar load ratio* have also been used as a method of estimated system annual performance. (10) These correlation plots are generally based on certain experimental or computed data points. They can also be used to roughly approximate monthly performance. Figure 26.8 presents solar fraction as a function of solar load ratio for a solar water heater employing a single pane selective surface collector. The correlation line shown is based on data points for annual performance obtained from F-Chart calculations for four locations in Alaska--Annette, Matanuska, Bethel and _ Fairbanks. These calculations were performed for a solar water heater with a 60 ft? single pane flat collector tilted at latitude angle. Freeze protection was achieved with an antifreeze loop and heat exchanger. The thermal load corresponded to a daily consumption of 60 gallons per day with a temperature rise of 90°F. The calculation indicates that in Matanuska (having climatic conditions similar to Anchorage), the annual. solar fraction would be approximately 47% of the thermal load. The correlation line in Figure 26.8 can be used as a means of approximating annual solar fractions of other locations for which incident solar radiation data is available--or, as an even broader approximation, estimating seasonal as well as than annual solar fraction. (Summertime solar fraction tends to be larger than indicated by the correlation of annual data, and wintertime solar fraction, less, because of ambient temperature effects.) Using Matanuska as an example, the annual solar load ratio is calculated as follows: Annual Incident Radiation = 400,000 Btu/ft? yr —. (Equivalent to an average daily value of 1100 Btu/ft? day) Thermal Load = 5224 kWh = 17.8 million Btu/yr Solar Collector Area = 60 ft? Therefore, the annual solar load ratio (SLR) is: Annual SLR = S00 Cae = 1.35 and Annual Solar Fraction = .47 per Figure 26.8. *Solar load ratio is defined as the ratio of incident solar radiation to load. Incident solar radiation for Alaska can be obtained from sources such as the National Climatic Atlas (11) (horizontal insolation) and Alaska Solar and Weather Information (12) (horizontal and tilted) in addition to F-Chart printout. 633 —— Solar Fraction 1.0 8 *System Data © 60 Ft2 Single Pane Selective Black Collector @ Latitude Tilt @ Anti-Freeze with Heat Exchanger © 60 Gallons/Day @ 90°F Rise Fairbanks Bethel Matanuska F-Chart Annual Performance*™ 5 1.0 1.5 2.0 2.5 Incident Solar Thermal Load Solar Load Ratio = FIGURE 26.8 SOLAR FRACTION CORRELATION 634 a During summer when daily incident radiation for Matanuska is in the range of 1400 Btu/ft? day (a 25% increase over the annual average), the solar load ratio would increase to about 1.70 and the solar fraction would be about 60%. However, in mid-winter when incident : radiation is about half the annual daily average, the solar load ratio i would drop to .7 and the solar fraction would be 30% or less. 26.2.2.3 Installation and Operational Considerations Solar water heaters achieve best performance when facing due South and tilted at latitude angle with respect to horizontal. Precision is not critical--departures of 10° to .15° in either azimuth or tilt will a a reduce annual solar fraction by only a few percent. In’ many cases, it may be preferable to accept the non-optimum orientation of an = existing. roof rather than construct .a complicated mounting ik arrangement. However, it is important that the collector not be shaded by trees or other structures at any time of the year. .- (Various simple sun locating devices are valuable in determining whether obstructions in Southern exposure, will shade thé collector at any time of the year.) The installation of solar water heaters requires care with respect to attachment of the collectors to the roof and the. plumbing connections i to avoid leaks in either the roof or the solar piping. Also, the \ placement of certain components: like air. vents and temperature sensors and the sloping of piping for drainback and draindown systems is critical. However, do it yourself installation of . prepackaged solar water heaters is quite feasible (and can result in [ substantial savings) for someone with carpentry and plumbing skills and the ability to follow installation instructions. oe 5 On another level of difficulty, it is possible to build the collectors { largely from raw materials (glass, copper sheet and tube, etc.).(13) A pilot program involving user fabricated systems conducted by Arizona State University resulted in extremely low out of pocket costs for the users.(14) However, these participants did have the benefit 17s of a short course in solar water heater construction and installation a provided by Arizona State. : 3; Maintenance of a well designed, well installed, solar water heater can be relatively simple and limited to periodic checks for leakage and to assure the controller turns the system on and off in accordance with the position of the sun. The major item of predictable maintenance is the annual checking and perhaps periodic replacement of. glycol based heat transfer fluids, the degradation of which was considered the most serious long range system problem by a DOE task force which reviewed the status of active solar systems.(16) Typical malfunctions experienced in solar systems were: (15) e controller failure -- widespread in early stages of the industry but improving in the last few years; 635 — e freeze-up -- partially a problem in early draindown systems due both to poor line slope and draindown valve malfunction; e piping leakage. 26.2.3 Environmental Considerations Solar hot water systems cause little adverse environmental impact other than aesthetics where the system is prominently, and perhaps awkwardly, installed. -Positive environmental impact is achieved from reduce emissions due to displaced fossil fuel. 26.2.4 Technology Status Solar water heaters have enjoyed commercial success in Israel, Australia and Japan for over 3 years. Typically the systems were designed for warm climates without provisions to prevent freezing. A thriving market also existed in Florida in the 1930's but collapsed due to the introduction of low cost natural gas. The current solar industry (water heaters and other technologies) in the United States received its start in the early 1970's due to the oil embargo. For several years, a variety of government programs were in -place designed to further R&D and serve as incentives to commercialization. At present, most active government support programs have been phased out. However, federal tax credits (40% of installed cost up to a maximum of $4,000 credit) and state tax credits still provide a very significant incentive. The United States solar hot water industry has grown: every year since the early 1970's. In 1981, over 92,000 solar hot water heaters were installed with over 88,000 of these being in single family residences.(1) Although areas with warm, sunny climates such as California and Florida predominate, there is also considerable activity in the Northeast states where freeze protection is critical. The solar water heating industry is still in its early stages. A large number of manufacturers exist. The June 1981 issue of Solar Age lists over 200 firms that offer complete solar hot water systems including one or more major components of their own manufacture. (17) (The list excluded firms that were packagers only.) The level of installation experience is still relatively low--a large majority of solar installers contacted in a recent survey indicated they had installed less than 25 svstems.(1) However, the industry is tending to mature and at both manufacturer and installer level. Fewer companies are capturing larger shares of the market. Constraints to the widespread use of ‘solar water heaters include: 636 @ marginal economics--as discussed in the next section, solar water heaters . currently require substantial. economic incentives (tax credits) to be financially attractive. @ availability of adequate unshaded southern exposure where collectors can be conveniently attached. e availability of adequate solar insolation. Anchorage, Alaska receives only about one-half the annual incident solar radiation (on a latitude tilted surface) of Phoenix Arizona; and only about one quarter in mid winter. Low solar radiation degrades economic performance--unless counterbalanced by high conventional fuel costs. Arthur D. Little, Inc. surveys of manufacturers indicate there is an emerging consensus on solar water heater design for the present and near term. Most manufacturers of systems for. northern areas favor antifreeze systems--with some increasing interest in drainback design. Typically, a single pane glass collector with selective black copper absorber is favored. Future directions which may improve system performance/cost tradeoffs include: | e low cost plastic collectors, @ seasonal collection systems designed for summer-only operation , @ owner assembled kits, e improved integral storage and thermosyphon systems. Field monitoring’ problems conducted’ by government agencies or contractors and by utilities have been very valuable in charting the progress of equipment development.(15,18,19) The development in the Lower 48 appears to be proceeding through three stages: ‘e solving installation problems; in early programs, installation problems obscured all other issues but now they have been largely resolved; “e overcoming persistent failure in key components, particularly controllers; the reliability of controllers appears to be improving; and e evaluating generic in terms of performance and operational reliability system types; installation and component performance are approaching a point where the intrinsic differences in systems can be accurately defined. 637 26.3 Economic Implications 26.3.1 System Costs Equipment First Cost Installed costs of solar water heaters vary widely with installer and manufacturer. However, in addition to significant random fluctuations, there are some systematic variations that can be identified. For example: e Economy of scale is appreciable. Cost per unit area tends to decrease with area since certain cost components such as controls, piping, and some installation tasks are primarily fixed costs with little or no size dependence. e Systems designed for northern regions tend to be more costly than those in southern regions due to the requirement for freeze protection. Also, the installed costs for southern systems may reflect easier installations (single story houses) and generally more experienced installers. Figure 26.9 presents installed cost per unit. area as a function of collector area for northern and southern systems. The range of installed costs shown are based on data from a number of sources including surveys of manufacturers and users, experience from government and utility programs and information supplied by regional solar centers (19,20,21). The data apply to systems installed in the Lower 48 in the last few years corrected to 1982 using a cost escalation factor of 10% per year. As an additional point of reference, a recent survey indicated a nationwide average cost of $3235. for 61 sq ft systems installed in 1981 $53/sq ft(1). The corresponding 1982 cost is $58.50 which falls approximately midway between the Northern and Southern curves. Cost reductions of up to 30% might be possible in the next ten years due to reductions in manufacturing costs, reduced installation costs, and reduced mark-ups throughout the distribution chain. Possibly even greater reductions might result from the new developments mentioned previously. However, experience in the last few years has not indicated any significant cost reductions have occurred to date. The breakdown of installed solar water heater cost, for systems in the range of 50-70 sq ft can be approximated as follows (22): Equipment and Materials. 60% (solar package, tanks, piping, etc.) Direct Labor 15% Overhead and Fee (on above) , 25% 100% 638 At $.039/kWh, the annual fuel cost for the electric auxiliary backup is $144, The annual fuel cost for a similar (but nonsolar) conventional electric water heater meeting the total load would be $240. Therefore, the savings attributable to solar is $96. Figure 26.9 indicates that the cost per unit area of a 60 sq ft solar water heater is $61 sq ft for Northern systems in the Lower 48. In Anchorage, this is estimated to cost $79.60 for Anchorage (Table 26.1), or a total installed system cost of $4776. It should be noted that since direct.labor plus overhead and fee comprise 40% of the total--substantial savings might be achieved by. the do-it-yourself route. Lastly, the annual maintenance costs estimated at $50/yr for the Lower 48 might be expected to be greater in Alaska due to higher labor and material costs. However, considering the fact that the $50/yr may be overly conservative since it is based on data a few years old when installers were less experienced, it is recommended that the $50/yr maintenance be applied directly to Alaska without further escalation. Much of this $50/yr could be eliminated by owners willing to do their own maintenance. In summary, the cost estimates (Tables 26.1 and 26.2) for the 60 sq ft Anchorage solar hot water heater are: ' First Cost:of Installed Solar Water Heater $4,776 (without tax credits) Annual Maintenance Cost : $ 50/yr Annual Fuel Cost for Solar-Electric Auxiliary ~ $144/yr Annual Fuel Cost for Conventional Electric $240/yr Water Heater Savings Due to. Solar $ 96/yr The above results suggest that the economics of solar water heaters in Anchorage are not very. attractive. Tax credits have not been included in the above. tabulation. However, after including the 40% federal tax credit--and even assuming an equally generous state tax credit, the equipment first cost would still be about. ten times the first: year fuel.savings. These economics tend to be much less favor- able than in many locations in the Lower 48 due to the combined effects of: e lower solar insolation e higher equipment costs e lower fuel costs 639 100 80 o Northern” Systems 60 SSSSSSSS=S___=S=——_— SSESSSSH_]_-_—-=—= N S-_S-_-_-_==-l=—™= ~ SSSSSSSS_SS_===S== uw eee > jSSSSSS===SS== oe SSS SSS 28 . Oo] O}< 40 “Southern’’ Systems 20 Based on Collector Aperture Area 20 40 60 * 80 2 A — Collector Area, Ft Source: Arthur D. Little, Inc. FIGURE 26.9 1982 SOLAR WATER HEATER INSTALLED COSTS FOR LOWER 48 640 7. ba Annual Maintenance Costs Maintenance costs for solar water heaters are not well defined--and those data which are available might be expected to be biased upward by the relative experience of installers. Some data points that do exist are: : - @ New England Electric experienced an average of 1.4 service calls per year with 80 antifreeze systems during 1977 and 1978.* They used an average cost. of $35 per call--or an r annual maintenance cost ‘of $49 per yr in their economic 5 studies. The split between labor and materials was roughly: bi labor--65%; materials--35%. (23) , e Long Island Lighting Company experienced a system average of $50-$75 for service calls during their program which began Vo in 1978 and ultimately involved 600 drainback systems in 1981. c However, it was felt these costs might be higher than ' necessary since many of the calls were for very minor problems (like adjustment of mixing valves), possibly encouraged by the terms of their warranty.(24,25) Typical Economic Example | The example chosen is a 60 sq ft flat plate solar water heater b installed on a single family residence in Anchorage, Alaska. The : collector uses the single selective surface design tilted at latitude 3 angle (i.e., 61°). An antifreeze loop provides freeze protection, | integrated with a single tank containing an electric auxiliary. This system corresponds to the schematic of Figure 26.1. Its performance and cost are defined by Figures 26.8 and 26.9 respectively. i The average daily water consumption is taken as 65 gallon/day and we the temperature rise 90° (i.e., 50°F to 140°F). This results in a - thermal load of 5224 kWh. Assuming an annual storage tank i efficiency of 85%, the total heating required is 6145 kWh. The results of Figure 26.8 indicated that the solar water heater would supply 47% of the thermal--or an annual solar contribution of 2450 kWh. The electric back-up would be required to supply the difference between the total heating requirement and the solar ‘ contribution or 3695 kWh. ‘ 38 bo *The New England Electric program was an early program which “ experienced extreme difficulties due to inexperience of installers and manufacturers. However, the service call data shown does not include chronically malfunctioning systems which were later replaced. 641 TABLE 26.1 CAPITAL COST SUMMARY TECHNOLOGY Solar Domestic Hot Water BASIS Location Anchorage Year 1982 constant dollars CAPACITY Input Not Applicable Output 2,450 kWh/year ESTIMATED USEFUL LIFE 15 years CONSTRUCTION PERIOD 1 week CAPITAL COST Equipment and Materials $2,525° Direct Labor : 1,015 Overhead and Fee 1,236 TOTAL CAPITAL INVESTMENT $4,776 642 TABLE 26.2 OPERATING COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output OPERATING FACTOR Operating Operating Costs VARIABLE COSTS Electricity (backup) TOTAL VARIABLE COSTS FIXED COSTS Maintenance Capital Charges, @ 9.5% of Capital Investment TOTAL FIXED COSTS TOTAL ANNUAL OPERATING COSTS TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST -Solar Domestic Hot Water Anchorage 1982 constant dollars Not Applicable 2,450 kWh/year Not Applicable Annual Unit Cost Consumption Annual Cost 3.9¢/kWh 3,692 kWh $144 144 50 454 504 $648 2,450 kWh $0.26/kWh 643 The effect of lower fuel costs is compounded by the fact that Alaskans generally use the lower cost fossil fuel (gas) water heating rather . than electricity. This preference for gas is even more pronounced than in the Lower 48. In addition Alaska has lower state tax incentives than most of the states in the Lower 48, The relatively low solar insolation is generally characteristic of all Alaska. Higher equipment costs also apply to all Alaska, sometimes much higher those assumed for Anchorage--except where the system is owner installed. However, many areas have substantially higher fuel costs than those assumed for Anchorage which could make a ‘substantial difference in the annual dollar savings attributable to solar. Perhaps the greatest opportunity for savings is the owner assembly of low cost summer only systems. Batch or breadboard type ‘of integral collector storage designs, can be built for as low as $300.(26) These systems would clearly not achieve the same savings as an equal area all year system. However, since solar radiation is highest in the summer months, it is reasonable to expect that’ the annual savings would be 60% or more of that achieved with an all year system. Therefore, when tax credits are considered, the out-of-pocket first costs might be 2 to 3 times the first year energy savings. 26.3.2 Socioeconomic Factors The widespread use of solar water heaters in Alaska could result in new permanent employment and increased purchase of goods in-state. A thriving solar industry would likely involve in-state manufacture of flat plate solar collectors from commodity items such as _ glass, aluminum extrusions, insulation and either copper sheet and tubes--or prefabricated precoated absorber plates. Pumps, controls, valves, tanks, etc. would probably be imported from out of: state. Also, evacuated tube collectors would probably be imported. Solar heat technologies are generally fairly labor intensive and new permanent employment could be expected from manufacture of flat plate: collectors, system packaging, distribution and installation. System installation would be expected to create the greatest additional employment--and training specific to solar system installation as well as plumbing, carpentry, and electric work would be required. Solar installation training in the Lower 48 has been offered by manufacturers, utilities, government organizations, and _ private vocational schools. Courses have been conducted both in classroom fashion generally supplemented by field projects and by correspondence. 644 _ 26.4 Impact 26.4.1 Effect on Overall Energy Supply and Use Residential water heating, assumed to be approximately 21 million Btu per household per year, currently amounts to about 1.7% of total end use energy consumption in Alaska based on the following: e residential use is 12% of total end consumption ; (27) use energy e residential thermal energy consumption averages about 150 million Btu per household in the more populated regions. (27) Of this quantity, roughly 14% is hot water; e therefore, residential hot water, as a fraction of total end use energy consumption, is about 12% x 14%, or 1.7%. If 100% penetration of 50% solar systems were achieved, the energy savings would be .85% of total energy use. Since economics would appear to preclude any very large penetration in the near term (i.e., next ten years), the actual energy saving is likely to be a small fraction of 1%. Although this impact is small when viewed as a percentage of total energy use, it could contribute to new industrial growth and employment opportunities as mentioned in the previous section. , 26.4.2 Future Trends As previously noted, economic considerations are currently the primary constraint to widespread use of solar water heaters. Increased energy costs will help solar water heaters. However, even with increases of the order of two or three over current rates in the more populated areas, high labor and equipment costs and relatively low solar availability in Alaska are likely to preclude wide use of conventional professionally installed,. year-round systems. Low cost summer only systems may provide a more economically sound alternative especially where the user is willing to assemble and install the system. 645 10. 11. 12, REFERENCES \ Mihlmester, P. and H. Bernstein, "National Sales Trends, What's Up, What's Down," Solar Age, December 1982. "Methods of Testing to Determine the Thermal Performance of Solar Collectors," ASHRAE 93-77. Manufacturer's Product Data (a) DayStar 1600 Solar Collector Data Sheet D785-1600/TLS 6179 (b) Owens Illinois Sunpak Product Data SP-1080-80 (c) Sunmaster Solar Energy Collection Systems Featuring Sunpax Evacuated Tube Receivers, Sunmaster Corporation, 1979 (d) "SLATS--A New Concept in Solar Collectors", May 1977. Available from George Kannapel, Suntec Systems, St. Paul, Minnesota. ‘ Winston, R. "Principles of Solar Concentrators of a Novel Design", Solar Energy, Vol. XVI, No. 2, 1974, pp. 89-95. Pendelton, R. L., “Evaluating a Solar Energy Concentrator," ASHRAE Journal, November 1976. Davidson, J. H. and A.J. Wendt, "The Solar Linear Array Thermal System," Solar Age, February 1978. "Methods of Testing to Determine Thermal Performance of Solar Domestic Water Heating Systems," ANSI/ASHRAE 95-1981. "Design Methods and Analytical Tools," Solar Age, August 1980. Beckman, W.A., et al., "Solar Heating Design by the F-Chart Method," Wily InterScience, New York,. 1977. "Predicting the Performance of Solar Systems," U.S. Army Construction Engineering Research Laboratory Report CERL-1R-E-98, January 1977. Climatic Atlas of the United States, U.S. Department of Commerce, June 1968. Alaska Solar and Weather Information, Western Sun, Portland, Oregon. 646 13. 14. 15. 16. 17. 18. 19, 20. 21. 22. 23. 24, Morgan, E., "Five Solar Water Heaters You Can Build," Popular Science Solar Energy Handbook for 197 ’. Mumma, S. A., "A Program to Involve the Homeowner in the Fabrication and Installation of a Solar Water Heater, ASHRAE Paper LA-80-9, No. 3. "Survey of United States Domestic Solar Water Heater Designs and Experiences," Final Report to Solar Energy Office, National Research Council, Ottawa, Canada, prepared by Arthur D. Little of Canada Limited, August 1982.. Blue Ribbon Committee Assessment of Field Status of Active Solar Systems, Institute of Public Administration, January 1982. The Solar Age Domestic Hot Water Directory, Solar Age, June ‘1981. Meeker, J. and L. Boyd, "Domestic Hot Water Installations -- The Great, The Good and The Unacceptable," Solar Age, October 1981. 1980 Survey and Evaluation of Utility Conservation Load Manage- ment and Solar End-Use Projects, EPRI EM-2193, Volume 2, Solar End-Use Projects, Volume 2, Project 1940-1, Final Report, January 1982, prepared by. Energy Utilization Systems, Inc. Arthur D. Little, Ine. unpublished data from surveys of manufacturers and users, 1980. Telephone conversations October 1980 relative to solar’ water heater costs to following regional centers: - Northeast Solar Energy Center - Western Sun - Mid America Solar Energy Center - Southern Solar Energy Center "Installation Q&A", Solar Times, October 1980. Gowing, R. N., Two Year Field Test of 100 Solar/Electric Domestic Water Heating Installations in Massachusetts, Rhode Island, and New Hampshire, New England Power ervice Company, 79-010-1 and 2, March 1979. : Avril, F. P. and S. Hooks, "Solar Energy and Electricity are Working Together on Long Island," Long Island Lighting Company, November 1981. 647 25. 26. 27. Telephone conversation with Mr. Fred Avril of Long Island Lighting Company, April 16, 1982. Gruber, J. and R. Home, "A Low Cost Batch Heating," Solar Age, May 1982. State of Alaska Long Term Energy Plan, 1982 Report, Division of Energy and Power Development, State of Alaska. 648 27.0 TRANSPORTATION END-USE SYSTEMS - OVERVIEW 27.1 Introduction - Transportation in Alaska Over the past ten years, saving fuel in the transportation sector has been a high priority national goal. Therefore, technologies for improving energy efficiency have been researched, documented, and, in many cases, adopted for use on highway vehicles, aircraft, railroads, and marine craft. , The situation facing Alaska differs substantially from that of the rest of the country. Today's transportation issues in Alaska are (1) adequacy of transportation systems, (2) availability of reasonable cost transportation, and (3) reliability of transportation systems and equipment. Energy conservation has been a secondary issue to these more fundamental needs and concerns. In the transportation sector, essentially all research, development, and manufacture of the vehicles and equipment takes place outside of Alaska. The majority of the vehicles (automobile, rail, and air) are designed for a mass market and used under very different operating conditions from those in Alaska. Therefore, the State limited control over incorporating. technology into transportation vehicles which would be particularly suited to the Alaskan environment. However, the recent focus of the current transportation sector nationally on energy conservation has made. vehicles and options available which are significantly more fuel efficient than those. prior to 1973. Through awareness of what influences the fuel efficiency of a transportation vehicle, the State of Alaska can promote the specific purchasing and usage of a more efficient transportation fleet. In 1980, of the 204 trillion Btu's of delivered energy consumed in Alaska, 89 trillion Btu's (44%) were consumed by the transportation sector.(2) This represents approximately 650 million gallons of transportation fuels. By sector, aviation consumed 44%, highway vehicle 39%, off-highway (including rail) 10%,.and marine 7%. For comparison, in the U.S., the transportation sector consumes 26% of the total energy consumed (6% of .the petroleum).(3) The breakdown by transportation mode is 8% aviation, 76% highway, 4% off-highway including rail, 9% marine, and 3% military. The most significant differences between the Alaskan consumption pattern and the national consumption pattern are the large percentage of fuel spent for aviation in Alaska (44%, compared to 8%). Correspondingly, the Alaskan highway fuel use percentage is slightly more than half the national average. 649° th erN eens ery7Av2 yaya types TAS TALS AT PE WAZ CLN TAT N NEA VIS Ne ‘Percent of Heat Supplied AX X BALK SSK Brake Horsepower O Miles Per Hour Based on average performance of gasoline engines. Source: Reference 10 FIGURE 27.1 ENERGY DISTRIBUTION INA | 4-STROKE GASOLINE ENGINE 650 Consumed by Engine Available to Propel Vehicle 27.2 Summary of Energy Technologies for Highway Sector Over the past ten years, conserving energy in the highway sector has been a national goal for the United States. Consequently, a significant amount of research has been performed by both the vehicle manufacturers and the U.S. Department of Transportation to identify the potential areas for efficiency improvement in the vehicle (both passenger cars and trucks). As a result of work done at the Department of Transportation in conjunction with Arthur D. Little, Inc., an accepted useful categorization for the technology areas for a vehicle has been developed.