HomeMy WebLinkAboutVenetie Village Council Biomass Feasibility Study 02-20-2012-BIOBIOMASS HEATING FEASIBILITY Level 2 Study
Venetie Alaska
i
Venetie Village Council
Level Two Biomass Feasibility Study
Alaska Wood Energy Associates
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 2 - Final 20 Feb 2012
Table of Contents
SECTION 1 3
1.1 GOALS AND OBJECTIVES 3
1.2 PROJECT SCALE 4
1.3 RESOURCE ASSUMPTIONS 5
1.4 FINANCIAL METRICS 6
1.5 LEVEL 2 SUMMARY 7
1.5.1 RECOMMENDED SCENARIO 9
1.5.2 GARN IN A BOX 9
SECTION 2 9
2.1 EXISTING FIELD CONDITIONS 9
2.2 PRELIMINARY SYSTEM INTEGRATION PLAN 12
2.3 SCHEMATIC DESIGN DATA 15
2.3.1 PIPING: 15
2.3.2 BOILERS: 17
2.4 RESOURCE ASSESSMENT 17
2.5 HARVEST SYSTEMS 17
SECTION 3 18
3.1 GENERAL 18
3.2 RECOVERED HEAT 18
3.3 WOOD HEAT 19
3.4 SUPPLEMENTAL HEAT 21
3.5 DISTRIBUTION 21
3.6 INTEGRATION OF RECOVERED HEAT 22
3.7 DESIGN CONSIDERATIONS FOR THE USE OF STICK FIRED BOILERS 23
3.7.1 SIZING, BOILER CONTROL, AND UTILIZATION RATE. 23
3.7.2 END‐USER ISSUES. 24
3.7.3 MATERIAL HANDLING. 26
3.7.4 EMISSIONS CONTROLS/EFFICIENCY. 26
3.7.5 MAINTENANCE. 27
3.8 SITE ISSUE. 27
SECTION 4 28
4.1 LIMITS 28
4.2 METHODOLOGY 28
4.2.1 ENERGY SAVINGS. 28
4.2.2 RECOVERED HEAT. 32
4.2.3 COST ESTIMATES. 32
4.3 BUSINESS STRUCTURE 32
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 3 - Final 20 Feb 2012
APPENDIX A. WILDLAND FIRE HISTORY NEAR VENETIE 34
APPENDIX B. PHOTOS OF HARVEST EQUIPMENT 35
APPENDIX C. BOILER SITE AND PROPOSED HEATED BUILDINGS 37
SECTION 1
1.1 Goals and Objectives
The objective of this Level 2 Cost and Feasibility Study report is to document the progress and
findings at this stage of the project. The scope of a Level 2 analysis is to provide enough cost
and performance data to allow the village to determine if the project or projects included should
proceed to investment grade studies, design, and eventual implementation. The Level 2 Study
looks at both small district heating “mini” plants as well as individual building-level biomass boiler
applications.
A “mini plant” for the purposes of this report is simply one or more stick-fired boilers in a single
building serving two or more buildings, by piping the heat from the plant to the buildings. The
term “District Heating Plant” (DH Plant) denotes a larger plant, using chip-fired boilers, and
generally serving five or more buildings. The location and size of the buildings in Venetie are
such that no DH plant is feasible here; there is not enough economy of scale to support the high
first costs associated with DH plants.
The study is complete based on current knowledge; the author of the study has been to Venetie
twice for the purpose of doing site work for this study. Should the project progress to an
investment grade audit, further site visits will likely be required. The study is based on data from
the village, from the site visits, from the State, and other professionals with experience in bush
Alaska. As additional information comes to light in the design phase, this is expected to result in
changes to the savings and/or cost projections.
Some of the examples herein are taken from a similar analysis performed for Fort Yukon, which is
currently 90 percent done with an investment grade study and more than 35 percent complete
with design.
The work on this and other projects in bush Alaska is being done by a team, Alaska Wood Energy
Associates. This team consists of people from a number of different companies, representing
various skill and knowledge sets. The feasibility study is being performed by Greg Koontz of
efour, PLLC, Seattle, WA. Greg will also provide design services specific to the boilers.
Obviously, a primary concern for any biomass fired plant is availability of the wood resources.
This issue is the responsibility of Bill Wall of Sustainability, Inc. The means and methods of
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 4 - Final 20 Feb 2012
procuring the wood and processing it will be documented in detail elsewhere; they are
summarized briefly in Section 2.4 below. For that reason, this report does not address biomass
supply in the same level of detail as it does the heating systems.