(10) The seven distinct areas are: Engines . Transmission » Tires Fuels and Lubricants Weight Aerodynamics . Accessories NIAupwnr An additional area which also represents potential for energy savings is driver awareness and education to promote (1) fuel efficient operation, (2) purchase and/or specification of fuel efficient vehicles, and (3) car pooling in metropolitan areas. This overview of energy technologies for highway vehicles is framed around these seven major categories. Engines For highway vehicles in the foreseeable future, the predominant choice of engine will be the conventional internal combustion engine either spark ignited (using gasoline) or compression ignited (using diesel fuel). There are several technologies available, developed by the engine manufacturers, which improve either the overall engine efficiency or cause the engine to operate in higher efficiency regimes for more of the duty cycle. Figure 27.1 shows the energy distribution and efficiency for the average performance of a gasoline engine. For example, optimized speed control of the engine using electronic microprocessor control systems will ensure that the engine is operating at the most efficient point in its engine map for the instantaneous road load and speed condition. Typical components of a full engine control system are shown as Figure 27.2. Use of an equivalently powered diesel engine instead of a gasoline powered engine will increase the overall engine efficiency for the vehicle's operation (Figure 27.3). This is the most significant fuel saving opportunity for the engine category. A vehicle with an indirect injected diesel engine in place of a gasoline engine can improve fuel efficiency 25% in miles per gallon (mpg). The direct 651 D> cs SINGLE-POINT THROTTLE BODY MULTI-POINT FUEL MANIFOLD © TEMPERATURE (| ( FUEL PUMP map (—Y (== mS SENSOR — \ FUEL FILTER —= ey" EGR . OXYGEN SENSOR VALVE y « aw CRANKSHAFT SENSOR : ) \ AUTOMATIC A A mae FIGURE 27.2 TYPICAL CONTROL SYSTEM COMPONENTS FOR GASOLINE ENGINES Source: Reference 11 sO meni, Diesel and Gasoline Engine Efficiency LESS GASOLINE EFFICIENT Pounds Fuel Used Per Unit of Work Output FUEL-EFFICIENT DIESEL MORE EFFICIENT Source: Reference 12 FIGURE 27.3 COMPARISON OF DIESEL AND GASOLINE FUEL EFFICIENCY 653 FORWARD CLUTCH TORQUE CONVERTER THIRD CLUTCH SECOND CLUTCH POWER TAKEOFF DRIVE GEAR FOURTH CLUTCH FIRST AND REVERSE CLUTCH SPEEDOMETER ORIVE GEAR GOVERNOR LOCKUP CLUTCH CONTROL VALVE BODY OIL SCREEN Source: Reference 13 FIGURE 27.4 ALLISON MT 640 FOUR-SPEED AUTOMATIC TRUCK TRANSMISSION WITH LOCK-UP TORQUE CONVERTER - 654 “ injected diesel (one injector per cylinder), although not commercially available yet, has the potential of improving fuel efficiency 40% in mpg over the equivalent gasoline engine. Although the use of diesel engines represents the best potential for fuel savings, it has inherent disadvantages in the Alaskan climate. The diesel is difficult to start in cold weather due to waxing of the diesel fuel and the problems in getting compression ignition in a cold engine. This would limit its reliability and usefulness in Alaska. Many heavy duty diesel trucks are left idling overnight in the cold climate to avoid the cold starting problem. Currently, "starting aids" are becoming commercially available such as plug-in engine block heaters, improved glow plugs, and fuel additives. Other power plants which may become available to a limited degree in 1990's are the electric vehicle and the heat engine/electric hybrid vehicle. In the electric vehicle, rechargeable batteries would be the power supply for this short range urban type vehicle; both the mini- car and mini-van. General Motors is currently developing these vehicles, but for minimally adequate operation, major battery improvements are needed. These vehicles will be, relative to their gasoline competitors, very costly and probably require a major purchase of a second or third battery set during the vehicle's life. Only if petroleum prices increase dramatically or if petroleum becomes unavailable will the electric vehicle represent a cost effective option. Similarly, the electric hybrid is in the development phase and projected to be a complex, costly power system capable of improving efficiency 25%. Transmission The transmission's role is to transmit the power from the engine to the drive axle, using appropriate gear ratios to match load and speed requirements with the engine speed and power output. There are two ways a transmission can affect the overall fuel efficiency of a vehicle: (1) the efficiency of the transmission of power from the engine to drive axle can be improved, and (2) the transmission can act as a controller causing the engine itself to operate at higher efficiency regimes. Vehicle manufacturers have been improving the automatic transmission efficiency through the use of lock-up torque converters in both automobiles and trucks. The lockup torque converter is used to eliminate fluid coupling slip losses in the hydraulic transmission by applying a mechanical clutch during reasonably steady state vehicle operation, but not during vehicle start-up in first gear. Figure 27.4 shows an automatic truck transmission with lock-up torque converter. The potential fuel efficiency improvement is approximately 5%. Many manufacturers are installing these transmissions on their current product lines. 655 Rear Axle eee se ee cooeseseeceeseossceneesseree 100 90 Accessories, 80 Aerodynamic Drag 70 40 30 Rolling Resistance Breakdown % of Energy Used by Vehicle 8 20 10 0 20 30 40 50 60 Speed — Miles Per Hour Source: Reference 13 FIGURE 27.5 ENERGY PARTITIONING DURING STEADY CRUISE OF 73,000# GVW TRACTOR — SEMITRAILER 656 In the development phase is a continuously variable ratio range transmission (CVT) which improves operating efficiency by perfectly matching the engine output to the required load and speed through the availability of an infinite range of gear ratios. Therefore, for the full range of vehicle operating conditions, the engine will be operating in. the most fuel efficient region possible. It has a potential fuel efficiency improvement of 5-10%. | For increased reliability due to simplicity of operation and improved efficiency, a CVT would be particularly applicable to snow mobile operation. This application is also in the development phase. Tires Rolling resistance accounts for a major portion of the energy required to propel a vehicle. For an example of the relative energy required to overcome rolling resistance for a trailer truck, see Figure 27.5. There are currently available two significant options for reducing rolling resistance on any highway vehicle (car, light, medium or heavy trucks): radial tires and optimal inflation pressure. Use of radial tires in place of bias ply tires can reduce rolling. resistance about 30% and thereby save 5-10% of the fuel used. Currently, almost 100% of U.S. produced passenger cars are equipped with radial tires, while only 60% of the light duty trucks and a smaller percentage of medium and heavy duty trucks are so equipped.(8) However, with the increased availability of radial tires for the medium and heavy duty trucks their usage will grow. Further, maintaining recommended inflation pressures in the tires will decrease rolling resistance and thereby improve fuel economy, although less dramatically than the use of radial, rather than bias ply tires. Fuels and Lubricants ' Alternate Fuels To shift the burden from conventional, petroleum based fuels, a variety of alternate fuels are being developed which can be used in internal combustion engines with only minor engine/carburetor modifi- cations. Utilizing alternate fuels which are derived from sources other than petroleum would offer several benefits to Alaska: e near term use of compressed natural gas, where readily available, provides a substantial cost advantage over gasoline, particularly in the Anchorage area where Cook Inlet gas remains a relative energy bargain; e if petroleum prices escalate significantly, then the production and use of alternate fuels from Alaska's natural -gas or coal resources could be cost effective; 657 Fuel Consumption — Gallons per Mile Vehicle Inertia Weight, Kg. Pr 500 1000 1500 2000 . @ Light Cars (20.67 #12) 09 ; A Mid Weight (22.86 ft) © Heavy (26.65 ft) 20 08 07 n c , & 15 Oo 06 3 z 3 3 .05 8 1 10 # 04 a g 8 03 S 3 5 02 01 0 1000 2000 3000 4000 5000 Curb Weight, Pounds Source: Arthur D. Little, inc. Source: Reference 14 FIGURE 27.6 VARIATION OF EPA COMPOSITE URBAN AND HIGHWAY. FUEL CONSUMPTION WITH WEIGHT FOR THREE DIFFERENT SIZE CARS. HAVING CONSTANT INDIVIDUAL AERODYNAMIC DRAG (FRONTAL AREA) 658 e for vehicle operation in Alaska's climate, gaseous fuels (i.e., CNG) would improve the cold starting capability and improve fuel efficiency during warm-up; and e alternate fuels generally effect a reduction in air pollution. Currently, with the motivation for using alternate fuels is economic; therefore, the promise of these fuels can only be realized where fuel distribution and storage facilities are already in place, e.g., ENSTAR Natural Gas Company's CNG fleet, or where the expected market penetration is sufficient to justify investment in new infrastructure. Lubricants Reduction of engine friction can improve the operating efficiency of an engine, as previously shown in Figure 27.1. In Alaska, poor cold weather characteristics of some lubricants can hinder cold starting and increase engine friction. There are now commercially available synthetic lubricants which reduce friction during cold start and engine warm up period (first 3-10 miles driven). Mobil-1 is an example of a low viscosity synthetic oil applicable for use in this climate. There is a potential of improving fuel efficiency 5% through the use of such lubricants. Weight Reducing the vehicle's weight reduces the energy required to propel the vehicle at a given speed. Figure 27.6 shows the effect of reducing weight on vehicle fuel consumption. If a substantial amount of weight can be designed out of a vehicle, then the engine can be resized for the changed load requirements. Utilizing a smaller engine appropriately sized for the light weight vehicle can gain more efficiency than simply utilizing a larger engine and operating it at a reduced power output. All current passenger cars have been designed in a weight conscious manner, including reduction in overall vehicle size (use of front wheel drive), substitution of lighter weight materials, and component redesign for optimizing size, weight, and performance. Medium and heavy duty trucks may now be purchased with lighter weight components such as aluminum cabs or trailers. Typically, for passenger cars a 10% reduction in weight results in a 3% improvement in miles per gallon. Aerodynamics Overcoming air resistance accounts for a major portion of the energy required to propel a vehicle as shown earlier in Figure 27.5. The amount of air resistance depends primarily on the vehicle's shape and speed, and wind conditions. As highway speed increases, the portion of the energy consumed to overcome air resistance increases substan- tially. Designing an aerodynamically shaped vehicle is, therefore, 659 14 Fan with Viscous Clutch 12 10 Horsepower a Horsepower Flex Fan 0 10 2 30 40. 50 Alternator — 55 amp. Horsepower 14 Air Conditioning c Power Steering - Curvel Full Load a n 906 psi 6 600 psi Time Averaged Horsepower Required © 3 Constant HP. Req. 0 10 20 30 40 50 60 Engine Speed (100 rpm) Engine Speed (100 rpm) sey Source: Reference 10 FIGURE 27.7. POWER REQUIREMENTS FOR ACCESSORIES IN STANDARD-SIZE CAR 660 most important for vehicles which will be used mainly for highway travel, a factor of somewhat less importance in Alaska than in the contiguous states. Vehicle designers and manufacturers are investing considerable effort in developing aerodynamic passenger car designs. Aerodynamic drag reduction devices are available for long haul trucks as an option or retrofit device. The fuel savings offered by such devices range from 3-8% with a reduction in wind resistance of 20%. Accessories All the accessories to the main powertrain require energy to perform their functions for a standard car as shown in Figure 27.7. As part of the extensive effort to improve vehicle fuel efficiency, even the minor gains available from designing and utilizing accessories with reduced power requirements have been sought and achieved. By reducing parasitic losses the overall vehicle efficiency has been improved through the use of such components as fan clutches, flexible fans, or electric fans. These are currently commercially available and widely utilized. In the future, the use of variable speed drives for accessories will further increase the efficiencies by providing the power required only for the actual accessory operating conditions. 27.3 Summary of Energy Technologies for Aviation Sector As in the highway transportation sector, increasing fuel costs over the past decade have caused the operators of aircraft, particularly large turbine powered aircraft, to be very energy conscious. As such, much work has been done by the aircraft and engine manufacturers, the operators and the FAA to design for fuel efficiency and minimize fuel consumption during operation. This overview discussion, therefore, focuses on the smaller aircraft which are more common in Alaska. The areas for improvement in the aviation sector divide into five energy technology categories: 1. Engines 2. Alternate Fuels 3. Weight 4. Aerodynamics 5. Pilot Awareness and Education Engines For the smaller aircraft in Alaska, the standard, air-cooled gasoline piston engine is predominant. There are several technological modifications to this engine which represent a potential for fuel efficiency improvement. 661 The first concept involves electronically controlling the fuel/air “mixture to achieve lean combustion and utilize improved fuel injection and timing to prevent detonation (engine knock). The _ potential energy savings is estimated at 7-10%. However, significant development work including engine redesign is necessary. As with any proposed aircraft modification, FAA certification would be required. This can be a lengthy and costly process. The second concept involves turbocharging the engine. The lower engine speeds and higher manifold pressures with a turbocharged engine can achieve 10% fuel savings. Propeller development and modifications to the engine cooling would also be required to obtain the full benefit of turbocharging. ‘ The third concept uses automatic mixture controls of engine fuel/air _ratio to achieve proper engine operation and minimize fuel consump- tion. These hydromechanical controls are not as sophisticated as electronic fuel control systems mentioned in the lean burn concept above. Automatic mixture controls could save significant fuel; however, questions remain on unit cost, reliability, and certification. Alternate Fuels For small gasoline fueled aircraft, the available alternate fuels of interest also. require alternate engines: diesel fuel for an advanced diesel engine and turbine fuel for a turboshaft engine. Advanced diesels, highly. turbocharged, have reported levels of specific fuel consumption 10-15% lower than conventional aircraft engines (5) but cost and weight penalties may preclude diesels from aircraft applications. Work on applications is currently in the development phase. For the turbine fuel, the turboshaft engine would replace conventional piston engines in the under 500 hp size class. The fuel consumption of a small gas turbine is high and the cost is also much higher than a conventional piston engine. The major advantages would be the use of cheaper, more easily produced turbine fuel, increased reliability, and increased operating life. This is currently a mature product for larger aircraft applications. Weight Reduction in aircraft weight reduces lift required and, therefore, thrust and fuel required. . The major current development effort to reduce airframe weight focuses on composite materials. Aircraft designers predict that 25% of the empty weight could be saved.(5) In addition to high cost of composites, there are technical issues to be resolved concerning inspection, testing, joining methods, new manufacturing processes, and reliability. 662 Aerodynamics The aircraft manufacturing industry believes there is at least 10% reduction in drag possible by developing more aerodynamic designs.(5) Tradeoffs between cost, payload, passenger comfort, and fuel consumption will impact whether significant further drag reduction programs are undertaken by the industry. Pilot Awareness and Education Pilot fuel management is the area of largest potential fuel savings in small aircraft. If pilots can be made aware of the savings possible when using proper fuel management procedures, and be educated in properly utilizing the appropriate procedures for the aircraft they operate and the missions they fly, there exists the possibility of reducing fuel consumption with little or no investment in hardware or modifications. (5) Most small general aviation aircraft incorporate a manual mixture control which enables the pilot to adjust the fuel/air ratio entering the engine from the carburetor or fuel injection system as differing flight and power conditions require. At cruise power conditions, excessively rich mixtures result in lost power and wasted fuel. Very lean mixtures cause a loss of power and under certain high power conditions, serious engine overheating. Because many pilots ere not cognizant of proper leaning techniques and are fearful of damaging engines by excessive leaning, it is suspected that overly rich mixtures are often used and fuel is wasted. A study conducted by Arthur D. Little, Inc., estimated that 4% energy savings in the general aviation fleet could be realized by adoption of better fuel leaning techniques by pilots.(5) Fuel savings can also be achieved by reduced cruise power and air speed. A 10% reduction in cruise power results in approximately a 10% fuel savings, while the resulting increase in total trip time due to the lower air speed probably will be nearly unnoticed. Ten percent reduced cruise power would save approximately 6% of the total general aviation gasoline consumption if universally adopted by the general aviation small aircraft fleet. However, since safety and reliability should continue to be aviation's primary concerns, it is important that programs on fuel saving techniques be taught in such a way that the fine safety record of general aviation be maintained or improved, and that increased conservation in no way degrades aircraft performance or reliability. It is not possible to identify the actual fuel savings potential of such a program; however, combined savings of approximately 10% of the aviation gasoline consumed yearly appear possible using modified pilot techniques. 663 27.4 Summary of Energy Technologies for the Rail Sector As an industry, North American railroads have been concerned about reducing fuel consumption and a number of areas of potential improvements have been identified. In many cases, the necessary technical developments have been carried out and the _ solution implemented. The broad categories for these energy technologies are: Engines Transmissions Wheels/Track Interaction Alternate Fuels Weight . Aerodynamics Systems oe ee Noonrwnr Engines The Alaska -Railroad uses conventional diesel-electric locomotives. Energy saving options for these conventional locomotives fall into five categories; thermal efficiency improvements, improved cold weather starting, improved control of dynamic braking engine speed, multi unit throttle control, and waste heat recovery. Thermal Efficiency Improvements Designers and manufacturers of new locomotives continue to strive for incremental improvements in the overall thermal efficiencies which result in minor decreases in comparable fuel consumption relative to locomotives of previous design. This saving can be realized as new fuel. efficient locomotives are purchased. . ’ Improved Cold Weather Starting In Alaska, as in other cold climates, diesel locomotives are left running continuously because in the 250 gallon cooling system there is no antifreeze, and a diesel engine is difficult to start in cold weather. Significant amounts of fuel are used in overnight idling; as much as 20% reduction in fuel used could be achieved if this idling is eliminated. Some starting aids are now commercially available. One aid consists of a small engine which runs continuously to keep the cooling water warm. The cost effectiveness of the initial investment (about $11,000 and one week downtime per locomotive) must be evaluated against the specific fuel savings available. This would appear to be an energy technology particularly useful in the Alaskan environment. 664 as Improved Control of Dynamic Braking Engine Speed This control, option is available in new locomotives and offers about one percent reduction in fuel use in the mountainous terrain where the Alaska Railroad operates.. As the train descends a slope and dynamic braking occurs, improved control of the. engine speed results in a lower rate of fuel combustion. Multi Unit Throttle Control In a train running multiple locomotives, power output required for climbing grades may be excessive for other operating conditions. The multiple unit throttle control "fuel saver" reduces the power output on some of the locomotives according to the power demand. An 8% reduction in fuel use has been estimated in mountainous terrain. The Alaska Railroad currently has 80% of its locomotives equipped with the "fuel saver," the other 20% being primarily in yard service. Waste Heat Recovery An energy technology which is currently in the developmental stage for locomotives is the recovery of the energy in the generated waste heat through the use of a Rankine cycle. As the energy in’ the waste heat is recovered and fed back into the main powertrain, as much as 8% reduction in fuel consumption .can be achieved. ‘This represents a medium to long term energy saving technology which in the future might be included in modern diesel locomotive design and manufacture. Transmissions On locomotives, significant amounts. (up to 25%) of fuel could be saved if the electric transmissions. were replaced with more efficient hydraulic transmissions. In addition to these fuel savings potential during locomotive travel, fuel used during some long idle times can be saved as the hydraulic transmissions have quick start-up capabilities. These transmissions are currently in use in Europe. Although the - efficiency advantage would be fully expected in Alaska, the cold climate here exacerbates the reliability and maintenance problems of the hydraulic unit. Wheels/Track Interaction This category includes the whole wheel/track interaction as a system. There are two important aspects of this interaction: (1) reducing rolling resistance of. the freight car which can reduce energy required, and (2) achieving increased adhesion for locomotives would allow increased load capacity reducing energy usage per net ton. 665 Reduction in rolling resistance can be achieved by the use of a radial truck. This design dllows movement between the two parallel axles of the truck under a freight car so that the wheels can accurately follow curves in the track. This reduces energy required to negotiate the curves as well as wear on the wheels. Several systems are commercially available including one that can be used to retrofit existing equipment. The use of low resistance bearing seals can achieve fuel savings on the order of 6%. This involves replacing standard roller bearing seals in freight cars with low resistance seals. Although these are commercially available, the reliability of the seals in cold weather has not been evaluated. Another method for reducing resistance on curves is the use of a commercially available curve lubricator. The lubricator reduces fuel use as well as. reducing wear on the wheel and axles. Some curve lubricators are currently used on the Alaska Railroad. The option available for improvement of the track contribution to the wheel track interaction is an improved track structure using such new products as welded rail, concrete ties, and geotextiles. These are commercially available and keep the track structure in better form thereby reducing rolling resistance. In Alaska, welded track is used inside all eleven ARR tunnels, and engineering fabrics are used under the ties to minimize frost heave problems. Currently, the Alaska Railroad is evaluating some test track using concrete ties. The greatest potential for fuel savings, however, from improved track structure is on flat terrain since in mountainous areas the additional fuel required for climbing grades is much greater than that required for overcoming rolling resistance. Improved designs for adhesion between the wheel and rail are currently available in new locomotives such as General Motors' SD-450. The improved adhesion allows larger cargo loads to be hauled by the same locomotive which can save almost 5% of fuel consumed per net ton of freight. Improved adhesion is particularly . applicable in Alaska's mountainous terrain and may be a worthwhile option in future . locomotive purchases. Alternate Fuels The Alaska Railroad currently runs on diesel fuel at about $1/gallon. The alternate fuels to be considered are off specification diesel fuel, nondiesel fuel or hybrid fuel as well as electricity for an electrified railroad. Alternate Fuels for Conventional Locomotives Off specification diesel fuel, nondiesel fuel or hybrid fuel such as a blend of No. 2 diesel with 10% heavy oil, have potential for reducing 666 "3B > the fuel costs for locomotive operations. Currently, Southwest ' Research Institute is conducting research on the potential usage and savings available from these alternate fuels for conventional locomotives. As yet, tests for using such fuels during cold weather have not been made. Rail Electrification Changing the source of energy from diesel to electricity results in several benefits to the overall railroad operation. e the railroad is no longer dependent on a source for diesel fuel; , e the required electricity can be generated by a variety of energy sources; e if hydroelectric power is available, there would be substantial saving in nonrenewable resources; e electric locomotives are more powerful and can withstand greater temporary overloads than diesels; e maintenance costs would be lower; e reliability would improve; and e quicker turnaround times are possible .because: there is no refueling. , Certain tradeoffs must be made to achieve these benefits. Most importantly, a significant capital investment to electrify the track and purchase the locomotives. In Alaska, there are some particular conditions which would cause the potential electrification to be very costly: e high cost for the catenary and support structures because of the curviness of the track; and e high construction cost for transmission lines because of the potential of thermal heaving of poles from the ground. Weight A reduction in the empty weight of the freight cars results in a reduction in fuel consumption, about 10-13% fuel saved per net ton of delivered cargo. Light weight design passenger cars are also available. Commercially, manufacturers have achieved lighter weight freight cars through the use of aluminum and fiberglass. However, 667 the durability of aluminum in very cold climates is not yet well demonstrated. Although the relatively small size of the ARR may inhibit. its ability to place a large enough order to obtain the most favorable purchase price, this option represents an investment with good payback potential. Aerodynamics Analogous to the aerodynamic discussion for highway. vehicles, streamlining the design of locomotives and freight cars can reduce air resistance and thereby reduce overall fuel consumption. For high speed routes over flat terrain, the use of more aerodynamic equipment has good fuel savings potential, whereas, in mountainous terrain, the potential for savings is marginal. Currently, manufacturers are doing development work on further aerodynamic improvements to railroad equipment. Systems Two key areas of current interest for saving energy in the railroad system are (1) fueling process for locomotives, and (2) -applications of solar energy. Fueling Process When locomotives are fueled almost 10% of the fuel is either spilled or otherwise unaccounted for. Automatic shut off nozzles, currently in use by the Central Pacific and Burlington Northern Railroads, can save as much as 7% of the fuel used. This represents an inexpensive method for reducing fuel use. In addition, a spilled fuel purifier can be used to recover spilled fuel. It is also in current use by the Norfolk and Western, and reported to recover 4-10% of the wasted fuel. Applications of Solar Energy Two applications of solar energy which are commercially available are for railroad signals and for caboose strobe lights. The Alaska Railroad is currently using solar energy for some of its remote signal locations. The usage eliminates the need to transmit energy to the remote location. Significant cost savings result from the use of such signals as the need for transmission lines is eliminated. , . One U.S. line, the Southern Railroad, is currently experimenting with a solar powered caboose strobe light. The energy supply must be reliable in this application to meet federal standards for this safety device. But only marginal savings are expected from this approach. 668 ¢ pore, oomey nd 27.5 Summary of Energy Technologies for Marine Sector Commercial marine activity in Alaska involves coastal shipping, ferries, off-shore oil support shipping, ‘industrial water movements and fishing. The energy technologies of interest are addressed below in three general categories: 1. engines (power plant) 2. alternate fuels 3., operations Engines The basic engine for marine use is the diesel engine. For the large diesel engine, efficiency improvements over current baseline would result from more fully utilizing the energy produced either through a waste heat recovery system or a bottoming cycle. Recovered waste heat could be used in the onboard heating, ventilation air conditioning (HVAC) system for deck machinery or electric power generation. These systems are commercially available; however, they -would only be cost effective systems for larger power plants. The other method for utilizing the energy produced, thereby improving specific fuel consumption, requires a bottoming cycle to extract energy from the waste heat stream and gear it back into the main powertrain. The bottoming cycle is still under development for marine applications. As for the waste heat recovery system, this would only be cost effective on larger marine power plants. Alternate Fuels The marine activity in Alaska in the largest part are dependent upon diesel propulsion. The smaller engines use marine diesel fuel, whereas the medium speed and larger diesels increasingly use a large proportion of heavy fuels. During the cold seasons, fuel handling becomes a distinct problem due to increasing viscosities. Kerosenes or lighter fractions often are used as diluents during cold weather. Only small craft utilize gasoline fueled engines. Coal and coal/oil slurries are now being evaluated for marine use. There has been little success to date in adapting diesel engines to such fuels and their main potential lies in steam plants. However, the nature and cost of such fuel installations appears to make them most applicable to deepwater ships. The advantage would be the utilization of a nonpetroleum based fuel, but the use of coal fired boilers would have a significant impact on the design of the ship, operating routes, and potential fuel availability. In the short to medium term, the potential for use of coal fired boilers will probably have little effect on Alaska's marine sector, in part because the majority of ships and craft in Alaska are in the small size category. 669 Operation - Slow Steaming (Fuel Consumption Optimization) The theory of "slow steaming" operation consists of minimizing the overall voyage costs.