The objective for the team and the village at this point is to use the study to choose a path
forward into final design and construction, or, failing that, to determine under what conditions the
projects would be pushed forward. This final report documents the L2 feasibility study that has
been completed.
1.2 Project Scale
In order to be successful, a DH plant must achieve a certain economy of scale. The capital costs
involved are quite large, and so the savings to the village must be on the same scale in order to
make economic sense. One of the reasons why a DH plant must be large to be feasible is that
they are projected to run on wood chips, not stick-wood. This is primarily a function of boiler
capacity, and the need for manual labor to feed stick fired boilers.
As shown below in several of the Figures, the largest stick-fired boiler being considered has a
firing rate of 925 kBTU/h (thousands of BTU per hour). The largest chip-fired boiler AWEA
recommends has a firing rate of 4,2650 kBTU/h - over four and half times larger. Chip-fired
boilers are automatically fed; stick-fired boilers are manually fed. One can imagine the manual
labor required to fire five stick-fired boilers four times a day each in cold weather.
At the same time, the equipment needed to harvest wood and chip it to feed a chip-fired DH plant
represents a very significant capital cost. Experience has shown that unless a village has about
six to eight large buildings (minimum) in a fairly tight cluster, a chip-fired DH plant is not
economically feasible. It follows that if there is no DH Plant to use chips, then it is not feasible to
make chips at all – so in villages with no feasible opportunity for a DH Plant, the study focuses
solely on stick-fired boilers, as in Venetie.
The Level 2 study for Venetie included only four building (three existing, one future); the spacing
of these buildings is such that they can be connected by hot water piping fairly economically.
As a result, the study looks at each of the four buildings separately, and then at three variations
on a “mini-plant”. The four buildings are: 1) the school (and associated shop/boiler building), 2)
teacher housing, 3) the washeteria/water treatment plant, and 4) the future clinic.
The school and the washeteria/water treatment plant are both included as individual buildings two
times in the model – once with recovered heat and once without. These two building currently
have the ability to utilize heat recovered from engine generators. However, in both cases, it
appears as though the heat recovery is not always utilized. Heat from the main village power
plant can be utilized by the washeteria, but this system currently appears to be inoperative. The
school has its own generators, and when they run, the school uses the recovered heat to
supplement the boilers – but the school does not always have recovered heat available. For
these reasons, the school and the washeteria are included as individual buildings twice – once
with no recovered heat available, and once with continuous recovered heat available.
The model uses PCE electrical data to generate a village electrical demand profile; this electrical
demand profile in turn generates a profile which calculates the amount of recovered engine heat
available to the washeteria for any given condition. However, no electrical demand data are
available for the school alone. The model therefore assumes that at each condition, the electrical
demand of the school is a fixed percentage of the total village electrical demand – that fixed
percentage is currently set at 50 percent. This implies that the school consumes ½ of the village
electrical energy at any given time. This assumption needs to be validated and/or changed as
new knowledge becomes available.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 5 - Final 20 Feb 2012
In addition to the individual building analyses, three variations of a single mini-plant are analyzed;
in all cases all four buildings would be served by the mini-plant. The variations are: 1) All four
buildings, with the School Heat Recovery active, 2) All four buildings, with the Washeteria Heat
Recovery active, and 3) All four buildings, with the no Heat Recovery active.
1.3 Resource Assumptions
In order to compare all the mini plants and individual boilers on an equal basis, some base level
assumptions about these costs had to made and used in the performance / financial model. The
assumptions shown in Figures 1.1 and 1.2 below are based on current estimates of recent fuel
costs in the villages, plus projections of the cost of obtaining stick wood.
Figure 1.1. Base Level Oil / Electrical Assumptions
The table lists four oil prices; this model is used in a number of villages, so it needs to be flexible.
The “low cost” is used when some buildings in the village get a better price than others (such as
schools in some villages). The “high cost” is what everyone else pays. The other two are the
cost of oil to the DH plant (in case it is different still), and the cost to the power plant (in case they
also get a different price). In this case, only one value was used, $8.50 per gallon.
Figure 1.2. Base Level Wood Assumptions (chip / pellet data grayed out)
The unit cost of cord wood has been increased to $300/cord since the preliminary report. This
includes the cost to procure the wood and feed the boiler.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 6 - Final 20 Feb 2012
The primary resources of concern in this study are the various energy sources, current and
proposed. Because the systems involved are closed piping loops, water and sewer use is almost
zero on an annual basis. Aside from filling and/or flushing the system, there is no water use and
thus no sewer use.