* This involves trading off. reduction in ship speed, which reduces fuel consumption and cost, with lengthening the overall voyage time, which increases cost for other ship operations (abor, insurance, maintenance, consumables, etc.) and cargo inventory costs. Calculating the appropriate desired ship speed and desired voyage time for each particular voyage and situation. is not ‘only possible, but also cost effective, and applicable to most operations with-the use of an on-board minicomputer. The specific climate in Alaska, along with the hydrographic conditions, may override optimum operating conditions based on fuel consumption and operating cost minimization alone. These phenomena include high tidal ranges, strong currents, ice conditions,. and extreme weather conditions. However, within the boundaries of these constraints the voyage cost optimization principle is applicable. In Alaska, the high cost of shipping operations make this analysis particularly valuable to such vessels as -coastal cargo ships, tug/barge combinations, offshore oil industry service ships, and fishing vessels. 27.6 Summary of Potential Improvements in Alaska's Transportation . Infrastructure as an Energy Technology As stated at the beginning of this chapter, the most critical transportation issue in Alaska is the availability of adequate, reasonable cost transportation. Conserving energy or improving the efficiencies in the transportation area is a secondary although important concern. , . The State of Alaska's Department of Transportation and Public Facilities (DOT/PF), has the charter of planning, designing, constructing, operating, and maintaining aviation, highway and marine facilities. This charter is in support of the state's overall transportation goal: to provide each resident with a safe, reliable, convenient and efficient transportation system that is socially, culturally, economically, and environmentally acceptable. In the future, it is possible that the responsibility for operating and maintaining the Alaska Railroad may also be within the DOT/PF. * Derived from the early days when the steam engine was dominant for marine use, the term "steaming" is used figuratively here to mean the rate at which the engine is operated. 670 —~ In support of this charter and goal, the DOT/PF has developed a six-year transportation improvement - program, ~ published in March 1982.(4) This program includes a comprehensive plan for land highways and marine highways system improvements, aviation related improvements, ports and harbor improvements and state transportation facility needs. Although improvements to the _ transportation infrastructure cannot be strictly categorized as an "energy technology ,".they are discussed briefly here in the overview, for two key reasons. The first reason is that the improvement programs. are critical to addressing Alaska's foremost transportation issue of the availability of adequate transportation. The second is that the improvements will, in fact, increase the overall efficiency of the transportation sectors consumption of energy: by improving the operating efficiencies; by enabling the use of some more advanced fuel efficient vehicle technologies; and by improving fuel efficiency of current vehicles being utilized. The overall six-year program has a projected cumulative total expenditure level of $5.1 billion.(4) By sector, this funding breaks down to 60% for highways, 20% for aviation, 9% for ports and harbors, 7% for marine, and 4% for | facilities. 27.6.1 Highway The highway system includes a network of rural highways, urban street systems, and remote intra-village roads. In the total state route system there are 8,500. miles of roadway.(7) In 1981, of the 260 thousand registered automobiles and trucks in Alaska, approximately 85% were registered in the Railbelt, 1I$ in the Southeast, and the remaining 4% in the Bush and the Interior. (2) . When contrasted with the U.S. total, some differences in the automobile and truck fleets are attributable to operating in the State of Alaska. For personal transportation, a larger proportion of these vehicles are light duty trucks rather than automobiles. There is also a higher proportion of four-wheel drive vehicles of light duty trucks, approximately 35% are front wheel drive in Alaska versus 12-15% in the Lower 48.(9) For the trucking industry, a larger portion of the heavy trucking (Class 7 and 8) is involved in operation within the local range (60% versus 50% for the U.S. total). Actual average miles travelled per year per vehicle in Alaska is only about half of the U.S. average, -which has an important effect on the economic tradeoffs for applicable technologies. (9 ) Overall, specific uses of and products carried by trucks in Alaska have. some significant differences compared to the U.S. average. Forty-five percent of Class 7 and 8 trucks in Alaska are used in .the construction business versus 18% in the total U.S. Another 8% of the heavy truck fleet carries petroleum in Alaska, compared with only 4% which carry petroleum in the United States (9) : 671 Much of Alaska's paved highway system currently exists in fair, poor, or very poor condition. Significant maintenance and reconstruction is planned to bring these highways to current design and. safety standards. Some highways were originally built at a substandard level before the adequate availability of funds from oil revenues. These require rebuilding to withstand the originally unanticipated truck traffic stemming from construction of the Trans-Alaska Pipeline. Other highways, routed and built on a. rush basis during World War II by military construction crews, require extensive rebuilding. In addition, there is a lack .of necessary or adequate intra-village road systems. The $3.06 billion designated for highway improvements over the Six Year Transportation Program will focus on (1) major rehabilitation to reconstruct highways to original design standards, and (2) upgrading highways to acceptable design standards which reasonably meet anticipated demand. Recent studies (1) have shown that travel over badly deteriorated pavement can increase fuel consumption by as much as 10%. The Alaska DOT/PF currently has a research program ongoing to develop a system to objectively measure the net energy impact of highway improvements. Upgrading of the pavement and design of the highway system will improve its overall energy efficiency. Other programs aimed at energy conservation in the highway sector are: ‘ e developing public transit operations (eight major locations have bus transit programs to date); e implementing. energy sensitive traffic control and flow; ® presenting public information/education programs on ‘energy conscious driving and use of bicycle or walking; and e operating a van pool program. 27.6.2 Aviation Alaska's aviation system consists of a network of 1,073 state, federal and privately owned facilities including land airports, seaplane bases, and heliports. There are 7,227 general aviation aircraft (Dec. 1981). This represents approximately one plane for every 55 people. The State of Alaska owns 411 of these facilities and actively maintains about 250 of these. For the DOT/PF, the focus for the aviation transportation area is the improvement of existing facilities. The international and regional hub airports do meet the essential applicable design standards. However, many rural airport facilities 672 are either below standards or simply unimproved landing strips. As many rural areas are dependent on air transport improvements needs to be made in order to insure safe, reliable operations. The $1.05 billion anticipated for aviation mode improvements will emphasize: e@ upgrading secondary airport runways to the minimum standard of 3,000' x 100', and improving their lighting and terminal facilities; and e expanding/enhancing the existing infrastructure at trunk airports needed due to increased economic activity. Over the long term, runway improvement in rural areas could affect the energy consumption efficiency of some of the small aircraft operation. Improved, extended runways could allow the use of the more modern type of aircraft which have more fuel efficient engines and better aerodynamics. Currently, the lack of a tail wheel on the modern planes as well as the. higher stall speeds make landing in many rural airports a dangerous or impossible proposition. 27.6.3 Ports and Harbors Ports and. harbor facilities play an important role in the economic development and, in fact, survival of most Alaskan _ coastal communities. In addition, Alaska's ports are their main link to world markets. Historically, waterborne commerce was the basis of Alaska's transportation system but aviation is gradually displacing much of the waterborne traffic. However, a significant amount of goods transport will continue to be water based. In addition, the fishing industry benefits from improved harbor | facilities as does the construction industry through transporation of materials to remote locations. The improvements to the ports and harbors existing infrastructure account for 9% of ‘the total six year budget, $474 million. In the central region, improvements will focus on _ small’ boat harbor improvement; increasing number of moorage spaces and dock construction. In the interior, due to the weather and physical constraints, the improvements focus on erosion control and beach protection for maintaining safe, accessible, and navigable waterways and coastal shores. In the Southeast, small boat harbor improvements and increase in moorage facilities will take place at major population centers. 27.6.4 Marine Highway System The Alaska Marine Highway System is an extension of the land based highway system. It is a state owned ferry .system which connects isolated areas in Southeastern Alaska, Southcentral Alaska, Kodiak, 673 and the Aleutian Islands with central and interior Alaska, -British Columbia, and Washington state. Two separate ferry systems, one for Southeastern Alaska, and one for Southwestern Alaska, make up Alaska Marine Highway. The Southeastern system has four main line vessels and three feeder ferries, servicing seventeen ports. The Southwestern system has two ferries which service twelve ports of eall. A total of $340 million is budgeted for improvements to the Marine Highway System. The three major project areas are: e the improvement, development, and refurbishing of shore facilities; e the major maintenance work necessary for the _ reliable operation of the system vessels; and . e improvements to the functioning of entire operations such as an improved reservation system. 27.6.5 Railroad The Alaska Railroad has been an operating agency of the U.S. Department of Transportation; it is currently being sold to the State of Alaska. The ARR consists of one single track main line of 419 miles and six branches totalling an additional 115 miles of track. (6) The ARR owns 63 locomotives, 1,702 freight cars, and 45 passenger cars. The Alaska Railroad is currently being purchased by the State of Alaska. The agency or department which will have the responsibility and authority to operate the railroad has not been named. In a July 1981 report prepared for the Alaska DOT/PF entitled "An Assessment of the Alaskan Railroad Ownership and Operational Alternatives," one of the final recommendations was as_ follows. "There is a need to establish a multi-year capital improvement program and programming process for the Alaska Railroad. Such a capital improvement program should include the modernization of the rolling stock."(6) Within this capital improvement program are undoubtedly areas for improvement which would address the area of energy consumption and conservation by the railroad operation. 674 10. il. 12. REFERENCES Staff Report Transportation Energy Conservation Program, System Evaluation Section, DOT/PF. State of Alaska Long Term Energy Plan, Department of Commerce and Economic Development, Division of Energy and Power Development, 1982 Report. G. Kulp and M.C. Holcomb, Transportation Energy Data Book, Sixth Edition, Oak Ridge National Laboratory, for Office of Vehicle and Engine Research and Development, Department of Energy, 1982. Six Year Transportation Improvement Program 1983-1988, State of Alaska, Department of Transportation and Public Facilities, March 1982. R.F. Topping, W.P. Lee, "Fuel Conservation in Recreational Aircraft," to Office of Environment and Energy, Federal Aviation Administration, Washington, D.C., September 1979. "An Assessment of the Alaskan Railroad, Ownership and Operational Alternatives," Bivens & Associates, Inc., and University of Alaska for Alaska Department of Transportation and Public Facilities, July 1981. "Annual Traffic Volume Report," 1980 Alaska Highways, Volume 1, State of Alaska, Department of Transportation and Public Facilities. Ward's Automotive Yearbook, 1982, Ward's Communication, Inc., Detroit, Michigan. Department of Commerce, Bureau of Census, 1977 Census of Transportation, Sept. 1979, Truck Inventory and Use Survey, Alaska. Hurter, D.A., et al," A Study of Technological Improvements in Automobile Fuel Consumption," for U.S. Department of Transportation, December 1974. Colello, R.G., et al, "Light Duty Motor Vehicle Fuel Injection Trends," August 1982. U.S. Department of Transportation, New Trucks Save Fuel, 1979. 675 13. 14, Hurter, D.A., et al," A Study of Technological Improvements to Optimize Truck Configuration for Fuel Economy," April 197 -. Colello, R.G., and P.G. Gott, "Automobile Body and Chassis Design Trends," July 1981. 676 ie 28.0 AIRCRAFT ENGINE MODIFICATION 28.1 Introduction and Summary 28.1.1 Technical Overview Energy savings are possible in Alaska's general aviation sector using techniques such as modification of aircraft and engines to provide better fuel economy. This chapter concentrates on smaller aircraft with gasoline-fueled piston engines. Large turbine-powered aircraft are not included since the economic advantages of instituting fuel saving techniques in these aircraft have been sufficient for some time to provide compelling incentives for the commercial operators of this equipment to have already implemented them on their own initiative. Various modifications to the standard, air-cooled aircraft piston engine are possible to achieve fuel savings. A turbocharged engine operating at reduced engine speed and increased manifold pressure may offer a 10% savings in specific fuel consumption. Lean combustion concepts with advanced fuel injection and timing may yield a 7% to 10% savings. Combining these two concepts may provide 10% to 20% savings if accompanying advances in engine cooling and a means for minimizing detonation can be implemented. Advanced diesels have reported levels of specific fuel consumption 10-15% lower than conventional aircraft engines although cost and weight penalties may preclude diesels from aircraft applications. Automatic mixture controls could save fuel if questions involving unit cost, reliability, and certification can be satisfactorily resolved. In the category of airframe modifications, experts predict a possible savings of at least 25% in aircraft empty weight by using composites, leading to large fuel savings. Lesser, but still significant fuel savings (10% to 20%), is also possible by drag reduction on current airframe designs. Realistically, fuel saving modifications do not lend themselves to retrofit in small single and twin engine aircraft since the modification cost could very well exceed the value of the aircraft, and the expected dollar savings over the remaining life of the aircraft would be significantly less than the cost of the conversion. Pilot awareness and education programs offer a large potential savings because, unlike the previous hardware modifications which apply mainly to new aircraft, this approach offers the opportunity for implementing fuel conservation techniques in the total fleet by reaching all pilots of all aircraft. Replacement of old aircraft with new models could also save fuel. Many of the technical improvements mentioned above are beginning to be implemented in the latest aircraft designs. Turbocharged aircraft 677 TABLE 28.1 REGISTERED GENERAL AVIATION AIRCRAFT IN ALASKA AS OF DECEMBER 31, 1982 * PISTON Single Engine 1-3 place 2928 4+ place 3536 ae 2 Engine | 1-6 = place 180 . 7+ place 166 ey 3+ Engine 7 | TOTAL 6817: . ~ ‘TURBOPROP Single Engine 10 2 Engine : : 1-12 place 23 : . 13+ place 16 nen 3+ Engine 3 TOTAL 52 TURBOJET 2 Engine 1-12 place 5 ‘ : ° 13+ place 1 . 3+ Engine 1 TOTAL «7 ROTORCRAFT Piston 86 Turbine 223 TOTAL 309 os OTHER 42 TOTAL 7227* on Source: Reference l. ee “6565 (or $9%) of these aircraft were in active use in 1981. 678 , : with retractable landing gear, designed for minimum drag and maximum speed and economy, could be used to a greater extent in Alaska if improved airports with paved runways were to become more available. The number of traditional Bush planes required for transportation into remote areas could be reduced. 28.1.2 Alaskan Perspective As of December 31, 1981, there were 7227 general aviation aircraft registered in Alaska. The U.S. Federal Aviation Administration (FAA) (1) estimates that 6465 were active, flying a total of 1,171,000 flight hours in 1981, resulting in an average utilization of 181 hours per aircraft. This hours-per-aircraft total is slightly lower than the national average of 191 hours which contradicts the generally held assumption that Alaskan aircraft are more heavily used than their counterparts in the Lower 48. This is probably because the severe Alaska weather seriously restricts winter flying; yet heavy activity in the summer, though not sufficient to raise yearly hours above the national average, gives the impression of very high general aviation activity. Table 28.1 gives a breakdown of the registered aircraft in Alaska by type. General aviation is defined as all aircraft in the civil air fleet except those that have been issued a certificate of public convenience to provide scheduled air transportation over specified routes and are used by large aircraft commercial operators. General aviation, therefore, offers such varied services as air taxi, air cargo, industrial, agricultural, business, personal, instructional, research, patrol, and sport flying. Referring again to Table 28.1, 6817 (94%) of the aircraft registered are piston powered, and 6644 (92%) are single engine or small (less than 7 seat) twin engine aircraft. The fuel consumption estimate for these aircraft is given in Table 28.2. An estimated 10.8 million total gallons of aviation gasoline were consumed in Alaska in 1981 by small piston-powered, general aviation aircraft. Therefore, each 10% reduction in aviation gasoline consumption would result in a savings in fuel costs of slightly over $2 million.* These savings would have to be balanced against the cost of implementing fuel economy programs to determine the net financial benefit to the state and its aviators. *Based on a quoted December 1982 retail price of $1.88 per gallon for 100LL aviation gasoline. 679 FUEL CONSUMPTION Average Hourly Fuel Consumption** Single Engine 8.1 (1-3 seats) Single Engine 11.0 (4+ seats) Twin Engine 26.3 (1-6 seats) Total Fuel Consumed TABLE 28.2 * ESTIMATE FOR SMALL Number of Active ke Aircraft 2606 3147 160 PISTON AIRCRAFT Average Hours Total Fuel Flown Consumption (gal) 181 3,820,000 181 6,270,000 181 760,000 10,850,000 * All single engine and multi-engine with less than 7 seats. RK Source: Reference (1). RAK Estimated by multiplying the total number of registered aircraft by the ratio of active to registered aircraft for the state (89%). 686 28.1.3 Significance of the Technology Advanced energy saving engine and airframe concepts, although technically feasible, are not currently. available in small general - aviation aircraft. Cost, complexity and regulatory requirements have slowed their development. Also, the general aviation industry has experienced a large drop in shipments due to the current recession (from nearly 18,000 aircraft shipped in 1978 to approximately 4,000 shipped in 1982).. This drastic downturn in the industry has further hindered the development of new concepts. It is probably unrealistic for the State of Alaska to attempt to accelerate these development efforts. As new engine and airframe designs are commercialized they. will become available for use in Alaska. However, these advanced aircraft, as well as the models available today utilizing efficiency improvements such as retractable landing gear, low drag aerodynamics, and turbocharging, will require paved runways and possibly better. navigational and radio aids in order to see increased use in Alaska. The number of older, less efficient Bush and float planes could decrease if. better airports become available. Programs aimed at teaching efficient fuel management to pilots have the potential for- most near term fuel’ savings with minimum investment. An estimated saving of 10% of the State's consumption of aviation gasoline is possible. The FAA already has avenues established (such as accident prevention seminars and flight instructor clinics) which could.be used to reach and: educate pilots. With the FAA's. cooperation, the State of Alaska. could initiate fuel saving programs for general aviation. that would have universal appeal. 28.2 Description of Technology 28.2.1. Operating Principles The fundamental parameters. affecting basic aircraft piston engine efficiency are: e compression ratio e fuel/air mixture ratio e manifold pressure e engine friction The effect of compression ratio and fuel/air mixture ratio on engine performance is expressed in the following idealized engine efficiency relation: . Ideal Engine Efficiency = 1-(1/CR) *} CR = compression ratio 681 k =c_/c. = Heat Capacity Ratio (increases.as the fuel/air ratio decreases) From this relation, it can be seen that the engine efficiency improves as the compression ratio (CR) increases and the fuel/air mixture ratio decreases (k increases). The compression ratio can be increased until the high pressure and temperature produced in the cylinder causes the mixture of air and fuel to auto-ignite, and detonation, or knock, occurs. The fuel/air ratio can be decreased, or leaned, until the mixture can no longer be combusted because of the scarcity of fuel. There are numerous hardware concepts that have been developed, primarily by the auto industry, to achieve improved fuel economy using lean combustion. With lean mixtures, less combustion energy goes into non-useful heating of the working gas and dissociation of the gaseous constituents in the cylinder, leading to improved engine efficiency. Stratified charge and hydrogen enrichment are _ two concepts designed to lower the lean limit by providing a richer combustible mixture near the spark plug while maintaining an overall lean cylinder mixture. Manifold pressure also has a significant effect on engine efficiency. At low manifold pressure (closed throttle) the reduced density of the intake air requires the engine to consume a significant amount of its available power to draw the combustion air into the cylinder. This negative work, called pumping loss, is minimized by operating the engine at high manifold pressure (open throttle). Also, at high manifold pressure a greater flow rate of air is drawn into the cylinder and the engine torque increases (at constant engine speed). Engine friction hurts efficiency since friction decreases useful engine output. Engine friction is generally proportional to the engine speed. For constant horsepower output, lower engine speed reduces the proportion of friction to useful combustion energy, and the engine efficiency increases. ' For these reasons, maximum engine efficiency occurs at operating points characterized by high torque and low engine speed (RPM). For a standard engine the maximum manifold pressure is the surrounding air pressure (which decreases with altitude for an aircraft engine). Higher manifold pressures can be achieved with turbocharging where exhaust gases are used to drive a turbine pump which compresses incoming air. Turbocharging can be used to increase the manifold pressure up to a practical limit determined by detonation, or knock limits. The high efficiency engine design parameters discussed above (high manifold pressure, high torque, and lean combustion) are embodied in the diesel engine. It is operated at the highest manifold pressure 682 available (atmospheric or higher if turbocharged), at extremely lean mixtures (one-tenth of the overall cylinder fuel/air ratio of the most lean spark ignited engine), and at lower speeds than spark ignition engines. Therefore, the diesel engine offers marked efficiency gains and has been investigated as a possible aircraft power plant. Table 28.3 shows a typical usage profile for a general aviation airplane. Less than 13% of the trip fuel is consumed during taxi, takeoff and landing. Cruise mode (55% to 75% of maximum rated power) is the principal fuel consuming mode of operation. Table 28.4 shows that during cruise the minimum achievable fuel consumption of production aircraft engines is equivalent to reported values for automotive diesel engines and substantially better than conventional spark engines. Therefore, significant fuel economy gains at the typical 65% rated power, 6000-foot cruise condition cannot, be achieved by adaptation of present automotive spark or diesel engine design to aircraft applications. Rather, design changes specifically related to aircraft engine operation during typical aircraft missions are necessary. Table 28.4 assumes that the aircraft engine is operating at the fuel/air ratio that provides highest efficiency. To achieve this, non-turbocharged (normally aspirated) aircraft piston engines require leaning (decreasing) of the engine fuel/air ratio at cruise power at altitude for two reasons: (1) to maintain efficient engine operation because of the decreased density of the ambient air; and (2) because aircraft engines require rich mixtures and excess fuel at high power settings (above 85% of maximum power) to provide necessary engine cooling. Therefore, aircraft have a manual mixture control in the cockpit which enables the pilot to adjust the fuel/air ratio entering the engine from the carburetor or fuel injection system as differing flight and power conditions require. Best fuel economy is achieved at a mixture slightly lean of that which produces peak exhaust gas temperature (EGT). If the mixture is leaned further, the probability of an ignitable mixture residing in the zone around the spark plugs is reduced. At the "lean limit" irregular ignition occurs since the spark is unable to ignite the combustible mixture reliably. Therefore, correct mixture adjustment is important for both economy and engine life. Because many pilots are not cognizant of proper leaning techniques and are fearful of damaging engines by excessive leaning, overly rich mixtures are often used and fuel is wasted. 28.2.2 Technical Characteristics Increased Manifold Pressure Increasing manifold pressure (which produces increased torque) while reducing engine speed to maintain a given power level is one means of increasing fuel efficiency. In a naturally aspirated engine the 683 TABLE 28.3 FUEL CONSUMPTION BY AIRCRAFT OPERATIONAL % of Max Power Landing, Take Off 40%, 100% Cruise @ 6000 ft. 652% Source: Reference 3 TABLE 28.4 BASELINE ENGINES Production Aircraft Engines Naturally Turbo Aspirated Charged (Ref 3) (Ref 4) Horsepower (2200-2700 RPM) — 285 210 Compression Ratio 8.5 7.5 Fuel Consumption @ Cruise .43 +42 ‘ibs fuel’ , “hp-hr 684 MODE % of Trip Fuel 13% 87% Automotive Engines Naturally Aspirated Diesel (Ref 2) 200 150 7.5 20.5 a) 43 ro i \ manifold pressure is limited by the surrounding air pressure and therefore available power decreases with altitude. At normal cruise conditions, full throttle (maximum available manifold pressure) is usually required to attain desirable cruise power. Increases in manifold pressure for these engines is therefore not possible. However, a turbocharger can pressurize the manifold above ambient and yield potential for fuel savings. To operate with higher manifold pressures at constant horsepower output, a propeller is required that will convert the engine power to propulsive thrust at low rotational speed. Also, engine cooling requires additional attention to maintain the proper valve and cylinder head temperatures. With a low speed propeller and improved engine cooling such as exhaust port liners, it is conceivable that a turbocharged engine could be raised to high enough manifold pressure to achieve 65% power with the engine speed reduced to the range of 1,200-2,000 RPM. A 10% energy savings could be ultimately achieved. This modification would require substantial revision of the propeller and/or the addition of an engine speed _ reduction. Limited discussions with a propeller manufacturer indicated that there is no obvious technical limitation to developing a variable pitch propeller that could run efficiently over a large engine speed range and allow operation below 2,000 RPM. However, development cost would be high. Lean Mixtures Brake specific fuel consumption (BSFC) is a measure of the fuel consumed by an engine (in pounds) per unit of engine output (brake horsepower-hour). Published test results on production and prototype lean burn engines indicate that a brake specific fuel consumption of .40 lb fuel/hp-hr at 65% power (cruise mode) is the best achieved today in a high speed (over 1,000 RPM) lightweight engine (Table 28.5). This represents about a 7% to 10% energy saving over reported conventional aircraft engine data.(3) Substantial engine redesign would be required to achieve this saving, including improved fuel injection and variable ignition timing, a feature not presently used on standard aircraft engines. Diesel Engines Most current automotive diesels do not offer fuel economy savings over the baseline aircraft engine. Work on special prototype highly turbocharged aircraft diesel engines resulted in a reported .38 BSFC in the cruise mode.