The fuels of concern, then, are No. 1 oil and electrical energy (current costs), and stick wood
(proposed costs), as seen in Figures 1.1 and 1.2. As will be seen in the following sections, even
if this project is implemented, oil and electrical energy will still be required in the village.
It was noted that the washeteria and the school can benefit from “free” recovered heat from the
respective associated generators. In this study, that heat has been treated as free; that is, the
owner of the generators does not charge for the use of this recovered heat. If this is not the case,
the charge for the recovered heat can be included in a revision to this study. This would
obviously negatively affect the economics of any boiler plant utilizing recovered heat.
Finally, figure 1.3 shows the properties of the stick wood that were used in the model. These
tables show the mix of species expected to be harvested in the area and the expected moisture
contents; this input is used to calculate the composite stick-wood properties.
Figure 1.3. Stick-wood Properties
1.4 Financial Metrics
There are any numbers of financial metrics that can be employed to evaluate a project. Many of
these require that the source and means of financing the project be known. Many require
knowing the expected interest rate that money could be borrowed at, and even the rate of return
the client would expect to achieve if they invested the capital elsewhere (not in the project).
In the case of potential projects in Venetie, much of this information is not known at this time.
The exact funding mechanisms are not known. The in-kind participation of the village, if any, is
not defined, and therefore the value of it cannot yet be determined. Finally, forward-looking
interest rates are not very predictable at this point in time.
At the same time, this study is not an investment grade study. As such, it does not seem justified
to make assumptions about all of the relevant financial variables. For all these reasons, this
study has used a single financial metric to evaluate each potential heating plant – both as a
stand-alone investment and as a way to compare different technologies and combinations of
buildings.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 7 - Final 20 Feb 2012
Net simple payback (NSP) as used herein is defined simply as the implementation cost of the
project divided by the value in dollars of the annual year one energy savings. Year one savings
are specified; it is assumed that resource rates will change year to year (or faster).
All financial summaries used in this study use NSP as the sole financial metric for evaluating
each option. There a number of variables which do not factor into the NSP as defined herein;
perhaps the two most relevant here are the labor cost and the maintenance cost.
It is not known how the plants would be manned. Stick-fired boilers require up to 12 loads of
wood per day, manually fed, at peak load. However, in such cold weather, the operator(s) may
choose to simply feed the boiler during the day, and let the oil-fired boilers take over at night.
The level of unknowns regarding the amount of labor that would actually be applied to each plant
makes estimating labor costs difficult at best. Similarly, judging the maintenance costs of the
boiler plants in the harsh climate of the interior of Alaska presents an issue. The Garn boilers are
very simple, with not much to break. Nevertheless, they have moving parts to maintain, and
possibly fail.
Labor and maintenance costs are annual, and thus deduct directly from the energy savings
(lengthening the NSP). The point being made above is simply that the range of possible values
for annual labor and maintenance costs is so wide that they should not be used to make financial
decisions as a part of this study. Instead, as the project is developed in each village, the
decisions on boiler technology and plant size should go hand-in-hand with discussions of how the
boilers will be operated and maintained so that the true cost can be determined prior to making
the investment.
1.5 Level 2 Summary
A mathematical model was constructed to model the performance of the various mini plants and
individual boilers in the village, and to compare their financial performance. The model is
discussed in more detail in Section 4, and a sample of the calculations involved is shown in
Appendix A. To date, it is not known for sure how this project would be financed, so the financial
model does not include the cost of money.
At this point, as noted above, the key financial metric is net simple payback (NSP) at current
costs. Figure 1.4 shows a summary of the results of the model using the Base Level resource
assumptions.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 8 - Final 20 Feb 2012
Figure 1.4. Project Summary
Some of the paybacks are very long, and it can be seen that in all these cases, the boiler
proposed is the smallest available; the Garn WHS 1500. These long paybacks indicate that the
associated building is small, and does not consume much oil on an annual basis. Even with the
smallest available boiler, the cost of a biomass boiler installation is significant. Unless there are
substantial potential annual dollar savings in the form of oil displaced, a given building may not be
able to support a biomass boiler installation.
This is the main reason we group buildings into mini plants – it improves the economics of the
projects. In the table above, items 1 through 4 are the four individual buildings, modeled with the
assumption that the school and washeteria get no recovered heat. Item 7 is the water treatment
plant again, this time assuming it does get recovered heat. Item 6 and 7 are the school and
washeteria with their respective heat recovery systems available. In item 6, the (already very
short) payback for the school gets even shorter, as one would expect with “free” recovered heat
available. Item 7 shows no payback at all; this is because it appears that the recovered heat
available from the village plant is so large, that it is capable of displacing all of the oil for the
washeteria – thus there is no point in installing a biomass boiler under those conditions.