(5) This is a notable achievement since it comes close to the .36 BSFC of a very large stationary turbocharged diesel engine operating at the highest compression ratio reported in the literature. The ‘aircraft diesel concept involves a secondary combustion chamber in the exhaust gas ahead of the turbine of the 685 Standard Aircraft Engine (Ref 3) TABLE 28.5 ALTERNATIVE ENGINE CONCEPTS Lean Burn (Hydrogen Enrichment) (Ref 5) Avco Rotary (Ref 5) Stratified Charge (Ref 2) High Speed (Lightweight) Diesel (Ref 2) Hyperbar Diesel Co. (Ref 5) RPM % POWER 2300 65 LEAN BURN 65 2400 75 2000 65 DIESEL ENGINES 2000 65 686 BEST BSFC 43 +43 -41 -40 43 - 38 % SAVING IN CRUISE(65% POWER) Baseline 12% turbocharger to provide additional turbocharging capability. It is, however, in the preliminary stages of development and may not see commercialization for some time. Turboprop Conversions Engine conversion from piston to turboprop is available for certain Bush type airplanes. For instances, a Cessna 185 can be outfitted with a 420 horsepower Allison 250 turbine engine, replacing its standard 300 horsepower piston engine. Turbine engines offer performance advantages, long life, the ability to use cheaper turbine fuel, and reliability (especially in cold climates where starting piston engines can be a problem). However, fuel consumption of small turbines is quite a bit higher than piston engines. The Allison engine in the Cessna 185 uses 50% more fuel than the engine it replaces. Cost is also a factor. The above conversion costs $150,000; a new piston engine costs less than $40,000 and a rebuilt engine costs considerably less. Therefore, turbine conversions neither save fuel nor offer reasonable economics for most operators. Automatic Mixture Controls An increase in cruise fuel consumption of 15% to 20% results when the mixture control is improperly set to the full rich position by the pilot while at cruise power. Therefore, automotic mixture controls could be beneficial to achieve minimum fuel consumption. The automatic mixture control schedules proper engine fuel/air ratio for the flight condition, and therefore relieves the pilot of the need to manually lean the engine mixture with each power and altitude change. Automatic mixture controls are not new but are used only in a few expensive, sophisticated aircraft. A reliable, simple control, inexpensive enough to be installed in smaller aircraft, could save significant amounts of fuel as well as protecting engines from possible damage due to incorrect leaning. Summary of Engine Modifications Turbocharged engines operating at reduced engine speed and increased manifold pressure and torque may offer a 10% fuel saving. Improved engine cooling may be required in order to operate these turbocharged engines at high manifold pressures; and a low speed propeller (or drive train speed reduction) will be needed to provide required power at cruise. Lean combustion with advanced fuel injection and timing may yield a 7% to 10% fuel savings. If advances in engine cooling and means for minimizing detonation can be implemented, the lean burn engine could be combined with the high manifold pressure-low speed concept, possibly achieving an overall savings in the range of 10% to 20%. 687 requirements and be subjected to a tortuous series of tests in order to obtain certification. The relatively low production rate of new aircraft limits the rate at which new designs can be introduced at a cost that provides a reasonable payback to operators. The government, primarily NASA and FAA, provides some _ support; however, the current oil glut has taken the urgency out of these programs. Substantial investments in research and development would accelerate the introduction of more fuel efficient aircraft; however, the payback in terms of energy saved per dollar invested would be very low. The new technologies are also generally not suitable for retrofit on older aircraft. With few exceptions, the cost of the hardware and the certification that must be secured from the FAA for modifications far exceeds the potential cost savings over the remaining life of the modified . aircraft. For example, we estimate replacing the conventional engine in a Bush type single engine aircraft with an advanced turbocharged engine with lean burn and low speed propeller to eventually cost $50,000 (Table 28.6) if the technology is developed. Assuming the aircraft is operated for. 200 hours per year(Table 28.7), the cost savings, assuming 1982 fuel prices, is $827, leading to a payback period of 60 years. Even if the operator waited until his engine needed replacement, the cost differential for substitution of the improved engine would not provide payback in a reasonable period. Assuming the added cost of substitution to be $20,000, the payback would be 24 years. The useful life of an aircraft engine is about 1400 hours, or seven years at 200 hours usage per year. Therefore, simple payback offers no incentive for the operator to make the switch. Unlike the turboprop, the conversion would not offer other advantages such as increased power, better reliability, startability, ete. Therefore, universal re-engining of older Bush aircraft is not really-a viable proposal. Replacement of aircraft of older, less efficient design that are in heavy commercial use (significantly higher than the 181 hour yearly average) with the more efficient aircraft available is an alternative that should be further investigated. Commercial operators would probably have sufficient financial incentive to replace these, aircraft if the operating environment was suitable to using more fuel efficient aircraft incorporating improved aerodynamics, retractable landing gear, and turbocharging. Such aircraft would require good quality paved runways and possibly better navigational/radio aids. Because of the Alaskan environment, there will continue to be a need for some number of relatively inefficient Bush and float planes. The only practical fuel saving alternative for these aircraft is to train pilots to operate them in the most efficient manner possible. The same is true for privately-owned aircraft that are flown relatively few 688 Advanced diesels have reported fuel consumption levels amounting to a 10% to 15% fuel savings over conventional aircraft engines. The cost and weight penalty of incorporating a highly turbocharged diesel into a light aircraft may, however, eliminate the diesel from further development. Turboprop conversions, while providing increased performance and reliability, do not offer fuel savings. Installation and maintenance costs are also very high making the overall payoff questionable. Automatic mixture controls may provide some fuel savings. However, if properly trained, the pilot can perform the same function without the need for additional on-board hardware. 28.2.3 Environmental Issues As an energy conservation concept, the modifications discussed in this chapter will affect the environment only to the extent that they reduce fuel use and lower total emissions to the atmosphere. 28.2.4 Commercial Status Some of the aircraft improvements discussed above will gradually be implemented into new aircraft as the industry, and _ possibly government agencies such as NASA, commit the necessary resources into developing and certifying them. It should be noted that the yearly production rate of general aviation is fairly low. The best year for production was 1978 when nearly 18,000 aircraft were shipped.* A production base this small makes it difficult to finance research and development programs for new concepts such as those listed above, especially when the design and certification of new aircraft models using established technologies is so costly. It is also unrealistic to expect that demand for a relatively small number of energy saving aircraft by a limited group such as Alaskan general aviation would be adequate incentive for the industry to accelerate its development efforts. The development costs will be high and can only be justified on a national (or international) level. 28.3 Economic Implications 28.3.1 Costs The costs of develop and certify the engine modifications discussed in Section 28.2 for introduction into general aviation will be high. Airworthy hardware must conform to a _ stringent set of FAA *The current recession has severely impacted the industry to the extent that shipments have dropped to around 4,000 aircraft in 1982. 689 7 4 ABLE 28.6 CAPITAL COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD CAPITAL COST Equipment and Materials Direct Labor TOTAL CAPITAT, INVESTMENT Advanced Turbocharged, lean-burn engine with low speed propeller (retrofit) Anchorage 1982 constant dollars Not applicable 200 operating hours 7 years 1 month $47,500 2,500 ‘$50,000 690 TABLE 28.7 OPERATING COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output OPERATING FACTOR Operating Costs VARIABLE COSTS Avgas, gallons TOTAL VARIABLE COSTS FIXED COSTS Capital Charges 9.5% TOTAL FIXED COSTS TOTAL ANNUAL OPERATING COSTS TOTAL ANNUAL OUTPUT TOTAL LIFE CYCLE COST Advanced turbocharged, lean-burn engine with low speed propeller (retrofit) Anchorage 1982 Not applicable 200 operating hours Not applicable Annual Unit Cost Consumption Annual Cost $1.88 2500 $4,700 $4,700 $4,750 $4,750 $9,450 200 hours $47.25/hour (versus $27.26/hour @ 2900 gal/yr without retrofit) 691 hours and whose owners could not afford, or economically justify, newer models, Seminars and written material could be used to teach good fuel management techniques to the pilots of these aircraft. 28.3.2 Socioeconomic Factors Because of strict aircraft maintenance requirements, a network of maintenance facilities with licensed personnel already exists in Alaska. If energy saving engine modifications were to be made to existing aircraft at an increasing rate, the demand on these facilities and personnel would create new employment opportunities for trained and licensed mechanics. The socioeconomic benefits of replacement of older aircraft with newer, more efficient models, however, would mainly be transferred out of state to areas in the Lower 48 where aircraft are produced. : Improved airports, which might foster introduction of fuel efficient models in Alaska, would provide small indirect benefits in terms of construction and operation jobs at these facilities. The Bush-type aircraft operated in large numbers in Alaska are not as fuel efficient as models currently available which have features such as retractable landing gear, turbocharging, etc. However, these aircraft are not capable of operating out of airfields with poor quality runways and lose efficiency when equipped with floats or skis. The construction of new airports with hard surfaced runways could bolster aviation and lead to the modernization of the general aviation fleet. The higher speed of the new aircraft and the expected increased activity would justify the installation of more radio and navigational aids. These improvements would lead to a demand for more skilled personnel in the workforce to man airports and operate and maintain equipment. 28.4 Impact 28.4.1 Effect on Overall Energy Supply and Use The small piston-powered, general aviation aircraft uses very little of the gasoline supply (about 1% in Alaska, 0.5% nationwide). However, manufacture of aviation gasoline (avgas) may affect total motor gasoline output to a greater extent than the percent of total production may imply. This suggests that conservation of aviation gasoline through more efficient use is an important means of easing the burden of producing this specialized product in existing refineries. Aviation gasoline is blended to a different set of quality specifications than automotive gasoline. The most significant of these quality differences is gasoline octane. In the case of 80 octane aviation gasoline, octane is similar or slightly lower than leaded regular automotive gasoline and should present no refining problems in meeting octane specifications. In the case of 100 octane and 100LL (iow lead) aviation gasoline, however, octane is as much as 10 to 15 octane numbers higher than premium automotive gasoline and is 692 extremely difficult to produce in the refinery because operating costs and energy requirements are significantly higher than for premium automotive gasoline. With current trends toward increased production of unleaded automotive. gasoline and increasing automotive octane requirements, high octane aviation gasoline will become increasingly more difficult and expensive to produce in the future. It is interesting to note that.the general aviation industry has been told by the refiners that 80 and 100 octane avgas will eventually be phased out of production so that one grade only, 100LL, will have to be used in all aircraft. Apparently refiners have opted to discontinue fuel types that are easier to produce for the sake of simplified distribution. This requirement has led manufacturers to modify engine designs that originally used 80 and 100 octane fuel, so that all piston aircraft now produced use 100LL fuel. In addition to octane, there are several other specifications that can present additional refining problems. All aviation gasoline is blended to lighter distillation, lower vapor pressure, lower sulfur and freeze point specifications than automotive gasoline. These specifications, coupled with the high octane requirement, result in aviation gasoline being blended primarily out of light high octane components such as aromatics, alkylate and isomerate which will be in greater demand in the future fot petrochemicals and gasoline blend components. All of the specifications: can limit refinery flexibility and capacity for producing aviation versus automotive gasoline and further increase refinery production costs and energy use. Therefore, conservation of avgas could have a positive impact on Alaska's energy outlook. Reduced distribution and storage requirements would also be of economic benefit to the state. 28.4.2 Future Trends Although some retrofits are possible, fuel saving engine modifications will gradually be introduced into: general aviation aircraft through new models. The rate will be determined by factors such as the price and availability of fuel, and the economic health of the general aviation industry. : : Another option which may impact fuel savings in a manner similar to engine improvements is modification of airframe components. This primarily involves drag and ‘weight reduction which allows engine power, and hence fuel consumption, to be lowered. Aircraft designers are looking toward composite materials to reduce weight in airframe structures, landing gear, and propeller blades. Experts predict a possible saving of at least 25% in empty weight by using composites.(6) However, before these materials can be introduced into general aviation, much more development work is necessary, Also, cost must drop sharply from current levels for composites to be practical in small aircraft. 693 In the area of drag reduction, one manufacturer believes that a 30% increase in speed (and miles per gallon) is available in current aircraft designs by drag reduction. Others estimate the possible savings to be nearer to 10%. These predictions represent a fuel savings in the range of 10% to 30%. Some of the improvements possible are semi-cosmetic (reduced parasitic and profile drag caused by airframe bumps, rivets, scoops, etc.), and others are non-cosmetic (new wing design, engine cowling design, flap design, incorporation of retractable landing gear into more aircraft designs, etc.). It is anticipated that tradeoffs involving first cost, payload, passenger .comfort, and fuel will determine the extent to which significant airframe modifications to reduce drag will be undertaken by the industry in the years ahead. A strategy with perhaps a more immediate fuel-saving payout in general aviation would be to educate pilots and increase their fuel-consciousness. A study conducted by Arthur D. Little, Inc., for the FAA estimated that 4% average energy savings in the general aviation fleet could be realized by adoption of better fuel leaning techniques by pilots.(7) Interviews with aircraft and engine manufacturers, owners and operators of large fleets of general aviation aircraft, certified flight instructors, and FAA personnel revealed the general consensus of opinion was that pilot fuel management is the area of largest potential fuel savings in small aircraft. Many of the pilots in general aviation today were trained at a time when fuel was inexpensive and fuel. management not very important. If pilots could be made aware of the savings possible when using proper fuel management procedures and be educated in utilizing the appropriate procedures for the aircraft they operate and the missions they fly, there exists the possibility of reducing fuel consumption significantly with little or no investment in hardware or modifications. Fuel savings can also be achieved by reduced cruise. power (and air speed) in much the same way as lowering highway speed limits for automobiles increases mileage. Small general aircraft are characteristically designed to cruise at 55% to 75% engine power. A 10% reduction in cruise power results in approximately a 10% fuel savings, while the resulting increase in total trip time due to the lower air speed probably will be nearly unnoticed. 10% reduced cruise power would result in a general aviation fleet yearly fuel savings of approximately 6%. Various programs are possible to increase pilot awareness of potential fuel savings and provide training for the implementation of fuel conservation techniques. An era of inexpensive fuel and misconceptions concerning proper engine operation have caused the average pilot to be less than fuel conscious; but current fuel prices of nearly two dollars a gallon have made fuel a significant portion of aircraft operating cost. In today's environment, therefore, a 694 conservation program may be well received. However, since safety and reliability should continue to be aviation's primary concerns, it is important that fuel saving techniques be taught in such a way that the fine safety record of general aviation would be maintained or improved, and that increased conservation in no way would degrade aircraft performance or reliability. Various fuel conservation programs aimed at the pilots of small general aviation aircraft could be initiated by the State of Alaska in cooperation with the FAA, aircraft manufacturers, and fixed based operators. It is not possible to identify the actual fuel savings potential of each program; however, combined’ savings of approximately 10% of the aviation gasoline consumed yearly appear possible using modified pilot techniques. As fuel prices increase and as older pilots give up flying, the aviation community will become more energy conscious and better fuel utilization will be achieved. Pilot training will emphasize fuel management and energy awareness. The FAA and other organizations will increase their effort in fuel conservation education. 695 REFERENCES Federal Aviation Administration, Census of U.S. Civil Aircraft, U.S. Department of Transportation, Washington, D.C., December 1981, Arthur D. Little, Inc., A Study of Technological Improvements in Automobile Fuel Consumption, Volume II: Comprehensive Discussion, U.S. Department of Transportation, Report No. DOT-TSC-OST-74-40-II, December, 1974. Rezy, B. J., et al., Concepts for Reducing Exhaust Emissions and Fuel Consumption of the Aircraft Piston Engine, Teledyne Continental Motors, SAE Paper 790605, 1979. Mooney Aircraft Corporation, M20K 1979 Pilot's Operating Handbook, Kerrville, Texas, 1978. Willis, E. A., General Aviation Internal Combustion Engine Research Programs at NASA-Lewis Research Center, NASA TM North, D. M., "Industry Seeks Lighter Aircraft Weight," Aviation Week and Space Technology, May 21, 1979. Federal Aviation Administration, Ener Conservation Potential of General Aviation Activity, U.S. Department of Transportation, Report FAA-EE-79-20, Washington, D.C., September 1979. 696 7» i on 1. - ko 29.0 RAILROAD ELECTRIFICATION 29.1 Introduction and Summary 29.1.1 Technical Overview Railroad electrification involves replacing the current fleet of diesel-electric locomotives by electric locomotives which draw power from wayside electric wires. Electrifying a railroad thus changes the source of energy from diesel to electricity. In Alaska, this means switching the Alaska Railroad (ARR) from petroleum based energy to natural gas, coal and, possibly in the future, hydro power. As shown in Figure 29.1, the electric power is transmitted to the locomotive from a contact wire supported by a structure called a catenary. The locomotive draws the power by contacting the catenary with what is called a pantograph. Modern catenaries and pantographs allow such power transmission at a very high speed (e.g., at more than 200 mph in a test run of the French Supertrain, Tres Grand Vitesse, or TGV). There are many advantages of electrifying a rail line. First, since electricity can be generated from a variety of energy sources the dependence of the railroad on diesel fuel is eliminated. Second, an electric locomotive generally has a higher horsepower than that of a comparable electric locomotive. Third, the electric locomotive is cheaper to maintain and it is more reliable and available. Fourth, since electric locomotives have overload capability built in, fewer of them are needed than diesel locomotives in terrains which require this temporary overload feature. Finally, since refueling is not required, the turnaround time of an electric locomotive is shorter. However, all these can be achieved only after a substantial capital investment is made to electrify the track and to acquire electric locomotives. Also, electrification may entail modifications to the bridges and tunnels along the route in order to provide sufficient clearance for the overhead electrical wires. Thus, it is difficult to draw informed conclusions regarding railroad electrification without first performing a site-specific analysis for the actual rail route in question. This chapter does not present such an analysis, but rather discusses the technical issues which would have to be examined in more detail. 29.1.2 Alaskan Perspective There is only one railroad in Alaska: the Alaska Railroad, which presents a unique set of opportunities and problems to electrification. The tremendous potential for hydroelectric power generation may make electricity cheaper than an equivalent amount of diesel fuel in powering the railroad. On the other hand, the severity of weather, the nature of terrain and the lack of accessibility may make the track electrification construction a very expensive project. For example: 697 FIGURE 29.1 Contact Wire *Pantograph THE PRIMARY COMPONENTS OF RAILROAD ELECTRIFICATION 698 e due to the high degree of curves on the track, the catenary cost will be more than that on an average track. This is because more support structures (poles) will be required to provide the necessary curve in the catenary; e the transmission line to carry power from generating stations will be expensive to construct because of the terrain; and e the problem of poles heaving from the ground due to the repeated cycle of freezing and thawing will need to be addressed by special construction techniques. Also, some of the usual advantages of the electric locomotive will not be available in the Alaskan environment. For example, in a flat terrain, it is generally possible to replace a fleet of diesel locomotives (more accurately called diesel-electric locomotives, because the diesel engine actually operates a generator to produce electricity which in turn powers motors which run the locomotive) by a fewer number of higher horsepower electric locomotives. However, in Alaska's mountainous terrain, the number of locomotives required to haul a train is determined by the traction capability of locomotives and not by the horsepower. Since a modern diesel-electric locomotive provides about the same traction as an electric locomotive, the replacement may have to be one to one. Further, each locomotive will need to use two pantographs, the first one to remove ice from the contact wire and the other to draw power. Finally, the increased power requirement in a mountainous terrain means that heavier than normal power stations will be required. 29.1.3 Significance of the Technology There are no obvious technical problems to electrifying the Alaska Railroad. However, the expenses involved in electrifying the track and acquiring electric locomotives have to be carefully studied to determine the economic feasibility of the project. The following elements have to be considered in determining the cost of electrification: (1) e First Costs - catenary; - substation/switching station; - signalling system modification ; - communication system; - transmission line/taps; and 699 - eleetric locomotive cost, net of and credit for selling the existing diesels. e Recurring costs - electric energy; - electric locomotive maintenance; and - catenary maintenance. On the credit side will be the annual cost saving of diesel locomotive maintenance and diesel fuel. The latter amounts to about 3.5 million gallons per year, or less than one percent of Alaska's total transportation sector liquid fuel use. 29.2.1 Operating Principles Locomotive The first difference between diesel-electric and electric railroad operation is the locomotive. A diesel-electric locomotive carries its own motive power and fuel supply, while an electric locomotive derives its power from electricity generated at a power station. Instead of a diesel engine and alternator, an electric locomotive uses a transformer that steps down the line voltage to that acceptable to the power conditioning equipment. This power conditioning equipment, the traction motors, and other ancillary equipment resemble corresponding equipment on a_ diesel-electric locomotive. The performance characteristics of an electric locomotive are described . later in Section 29.2.2. Power Supply The second major difference between the diesel-electric operation and an electrified railroad is the power supply system. An electric locomotive receives power from a catenary suspended directly over a track. The locomotive picks up the current with a pantograph, which is an articulated structure mounted on top of the locomotive. The return current flows through the rails. The catenary is supplied power at 25 or 50 kV for substations spaced 20 or 40 miles apart, respectively. The selection of voltage is dependent on the clearances available. Although a higher voltage usually results in lower cost, greater clearance required in tunnels and bridges for adequate insulation may push its cost higher. Sometimes it may be possible to use both 25 kV and 50 kV system if dual voltage locomotives are used. In such a case, a 50 kV system will be used along most of the track, with 25 kV system used in sections with tunnels and bridges. 700 sm Catenary The catenary is the most important and expensive element in electrification. Therefore, it is described in. some detail in the following paragraphs. Additional information is contained in References (2,3,4,5). The generic name "catenary" is applied to systems which. have two or more wires. A two-wire system is commonly referred to as. simple eatenary, and a system which has three wires in vertical alignment is referred to as compound catenary. Other types of catenary take their names from particular features, for example: stitched, inclined, ‘or tangent chord. In stitched catenary, shown in Figure 29.2, an additional wire is clamped to the messenger a fraction of the span on either side of the supporting structure, and used to. support the contact wire under the structure. Contact wires are usually drawn from copper or bronze, although steel (or an aluminum-steel combination) has been used in the past and new developments are being evaluated.(2) A considerable variety exists in the size, material and stranding of the cables used for the messenger wires. The use of steel has largely been discounted in modern ac catenaries because of its high electrical reactance. Bronze is less economical than conductors made from a stranding of plain copper or copper and copper-clad steel which is now in common usage. In addition aluminum and aluminum-clad_ steel or steel reinforced aluminum conductors are in use. Aluminum cables are suitable for ground, rail return, or feeder wires as circumstances dictate. The contact wire in a catenary is supported by the messenger through droppers. The complete network of messenger and contact wire is energized. A typical dropper is shown in Figure 29.3.. For a single track electrification, the messenger and the contact wires are attached to the supporting poles by a cantilever such as that shown in Figure 29.4. Finally, a typical support structure for use is in a ‘tunnel is shown in Figure 29.5. Extremes in temperature, such as those found in Alaska, cause thermal expansion and contraction in the catenary messenger and contact wire. With a fixed dead-end catenary system in which the catenary is anchored at each end, thermal extremes can produce undesirable variations in contact wire height, resulting in pantograph bounce. Additionally, excessive tension can be developed in the fixed dead-end system at low temperatures requiring use of higher strength contact wires with lower conductivity and placing greater tension requirements on all supporting structures. 701 CATENARY ‘(MESSENGER } DROPPER .. CONTACT WIRE A SIMPLE CATENARY EQUIPMENT (LEWEL CONTACT) STITCH WIRE B STITCHED CATENARY ECU:PHENT (LEVEL CONTACT) AUXILIARY CATENARY € COMPOUND CATENARY EQU:PMENT (LEVEL CONTACT} Source: Reference (5). FIGURE 29.2 TYPES OF CATENARIES 702 aluminium. alloy sleeve aluminium conductor Steel -reinforced nylon sleeve 4mm dia. stainless- steel wire nylon sleeve rubber packer Source: Reference (4). FIGURE 29.3 A TYPICAL DROPPER 703 galvonised malleable. cast-iron fittings aluminium conductor steel-reinforced Insulators galvanised-stee! tubes stainiess- steel wire, 0-9m normal Source: Reference (4). FIGURE 29.4 A TYPICAL CANTILEVER TO SUPPORT MESSENGER AND CONTACT WIRES ' 245mm minimum buty!-rubber-covered glass-fibre insulating arm ingle or double contact wire adjusting screw Source: Reference (4). FIGURE 29.5 ATYPICAL RESTRICTED TUNNEL SUPPORT 704 This problem is addressed by a constant tension (balanced weight) system. This is considered to be the only system capable of meeting the commutation requirements of electric locomotives over a wide temperature range. In the constant tension system, each full tension section is fixed at mid-span against longitudinal movement. To minimize variations in catenary height caused by thermal expansion and contraction, a constant tension is maintained on the catenary by a series of pulleys and counterweights at either end of each tension length. Catenary support brackets at each structure have adequate movement to permit horizontal dislocation due to expansion and contraction. The catenary may be divided, by electrical section breaks, into lengths which extend to as much as 20 to 40 miles or more. These section breaks, which can be connected to one another by switching stations, are required because adjacent sections are fed from different electrical phases to balance the load on the transmission system. The existence of these breaks also eases the problems of maintenance by making it possible to isolate a catenary substation on a_ section without affecting the rest of the system. The use of more than one electric voltage (25 kV and 50 kV) would also require presence of such sections. Supporting Structure The supporting structure of a catenary is designed to withstand the loads imposed by not only the catenary but also by feeders, return conductors, ground wires or signal power wires which may be installed. Circuits sensitive to electrical interference are located as far as possible from the power conductors, preferably on separate pole lines or underground. Allowance must be made for wind and ice loading on the wires and poles, and for the increase in the radial forces due to wire tensions where the track is curved. This is particularly important in a track system with many curves such as that of the Alaska Railroad. Poles are checked for deflection resulting from changing loads. A limit is usually imposed on such deflections, since this affects the maximum pole spacing. Another important consideration in Alaska is frost heaving of poles. Special construction techniques may be required to address that problem. There are many variations in the types and construction of supporting structures. The widespread use of wood poles for utility distribution lines has generated extensive data on performance and application, but they often need to be guyed in order to offset their tendency to bend and creep. In addition, natural timber pole is increasing in cost at a greater rate than its competitors. Alternatives as laminated and built-up glued sections are available. H-Section steel poles are commonly used especially where long life and low 705 maintenance are required. Galvanizing is preferred to _ initial painting. Weathering steels are finding extensive use in catenary supports. Reinforced concrete poles have been used, some initial designs had an inadequate cover of porous concrete and fell into some disfavor. Under suitable economic conditions, the modern reinforced, prestressed, centrifugally-cast tubular pole may be successful. Multiple tracks are frequently wired by means of cross catenary construction having either side-guyed or self-supporting poles. Portal structures are an alternative. Comparative economics may show only marginal cost differences between these support systems and each situation should be critically examined with respect to foundations, structures and supporting hardware. Pantograph The trend to higher system voltage with correspondingly lower currents has enabled the strength and weight of the pantograph to be reduced, largely because the required lineal contact between wire and pantograph is less. At the same time, the move to higher speeds has increased the tendency to create oscillations in the catenary. For train speeds at which the frequency of the forces on the catenary coincides with a natural frequency of the system. In these circumstances, large oscillation may develop. (6,7) Because the catenary system has relatively little natural damping it is found in practice that viscous velocity damping, applied to the pantograph,. will extract sufficient energy from the system to enable satisfactory current collection to be obtained. This is particularly true for multiple pantograph consists when pantographs other than the leading one move onto a wire which is already vibrating. The shape and size of the pantograph can have considerable bearing on the costs of electrification since the wider the pantograph the greater may be the spacing between the structures which support the catenary. As an isolated railroad, the Alaska Railroad could select a wide pantograph at relatively little extra cost, limited only by the maximum width admissible through tunnels or under overhead structures within its own track system. The contact surface of the pantograph can be of steel, copper or metalized carbon. In practice, the latter is found to give very long life and is least abrasive to the contact wire. Signal and Communication System Signaling and communication systems designed and built for diesel operation are almost invariably not satisfactory for use under electrified operations. The overhead open wiring used for communications and signal interconnections is adversely affected by 706 the catenary system. The electrostatic field will produce an undesirably and usually unacceptably high potential in these conductors. The electromagnetic field produced by the traction current flowing through the catenary and return paths induces noise levels sufficiently high to make signal operation unreliable if not impossible. This situation is particularly aggravated by the thyristor control systems used in modern high-powered commercial frequency locomotives. The current drawn by these units is very rich in harmonics, making effective filtering extremely difficult. As the principal return path for the traction current is through the rails, a means must also be evolved to transfer this traction current between insulated rail sections. Correcting these problems is relatively straightforward, but it requires a major revision and reconstruction of the existing signaling systems. The effects of electrostatic and electromagnetic fields can be greatly reduced by use of properly shielded cables which are buried in shallow trenches adjacent to the rails. It is necessary to use AC track circuits operating at a frequency which is not a multiple of 60 Hz, so that frequency selection can be readily achieved greatly reducing the effect of traction current fundamental and harmonic frequencies. 97 or 100 Hz are those frequencies usually selected for operation on 60 Hz systems. Return traction current is transferred between insulated track sections by means of an impedance bond, is a transformer consisting of two identical windings, each center tapped. One winding is connected between the rails of one block and the other winding between the rails of the adjacent block. The center taps are connected together. The transformer has windings so arranged that equal traction currents flowing in each rail are self-concealing, and this current is transferred between the blocks through the center tap connection. Each transformer winding presents a high impedance to the 100 Hz signal voltage and does not act as a short circuit between the rails. Grade crossings generally use overlay track circuits. Another factor which must be considered in the costing of signal modifications is the somewhat shorter block length which can be reliably employed using AC track circuits. Shorter blocks are necessary because of the higher track reactance occurring with these higher signal carrier frequencies. This problem is corrected by using shorter sections where long blocks are required and combining these sections to appear as a single block. These are known as cut sections. 29.2.2 Technical Characteristics A conventional diesel-electric locomotive has a low-speed tractive effort limited both by its adhesion and traction motor current limit, with the adhesion usually being the principal limitation. Its high speed operation is limited by the horsepower output capability of the 707 diesel engine. Significant short-term overload capability does not exist in this type of configuration by virtue of its basic design. The tractive effort control systems on these units have historically remained relatively simple in concept, and for this reason individual wheel-slip control has not been utilized to any great extent. Improved control systems, which are currently being developed and installed, permit higher low-speed tractive effort as well as reduced dependence upon rail sanding at higher speeds. Therefore, one must avoid attaching too much importance to the high adhesion achieved by -electric locomotives, as is already being achieved in modern diesel electric locomotives. The modern thyristor-controlled electric locomotive has substantially different performance characteristics.(1) Because it does not carry its prime mover, it is possible to design this unit with substantially higher horsepower without exceeding acceptable axle loads. While a typical 3,000-hp diesel-electric unit has a ratio of 6.5 hp per 1,000 lbs, the latest European electric locomotives are reaching levels as high as 40 hp per 1,000 lbs. Present United States designs are in the range of 14-22 hp per 1,000 Ibs. With improved individual wheel-slip control which is readily achieved on electric units, it is possible to operate in the range of 25% adhesion. The Swiss built’ type Ae6/6 locomotive is actually dispatched at 28% adhesion without sand. These tests have shown that starting adhesions on the most sophisticated designs of electric locomotives in the range of 30-35% are possible under excellent rail conditions. With the higher adhesion available on electric units, it is possible to utilize the overload capacity which can be designed into the locomotive. By proper sizing of transformer, thyristor bridges, and traction motors it is possible to achieve a one-hour overload rating of 5-10% above the continuous rating. This permits more rapid acceleration of a train from a stop and the climbing of shorter grades at higher speed without resorting to excess motive power for the longer continuous ‘sectors of a given run. Most recently built diesel locomotives are in the 3,000 hp range and of either 4- or 6-axle configuration. Units of this type deliver approximately 2500 rail horsepower. The smallest electric locomotive being built or considered in the United States at the present time has a minimum capability of 5100 rail horsepower continuously with a one-hour overload potential of 5400-5600 rail horsepower. To utilize this horsepower, the units are necessarily electromotive of 6-axle design. An experimental unit, the GM10B is being built by the Electromotive Division of General Motors to produce 8500 rail horsepower continuously in a 6-axle configuration and an even larger unit in the 11,000 rail horsepower range has been suggested by the Swedish manufacturer ASEA. : 708 Because of the mountainous terrain of Alaska, the tractive effort of a locomotive is more of a limiting factor than the horsepower. Since a modern diesel-electric locomotive can provide as much traction as an electric locomotive, the replacement of the diesel locomotives may have to be on a one to one basis. 29.2.3 Environmental Issues The environmental impact of electrification is in three areas: atmospheric emissions, noise, and _ aesthetics. Although engine exhaust will be eliminated, central station emissions will increase if the burden is to be carried instead by fossil fuel. Somewhat less centrally fired fuel would be required than diesel fuel (on a Btu basis), since large-scale power plants are more efficient than mobile diesels. However, the emissions would be concentrated in the area of the power plant, and their severity would be a function of the fuel burned. If clean burning natural gas or hydroelectric power were to be a major portion of the power generating mix, the net overall atmospheric benefit of railroad electrification would be decidedly favorable. If the fuel were coal, the impact might be unfavorable but would depend on the degree of stack gas cleanup employed. The electric locomotives are _ significantly quieter compared to the diesel-electric locomotives, but the overhead catenary structures of an electrified track are an aesthetic problem. 29.2.4 Commercial Status In the United States, a very limited demand for electric locomotives exists. For this reason, it has not been economically appropriate to make substantial investments in new tooling or radical designs. The General Electric E60C and the General Motors GM6C (Figure 29.6) are essentially diesel-electric locomotive frames with trucks and axle-hung traction motors similar to those used throughout the country.(8) Each truck is supplied from a single thyristor bridge set. Torque is controlled on individual motors. In addition to the basic economics, there is a sound technical reason why a typical diesel-electric type locomotive frame is used. In Europe, as discussed above, train dispatches are invariably much lighter than in the United States. U.S. power requirements are much more severe, and the basic European design as implemented today would not be satisfactory for U.S. freight operations. The European locomotive design philosophy is quite different from that in the United States. Probably, primarily because of the nature of the traffic, it has been customary for European trains to be considerably lighter and shorter than those in the United States. Also, mountain grades reaching 3.5% are regularly encountered. The trend is, then, to run more smaller trains at relatively close intervals. For this type of service lighter high horsepower 709 Source: Reference (9). FIGURE 29.6 GM10B Demonstrator TYPICAL ELECTRIC LOCOMOTIVES 710 Amtrak E60CP Locomotive FIGURE 29.6 = (Continued) ad etl TABLE 29.1 ELECTRIC LOCOMOTIVES AVAILABLE 1 cmnswine «eke ity, «SEAL MRO anemia? toe Continuous Rail HP 5,100 5,568 8,450 11,000 5,400 6,450 4,560 7,745 1-Hr Rail HP 5,600 6,100° 8,900° 12,150 6,000 6,780 4,830 7 Tractive Effort, Continuous, Ibs 77,000 88,000 99,000 117,000 39,900 48,180 34,800 46,084 Tractive Effort, 1-Hr., Ibs 80,000 92,000 127,000* 150,000° 47,840 52,500 37,000 = Tractive Effort, Start, Ibs 120,000 126,000 137,500 177,000° 72,800 70,400 61,700 35,000 Running Adhesion, % 22 . 25 32 30 28 30 22 18 Starting Adhesion, % 33 35 35 34 30 40 36 = Locomotive Weight, Ibs 360,000 365,000 394,500 520,000 205,700 176,000 169,000 256,000 Wheel Arrangement c-c cc B-B-B D-D c-c B-B B-B cc Wheel Diameter, inches 42 42 51 - 51 50 49 51 45 Catenary Volts, kV 25/50 25/50 11-25 25/50 15 15 15 4 Catenary Frequency, Hz 60 60 25/60 60 16-2/3 16-2/3 16-2/3 25 Control Type Thyristor Thyristor Thyristor Thyristor AC Thyristor Thyristor Thyristor Continuous Bridge No. 3 step Yes Yes Yes No, LTC Yes Yes Yes Motors per Control 3 . 1 1 1 1 2 1 _ Anti-Slip Control Field shunt Differential Presductor® Presductor™ Field shunt Field shunt Presductor® - Wheel-slip "Arthur D. Little, Inc., estimates. — Not specified. Notes: | 1, The numbers given below are design estimates. 2. The Re 3and Rc 4 are similar but geared for passenger service. The Re 4 is adapted for 11 kV, 25 Hz operation. Source: Reference 8 © v c a a locomotives are economical and practical. To obtain the necessary tractive effort with these lighter locomotives, typically 20 metric tons per axle, it is necessary to adopt more complex kinematics and control than is necessary on the heavier U.S. diesel-electric designs. In the SBB (Swiss Federal Railways-Schweizerische Bundesbahnen) applications, the Swiss have developed superior truck kinematics to provide uniform weight distribution among the several locomotive axles. As the SBB locomotives are all ac operated, only limited individual motor anti-slip controls can be used. Work has been done, however, on the later rectifier and thyristor designs to provide a more advanced anti-slip control system, and this combination of systems provides remarkably high adhesions. The SBB regularly dispatches their AC locomotives at 28% running adhesion without sand. In Sweden, the trend has also been towards lighter high horsepower units. West Germany, France, and the United Kingdom are all following this same practice. Locomotives with specific power in the range from 28-40 hp/1000 lbs are now available in the latest designs. This compares with the GE E60C, a U.S. built electric locomotive presently in production, at 14.2 hp/1000 lbs. This must also be compared with a typical U.S.-built diesel-electric locomotive which has a specific power in the range of 6-8 hp/1000 lbs. In the Swedish designs, those produced by ASEA, a high level of sophistication has been reached in the anti-slip control systems. With individual thyristor bridges for each traction motor or pair of traction motors, it is possible to control wheel slip before it becomes serious and properly limit and apportion torques to the various traction motors. Table 29.1 gives the basic information on several locomotives now available or in concept. This table is far from inclusive, but it does represent a good cross section of the available equipment with particular emphasis on U.S.-built units and a comparison to several available European electric locomotives. There are two major manufacturers of catenaries: Ohio Brass, Mansfield, Ohio; and BICC, London, England. The major designers of catenaries are: Gibbs & Hill, New York; Electrotrak, Union, New Jersey; and International Engineering Co. (IECO). 29.3 Economic Implications 29.3.1 Costs Estimating the cost of electrification requires a detailed study of the track to be electrified, the labor costs in the region and the energy costs. For this project, such an extensive study was not carried out. Thus, although Arthur D. Little, Inc. has performed several 713 studies of this nature,(1,8,11) the estimates provided in this section are based largely on general experience with electrification projects (12,13) and may not be accurate for the Alaska Railroad. Estimating the cost of railroad electrification involved the following assumptions: e The system will be 50 kV, 60 Hz in most places, with 25 kV in sections with many bridges and/or tunnels. e The yards are not electrified. Thus, the mileage to be electrified is as follows: (14) 478.3 miles of mainline track 47.5 miles of branch line track 525.8 total route miles. 39.6 miles of sidings, spurs & passing 565.4 total track miles. e The transmission line is 230 kV. Catenary Cost The cost of catenary is estimated to be $75,000-146,000 per track mile. For Alaska, it is more likely to be the upper bound because of @ more poles on curves, and e problem with pole uprouting. Thus, an estimated $140,000 per track mile was used, totalling $79.1 million for the 565.4 track miles of the Alaska Railroad. Substation Cost For a 230 kV, 60 MVA substation, the cost is estimated to be $2 million. Since one substation is required every 40 miles, the total number of substations will be 14, and the total cost $28.0 million. Switching Station The cost of switching stations is estimated to be $9,000 per route mile. Thus, the total.estimated cost is $4.7 million. Signaling System The cost of modifying the signaling system on a single track with- sidings is estimated to be $75,000-100,000 per route-mile, or $52.5 million overall. 714 Communications At $5,000 a route mile, the communication costs are estimated to be $2.6 million. Transmission Line For a 230 kV transmission line, the investment per mile is typically $110,000-250,000. We have assumed the cost in Alaska to be $200,000 per route mile (considering the difficulties in building a line in a terrain such as that in Alaska, even this number may be on a low side). The cost is then $105.0 million for the 525.8 route miles, assuming that the power plant is located along the route. ' Taps The taps on a 230 kV transmission line are expected to be $100,000 each. For a 14 substation system, this will translate to $1.4 million. Civil Reconstruction Cost . This cost includes the cost of modifying bridges and tunnels to accommodate electrification components. Arthur D. Little, Inc. has developed estimates for civil reconstruction in various categories: (8) e rail traffic under bridges, e rail traffic under other structures, e rail traffic through tunnels, e rail traffic over bridges, and e@ multiple span bridges. However, this cost cannot be estimated realistically without a thorough study of the route to be electrified. Locomotive Cost As mentioned earlier, an electric locomotive is usually of higher hp than an equivalent diesel locomotive. In a mountainous terrain, however, the tractive effort of a locomotive is more important than its power and a modern diesel-electric locomotive with slip control system can provide as much tractive effort as an electric locomotive. Thus, each existing diesel-electric locomotive will need to be replaced by an electric locomotive. This preliminary estimate may be too high for the following reasons: 715 TABLE 29.2 THE PRESENT FLEET OF ARR LOCOMOTIVES LOCOMOTIVES - DIESEL DESCRIPTION Characteristics No. Units Mfg. Horsepower _ Ton-weight {| Built/Rebuilt 1930 1981 GP-40-2, road EMD 3000 132 1975 6 6 GP-40-2, road EMD 3000 132 1976 5 5 GP-40-2, road EMD 3000 132 1978 4 4 . GP-40-2, (reblt GP35) EMD 3000 ~ 132 1964/1980 1 1 GP-35, road EMD 2500 132 1964 3 3 ‘E-8, passenger EMD 2400 158 1956/1974 -- 2 GP- 7, road EMD -1600 128 1951/1977 10 10 5 FP- 7, passenger EMD 1500 128 1951 13 13 ° RS- 3, switcher ALCO 1600 115 1953 12 12 RS- 1, switcher EMD 1600 | 115 1953 5 5 300 HP switcher GE 300 45 1944 4 4 Total locomotives * - Locomotives in~service 38, shopped for heavy repair 9, stored serviceable 13, ‘stored non-serviceable 4, leased 1. Source: Reference 14 e the present fleet of road locomotives (shown in Table 29.2 may not incorporate slip control and therefore have lower tractive effort than an equivalent electric locomotive; e the lower maintenance and turn around time of an electric locomotive means that fewer locomotives will be required to do the same job. If the current fleet of 21 road locomotives above 2400 hp is replaced by a fleet of an equal number of electric locomotives, each costing about $2 million, the cost of purchasing them will be $42 million. Also, if the 23 road and passenger locomotives in 1500-1600 hp range are replaced by an equal number of less powerful electric locomotives, each costing, say, $1 million, an additional $23 million will be required. The 21 switchers would not be replaced because it is -assumed that the yards will not be electrified. If one assumes that the 21 larger diesel-electric locomotives (above 2400 hp) can be sold for, say, $10 million, and the 23 smaller (and considerably older) diesel-electric locomotives (in 1500-1600 hp range) fetches $3 million, then the net cost of replacing the current fleet by that of electric locomotives will be: $42 + 23 - 13 = $52 million. The construction and locomotive costs are summarized in Table 29.3. Annual Costs The annual cost to maintain the catenary and substations is usually $1800-$3000 per track mile. Given Alaska's terrain difficulties, the number would probably be at the top of the range and would total $1.69 million per year. The cost to maintain an electric locomotive is usually 25%-40% the cost of maintaining a diesel-electric locomotive. Approximately $2.51 million was spent on ARR locomotive maintenance in 1981,(15) indicating that perhaps $0.75 million would be required to maintain an equivalent electric locomotive fleet. Energy purchase represents the other major recurring cost. Currently, the annual diesel requirement is about 3.5 million gallons. At $1.38 per gallon, this is about $4.83 million per year. As explained in Section 29.4, the railroad will need 39.2 million kWh when fully electrified. This would translate, at 9.2¢/kWh, to $3.60 million per year. Thus, about $1.23 million per year will be saved in energy cost. This is summarized in Table 29.4. This preliminary analysis indicates that electrification may not be a very good investment. A larger difference between the cost of electricity and diesel can, however, make the investment more attractive. Also, the load leveling benefit on central power 717 TABLE 29.3 CAPITAL COST SUMMARY TECHNOLOGY _ BASIS Location Year CAPACITY Input Output ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD CAPITAL COST Equipment and Materials Direct Labor Indirect Costs Home Office Costs Contingency TOTAL CAPITAL INVESTMENT 718 Railroad Electrification Railbelt 1982 constant dollars 565 miles Not applicable 400,000,000 revenue ton-miles/year 30 years 5 years $230,000,000 55,000,000 14,000,000 27,000,000 $326 ,000,000 TARLE 29.4 OPERATING COST SUMMARY TECHNOLOGY Railroad Electrification BASIS Location Railbelt Year 1982 constant dollars CAPACITY 565 miles Input Not applicable Output 400,000,000 revenue ton-miles/year OPERATING FACTOR Not applicable Annual Operating Costs Unit Cost Consumption Annual Cost VARIABLE COSTS Electricity (kwh) 9.2¢/kwh 39.2 million $3.60 million Direct Fuel Credit, gallons $1.38/gal 3.5 million (4.83) TOTAL VARIABLE COSTS ($1.23) million FIXED COSTS Maintenance, Labor and Materials Catenary, Substations 1.69 Electric Locomotive 0.75 Diesel-electric Locomotive (credit) (2-051). Capital Charges (9.5% of total capital investment) 30.90 TOTAL FIXED COSTS 30.83 million TOTAL ANNUAL OPERATING COSTS 29.60 million TOTAL ANNUAL OUTPUT 400 million ton-miles TOTAL LIFE CYCLE COST (net incremental cost) $0.074/ton-mile 719 generation, if most of the operations are conducted in off-peak hours, has to be taken into account in estimating a realistic charge for electricity. Finally, the cost differential will depend on whether a new electricity generation facility is required or not. . 29.3.2 Socioeconomic Factors Electrification of the Alaska Railroad will have several effects on the socioeconomic framework of the region. First, extensive construction labor will be required while the line is being electrified. This work will be temporary and require workers to build transmission line. feeder lines, substations, catenary, signaling and communications system. Additional labor will be needed for raising, changing bridges, tunnels and other structures. Once electrification is completed, some job categories associated with _ current railroad service will change. Additional labor(15) -- perhaps 30-40 people -- will be required to maintain track and perform electric repairs. All other jobs will decrease by a roughly equivalent number, since electric locomotives are simpler and run cleaner than diesel-electric locomotives and hence are easier to _ maintain. Similarly, the number of employees required to fuel diesel engines will decrease; their job will be to fuel only the yard locomotives. The lower operating cost and higher reliability of the electrified railroad may lead to additional business for the railroad. Also, lower shipping costs may make the bulk commodities produced in Alaska (such as coal) more price competitive. 29.4 Impact 29.4.1 Effect on Overall Energy Supply and Use Railroad electrification will create a shift from consumption of rail diesel fuel to that of electricity. Each gallon of diesel would -be replaced by 10.4 kWh of electricity at the power line,(16) taking into account the conversion efficiency of diesel-electric locomotives, the absence of idling, fuel spills and other losses incurred by diesel- electric and electric locomotives. Also, for every kWh purchased by a rail system, about 1.076 kWh has to be generated.(17) (This number will, however, be strongly dependent on where the electricity is generated relative to the railroad.) If the activity of the ARR were to remain at its 1981 level of about 400 million revenue ton-miles,(14) 3.57 million gallons of diesel fuel (15) would be replaced by 39.2 million kWh of electricity annually. Assuming a five-year period to complete the electrification process, the shift from diesel to electricity would probably take place in two stages of equal magnitude. Thus, of the 565.4 miles of track which can potentially be electrified, 280 miles will be. electrified in, say, 3 years, and the rest in two years after that. 720 7 29.4.2 Future Trends Railroad electrification, has failed to make any inroads in the U.S., while the rest of the railroads in the world are being rapidly electrified. Thus, there is practically no market for railroad electrification hardware in the U.S. and most of the technological advances are being made in Europe. Two such trends are: e the use of a 3-phase AC drive instead of a DC drive. This would save on commuters and other components associated with a DC drive. Brown Boveri has already built a number of such locomotives for the German railroad - Deutches Bundesbahn; and @ the change from phase back rectifier to chopper control. This will improve the power factor and reduce harmonic currents which interfere in communication. Other relevant factors relate to any electrification decision: e the shift of control of the railroad from the Federal to the state level; e the growth in population; e@ the desire to reduce dependence on conventional liquid fuels; and e the probable growth in the use of hydroelectric power in the Railbelt if the Susitua Project proceeds. 721 10. 11. REFERENCES Schwarm, E.G., Factors Affecting Railroad Electrification as Applied to Conrail, prepared by Arthur D. Little, Inc.,- for the United States Railway Association, Contract No. USRA-C-50124, November 1975. Bamford, T., et al., "Catenary State-of-Art," Bulletin 646, American Railway Engineering Association, 1974. Anon., "Railroad Electrification: Catenary System for a 25 kV Industrial-Frequency Electrified Railroad," Edison Electric Institute, New York, NY. : Goldring, A. C. and Suddard, A. D., "Development of Overhead Equipment for British Railways 50 Hz A-C Electrification Since 1960," Proc. IEEE Vol. 118, No. 8, August 1971. Thomet, M. A., "Catenary Tension and High Speed Power Collection," ASME/IEEE Joint Railroad Conference, 1976. Levy, S., et-al., "Railway Overhead Contact Systems, Catenary- Pantograph Dynamics for Power Collection at High Speeds," paper No. 68-RR-2, ASME/IEEE Joint Railroad Conference, Chicago, IL, 1968. Morris, R.B., "The Application of an Analogue. Computer to a Problem of Pantograph and Overhead Line Dynamics," Proc. Inst. Mech. Eng., Vol. 179, Pt. 1, No. 25, 1965. Schwarm, E.G., Engineering Cost Data Analysis for Railroad Electrification, prepared by Arthur D. Little, Inc., for the Department of Transportation, Transportation Systems: Center, Contract No. DOT-TSC-1156, October 1976. Baker, P.H., et al., "One Hundred Years of Progress in Railway Mechanical Engineering-Electriclocomotive Development," Chapter 5 in "Railway Mechanical Engineering, Car and _ Locomotive Design, A Century of Progress," ASME, New York, NY, 1979. Ephraim, M. Jr., "The AEM-7 - A New High Speed, Lightweight Electric Passenger Locomotive," ASME Paper 82-RT-7, 1982. Schwarm, E.G., Ener Costs for Railroad Electrification, prepared by Arthur D Little, Inc., for the Department of Transportation, Transportation Systems Center, DOT-TSC-1156, May 1977. 