Items 9, 10 and 11 are three variations on the mini-plant; as expected, the one with the largest
amount of “free” recovered heat has the best payback. However, in all three cases, the NSP is
very good.
One conclusion that is drawn from the study to date is that the projects are extremely sensitive to
the cost of oil, and less so to the cost of wood and electrical energy. Another is that economy of
scale is important to make these projects feasible.
In terms of recommendations, it appears that any of the mini plants have very favorable
paybacks. The village must determine if they want to proceed to an investment grade study and
preliminary design based on these results, but the Level 2 results certainly appear to support
proceeding at the current resource price levels.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 9 - Final 20 Feb 2012
1.5.1 Recommended Scenario
The scenario on line 10 captures the most heat from the village electric generators, displaces the
most diesel fuel in the village, uses the least amount of wood and has the highest overall dollar
savings. This report recommends this scenario with two Garns in a Box (described below). In
addition, since the VCC owns and operates the powerhouse and proposes to own and operate
the district heating system, then the business model of utilizing the heat from the generators
becomes much simpler and develops a vertically integrated enterprise that allows greater overall
business efficacy.
1.5.2 Garn in a Box
Since the preliminary version of this report was submitted, Garn, the recommended stick-
fired boiler manufacturer, has begun to develop a new product. Informally, it is known as the
“Garn in a box”. As the name implies, it is a Garn boiler that comes as a complete package.
Currently, a Garn boiler requires a primary pump, a heat exchanger, and some piping and
controls in order to operate, as well as separate electrical connections for each piece of
equipment (the additional required piping and equipment beyond the heat exchanger is
considered part of the distribution system, not part of the “boiler plant”). In the new configuration,
Garn will pre-pipe and pre-wire all the parts and pieces that make up the “boiler plant”. In
addition, the boiler and balance of plant will be enclosed in a container that is meant to act not
only as the shipping container but the as the boiler “building” as well.
On site, the end-user need only construct a concrete slab, and bring the power and distribution
piping to the slab. Once the Garn is placed on the slab by a lift truck or similar, the piping and
power can be connected in a day, or two at the most. They are also meant to be able to sit side-
by-side with no clearance between. The only site-erected “building” required is an enclosed
storage or lean-to for wood, and perhaps an enclosed walkway from the storage to the “boiler
room” doors (for multiple-boiler applications).
Currently, Garn is planning to offer this only in the WHS 2000 size. However, this works very well
for Venetie – that is the recommended boiler size for the mini-plants; each consists of two WHS
2000s.
The units are scheduled to go into construction in 2012. The estimated cost of the units is
relatively high, compared to just buying the boiler. As more are made, the cost may come down.
In the preliminary report, the cost estimate assumed that all Garn WHS 2000s were bought
separately (as were the other sizes), and piped and wired on site. And, for now, this appears to
be perhaps slightly less expensive. However, there are advantages to having much of the work
done off-site. For this final report, therefore, all of the Garn WHS 2000s are assumed to be the
pre-packaged models. The prices are higher; but the net simple paybacks are still very short.
Pricing the models this way allows the village the flexibility to purchase either variety of the boiler.
By the time an investment grade study / concept design is done, more will be known about the
“Garn in a box”, and this new information will help inform any future equipment decisions.
Section 3 below includes a discussion about Garn boilers, and the design considerations that
apply to stick-fired boilers. All of that information is valid regardless of how the Garn boiler is
configured or packaged.
SECTION 2
2.1 Existing Field Conditions
At this time, there are four buildings included in the study. Each of these four buildings has been
evaluated for the installation of a dedicated biomass boiler, and evaluated as inclusion in a “mini
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 10 - Final 20 Feb 2012
district heating plant”, where two biomass boilers located in a single building would serve all four
buildings. Heat from the boiler “plant” would be piped to each building served Economy of scale
is important in biomass boilers, so mini-plants generally have a better payback than the dedicated
building boilers, as the summary in Section 1 indicated.
The four buildings included in the study at this time are: 1) the school (and associated shop/boiler
building), 2) proposed teacher housing, 3) the washeteria/water treatment plant, and 4) the future
clinic.
All of these buildings are served by boilers; this is important – connecting a new external boiler to
an existing one in a building is generally very simple (and thus relatively inexpensive). Even
connecting to a building with oil forced air furnace can be easily done, by adding a heating coil in
the furnace ductwork.