722 12. 13. 14, 15. 16. 17. Schwarm, E.G., "Comparative Costs for Railroad Electrification," presented to Transportation Research Board, Rail Electrification Systems Committee, July 1982. Schwarm, E.G., "Summary of Recent Estimates and Bids for Railroad Electrification Project Elements," presented to AREA Committee 33, Economic Subcommittee, March 1982. The Alaska Railroad, Fiscal Year 1981, Annual Report. Conversation with Mr. Michael Sudol, Chief Mechanical Officer, The Alaska Railroad. White, R.K., et al., Railroad Electrification in America's Future: An__ Assessment of Prospects and Impacts, Report No. DE-AT03-76CS-50115, Department of Energy, January 1980. Henderson, C., Ener Study of Rail Passenger Transportation, SAN-1176-T2 (Vol. ie prepared by SRI for os. Department of Energy, August 1979. 723 30.0 ALTERNATE FUELED HIGHWAY VEHICLES 30.1 Introduction and Summary 30.1.1 Technical Overview The vast majority of highway vehicles are currently powered by gasoline or diesel fuel, both refined from crude oil. Alternate fuels are defined as fuels from sources other than petroleum. Alternate fueled highway vehicles have been used to some extent in Europe, Japan, South America, and rural areas of the United States for many years. However, significant and widespread interest was not generated in the United States until the late 1960's. At that time, the impetus was the alleviation of air pollution through reduced exhaust emissions. Later, tight petroleum supplies caused by the Arab Oil Embargo of 1973 and soaring costs of petroleum fuels in the late 1970's fostered increased research, development, and production of alternate fueled highway vehicles. Alternate fuels used in highway vehicles are: e gasohol, which is a blend of 90% unleaded gasoline with 10% ethanol. e@ alcohol, which includes: - ethanol, presently produced by fermentation using sugar and grain or by catalytic hydration of ethylene - methanol, presently produced from natural gas (previously from coal). @ propane or liquefied petroleum gas (LPG), which is a by-product formed during the refining of natural gas or gasoline. e methane, which is used in the form of compressed natural gas (CNG) or liquefied natural gas (LNG). Although the term "methane" is often used interchangeably with "natural gas," methane can also be derived from coal, biomass, waste products, and other renewable sources; and e synthetic fuels, which is a generic term referring to any fuels produced from feedstocks other than crude oil and natural gas. The most significant liquid "synfuels" are shale oil, coal liquids, and alcohol from biomass. Several other alternate fuels are under investigation as long range candidates for highway vehicles. Among these are: 725 @ vegetable oils, which include sunflower seed, cotton- seed, soybean; and peanut oils. . e hydrogen, which can be used directly as a vehicle fuel or through dissociation of methanol into hydrogen and carbon monoxide with a system on board the vehicle. e solid fuel, which is primarily coal in the U.S., but can include any other combustible solid such as peat. The alternate fuels listed above can be used, with varying degrees of success, in the two types of engines that presently power highway vehicles: | e spark ignition (gasoline) engines @ compression ignition (diesel) engines Spark ignition engines power most cars and light trucks (vans, pick-ups, and utility vehicles). Compression ignition engines are installed in most heavy trucks (tractor-trailers) and buses. Medium trucks (straight-trucks, delivery and hi-cube vans, etc.) and school buses, are equipped with a significant percentage of both types of engines. The current driving force for commercial use of alternate fuels such as propane and methane can be lower cost. However, the benefit is limited to localities where fuel distribution facilities are in place. Thus, natural gas company fleets, propane delivery fleets, or farms where significant amounts of propane (LPG) are stored are generally the benefactors and users of alternate fuels. National, state, and city .governments have encouraged the more general use of alternate fuels to reduce dependence on petroleum imports, or to reduce air pollution. Some noted examples follow. e Brazil's Proalcohol Program was initiated in 1975 and continues today to encourage the use of ethanol powered vehicles. The benefits to Brazil include improved balance of payments, support of sugar prices for local agriculture, reversing the urban migration trends, decentralizing its industrial area, and aiding the manufacture of domestic industrial products. e California is encouraging the retrofit of old vehicles and purchase of new vehicles operating on methanol. The state allows a $1,000 state income tax credit for the conversion of existing vehicles to methanol. The California Energy Commission (CEC) subsidizes orders for methanol vehicles from state agencies, cities, and municipalities. The CEC also 726 provides interest free loans to gas stations that install methanol tanks and pumps. California believes the benefit to the state is reduced air pollution. e In the city of Tokyo, taxi cabs are powered by propane. The benefit to the city is improved air quality. 30.1.2 Alaskan Perspective After many years of alternate fuel use and technical investigation, petroleum-based fuels are generally the first choice for highway vehicles. Petroleum fuels are likely to remain dominant for the remainder of this century. However, as the world petroleum situation changes, alternate fuels will gradually emerge. In the long term, Alaska is in a position to benefit from the emergence of alternate fuels, because the state possesses abundant natural gas and coal resources, as well as oil. In the near term, alternate fuels such as compressed natural gas can yield local cost advantages. Gaseous fuels also offer vehicle performance benefits in cold climates. These benefits include: e improved cold starting, and e improved fuel efficiency during warm-up (vehicle operation during the first five to seven miles following a cold start). Two examples that demonstrate Alaska's potential to benefit from alternate fuel usage follow. The first example is described in a recent investigation by the U.S. Department of Transportation's Maritime Administration, which assessed the economic merits of Prudhoe Bay natural gas as a vehicle fuel on the U.S. West Coast. (1) The mode of delivery to market was assumed to be _ marine transportation. The analysis focused on the economics of Alaskan gas as a transportation fuel in the forms of methanol and compressed natural gas (CNG). The economic merits were assessed by comparing the market value against the cost at the point of delivery to West Coast markets. These results suggest that Alaskan natural gas could be profitably sold to West Coast fleet users in the Lower 48 states as well as in Japan. A second example is represented by the plans of Placer Amex, Inc. (San Francisco) to construct a 7,500 ton per day methanol plant at Beluga.(3) The plant would produce methanol from 8.7 million tons per year of subbituminous coal. The product would be shipped to the West Coast of the Lower 48 states to be sold as vehicle fuel. The U.S. Government's Synfuels Corporation turned down Pacer Amex's initial application for financial aid because of insufficient equity. However, the company was invited to submit a new proposal. 727 The future of alternate fueled highway vehicles will impact directly upon the Alaskan economy. Moreover, the present use of alternate fueled vehicles directly involves Alaska. An example is the progress made by the ENSTAR Natural Gas Co. in Anchorage: to save costs by powering cars, light trucks, and fork lifts with compressed natural gas (CNG). Seventy-seven vehicles, more than one-third of the company's fleet, have been converted to dual fuel systems. Gasoline provides the primary energy source, and CNG is the second source. Natural gas presently costs about one-half the gasoline equivalent (where 110 cubic feet of gas equals one gallon of gasoline) in Anchorage. ENSTAR is presently converting vehicles for several other local companies to CNG, as well as vehicles of the municipal agencies. ENSTAR also operates a CNG fueling station in Anchorage to service the public. : 30.1.3 Significance of the Technology Today's U.S. transportation sector consumes more than half the domestic petroleum supply. About the same ratio also exists for the State of Alaska, with more than half of the State's petroleum consumption dedicated to transportation. The nation's highway vehicle fleet is almost totally dependent on petroleum-based fuels. Since imports constitute nearly half of all petroleum consumed in the U.S., there is continuing interest to develop effective alternate fueled highway vehicles. Alcohols (methanol and ethanol) and gaseous fuels (methane and propane) are well suited to spark ignition (gasoline) engines because of their high octane ratings. These fuels are currently used in various fleets and locations in the world where gasoline costs are particularly burdensome or where air pollution problems are severe. These same alternative fuels are particularly poor for compression ignition (diesel) engines because of their very low cetane ratings. Gaseous fuels can offer better cold weather starting and cold engine efficiency than gasoline - an important benefit for Alaska. Alcohols have more severe cold starting problems and offer worse cold engine driveability than gasoline. Distribution systems and public fueling facilities are not generally available for the alternate fuels in the U.S. This limits their near term use to fleets, where central fueling stations can be provided. Generally, fleets decide to convert to an alternate fuel based on life cycle cost advantages. In the U.S., such opportunities have been generally limited to natural gas company fleets or propane delivery fleets. Natural gas (methane) and propane have several important uses other than providing fuel for transportation systems. Future availability of these fuels face similar uncertainties as petroleum-based fuels. 728 Therefore, in the near term, the gaseous fuels may be important alternatives to petroleum because of the cost advantage of natural gas and propane. However, in the long term alcohols will likely dominate as the petroleum alternative. Ethanol can be produced by fermentation of sugar, grain, or other agricultural products. It is readily produced in developing countries with high agricultural population and corresponding infrastructure. Currently, 25% of the cars sold by Volkswagen in Brazil have ethanol fueled engines. Paraguay has started a national alcohol program. Indonesia is following close behind. Methanol is more easily produced in industrialized countries from coal (also natural gas). Alcohol cars are fuel efficient and burn cleaner than gasoline, improving air quality. Ford Motor Company predicts that alcohol powered cars will represent 10 to 15% of the new car fleet by the year 2000.(4) Hydrogen is an excellent fuel, but will not emerge as a fuel for highway vehicles for the foreseeable future. The problems include storing hydrogen in the vehicle, the high costs of producing hydrogen fuel, and building a distribution system. The explosive nature of hydrogen presents a major safety problem. Concerning solid fuels, a U.S. Department of Energy sponsored investigation concluded that "highway transportation is not amenable to solid fuels utilization due to severe environmental packaging, control, and disposal problems."(5) Therefore, gaseous fuels and alcohol fuels are the significant alternatives to petroleum based fuels to the year 2000. However, alcohol fuels will likely emerge as the winner in the long term. 30.2 Description of the Technology 30.2.1 Operating Principles Gaseous fuels for highway vehicles are considered to be: e natural gas (primarily methane) - compressed natural gas (CNG) - liquefied natural gas (LNG) e liquefied petroleum gases (propane) e synthetic natural gas (SNG) which can be produced from coal, peat, and renewable sources such as municipal wastes and biomass (biogas). Here, synthetic natural gas is defined as a gas whose major component is methane. 729 Source: Reference (7). FIGURE 30.1 18 GALLON LNG TANK ASSEMBLY BY BEECH AIRCRAFT 730 Highway vehicles that operate on either natural gas or liquefied petroleum gas exhibit similar characteristics. Engine conversion equipment and vehicle modifications required to convert to either gas are very similar. Therefore, the operating principles and technical characteristics of natural gas and liquefied petroleum gas will be considered under one category, i.e., gaseous fuels. Natural gas can be converted to methanol. Methanol can be further converted into gasoline through the Mobil process. The synfuels, e.g., shale oil and coal liquids are generally refined into alcohol, gasoline or diesel fuel before being burned in highway vehicles. Methanol and ethanol (both alcohols) are alternative fuels that are effectively burned in highway vehicles. Therefore, alcohol fuels is the second major category of this report. 30.2.1.1 Gaseous Fueled Vehicles Spark Ignition Engines The gaseous fuels are particularly suited to vehicles powered by spark ignition engines. The significant property is their high octane ratings (110 to 120 research octane number) which permit the use of high compression ratio engines that run without autoignition or "knock." According to basic thermodynamic principals, high compression ratios mean high efficiencies; Table 30.1 illustrates the significant properties of gaseous fuels compared with gasoline. Dual fuel engines, using both gasoline and gaseous fuel, are limited in their ability to operate with maximum efficiency with either fuel. For instance, compression ratios needed to obtain the lowest fuel consumption using gaseous fuels (12:1 to 14:1) are higher than could be tolerated by gasoline (8:1 to 10:1). Likewise, gaseous fuels require advanced spark timing to compensate for slower flame speeds relative to gasoline. Finally, the additional weight of two fuel systems causes decreased fuel economy at all times. Thus, the most efficient use of gaseous fuel would require an engine modified for one particular fuel. Vehicles powered by spark ignition engines can be readily converted for gaseous fuel operation. Lack of widespread refueling systems for the vehicles generally dictates that gasoline be maintained as a backup fuel, and some potential for fuel conservation is wasted. Among the disadvantages of vehicles having this dual fuel capability is the added weight and space required by gaseous fuel tanks. Figure 30.1 offers an illustration of an LNG tank installed in a car trunk. LNG must be stored at very low temperature. Table 30.1 shows that temperature to be -259°F. Therefore, the LNG tank shown consists of a double wall container with stainless steel pressure vessel and carbon steel outer vessel. Insulation is provided by a 731 oEL Composition Physical state during storage Boiling Temp. @ 1 atm Press., °F TABLE 30.1 SELECTED PROPERTIES OF GASEOUS FUELS AND A TYPICAL GASOLINE NATURAL GAS CNG LNG Primarily methane (CH, ) but can contain up to 20% CoHg hydro- carbons. Gas... Liquid ~259 COMMERCIAL _PROPANE _ Approximately 75% Propane (C,H) and 25% C,H 6 hydro- - carbons. Liquid (under pressure) BIOGAS “65-75% meth. 24-34% carbon dioxide. Gas GASOLINE Mixture of C, to Ci, hydro- carbons. Liquid 80-450 SEL Vapor Pressure Psia @ 100°F, psia Lower Heating Value: Btu/1b Btu/gal Octane Ratings: Research Motor Source: Reference 6. TABLE 30.1 SELECTED PROPERTIES OF GASEOUS FUELS AND A TYPICAL GASOLINE (Continued) NATURAL GAS COMMERCIAL CNG LNG PROPANE BIOGAS GASOLINE 189 9-12 21,300 21,300 19,750 9,300 avg. 18,920 avg. 19,760 76,300 82,870 13,125 115,400 @ 2,400 psi @ -259°F @ 2,400 psi 70°F 1 atm 120 120 110 120 91-100 120 - 97 120 82-92 peL AIR CLEANER CONVERTER/ REGULATOR TANK RELIEF Source: Reference (7). FIGURE 30.2 LM MIXER SCHEMATIC OF LMG CONVERSION COMPONENETS STOCK FUEL GAUGE ELECTRICAL CONNECTOR QUICK-DISCONNECT TANK FILL BRACKET multilayer aluminized mylar separated by nylon netting. The insulation is wrapped in the space between the pressure vessel and outer shell, which is also evacuated. This tank holds 18 gallons of LNG at a pressure of 25 psig. It weighs 108 lb empty and 171 Ib full. CNG tanks do not require the insulation of LNG tanks, but are pressure vessels normally operating at about 2,400 psi. Typical vehicle installations use two CNG tanks which weigh a total of 250 to 350 lb, providing energy equivalent to about 5 gallons of gasoline. Lighter weight CNG tanks have been developed, but are not yet widely used. The weight of these advanced tanks is about half that of ordinary 2,400 psi CNG cylinders. Their construction consists of an inner shell of steel or aluminum wrapped by a layer of reinforcing fibers such as fiberglass or Kevlar. LPG can be contained as a liquid at ambient temperature (70°F) under moderate pressure (150 psi). Therefore, the fuel tanks do not weigh as much for a given volume as for CNG or LNG. Figure 30.2 is a schematic showing all the components required to convert a gasoline vehicle to LNG. Similar components are required for CNG or LPG. The major components, in addition to the fuel tank, are the converter, regulator, and the mixer (carburetor) which are shown in photographs on Figure 30.3. The mixer (carburetor) provides gaseous fuel to the engine at a regulated pressure and correct demand rate. In a dual fuel engine (gaseous fuel and gasoline), the mixer is mounted in series with the gasoline carburetor, usually in the position of the air cleaner. In a single fuel installation, the mixer replaces the gasoline carburetor. The converter regulator unit regulates the gas pressure from tank pressure to mixer inlet pressure. LNG and LPG vaporize in the fuel supply line from the tank to the engine. For LNG and LPG the converter regulator also warms the cold vapor by heat exchange with the engine cooling water. Compression Ignition Engines Compression ignition (diesel) engines impose different requirements on gaseous fuels than spark ignition engines. First, since there are no spark plugs in a diesel engine, the fuel must be able to ignite at temperature reached by the compression of the air within the engine's cylinders. This self-ignition must occur under a wide range of load and speed, including starting. Second, the fuel must be able to lubricate the high pressure pumps used to inject the fuel into the combustion chamber. Engine operation also imposes other fuel requirements, such as low temperature fluidity. The most basic fuel requirements, however, are combustibility and lubricating quality. 735 A. MIXER B: CONVERTER REGULATOR Source: Reference (7). FIGURE 30.3 THE MIXER AND CONVERTER REGULATOR, SHOWING UNDERHOOD LOCATIONS 736 For compression ignition engines, fuel ignition quality is measured in a laboratory test engine by the cetane method. Cetane number is a measure of ignition delay. The shorter the delay, the higher the cetane number assigned to the fuel. Since octane rating is a measure of a fuel's tendency not to auto-ignite under compression, the high octane numbers of gaseous fuels mean very low cetane numbers. Therefore, gaseous fuels are unsuited for direct use in compression ignition engines. However, they can be burned with chemical or thermal ignition aids. Chemical ignition aids include fuel additives or, alternatively, a "pilot" injection system may be used. With the pilot system, a second injector in each cylinder provides a small amount of diesel fuel where it ignites and burns. The gaseous fuel is, in turn, ignited by the pilot combustion. The gaseous fuel can be injected into the combustion chamber or inducted via the inlet valve similar to spark ignition engines. The latter approach is called fumigation. Fumigation is the favored approach for retrofit diesel engines, since it requires the simplest engine modification. The retrofit hardware is similar to that used for spark ignition engines. The diesel fuel pilot injection accounts for 25 to 35% of total engine fuel at full load in the case of fumigation. Only 5 to 10% diesel is sufficient for dual injection systems where the gas is injected directly into the cylinder along with the pilot fuel. As power decreases from full load, the pilot charge accounts for an increasing percent of the total fuel. This variable pilot/gaseous fuel ratio requires a sophisticated control system. For this reason, natural gas in compression ignition engines has been limited to engines operating at constant speed and load for long period -- such as stationary or marine engines. 30.2.1.2 Alcohol Fueled Vehicles Spark Ignition Engines Alcohol fuels (methanol and ethanol) are readily burned in spark ignition engines. The research octane number of ethanol and methanol are about lll to 112, compared with gasoline research octane numbers of 91 to 100. Thus, alcohol fueled engines can be built with higher compression ratios than gasoline engines with the corresponding improvements in thermal efficiency. Ford's methanol-fueled cars have a compression ratio of ll:1 compared with 8.8:1 for the gasoline counterpart. Table 30.2 lists the significant properties of alcohol fuels compared with a typical gasoline, a typical No. 2 diesel fuel, and gasohol. 737 8éL TABLE 30.2 SELECTED PROPERTIES OF ALCOHOL FUELS, A TYPICAL GASOLINE, AND A TYPICAL NO. 2 DIESEL FUEL NO. 2 CHARACTERISTIC ETHANOL - | METHANOL GASOHOL DIESEL FUEL GASOLINE 90% UNLEADED GASOLINE MIXTURE OF MIXTURE OF CHEMISTRY CjH,0H CH,OH 10% ETHANOL HYDROCARBONS HYDROCARBONS Approx. Specific Gravity @ 60°F Boiling Point °F 375-630 Lower Heating Value Btu/1b Btu/gal 18,700 117,000 18,400 130,000 18 ,000 112,900 170 400 250 600 200 Heat of Vaporization 465 kJ/kg 6EL TABLE 30.2 SELECTED PROPERTIES OF ALCOHOL FUELS, A TYPICAL GASOLINE, AND A TYPICAL NO. 2 DIESEL FUEL (Continued) NO. 2 CHARACTERISTIC ETHANOL METHANOL GASOHOL DIESEL FUEL GASOLINE 90% UNLEADED GASOLINE 10% ETHANOL MIXTURE OF HYDROCARBONS MIXTURE OF HYDROCARBONS TRY Cc CHEMISTR C,H,0H HOH Vapor Pressure @ 100°F psi Octane Number Research Motor 111 92 112 91 0 to +2 Above Gasoline Not Applicable Cetane Number Below 15 Below 15 Below 15 40-60 Stoichiometric A/F Ratio Source: Reference 8. The most common use of alcohol for highway vehicles is gasohol, which is defined as a blend of 10% anhydrous (200 proof) ethanol and 90% unleaded gasoline. This fuel is used in spark ignition engines designed for lead-free gasoline without any engine modifica- tions. Other blends of ethanol and gasoline have been used. In Brazil, blends of up to 20% ethanol are used. Blends higher than 20% ethanol generally cannot be used without engine modifications. . The effects of blending 10% methanol with gasoline are similar to the use of gasohol in spark ignition engines. Spark ignition engines can burn straight alcohol. The alcohol fuel can be either anhydrous or a water/alcohol mixture. The lower limit of alcohol content is generally considered to be 160 proof (80% alcohol and 20% water). The engine can be calibrated for only a narrow range of alcohol to water ratio. Conventional fuel metering systems cannot tolerate a wide variation in this ratio. Alcohols are completely soluble in water, whereas gasoline and diesel fuels are essentially insoluble in water. Therefore, if an alcohol blend with gasoline or diesel fuel is contaminated with water, the water will be absorbed initially by the alcohol. Further amounts of water, however, will cause the separation of the blend into its components. U.S. automobiles beginning with the 1981 model year are generally equipped with an exhaust oxygen sensor with feedback control to the carburetor. These systems can tolerate a variation in alcohol to water ratio since the air-fuel ratio is continuously measured and adjusted. Spark ignition engines will require several modifications to burn straight alcohol with good fuel economy, performance, and reliability. If the vehicle is not equipped with an oxygen sensor with feedback to the carburetor, which is the case for most pre-1981 model year cars and all trucks, the carburetor must be recalibrated for alcohol. The materials in the fuel system may need to be changed. Alcohol is a strong solvent that attacks different classes of materials than gasoline does. Alcohol will cause cork gaskets to shrink and some plastic materials to crack. Gums and other material that deposit in fuel tanks used for gasoline will be dissolved by alcohol. Thus, fuel filter clogging may occur if a gasoline vehicle is converted. Methanol will also cause excessive corrosion of some metals currently used in engines. Alcohols have a higher latent heat of vaporization than gasoline: four times higher for methanol and 3 times greater for ethanol. Thus, more heat must be applied to the intake manifold to vaporize the 740 alcohol. For this reason, alcohol fueled engines are more difficult to start in cold climates, which is a disadvantage for Alaskans. They are also more likely to surge and misfire at low ambient temperatures. An increase in compression ratio can be accommodated by alcohol fuels. Alcohols have high octane ratings compared with gasoline. The engine will operate satisfactorily without an increase in compression ratio. However, potential fuel economy benefits will be lost. In some cases, lubrication of the valves and upper piston rings may require modification. Alcohol fuels act as a solvent and wash the lubricant film from these critical areas. If the engine is not equipped with hardened valve seats, as is the case for leaded gasoline fueled engines, hardened valve seats may have to be provided. Compression Ignition Engines Table 30.2 shows that alcohol fuels have very low cetane ratings, with cetane numbers below 15. Thus, alcohols cannot be directly used as fuels for compression ignition (diesel) engines. The minimum cetane number required by the high speed diesels used in highway vehicles is 40. Several techniques have been investigated to utilize alcohols in diesel engines. These techniques include: e fuel modification with ignition improvers, e alcohol/diesel fuel blends, e mechanical emulsification of alcohol with diesel fuel using an emulsifier installed on the vehicle, e alcohol fumigation, e dual injection systems, and e ignition assisted diesels (spark plug or glow plug added) These techniques are currently in the research and development phase. No clear winners are currently discernable for future production engines. 30.2.2 Technical Characteristics 30.2.2.1 Gaseous Fueled Vehicles Several factors made gaseous fuels attractive as alternate fuels for highway vehicles in Alaska. These fuels are introduced into the engine as a gas. Therefore, the engine lubricating oil is not diluted 741 ENERGY DENSITIES OF GASEOUS FUELS AND GASOLINE TABLE 30.3 Fuel Gasoline LPG LNG CNG (@ 2,400 psi, 70°F) Lower Heating Value (Btu/gallon) 115,400 82,870 76,300 19,760 742 1. Relative Energy Density 00 ~/2 +66 017 as happens with gasoline, which is deposited on cold cylinder walls during cold starts or cold weather operation. This suggests the possibility of extended oil change intervals in Alaska's cold climate. Gaseous fuels readily mix with air, providing good cylinder to cylinder air/fuel distribution. This results in smoother engine running during warm-up. Also, gaseous fuels do not have to vaporize in the intake manifold, as gasoline does, to provide a combustible mixture in the engine. (LPG and LNG vaporize in the fuel line from tank to engine.) Thus, cold starting is greatly improved over gasoline-fueled engines. Studies conducted on two natural gas fueled vehicles have demonstrated a 10% improvement in energy efficiency during short trips (less than five miles) at low temperature (20°F).(6) The reason was that the engines operated more efficiently during the cold-start and warm-up portion of driving. The test results further showed no difference in energy efficiency compared with gasoline during fully warmed-up engine operation. However, the energy density of the gaseous fuels is low compared with gasoline. Table 30.3 shows the energy densities of gaseous fuels compared with gasoline. The relative energy density translates directly into relative vehicle range. For the same volume fuel tanks, an LPG powered vehicle will offer about 72% of the range of its gasoline counterpart. An LNG powered vehicle offers 66% of the range, while a vehicle burning CNG goes only 17% as far. Another disadvantage of gaseous fuels concerns engine peak power. Because of the fuel's gaseous state, its use results in a moderate reduction in engine power due to air displacement in the intake system. In other words, the engine's volumetric efficiency is reduced. A number of dual-fuel vehicle fleet operators have reported power reductions with methane ranging from 10 to 20% relative to vehicle operation on gasoline.(10) These losses are related to lean-air fuel mixture calibrations and the use of non-optimum spark advanced as well as reduction in engine volumetric efficiency. Thus, part of the 10 to 20% power loss can be regained by spark advance and fuel enrichment. But, the power loss attributable to volumetric efficiency loss requires major changes such as increased valve sizes. The safety of CNG, LNG, and LPG systems is an important issue. Accident related, gasoline fueled automobile fires occur at a rate of only one per 100 million vehicle miles. Insufficient mileage has been accumulated to determine the rate for gaseous fueled vehicles with a reasonable degree of statistical significance. 