There is one potential complication in hooking up the buildings to a biomass-fired boiler or boilers;
that exists in the washeteria, and is explained in detail in Section 3.7.2. It has to do with the
clothes dryers in the building, and their specific requirements.
The table in Figure 2.1 below lists the existing heating equipment in the subject buildings, with
some comments on condition.
Figure 2.1. Existing Village Boilers
The Clinic is the only building on the list that has no back-up heat. The boilers themselves are
generally in “medium” condition, but the mechanical rooms themselves are in various stages of
repair. The school boiler room piping is well insulated, and labeled, while the teacher housing
piping is almost completely uninsulated. The washeteria piping is generally insulated, with
frequent gaps in the insulation.
The mechanical room / piping for the teacher housing is shown in Figure 2.2 below:
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 11 - Final 20 Feb 2012
Figure 2.2. School Boiler Piping/Header
The washeteria/water treatment plant is adjacent to the village power plant. Like many villages, a
system has been installed to recover heat from the engines. In many villages, this system is
turned off, or does not work, and this appears to be the case in Venetie. Were it operating and in
good condition, it appears that it could provide all or nearly all the heat required in the water
treatment building.
Normally, the heat from the engines would be rejected to the atmosphere by running the cooling
water through radiators. The heat recovery system is installed in series with, and upstream of the
radiators; it has the ability to extract any amount of heat from zero to 100 percent of the heat in
the cooling system. A 3-way control valve controls the flow of the cooling water through a flat
plate heat exchanger. One the “other side” of the heat exchanger is the water treatment plant
heating water. The heat exchanger transfers heat from one loop to the other, cooling the engine
cooling water while heating the building heat loop. If the heating loop does not extract all the
engine heat (in summer, for instance, it might need no heat at all) the remaining heat in the
engine cooling loop is routed to the radiators for final cooling. Figure 2.3 below shows the key
elements of this system, which appear to be in reasonable condition.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 12 - Final 20 Feb 2012
Figure 2.3. Flat Plate Heat Exchanger / 3 – Way Valve / Radiators
2.2 Preliminary System Integration Plan
Hot water will be piped from the external biomass boiler(s) into the existing building mechanical
room. Once in the building mechanical room, the new hot water distribution piping will be tied into
the existing hot water supply and return lines that feed the existing boilers. Typically, four 2-
position, 2-way automatic isolation valves will be installed in the piping, as shown in Figures 2.5
and 2.6 below. The position of these valves will determine whether the heat comes from the
building oil-fired boiler, the biomass boiler, or both. The existing building pumps will continue to
serve the building heating load.
The valves that control the origin of the heat will be controlled by the existing building controls
where they exist, or by a small dedicated control panel if needed. If this proves too costly for very
small installations, the switchover can always be done with manual valves, but this relies on an
operator being present.
Figure 2.4 below shows a schematic drawing of a typical installation for two oil fired boilers. In
this scenario, each boiler is sized for 100 percent of the load; the boilers are manually alternated
so that they get roughly equal run time. In all cases (figures 2.4, 2.5, and 2.6), light solid lines
indicate existing equipment and piping, dark solid lines depict new equipment and piping, and
light dashed lines show the water flow through the system. For convenience, it is assumed in all
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 13 - Final 20 Feb 2012
cases that Oil Fired Boiler – 1 is the active boiler, and boiler 2 has been isolated using the
associated manual isolation valve. HWS is hot water supply to the building, HWR is hot water
return from the building.
Figure 2.4. Typical Oil-fired Heating Plant
Figure 2.5 shows the initial configuration of the combined oil and biomass heat, with the biomass
boiler providing all the heat. Closed “auto” (automatic) valves are solid, open valves are not
“blacked in”.
Figure 2.5, Combined Oil and Biomass Heat, all heat from Biomass
In the event that the biomass heat cannot meet the building setpoint for any reason, the two
systems can operate in series. The lack of adequate heat output from the biomass boiler could
range from small (an extremely cold day) to total (a boiler failure or someone forgot to fire the
boiler in time), but the operation would remain the same – the oil-fired boiler would simply add
enough heat to maintain setpoint, whether this is 1 percent or 100 percent of Load. This is shown
in Figure 2.6.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 14 - Final 20 Feb 2012
Figure 2.6, Combined Oil and Biomass Heat, Boiler Heat in series with Biomass Heat
The scenario gets only slightly more complex if recovered heat is to be used also. In this case,
the existing heat exchanger (HX) remains in place, and the piping on the “hot side” remains as is
(the engine cooling loop is rejecting heat to the heating loop, so the engine cooling loop is the
“hot side” of the HX, and the building heating loop is the “cold side”). On the cold side, the piping
is re-arranged as shown in Figure 2.7 below:
Figure 2.7, Combined Oil and Biomass Heat, Boiler Heat with Recovered Heat
The stick-fired boilers proposed for use on this project incorporate a hot water storage tank (see
2.3 below, and Section 3 for more detail). This water in this tank is normally kept between 120
and 200 deg F. The existing 3-way control valve (upper right of Figure) would be used to
modulate the flow of heat through the heat exchanger, and thus keep the boiler storage tank as
close as possible to 200 deg F (its maximum charged temperature).
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 15 - Final 20 Feb 2012
However, as it gets colder outside and the building heating load increases, tank temperature will
begin to drop as more heat is extracted from the tank. When the temperature drops below 120
deg F that is the point at which the recovered heat can no longer maintain the building heat. At
that point, the boiler operator will have to begin to fire the biomass boiler. The recovered heat will
continue to be used, however.
The HX is in series with, and upstream of, the biomass boiler, which in turn is in series with and
upstream of the oil-fired boilers. This determines the order in which the heat sources are used:
recovered heat first, then biomass heat, and then only when both of these sources are fully
loaded will the oil boilers be used. A failure of either the HX or biomass systems is fully backed
up by the existing oil boiler(s).
2.3 Schematic Design Data
The subsection above shows how a proposed biomass boiler would be interconnected with the
existing boilers in each building. Note that a single boiler (or two or three in combination) can
serve multiple buildings. The piping from the boilers simply branches out to each building – all
the connections would still look like those in 2.2 above. The subsections below contain
preliminary design information on equipment proposed for this project. More detail about all
these elements can be found in Section 3.
2.3.1 Piping:
If the proposed biomass boiler installations and plants proposed herein had to rely on traditional
rigid “arctic” piping, the payback would be significantly extended. Instead, the basis of design is a
flexible, pre-insulated system that uses a plastic carrier pipe (carrier pipe is the pipe that “carries”
the water, or the inner pipe). The carrier piping is constructed of cross-linked polyester, or PEX.
The outer pipe (or casing), which protects the insulation, is made of corrugated plastic to make it
flexible.
The piping comes on rolls that are dozens or hundreds of feet long. It is much lighter than steel
or copper pipe, and thus cheaper to ship into the villages. Standard easy-to-install fittings are
used anywhere in the piping to connect end-to-end, tee, or elbow as required. The piping can be
obtained as a single pipe within a casing, or in the smaller sizes, supply and return can be
combined into a single common casing. Figure 2.8 shows an example of a PEX system:
Figure 2.8, PEX Piping, Two Carrier Pipes in a Single Casing
Figure 2.9 shows the sizes of PEX piping available from Rehau, the manufacturer that
the preliminary performance modeling is based on.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 16 - Final 20 Feb 2012
Figure 2.9, Preliminary Piping Schedule.
The insulation rating, or R value, of the piping depends on the depth of bury, and the soil
conditions. The performance model has all these factors built in. The standard installation detail
is quite simple. A shallow trench is dug, between three and five feet deep. A minimal amount of
bedding is put down, and the piping is unrolled into the trench. The trench is back-filled to about
6 inches above the top of the casing pipe. Then a 4’ x 8’ sheet of “blueboard” (rigid insulation) is
placed lengthwise in the trench above the piping to increase the R value of the total assembly
(blueboard not shown in Figure 2.10). Then the rest of the trench is back-filled as shown in
Figure 2.10; the final assembly can support road traffic, if need be.
Figure 2.10, Preliminary Piping Installation Detail.
Within the buildings, piping will be steel or copper, with standard fiberglass insulation.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 17 - Final 20 Feb 2012
2.3.2 Boilers:
Garn manufactures the proposed boilers; Dectra Corporation, located in St Anthony, Minnesota
owns Garn. A number of Garn Boilers are already installed in Alaska. Some Design
Considerations specific to stick-fired boilers are presented in detail in Section 3. Figure 2.11 is a
schedule of capacity and physical characteristics of the Garn boilers.
Figure 2.11, Garn Boiler Characteristics.
The model 3200 is quite a bit larger than the other two models, in every way. It is expensive to
ship to Alaska, and therefore has not been widely used in bush Alaska. However, it is almost
always less expensive to install one boiler with X capacity than two boilers with ½ X capacity
each, so the model 3200 has been included in the study.