743 744 Source: Reference (9). A CNG FUELING STATION FIGURE 30.4 Major safety concerns under normal operating conditions include: e fuel leakage, e boiloff of LNG or LPG vapor, and e corrosive failure of CNG cylinders. Specific design steps to control these hazards include: @ venting the passenger compartment and/or the space used for fuel storage, e the use of odorants, and @ gas sensing devices. Hazards associated with collisions include: e fuel release upon impact, e tank rupture due to fire, and e intrusion of the heavy fuel tanks into the passenger compartment. There are currently no industry wide standards regarding the design, manufacture, installation, and performance of CNG fuel systems. The American Gas Association is actively working to develop such_ standards. Standards are especially important regarding aftermarket kits that are not integrated into the vehicle's original design and testing. In the current general absence of public fueling facilities for gaseous fuels, it is necessary that fleet operators install fueling systems for their vehicles. Two types of fueling designs are used, fast fill and time fill systems. Figure 30.4 illustrates a CNG fueling station of the Rochester Gas and Electric Corporation in New York. The illustrated station includes two fast fill posts. RG&E also maintains twelve time fill posts for overnight refueling. Reference 10 describes CNG fueling facilities as follows: "In the fast fill system, natural gas from a utility pipeline is routed through a dehydrating bed and a particulate filter into the 2,400 psi, the compressor discharges the gas at approximately 3,600 psi into a bank of high pressure cylinders. The compressor discharge is equipped with a high pressure 745 gauge and relief valve. The vehicle filling manifold at the storage bank outlet is comprised of fill-vent valves, flexible hoses and self- sealing quick disconnect coupling. The couplings on the filling manifold mate with couplings on the vehicles. When the fill-vent valve is closed after vehicle filling, the gas in the hose is vented about ten feet above the ground remote from the operator. To achieve a full charge, the final pressure in the vehicle cylinders is varied depending upon ambient temperature; the pressure can be controlled manually or by use of a temperature compensating regulator. A vehicle can. usually be filled in three to five minutes. "The time fill system is essentially the same as the fast fill installation except the bank of high pressure storage cylinders is not required. Also, the gas from the compressor is reduced in pressure to near the final filling value for the ambient temperature. Time filling is customarily accomplished over- night." 30.2.2.2 Alcohol Fueled Vehicles The factors that make alcohol fuels attractive are that alcohol fueled vehicles are: e fuel efficient, e low emitters of regulated exhaust pollutants, @ supported by resource bases that can support a large industry, i.e., coal, natural gas, grains, sugars, biomass, etc. : According to Dr. Winifred Bernhardt, Director of Energy Research and New Technology, Volkswagen's "Brazilian alcohol car is 20% more efficient on the average than its gasoline counterpart."(11) Volkswagen has extensive experience in producing ethanol-fueled ears. VW's current Brazilian production of these vehicles is about 5,000 cars per month. The higher efficiency of alcohol-fueled vehicles leads to lower energy consumption, but not lower volumetric consumption of the fuel. Alcohols have lower energy content per gallon than _ gasoline. Therefore, higher efficiency does not necessarily translate into higher miles per gallon. Table 30.4 gives the energy densities of alcohol fuels compared with gasoline. 746 TABLE 30.4 ENERGY DENSITIES OF ALCOHOL FUELS AND GASOLINE Lower Heating Value Relative Fuel (Btu/ gallon) Energy Density Gasoline 117 ,000 1.00 Ethanol 76,000 65 Methanol 57,000 -49 Thus, ethanol-fueled vehicles would achieve about 80% the miles per gallon of its gasoline counterpart if it is 20% more energy efficient. The reason is that ethanol contains only 65% the energy of gasoline per gallon. The relative energy density of methanol is lower. Methanol has 49% of the energy content of gasoline per gallon. The range of alcohol powered vehicles is less than gasoline fueled vehicles for the same size fuel tank. The ethanol fueled car achieving 80% of the miles per gallon of its gasoline powered counterpart, in turn, has 80% of the range for the same volume fuel tank. The methanol powered vehicle would have 60% of the range. Alcohol fuels give greater problems than gasoline for cold starting and driveability during engine warm-up. The higher heat of vaporization of alcohol fuels greatly increases their evaporative cooling effect. For equivalent engine power output, it takes four times as much heat to vaporize ethanol and six times for methanol compared to gasoline. Moreover, the low vapor pressure of alcohols at moderately lower temperatures (50°F) causes difficulty in starting. The engine requires the use of auxiliary starting devices, intake manifold heating, or volatile starting aids (e.g., gasoline, ether, etc.). Ford overcame the cold starting problem in a methanol test fleet by adding 5.5% isopentane (a very volatile aromatic) to the methanol. However, the isopentane is too volatile in a hot climate. Los Angeles will purchase 1,000 methane cars, and specifies that the fuel will be 90% methanol with 10% gasoline. The gasoline is a starting aid. In Alaska, this implies that a starting aid more effective than 10% gasoline would be required for methanol fueled vehicles, perhaps isopentane. Alcohols have greater flammability limits than gasoline or diesel fuel, which presents a potential safety hazard. For example, at 68°F, air saturated with methanol vapor contains 13% methanol, putting it in the combustible region. Precautions must be taken to shield this vapor from a spark or flame. 747 Methanol is a cumulative toxin that affects the nervous system, in particular the optic nerves. Vapors enter the body through the lungs and liquid can penetrate the skin. Therefore, care is required in the handling of methanol. 30.2.3 Environmental Issues 30.2.3.1 Gaseous Fueled Vehicles The exhaust emissions regulated by the U.S. Environmental Protection Agency, and more stringently by the California Air Resources Board for vehicles sold in that state, include unburned hydrocarbons (HC), carbon monoxide (CO), and nitrogen oxides (NO_). Gaseous fuels generally generate less exhaust emissions than gasoline. The basic factors for the advantage are: e improved fuel-air mixing and cylinder-to-cylinder distribution, and e improved operation at lean air-fuel ratios. It is difficult to draw general conclusions for exhaust emissions that hold for all vehicles, with all emission control equipment and at all levels of emission standards in California and the other 49 states over the past several years. However, gaseous fuels generally result in significant reductions in NO, and CO emissions directly out of the engine (upstream of the catalytic converter). Hydrocarbon emissions, on the other hand, are generally found to increase. The basic conclusion that can be drawn is that exhaust emissions generated by gaseous fuels are not very different in species than emissions from gasoline-powered vehicles. Therefore, the emission controls installed on the vehicles will generally meet emissions standards whether gaseous fuel or gasoline is burned. 30.2.3.2 Alcohol Fueled Vehicles Ford currently has a fleet of 40 Escort wagons powered by 1.6 liter, four cylinder engines converted for methanol and ethanol.(4) Ford is bidding to sell 1,000 similar cars to Los Angeles County. The cars offer very low exhaust emissions. Los Angeles County will buy 1,000 methanol powered cars. California's interests include: e protection against possible gasoline supply interruptions or soaring prices, and e low exhaust emissions. 748 Ford's methanol cars reportedly yield regulated exhaust emissions, shown in Table 30.5. The emission results with gasohol are mixed. Tests on a number of motor vehicles showed: (8) e -15% to +16$ unburned hydrocarbons e@ -28% to -7% carbon monoxide e +4% to +13% nitrogen oxides Also, an increase in aldehyde emissions was noted. Aldehydes are not currently regulated, but may be a future concern. 30.2.4 Commercial Status 30.2.4.1 Methane Powered Vehicles Over 250,000 vehicles in Italy, more than 20,000 in the United States, and lesser numbers in other countries are currently capable of operating on methane (natural gas).(10) Most of these vehicles have spark ignition engines that have been converted to operate on both gasoline and compressed natural gas (CNG). CNG and LNG conversion kits are currently available from several suppliers, shown on Table 30.6. Converting compression ignition (diesel) engines to use methane is difficult because of the very low cetane number of the fuel. Only experimental vehicles have been operated. No conversion kits are commercially available. Conversely, the high octane number of methane allows efficient, new spark ignition engines to be produced that are dedicated to methane; rather than the dual fuel, gasoline or methane systems. For example, Ford Motor Company is developing a two-passenger commuter car designed to obtain optimal efficiency from methane fuel.(13) The 1.6 liter, 4 cylinder engine's compression ratio has been increased to 13.6:1 to take advantage of methane's 120-octane rating. The compression ratio of the equivalent gasoline engine is 8.8:1. Higher compression ratio translates directly into higher efficiency. Ford uses the methane compressed to 2,500 psi. The requirement for higher pressure fuel tanks is a drawback for CNG-burning cars. A second disadvantage is limited range. Unlike other alternate fuels, however, distribution of natural gas is no problem. Ford is working with Outboard Marine Corp., to develop a small, home-use compressor that is now being tested in Several states. These units could be installed at homes to be attached to present 749 TABLE 30.5 EXHAUST EMISSIONS OF METHANOL FUELED VEHICLES Ford's 4) Methanol Results (grams per mile) Federal Regulated Standard Species (grams per mile) HC 24] co 3.4 NO. 1.0 x Source: Reference 4. 750 -1] 1.32 “44 = TABLE 30.6 SUPPLIERS OF CNG AND LNG CONVERSION KITS (10) CNG Dual Fuel Systems, Inc., Gas Service Energy Corp., Advanced Fuel Systems, Inc., Essex Cryogenics, Inc., Dual Energy, Inc., CNG Services, Inc., CNG Fuel Systems Incorporated of Canada, and Auto Italia of Italy. 751 LNG Beech Aircraft Corp., LNG Services, Inc. Essex Cryogenics, Inc., Kaiser Electro Precision, and Gibson Cryogenics. natural gas pipes. Fueling time with the small home compressor is four hours, from empty. - In comparison, a central filling station compressor could do the job in 15 minutes. Present price for the home compressor is $5,000. However, the price would come down significantly with high volume. 30.2.4.2 Propane Powered Vehicles Ford Motor Company has been building special-order propane truck engines since the mid-1960's. In 1981, Ford introduced propane powered Cougar/Granada models for sale in Canada. The propane . engine option was extended in 1982 to Cougar/Granada models. (12) Other countries have used propane powered vehicles for many years. Japanese taxi cabs are a noted example. Propane use is common where gasoline prices are high, e.g., Italy, the Netherlands, and Belgium. , Propane is an established alternative fuel and is used in current commercial vehicles. It is an option that is available from Ford Motor Company in the U.S. for cars and trucks. 30.2.4.3 Ethanol Powered Vehicles Vehicles fueled with 100% ethanol are currently in volume production in Brazil. Since 1975, Brazil has established a national alcohol program, focusing on ethanol fermented from sugar cane. Five car manufacturers operating in Brazil produce these _ vehicles: Volkswagen, Ford, General Motors, Chrysler, and Fiat. Today, 25% of Volkswagen's production in Brazil. includes ethanol-fueled cars. That amounts to about 5,000 cars per month. The remaining vehicles can operate on gasoline-ethanol blends, with up to 20% ethanol. Other developing countries are following Brazil's example. Volkswagen has begun deliveries to Paraguay, where a national alcohol program has been started. Their ethanol is made mainly from sugar cane, but also from cassava (manioc). Indonesia also has an alcohol program. Volkswagen believes Indonesia will be a substantial market for its ethanol-fueled cars. (11) 30.2.4.4 Methanol Powered Vehicles In the U.S. and other industrialized countries, a limited resource base and cost are deterrents which work against ethanol becoming a major fuel. For methanol, however, coal and natural gas are the resource base, and this base can support a major fuel industry. Thus, methanol appears to have a greater potential for competing with petroleum fuels than does ethanol. 752 my The State of California has taken the lead in the U.S. to encourage the manufacture and purchase of methanol-fueled cars. Ford Motor Company has participated in highly successful tests where a 40-car methanol fleet of Escorts have been used by selected California county and city agencies over the past year. Currently, Ford is bidding to sell 1,000 methanol-fueled liftback 4-door sedan Escorts to Los Angeles County agencies. Ford believes this contract will eventually lead to production alcohol burning engines. Volkswagen has also operated a test fleet of 37 cars in California. Most of the vehicles were fueled by pure methanol. By buying methanol-powered vehicles and establishing a distribution network for the fuel, California hopes to create a market for methanol-fueled highway vehicles. 30.3 Economic Implications 30.3.1 Costs 30.3.1.1 Gaseous Fueled Vehicles The cost of an aftermarket conversion to CNG, LNG, or LPG ranges from about $1,200 to $2,200 per vehicle.(10) Ford has a CNG powered commuter car under development (the "Shuttler"), but not yet available for purchase. The price premium estimated for the new car powered by CNG will be about $1,500.(13) Ford Motor Company is currently offering an LPG engine option on 4-door 1982 Cougar/Granada models. The price for this option is $750 to $900 for fleet buyers and about $1,500 retail. (14) These initial cost premiums must be balanced against fuel costs to determine whether a life cycle cost advantage exists. Maintenance and other costs are expected to be essentially equivalent to gasoline powered vehicles. The results of one analysis on the relative energy economy of gaseous fuels compared with gasoline are given on Table 30.7. The factors considered in the calculation are differences in vehicle weight, engine operating efficiency, and starting efficiency. The energy efficiency is the total energy required to power the vehicle per mile. Note that less energy is expended using CNG, LNG, or LPG, compared with gasoline. ENSTAR Natural Gas Co. operates a CNG station in Anchorage. Current price of CNG at that station is $0.62 per 110 cubic feet (ENSTAR's equivalent per gallon of gasoline), compared with about $1.60 for a gallon of unleaded gasoline in Anchorage. Based on a 753 Fuel CNG (reduced range) LNG LPG *1] of 3 starts are cold-starts Source: Refe TABLE 30.7 PERCENT CHANGES IN ENERGY ECONOMY FOR GASEOUS FUELED VEHICLES COMPARED WITH GASOLINE -6% rence 6, Engine Efficiency Increase +10% +10 754 Cold~Start Efficiency Increase* +3.3% 3.3 Total Relative Energy Economy 107.3% 107.3 105.3 $2,000 vehicle conversion cost and 12,000 miles per year travelled, the payback period is 3 to 34 years for CNG vehicles (Tables 30.8 and 30.9). No LNG is currently available for sale in Alaska. LPG is sold in Anchorage at about $1.00 per gallon by Petrolane Fuel Services and Vangas. Based on a $1,500 vehicle conversion cost and 12,000 miles per year, the payback period is about four years. 30.3.1.2 Alcohol Fueled Vehicles The recent cost of alcohol and petroleum fuels is shown in Table 30.10. In a previous section (30.2.2.2), it was estimated that the fuel economy of ethanol fueled vehicles in miles per gallon would be about 80% of its gasoline counterpart. Methanol vehicles can reasonably achieve 60% of the miles per gallon of a gasoline vehicle. Combining the mile per gallon estimates with the wholesale prices of Table 30.10, leads to the conclusion that alcohol fuel costs would be significantly higher in the U.S. than gasoline. Table 30.11 indicates the approximate size of that cost penalty for a typical pick-up truck. The life cycle cost disadvantage of alcohol fuels would not lead U.S. fleet operators or individuals to purchase alcohol-fueled vehicles. Financial incentives greater than those provided by the State of California are currently necessary if alcohol powered vehicles are to be encouraged. California allows a $1,000 state income tax credit for the conversion of existing vehicles to methanol. TABLE 30.11 ESTIMATED ANNUAL COST PENALTY RELATIVE TO GASOLINE IN THE U.S. Miles Per Year Ethanol Methanol 12,000 $700 $200 30.3.2 Socioeconomic Factors CNG and LPG vehicle conversions can offer some economic benefits to Alaska. Employment benefits are positive, but not expected to be great. ENSTAR Natural Gas Co. added about four additional employees to convert existing vehicles to CNG. Other local companies such as CNG Systems, Petrolane Fuel Service, and Vangas also employ Alaskans to convert vehicles. The primary benefit to Alaskans will be the lower cost of transportation fuel to those who convert to CNG or LPG. 755 TABLE 30.8 CAPITAL COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY Input Output ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD CAPITAL COST Equipment and Materials Direct Labor TOTAL CAPITAL INVESTMENT 756 Automotive CNG Conversion Anchorage 1982 constant dollars Not applicable 12,000 vehicle miles/yr. 10 years 1 week $1,200 800 $2,000 wa TABLE 30.9 OPERATING COST SUMMARY TECHNOLOGY Automotive CNG Conversion BASIS Location Anchorage Year 1982 CAPACITY Input Not applicable Output 12,000 vehicle miles/yr. OPERATING FACTOR Not applicable Annual Operating Costs Unit Cost Consumption Annual Cost VARIABLE COSTS Fuel (CNG) gallons of "gasoline 0.62 600 $372 equivalent" TOTAL VARIABLE COSTS $372 FIXED COSTS Capital Charges 9.5% $190 TOTAL FIXED COSTS $190 TOTAL ANNUAL OPERATING COSTS $562 TOTAL ANNUAL OUTPUT 12,000 miles TOTAL LIFE CYCLE COST $0.047/mile (equivalent to $0.94/gallon of gasoline) 757 TABLE 30.10 U.S. COST OF ENERGY FOR ALCOHOL AND PETROLEUM FUELS (Based on estimated July 1981 wholesale prices - taxes not included) Wholesale Price Cost of Energy Fuel $/Gallon $/Million Btu Methanol (from natural gas) 0.80 , 14.00 200-Proof Ethanol (by 1.77 23.30 fermentation) Regular Gasoline 1.03 8.80 No. 2 Diesel. Fuel 1.00 . 7.70 Source: Reference 8.” 758 The City of Anchorage currently operates a test fleet of 25 CNG-fueled vehicles. In addition to fuel cost savings, there may be an air quality benefit within the Anchorage bowl. The conversion of vehicles to CNG and LPG is relatively new in Alaska. ENSTAR began converting its own fleet in 1969, and currently operates 77 dual-fuel (CNG and gasoline) vehicles. However, ENSTAR conversions for customers has occurred for about one year only. 30.4 Impact 30.4.1 Effect on Overall Energy Supply and Use The total U.S. demand for petroleum fuels in 1980 was 10 million barrels per day. While significant amounts of methanol and ethanol were produced commercially during that period, U.S. production represented only about 27 million barrels of methanol and 5 million barrels of ethanol in a year. Moreover, all methanol produced in the U.S. is based on natural gas. Methanol could be a product produced from Prudhoe Bay natural gas, and sold in Japan or California (see Section 30.1.2). Ethanol is presently produced from ethylene and by fermentation using sugar and grain. Natural gas and ethylene feedstocks are not abundant enough in the U.S. to support a transportation fuel industry. The general national objective in the development of alternative fuel production is to add to the fuel supply by using more abundant or renewable, lower value, domestic resources such as coal or biomass. The Department of Energy has forecast maximum alcohol production from U.S. biomass resources, Table 30.12 shows the results of that forecast. Note that if the maximum scenario for alcohol is achieved, in the year 2000, 3,683 million barrels of methanol and 291 million barrels of ethanol could be produced from biomass per year. Note from Table 30.12 that feedstocks used for methanol production cannot be also used for ethanol production. The ethanol forecast of 1,286 million barrels and methanol forecast of 3,683 million barrels are not additive. Rather, the feedstocks used to produce methanol are not available to produce ethanol. Therefore, only 291 million barrels of ethanol are achieved along with 3,683 million barrels of methanol. Thus, methanol could satisfy about 26 weeks of the 1980 petroleum demand, and ethanol a further three weeks. Together they could supply over 50%, which would more than offset the 1980 U.S. crude oil imports of 38%. While these quantities may never be achieved, the forecast shows that methanol offers greater potential than ethanol. When the vast coal reserves of the U.S. are considered, even greater production of methanol can be achieved using coal as a feedstock. Therefore, methanol could be a future alternative fuel for highway vehicles of major national importance. Alaska could participate 759 TABLE 30.12 PROJECTED MAXIMUM ALCOHOL PRODUCTION FROM U.S. BIOMASS RESOURCES* Millions of Barrels per Year Wood Agricultural Residues Grains:, Corn Wheat Grain Sorghum Total Grains Sugars: Cane Sweet Sorghum Total Sugars Municipal Solid Waste Food Processing Wastes Total *The amounts in the table are "either, or." 1990 2000 eta [reenaes | exes | netans If, the resource to produce ethanol, the methanol potential is zero. Source: Reference 8. 760 is used through conversion of its extensive coal resources to methanol. The Placer Amex, Inc., proposal to construct a methanol plant at Beluga is an example. 30.4.2 Future Trends Methanol has the potential to become a major fuel in the U.S. because coal and biomass are the feedstocks and are available in sufficient quantities. Methanol is also a good spark ignition engine fuel. Thus, automakers, exemplified by Ford and Volkswagen, are investing time and manpower to develop methanol-fueled cars for production. Ford predicts that by the year 2000, 10 to 15% of cars sold in the U.S. will be fueled by pure methanol. (4) 761 10. 12. REFERENCES ICF, Inc., Alaska Natural Gas Development: An Economic Assessment of Marine Systems, U.S. Department of Transportation, Maritime Administration, 1982. Hederman, W.F., "Assessing a Potential Source of Non-Petroleum Vehicular Fuel: Alaska Natural Gas," Nonpetroleum Vehicular Fuels III Conference, Institute of Gas Technology, Virginia, 1982. "California Puts Methanol Plan Into High Gear," Chemical Engineering, November 1, 1982. "Ford Bidding on 1,000 Cars Using Alcohol," Ward's Engine Update, September 1, 1982. Mueller Associates, Inc., Solid Fuel Applications to Transportation Engme. U.S. Department of Energy, Division of Transportation nergy Conservation, Contract No. AC05-79CS56051, 1980. "Gaseous Transportation Fuels: A Study," Automotive Engineering, Volume 90, Number 8, August 1982. Fischer, F.L., "Introduction of a Commercial System for Liquid Methane Vehicles," Nonpetroleum Vehicular Fuels III Conference, Institute of Gas Technology, Virginia, 1982. Alternate Fuels Committee of the Engine Manufacturers Association, "A Technical Assessment of Alcohol Fuels," SAE Technical Paper Series No. 820261, 1982. Whitlock, D.J., "A Fleet of Twenty CNG Cars and Trucks in Upstate New York: Ten Months Experience," Non-Petroleum Vehicular Fuels III Conference, Institute of Gas Technology, Virginia, 1982. Assistant Secretary for Conservation and Renewable Energy, "State-of-the-Art Assessment of Methane-Fueled Vehicles," U.S. Department of Energy, Office of Vehicle and Engine R&D, DOE/CE-0026, 1982. "VW Sees Big Growth for Alcohol-Fueled Cars," Ward's Engine - Update, October 15, 1982. "Ford Brings Propane-Fuel Cars to U.S. Market," Ward's Engine Update, January 1, 1982. 762 13. "Ford Engineers Natural-Gas Engine for Minicar," Ward's Engine Update, January 15, 1982. 14, Zino, K., "Propane Power," Motor Magazine, December 1981. 763 31.0 SLOW STEAMING OF MARINE ENGINES 31.1 Introduction and Summary 31.1.1 Technical Overview The management of the fuel consumption of shipping powered by internal combustion engines must be based upon an understanding of the basic relationships between fuel consumption and costs and the environment in which the transportation system operates. The many variables which affect the problem make single generalized solutions impractical. However it is feasible to place into the hands of the operator the tools to make a determination possible for each specialized case or each particular ‘situation. "Slow steaming" is a technique by which a ship is operated at a speed lower than its conventional or design speed in order to-reduce fuel consumption and the attendant fuel costs for the voyage. In consequence of the lower speed, the voyage duration is extended and an increase in other voyage costs must be accepted. Frequently, these other costs do not increase as rapidly as the rate at which fuel costs are reduced and a net savings results. Fuel consumption may be optimized against a variety of criteria, as, for instance, costs in dollars per ton for a given voyage, unit cost per year, or similar parameters of interest to the operator. From the viewpoint of fuel minimization and fuel conservation, it must be noted that there exists no practical minimum for fuel usage, or fuel costs, alone. This would mean no travel at all since fuel consumption rate decreases continuously as the ship's speed decreases. However most other ship operating costs, as well as the inventory cost of the cargo itself, depend only upon the time for the voyage. The combined costs with ship speed as the variable, produce definite and quantifiable minima. Fleet operators such as the major oil companies have developed computer programs for the evaluation of best ship speeds. With the rise of oil prices in the past decade, fuel economy has become a major factor in the viability of marine transportation enterprises. The advent of the hand held calculator and mini-computer have made it possible to perform evaluations on the spot, by the actual operator of the vessel rather than at some distant home office. The cost components of owning and operating a ship or craft can be organized into three major groups. First there are costs which vary with ship speed or power. Fuel is the principal constituent of this category. Second are the costs which are a function of time. They include crew costs, maintenance, insurance, finance and _ such. Third, there is a small group of costs which are dependent upon neither and which may be related only to the voyage, such as canal 765 or lock fee and port charges.. These cost components can be modelled in term of ship speed, and the sum of components can be examined to find a minimum cost against a particular criterion. 31.1.2 Alaskan Perspective The hydrographic and climatic conditions existing in the principal sections of Alaskan waters demand, in some instances, that absolute ship performance rather than fuel consumption or operating costs be the principal measure. For example, the high tidal range and strong currents along the southern Gulf of Alaska coast determine when log tows can move, and they can move at low speed at best. Coastal traffic on the west coast may be affected by ice conditions in the Bering Sea; ice operations require particular modes of ship operation and power usage, and may involve random and unpredictable time delays. On an annual basis or over longer periods, such constraints can be addressed by statistical and probabilistic methods. Without going to such complex or sophisticated approaches, the operator can. obtain some guidance if he himself can revise the estimates for. the voyage as conditions and circumstances change during the course of the voyage. Intelligent control of the voyage, in the Alaskan context, is aided by the ability to develop fuel and cost information on board at any time and without reference to a control management point via long distance communications. The applicability of this type of analysis is limited in some instances by the area of operation and in others by the type of ship or operation. Although the principles of the analysis are valid for all craft, there is a lower size limit beyond which little is gained by such an analysis. Coastal cargo ships, tug/barge combinations, fishing vessels, and offshore oil industry service ships are amenable to a detailed examination. Highly specialized ships and ships operating in extreme conditions, whether in the north or along the southern coast, gain less from this management and control approach. Ships with fairly standard operating procedures and operating areas stand to gain the most. The high cost of transporting commodities and people within Alaska and from the Lower 48 to Alaska is reflected in the costs of Alaskan shipping operations. A further factor exists in the costs of transporting to and storing marine fuels in coastal areas outside of the major port areas. Marine operations with an improved understanding of fuel consumption factors will permit better estimates of future marine fuel requirements and. fuel distribution system requirements. 