2.4 Resource Assessment
A significant amount of the forests around Venetie have burned in the past 60 years as part of a
natural fire disturbed ecosystem (Appendix A). Burned areas typically regenerate to hardwoods
first and then as succession moves through time the stands are taken over by spruce either white
or black depending on the quality of the site and the return fire interval. Fires create a mosaic
pattern of forests in all successional stages. There are burned and mature stands within five
miles of Venetie. Since there have been recent fires (last 10 years), there is a lot of standing
dead that is being utilized as fire wood and could also be used as fuel for a district heating
system. Residents in Venetie have a lot of experience in gathering firewood and one fire wood
supplier gets up to 100 cords of wood annually for sale into the village using hand felling with
chain saws and transporting with a snow machine and sled.
There are three DH scenarios described in this report. Since the goal of the project is to displace
as much fuel oil as is financially feasible and ecologically sustainable, the largest system appears
to be the preferred. That scenario of two Garn boilers with heat recovery from the generators will
require approximately 190 cords of wood annually. We estimate ranges of dry standing burnt
spruce to be approximately 12-14 cords per acre in many stands. The community is interested in
using these stands first whenever possible. Approximately 14 acres will be needed to fully
support the largest DH system proposed.
This resource assessment is based on several trips to Venetie and field inspections of stands in
all conditions from burned to early successional hardwoods to mature stands. No formal
inventory has been developed nor has systematic forest stand cruises been conducted.
However, three different experienced foresters have been on the ground observing the forest and
all agree that the level of harvest for a DH heat system is fully sustainable within 5-10 miles of the
village.
2.5 Harvest Systems
A full analysis of harvest systems is beyond the scope of this report. However, a key to a
successful biomass program includes the integration of getting enough quality wood on an annual
basis securely stored and then fed to the boilers. There are three different general harvesting
systems that could work in Venetie: A. Hand cutting and loading with snow machine transport; B.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 18 - Final 20 Feb 2012
Hand cutting and mechanized loading and hauling; C. Mechanized cutting, loading, and transport.
Of course, there are multiple scenarios for equipment for systems B and C.
Photos of different types of equipment that could be utilized are in Appendix B. Deciding which
system to utilize will be an important decision for the community to make in the development of a
final business model. It is the recommendation of this report to develop and use system B
described below. Although system A yields more hand labor for potentially more people, this
may not be the most effective way to assure an adequate supply of wood annually and may not
be the most cost effective. System C could be too expensive in capital costs and maybe overkill
for the amount of wood required for the Venetie DH system.
Hand felling and limbing will take a significant amount of labor in itself for producing 190 cords
annually. Adding a tractor with a loading arm and a trailer for hauling (Appendix B) will make the
harvesting-transport system both affordable and more reliable. The issue with the tractor for
winter hauling will be maintaining the trail to the wood stock piled in the forest. The tractor will
have attachments for snow removal, both a plow and a blower. Although not a versatile a track
vehicle could be used instead of the tractor as the primary transport vehicle.
SECTION 3
3.1 General
There are often two general configurations of boiler plants examined in studies such as this;
single building applications (or very small groups of adjacent buildings – mini-plants) and district
heating (DH) plants. Likewise, two wood burning technologies are usually included; stick fired
and chip fired. An individual building boiler could be either a chip or stick-fired boiler, but only
chip boilers are large enough (in terms of heat output) to power a DH plant.
However, chip-fired systems are very expensive to install, and require a great deal of support
equipment (chippers, etc). Experience has shown that in order to get the required economy of
scale to make a DH plant financially practical, a minimum of about six to eight or more buildings
(depending on size) must be connected to the plant. Venetie, as noted in Section 2, has only four
buildings included in this study, so no DH plant has been proposed or evaluated for Venetie. In
the absence of a chip-fired DH plant, there is no economical way to make wood chips, so by
process of elimination, this study includes only stick-fired boilers.
In general, stick fired boiler systems are smaller, less automated, require less associated
equipment, and their lower heat output range makes them appropriate for smaller scale
installations (single building or small group applications), such as Venetie. They are simple and
robust; but their simplicity means that they are more labor-intensive to operate than chip fired
boilers.
Overall, the study considers three sources of heat: 1) heat recovered from power generator
engines, 2) heat from wood, and 3) heat from oil.