766 31.1.3 Significance of Fuel Consumption Control Technology In today's world economy, fuel costs may comprise over 50% of the annual costs of operating a ship. Therefore this cost factor has been studied both from the technical aspect and from the operational aspect by commercial, academic and governmental investigators. Probably the most dramatic demonstration occurred when larger tanker (VLCC: Very Large Crude Carriers) owners began "slow steaming". In most instances these ships were steam turbine ships with two boilers. By operation on one boiler at normal rating and efficiency, these ships could make between 70% and 75% of their normal cruising speed. The reduction in total costs per ton of cargo, or per ton-mile of cargo, saved these ships from going to the ship breakers prematurely. Eventually other factors conspired to make many of these ships obsolete and in the long term many went into permanent lay-up, other uses, or to the breakers after all. During the design phase of a proposed ship, detailed and far reaching economic comparisons can be performed for optimizing the ship for its intended purpose. However few ships or craft serve a single purpose and route during their lifetime; many must be designed for multiple roles initially. Furthermore, the costs of fuel and other operating costs will vary with time and place. Rarely are these cost changes over the 20 year life span of a ship in accordance with estimates made at the time of design. Continued economical operation of the vessel therefore requires continuous updating of the cost components for that particular vessel and continuous evaluation of best operating scheme. Present technology permits this capability to be placed on board the ship itself so that short-term variations of circumstances can be accommodated. The present day micro-computer readily can handle the necessary calculations, and can be programmed so that a minimum of training is required for the shipboard personnel who would use it. The costs of this information system lie in the accumulation of the basic cost factors for the particular ship and the minimal costs of the computing electronics themselves. The returns on this investment can be manifold. Today the diesel engine has replaced the steam turbine except in the case of very high powered ships. Most of the types of ships operating in Alaskan waters are diesel propelled. Some boats retain gasoline engines because of weight considerations, but the majority of craft of interest use the diesel engine as prime mover. Although this discussion is oriented toward diesels, the performance characteristics of gasoline engines are analogous. The steam types would result in very similar relationships with several minor changes. 767 31.2 Description of Technology 31.2.1 Operating Principles In basic terms, the resistance of a ship is proportional to the square of its speed. The resistance times the speed gives the horsepower required to propel the ship. This horsepower gherefore is proportional to the cube of the speed, that is, to V’. The fuel consumption rate of the propulsion engine is directly proportional to the power generated by the engine. The fuel consumption for a given voyage becomes proportional to the power times the duration of the voyage. The duration of the voyage, in hours or in days, is the distance divided by the speed of the ship or S/V, apd the fuel fuel consumption becomes proportional to S V"/V or to S V". . Other costs of the ship are either a function of time or are constant. The objective of slow steaming is to optimize the sum of the costs associated with each of these relationships to speed. The costs per voyage can be grouped to give in first approximation: Cl, =a v2 +b / V +c, in dollars per voyage The constant "a" defines the cost factor for propulsion fuel:of the ship including lube oil. "b" contains the cost per day of crew, including fringe benefits and subsistence, of maintenance and repair, of insurance, of consumables, of administration, and of all financial charges. "c" contains voyage dependent charges such as port, canal, pilotage and tug fees. In addition, "c" contains the in-port costs of the categories listed under "b" on the assumption that port time of loading and unloading cargo is independent of voyage duration or ship speed. Furthermore, both "b" and "c" must contain the costs of fuel required to provide the hotel services of the ship as this is a continuous requirement, in addition to which the generators and HVAC (heating, ventilating, air conditioning). These systems may require a separate grade of fuel from the main engines. Refinement of this general expression stems from three considerations. First of all, the speed-power relationships is not strictly one of P = v*. Each ship has an individual speed-resistance or speed-power curve which is not a simple parabola. It is a curve which may have several humps and hollows as wave phenomena affect the hydrodynamics of flow about the hull. However, for the portion of the speed-power Cur ys under considerations here, we assume that the relationship P = kV"" is a valid approximation. Second, the fuel consumption is not only a function of speed or shaft horsepower but also a function of engine efficiency. The translation of power requirements into fuel consumption requires that an efficiency "E" be added, that is, fuel consumption. becomes a function 768 of P/E. This efficiency in turn is a function of the proportion of full engine power that is demanded of the engine and can be stated in terms of the percent of engine power developed. A typical curve is shown in Figure 31.1. Third, the voyage transit time upon which many of the costs depend is not strictly only distance over speed, or S/V. The speed through the water of the ship, and the equivalent power provided by the engines, is modified by the effects of wind, currents and seas. As extreme example, assume that the practical speed for a tug and tow is four knots. If the tug is performing the tow against a four knot current, the speed of the tow over the ground is zero and the time and cost for the voyage tend to infinity. As another example, if the same tow has to anchor for six hours each at two narrows where currents are high, twelve hours of waiting time (though not of engine time) are added to the voyage. One method of dealing with these losses is to use the form T = S/(V-V,) where VE is an average speed loss determined from actual experience of the ‘barticular vessel and the particular locality. In addition, it is noted that in the relationship P = k ee the value of "k" will vary with different conditions of loading of the ship. The basic value of "k" is derived from the horsepower and speed achieved during trial conditions under favorable weather and sea conditions. The optimization for a round-trip voyage must consider that for a ballast leg the value of "k" is smaller than in the full load condition. This change in value of "k" also must be determined from basic data for the particular ship. In. the matter of engine fuel efficiency, each manufacturer has test bed data from which such information can be provided. As typical example the curves published by General Electric in references (1) and (2) and (3) have been used here. The curve shown in Figure 31.1 can be defined by the expression: = 1.006 - 0.303 (1.093 - CIN, pore where V is the ship speed in knots and Nis is the design or trial speed at full power. The fuel rate at © 100% power (maximum continuous rating) is 192 gms/HP-hr or .423 lbs/HP-hr. This is a conservative value; current large engines give somewhat better performance. 31.2.2 Technical Characteristics The algorithm for determination of optimum speed or for analyzing fuel usage and cost components is flexible. It can be modified to suit other propulsion systems and it can be elaborated to account for more detailed relationships. For example, the slight changes in propulsive 769 Specific Fuel Consumption, gms/HP-hr 190 200 210 220 oc ao = oO a 3 wu e o . Cc o Vv oc o a 27 %0 Source: GE Publications. FIGURE 31.1 90 2) 1,00 Effici SFC Cc ficiency, 100% / SF TYPICAL DIESEL FUEL RATE AND EFFICIENCY 770 efficiency (the efficiency of the propeller and hull as opposed to engine efficiency at various loads) has been neglected in the version presented in the previous section. An expression for propulsive efficiency could be added, or it could be combined with the expression for engine efficiency. Experience with marine propulsion systems has shown that, except for some esoteric forms, the curves of costs and efficiencies show rather flat inflection points whenever such minima occur; the lowest points on the curves are not sharply defined. In corollary, the optimum speed is not defined to a high degree of accuracy such as 0.1 knot. However the 1/2 knot or 1/4 knot differences which are identifiable make an appreciable difference in fuel consumption in high speed or in large ships. It is also noteworthy that the range of speeds under discussion is limited. Even in the cases of ships which show large economic savings by slow operation, the speed lies within about 25% of design speed although fuel consumption may drop up to 50%. The principal behavior of ships in this upper 25% of the speed range does not vary sharply and operations in that speed range can be extrapolated satisfactorily from design and trial performance. However in some instances it became desirable to improve engine and _ propeller efficiencies at a new reduced speed; propellers were designed to the new conditions and replaced the original. The new conditions were selected to give a torque and RPM point which provided good engine performance. Such performances are advantageous principally when the ship in question is one which operates under constant conditions. For ships where this type of optimization is used to permit adjustment to ambient conditions which may vary from voyage to voyage, such permanent changes to the propulsion system are not desirable. The intent here is to permit the ships to make a determination of best operating conditions rather than to determine desirable modifications of the existing ship. In this same context, even though an optimized speed may define an appreciably lower installed horsepower than is fitted in conventional ships, in most cases the owners and operators will retain the higher horsepower installation and carry around the excess weight and cost. Most owners and operators are of the opinion that even though a lesser power is used in normal continuous operation the higher installed horsepower is necessary to provide for contingencies of maneuvering, casualty avoidance, or emergency missions. Most of the benefits of slow steaming can be retained by optimizing the propulsion system for the slow speed and considering "full power" as the exception for which lower engine and propulsion efficiencies are acceptable. The Mississippi-Alabama Sea Grant Consortium has examined the Gulf shrimper fleet.(7) Twelve to fourteen percent of time is used in travelling to and from the fishing grounds. The speed during these 771 runs is adjusted to minimize fuel within the other constraints. 51% of time is employed in fishing and 70% to 75% of total fuel is expended during the fishing period. Larger tankers may expend 90% of time and close to 98% of fuel during their sea time. The ship used in the example given in Section 31.3.1 on the 800 mile one-way trip at 13 knots with no speed loss, uses 11% of the fuel in port and 89% in transit. She is in transit 30% of the time and in port 64% of the time. This randomly selected sample has a very high proportion of port time to sea time, which is a deterrent to optimization results at low speed. However the characteristics of the optimization process are adequately demonstrated by the results. In regard to the hardware involved, it is noted that the Swedish firm Jungner Marine of Solna now offers commercially a "Fuel Economy System". This provides a continuous display of power, speed and fuel consumption and gives Specific Fuel Consumption, tons per hours, and tons per nautical mile. The Texas Instrument T-18000 gives data on speed and remaining fuel. These and similar monitoring “systems provides engineering data to the operator but do not take into consideration the other economic factors of the marine transportation system. Lastly, in the sample given here, inventory costs have not been included. Their inclusion would require determination of the average . cost per ton of cargo carried and the interest rate at which the cargo would be charged. The final daily cost would then be added to the ship's own daily time-dependent cost factor. In the case of fishing vessels that operate on a definite short cycle, time taken in transit cuts into the fishing time. A mean value of fish harvested per day or per hour is a penalty imposed by a reduced transit speed. . Such a "cost" or loss of revenue due to the reduced speed should be. included in the time-dependent cost factor in order to optimize the economics. Both cargo inventory costs and lost fishing time costs would tend to increase the optimum speed of the ship and would reduce any indicated fuel savings. 31.2.3 Environmental Issues The environmental effects of a speed reduction program made rational by the sort of analysis provided here are small. There would result a reduction in the usage of marine power: with commensurate reduction in emissions. Most small boats would not be affected since their power use is dictated by factors other than fuel costs and economy; either a minimum speed for safety reasons or a high value on personal time (real or perceived) tend to drive speeds up. These small boats however provide a substantial proportion of noise pollution, although this would have to be quantified in relation to background noise. In remote areas they frequently are the only or predominant sources. 772 pe _ As a secondary effect, a reduction in the use of marine power reduces the magnitude of the fuel distribution system, or at least reduces its rate of growth. This results in some small improvement in casualty and fuel spill probabilities in the distribution system. Proper management of marine power and fuel usage principally is a matter of fuel conservation in the environmental context. In itself it produces no effects upon the environment other than potentially ameliorating existing effects originating from the marine power plants. There is no obvious reason why it should be opposed by environmental agencies or groups, and in many instances it can be demonstrated to be economically advantageous to the operators. 31.2.4 Commercial Status The principles of this type of analysis have been investigated for nearly a decade. The major oil company fleets have developed \ proprietary systems by which ship voyages are analyzed ashore and the masters are directed from shore as to speed and routing. J Shipboard equipment which gives continuous readouts of fuel consumption and fuel rates is now available. Individual papers (1,2,4,5) address the subject or touch upon it. Technically, the problem is well understood and solutions have been implemented on shore-based computers. We understand that most of the large shipping companies operate in and out of Whittier Bunker outside of Alaska. This includes both ship and tug/barge operations such as Sealand, Crowley, Dillingham, Foss and Tote and others. We also understand that these companies are active in controlling fuel economy from their home offices while the ships themselves are | generally equipped with fuel performance monitoring computers. The ‘ ocean-going tugs, for example, may have a fuel capacity of over 600 \ tons and may engage in voyages of two weeks individual duration. During such time, they maintain speed and fuel consumption according to information developed at the home office. The applicability of such analysis can be extended to a much wider spectrum of ships. It is possible to use any one of the hand-held engineering electronic calculators to achieve a satisfactory solution. A specialized celestial navigation calculator improves on the time and accuracy for celestial navigation. The calculating process is sufficiently simple that software and chip development poses no great problems or costs. In summary, there is no portable specialized hardware available for the purpose; if a demand arose it undoubtedly would be produced. Meanwhile it is possible to address the problem with any one of a large variety of general calculating devices although additional pencil and paper might be required, as well as a few minutes more of time. 773 TECHNOLOGY BASIS Location Year CAPACITY Dimensions ESTIMATED USEFUL LIFE CONSTRUCTION PERIOD CAPITAL COSTS Shipyard Price Construction Loan ‘4 Design, Inspection, Trials Owners Equipment TOTAL CAPITAL INVESTMENT TABLE 31.1 CAPITAL COST SUMMARY Small Supply Ship U.S. Construction 1982 Length Beam Draft Displacement Installed Horsepower Cargo Deadweight Design Speed 20 years l year+ $4,000,000 320,000 380,000 100,000 $4,800,000 774 180 ft 30 ft ll ft 1290 tons 2600 BHP (mcr) 700 tons 13.0 knots The principle inputs to the program must be developed beforehand. These are peculiar to the ship in question and must be generated by the owner or operator. Instructions to do this can be placed into simple cookbook form. 31.3 Economic Implications 31.3.1 Costs The costs of implementing this system itself are small and are distributed among the ship operators. If any economic savings result, i.e., where a reduced speed not only saves fuel and fuel costs but also results in a lower overall transportation or freight rate, the costs of any calculating equipment is small compared to annual potential savings. In approximate terms, if a ship proceeds at half power it may be proceeding at 3/4 of its full power speed. It will require about 4/3 of the time for the voyage as at full power. If a specific number of tons of cargo must be transported per year, for the reduced speed 4/3 as many ships will be required (or a ship 1/3 larger). Since the fuel used per ship voyage is 1/2 x 4/3 = 2/3, a fuel saving of 1/3 or 33% has occurred per ship. However, 4/3 ships are now required and 4/3 x 2/3 is 8/9 as the fleet fuel usage at the reduced speed which is the equivalent of an 11% saving in fuel. This presents a further aspect of the cost implications of slow steaming, and must be considered by fleet operators. Here also the proportion of fuel costs to the time-dependent costs of the ship influences the benefits to be derived. The type of operation of the ship is an essential ingredient in the determination. An evaluation of the potential fuel usage reduction depends upon data for the types and sizes of ships operating in the Alaskan area. Such data would permit generation of valid estimates which could provide guidelines for future planning. As noted above, a_ reduction in individual ship speed often necessitates larger ships or a larger number of equivalent ships. The individual ship savings are thereby partially negated. The increased tonnage has effects noted in the following section. Since the problem is both ship and route specific, it is most clearly demonstrated by a sample calculation. The vehicle chosen here is a small cargo vessel of the offshore supply ship type. The characteristics of this ship are shown in Table 31.1. Costs are shown in Tables 31.1 and 31.2. The notation which is used here and the assumptions which are included in the problem are shown on Table 31.3. 775 TABLE 31.2 OPERATING COST SUMMARY TECHNOLOGY BASIS Location Year CAPACITY OPERATING FACTOR : Operating Costs VARIABLE COSTS Fuel FIXED COSTS Crew Maintenance & Repair Insurance Stores & Consumables Hotel Load Fuel Administration Capital Charges TOTAL FIXED COSTS Smal Alas 1982 1 Supply Ship ka 218,400 ton-miles/day 345 op. days/year less port time and Unit Cost 0.09-0.27 $/1b $26,800/man-yr 272 $/t 9.5% TOTAL ANNUAL OPERATING COSTS TOTAL ANNUAL OUTPUT (The TOTAL LIFE CYCLE COST oretical) 776 empty voyages Annual Consumption 1400-2100 tons 2x 1l 235 tons Annual Cost 280-1270 thousand 592 80 $ $ $ 100 $ 30 $64 $ 40 $456 "$1,362 $2, 642-2, 632 $75,348,000 TM/yr ee TABLE 31.3 SLOW STEAMING EXAMPLE ASSUMPTIONS AND RELATIONSHIPS SHP Design Condition: BHP SFC, (Specific Fuel Consumption Fuel Efficiency E Port Time t Port Costs cy Daily Fixed Costs c, Fixed Voyage Costs Fuel Costs Voyage Cost 0.95 BRP = 1.287 v7*86 0.80 Installed BHP (Maximum Continuous Rating) 0.80 P mer 0.423 lbs/BHP-hr = 0.445 1bs/SHP-hr 192 gms/BHP-hr 1.006-0.303(1.093-(v/v_)7*8°)?+027 2 days loading + 2.5 days unloading 108 hr/voyage $1000/entry = $2000/voyage = Cc, $4984. 63/day c. ((S/V-V, ) +t) + Cr (BHP) (SFC, /E) (c,) (S/V-V, )) = cy) Ce + Cy + Cy CUT TABLE 31.3 Continued NOTATION ae Maximum Continuous Rated Horsepower PB Brake Horsepower (BHP) Py Shaft Horsepower (SHP) Vv Speed, in knots WA Trial or Design Speed Vy Speed loss due to wind, seas, current, hull condition SFC, Specific Fuel Consumption in lbs/HP-Hr at design cond. SFC Specific Fuel Consumption E Fuel Efficiency Ce Fuel cost, in $/ton Cc, Fuel cost, in $/voyage Ss Voyage distance in nautical miles ao Time in transit in hours t Time in port in hours Cp Other costs, time dependent, in $/voyage Cc, Independent costs, in $/voyage = 0.423 lbs/BHP-hr = 0.445 lbs/SHP-hr 192 gms/BHP-hr. Fuel Efficiency E = 1.006-0.303(1.093-(V/V.)7*86)?-017 Port Time: t = 2 days loading + 2.5 days unloading = 108 hrs/voyage Port Costs C; = $1000/entry = $2000/voyage = Cc, Daily Fixed Costs c, = $4984 .63/day Fixed Voyage Costs = c, ((S/(V-V,) + t) + G, Fuel Costs = (BHP) (SFC,/E) (cp) (S/V-V,)) = C,) Voyage Cost = C, + G, + C, 778 The results are shown in Figures 31.2 and 31.3. The first figure assumes the standardized finance cost of 9.5%; the second series of curves are based on a 75% loan at 12% for 12 years plus 25% equity at 20% per year. The time-dependent daily cost of the ship increases by 26% in the second scheme. The distance, fuel price, and speed loss used for each curve are shown on the right. The curve shown in solid line in Figure 31.2 has subsidiary curves in short broken lines which show the fuel cost component of the total cost per ton-mile. It is apparent from these results that at the low fuel costs shown this small and expensive ship should be running at 3} to 14 knots above design speed in order to achieve lowest overall freight costs. However such speed increases would result in fuel consumption increases up to 6 tons per voyage. On the other hand, a doubling of fuel prices results in an overall optimum speed of 12.0 to 12.5 knots, or a drop of } to 1 knot from design speed. This is equivalent to at least ‘a 1.4 ton saving in fuel per voyage of 800 miles. The fuel price of $201.60 per ton shown in the example is a composite of quotations for 80% Intermediate 180 CS and 20% marine diesel for the U.S. West Coast. It assumes that the marine fuels are delivered by large ocean-going barge directly to Alaskan ports which have waterfront storage and bunkering facilities where no handling or trans-shipment is required. Marine fuels also are available from Kenai and can be distributed from there by barge. The doubled price of $403.20/ton is appropriate for ships fueling at remote ports. Generalized prices for delivered #2 Distillate have been given as: Anchorage $ 1.31 per gal ( = 414 $/ton) Railbelt $ 1.38 per gal ( = 436 $/ton) Brush $ 1.90 per gal ( = 600 $/ton) Conversely, $201.60/t is the equivalent of 0.64 $/gal and $403.20/ton equals $1.28/gal. The latter is approximately the same as the price of #2 Distillate at Anchorage. As indicated by the curves of Figures 31.2 and 31.3 the higher fuel prices would result in further reduction of the optimum speed of the sample ship. At a fuel price of $600 per ton on an 800 mile voyage the optimum speed would drop to about 11} knots or 14 knots below the design speed of the ship. However halving of the daily costs (time-dependent costs) of the ship from $4984.63 to $2500 per day achieves the same result of dropping the optimum speed to 12.5 knots. This demonstrates the principal sensitivity of the solution upon the ratio of fuel costs to daily costs, rather than upon the absolute value of either. For larger ships and longer voyages, this ratio increases. For large tankers, for example, fuel costs may be 50% of total costs or more, and this ratio becomes 1.0 or more. The resultant drop in optimum speed is commensurately more dramatic. 779 osomp fp ‘ Finance: 9% % DESIGN SPEED o.ogob__ 0.06 0.05 0.04 FIGURE 31.2 11 SPEED 12 in KNOTS 13 (N.M./HRD SAMPLE COST CURVES STANDARDIZED FINANCING 14 Po sonrserrn 4400/403.2/.5 pi00/201 6/49 0.07 0.05 ~ 781 11 12 13 14 SPEED in KNOTS (N.M./HR) FIGURE 31.3 SAMPLE COST CURVES 800/403.2/1.0 800/403.2/0.0 800/201.6/0.0 ie ~7400/209 26/065 t 31.3.2 Socioeconomic Issues For a given annual tonnage to be carried between two ports, a speed reduction means more or larger ships. In this case of more equivalent ships, since the voyage time is extended, the number of port calls would be reduced commensurately. Therefore, the effects upon the terminals and the shoreside logistic activities would be essentially zero. Employment afloat would increase if equivalent ships were used but would not grow noticeably if somewhat larger ships were employed. Crew members are almost constant over the small changes in size of ship resulting from speed changes. If speed reduction was employed in such operations as fishing, where routine cycles are observed, some timing and schedule changes may -occur which affect local customs and habits. In the case of self-employed boat crews, longer hours may be -acceptable provided they result in operating cost savings. Cost factors used in an evaluation must be tailored to reflect the type and method of crew remuneration predominant in that segment of the industry. As noted in section 31.2.2, the type of operation also indicates the necessity for including inventory costs. If the ship operator is the entrepreneur, inventory costs are of direct concern to him. If a fishing boat operator fishes on his own account, selling his catch at some time after making port, an extension of transit time is equivalent to a loss in value of his cargo. For a boat owned and operated by a large seafood company, other factors may have significance and can be similarly incorporated into an economic balance. Lastly, if by speed alterations an enterprise can be made more profitable, marine activity will be stimulated. This obviously would enhance the economics of the Alaskan littoral by attracting further marine activity. 31.4. Impact 31.4.1 Effect on Overall Energy Supply and Use "Slow Steaming" implies the prerequisite of the ability to determine an optimal speed for a given ship and voyage. The techniques of finding such a speed in respect of the various costs and constraints of a particular situation are readily available and can be performed on board in order to adjust to short-term changes in condition. Slow steaming initially was applied in the tanker trades and resulted in both overall economies and fuel economies. For the larger ships, fuel savings of up to one third were achieved. If fleet capacity however is taken into account, the net saving even under such 782° individual ship economies is reduced to about 11%. As indicated in Section 31.3.1, no speed reduction and no fuel savings may be appropriate for small and expensive ships. The benefits are most apparent in the transport of low value bulk commodities in ships where fuel costs form a large proportion of the total operating costs of the ship. The techniques applied to any type of ship leads to illuminating results and is suitable both for commercial and public vessels. It may be expected that even with adoption of this technique throughout the marine industry in Alaska, fuel savings of less than 5% will accrue. Since marine transportation now uses about 3% of the annual fuel of the State, the reduction will be amount to a fraction of 1%. In this discussion diesel engines and intermediate fuels have been considered. Substitute power plants or substitute fuels would show similar trends. Inasmuch as the application of these techniques is a distributed process, it is not expected that the overall effects upon the marine fuel consumption would be either sudden or dramatic. They do permit the definition and the establishment of future fuel requirements in the context of the ship's operations. In many instances they permit a reduction of fuel requirements and usually they are a valuable guide to the economic health of a marine enterprise. As such, they add to the well-being of the marine industry of Alaska. In particular, the demands upon the fuel distribution system to points outside of the Railbelt and in outlying coastal areas would be reduced commensurately. This fuel distribution system, whether by sea, by rail, by road or by air, poses an additional fuel demand which can reinforce any savings achieved by the operating ships themselves. 31.4.4 Future Trends Fuel prices at present are stable. Nevertheless the relative costs of liquid hydrocarbon fuels may be expected to increase in the next decades. Currently, there is no economical substitute on the horizon for the marine diesel engine except possibly some form of coal-fired plant. The importance of fuel costs to shipping will remain a principal factor in the future of the industry and they will be of increasing concern to the individual ship owner and operator. The ability to evaluate and to control fuel consumption will be a significant factor in future shipping operations. With the potential of major activity in the offshore oil exploration and production areas, the supporting marine activities also can increase. Without the ability to analyze ship fuel consumption such an expansion can become more costly then necessary and can place inordinate strains upon the fuel distribution system, particularly if the areas under development are remote from established port facilities. Conversely, the availability of fueling points is weighed in the selection of shore bases for oil development activities. 783 The use of fuel is distributed over wide areas of the sea and shore and is not a source of concentrated emissions. Marine fuel usage is only a minor component in the environmental and ecological picture, but nevertheless one which will be desirable to maintain at controllable levels. Measures which keep such emissions from excessive growth will remain beneficial. The technology for improving the fuel economy of marine engines and the entire ships is progressing steadily. At present these improvements are incremental, and sometimes expensive. No sudden breakthrough to much higher efficiencies is visible at this time. The other side of the coin in regard to fuel consumption lies in the operational aspects. The techniques described here address those factors in regard to the economics which are the motivating force behind marine enterprises as well as all commercial activity. 784 q } REFERENCES Stott, C.W. and J.P. Casey, "A New Steam/Diesel Comparison," Marine Engineering/Log, August 1979, p. 33. Spears, H.C.K., "Steam Propulsion Into the Eighties," The Naval Architect (RINA), March 1979, p. 82. General Electric Company, advertisement Marine Engineering/Log, October 1979, p. 39. Buxton, I.L.,"Engineering Economics and Ship Design," 2nd edition, British Ship Research Association, 1976. Benford, H., "Bulk Cargo Inventory Costs and Their Effect on the Design of Ships and Terminals," Marine Technology (SNAME), Vol. 18, p. 344, October 1981. Private Discussions with U.S. Maritime Administration, Washington D.C., 1980-1981. "Fuel Studies to Aid Shrimpers" Linda Skupien Sea Grant Today, Vol. 12, No. 2, March/April 1982. 785