3.2 Recovered Heat
Heat recovered from an engine generator and used in a boiler system or DH Plant is “free” in the
sense that there is no marginal cost increase to reject that heat to a heating loop compared to the
rejecting it to the atmosphere. The heat comes primarily from the cooling jacket of the engine,
and must be carried away from the unit to prevent it overheating. In the absence of a co-located
building or heating plant, the heat is normally carried to a radiator, which cools the jacket water by
rejecting the heat to the atmosphere (a fan blows air over a radiator coil).
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 19 - Final 20 Feb 2012
Engine generators producing prime power are an ideal source of heat for any heating plant. They
run continuously, and the quality (temperature) of the heat rejected is almost identical to the heat
required by the boiler or plant. Generally, recovered heat is the lowest cost form of input energy
to the heating system, thus it is normally selected first and used to the fullest.
However, in some villages the owner of the power plant chooses to charge for the recovered
heat; in such cases the “waste” heat is “free” for them to provide, but not free to the end-user. In
this model, the recovered heat has been assumed to be free to the end user. This will need to be
confirmed in further studies, and, in the event that a cost for the recovered heat is to be charged
to the user, this charge will be incorporated into the financial and performance models.
Sections 3.6 below and 2.2 above describe how the recovered heat will be integrated into the
proposed biomass fired systems.
3.3 Wood Heat
As noted above, recovered heat (when available) is considered the primary heat source, because
it is generally the least expensive heat source. Wood heat is the thus considered the secondary
heat source. As with recovered heat, wood heat will always be used to the extent possible before
using the tertiary heat source (oil, in this case). The villages own significant amounts of this wood
resource in the surrounding lands, and AWEA believes it can be produced in a usable form (cord
wood) at a price significantly below that of oil, on a BTU basis.
Solid fuel boilers require more infrastructure than oil-fired boilers. They require space for wood
storage and processing. Stick fired boilers require space to cut to length and split wood to meet
the boiler specifications.
Given the remoteness of the Yukon Flats, any equipment installed must be reliable and well
tested. It is also desirable that any boilers used be standard units, “off the shelf” so to speak.
The use of proprietary or customized equipment increases the chance that if equipment failure
occurs, it will be expensive and/or time consuming to get it fixed.
For stick fired boilers, the team proposes to use Garn boilers. Garn is owned by Dectra
Corporation, located in St Anthony, Minnesota. A number of Garn Boilers are already installed in
Alaska. Figure 3.1 is a picture of a Garn boiler.
Figure 3.1, Garn Boiler
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 20 - Final 20 Feb 2012
Note that aside from cord wood, a Garn boiler can also burn clean construction waste, slab wood,
and densified wood products (briquettes, etc). However, neither construction waste nor slab
wood was considered to be a reliable resource at the sites considered. Densified wood products
are not available in the interior; in fact, one of the primary reasons to consider a stick-fired boiler
was minimize the processing required for the fuel.
As Figure 3.1 shows, the Garn boiler consists of a burn chamber (the chamber hatch-style door
can be seen in 3.1) surrounded by a hot water storage tank. Because of the water tank, the Garn
boilers are much larger than chip-fired boilers for a given output. It also means that Garn boilers
have two output ratings: 1) burn rate, and 2) storage capacity. The burn rate is the rate at which
heat is released when the chamber is loaded with stick wood per directions and fired. The
storage capacity is how much heat the tank can hold.
In order for the storage capacity to have any meaning, the maximum and minimum storage
temperatures must be specified. For the Garn, the tank is considered to be fully charged when
the tank water is 200 deg F. It is considered to be depleted when the tank temperature is 120
deg F.
Heating a building is a continuous process; heat is continually withdrawn from the tank by a pump
and heat is transferred to a building. Heating the boiler tank is a batch process. A discrete
amount of wood is burned in each “batch”. No more fuel is added until the previous burn is
complete. Fuel is not added continuously, thus the burns are a batch process. The storage tank
allows the Garn to bridge the gap between the batch process of burning a load of wood and the
continuous process of heating a building. Section 3.7 below offers a detailed description of the
implications of using stick fired boilers.
Figure 3.2 below is reproduced from Section 2, and shows the characteristics of the Garn boilers
included in the study.
Figure 3.2. Garn Boiler Characteristics
The material handling associated with stick fired boilers is much simpler than that associated with
chip fired boilers. In essence, the wood should be cut to length, and should fall within an
acceptable range of diameters. Larger diameter lengths may have to be split. Ash must be
removed from the burn chamber manually, and tube cleaning is manual as well.
The boilers must be fed manually regardless of how the wood is processed, but the actual
processing is a trade-off between simplicity and availability of equipment and more manual labor
versus more expensive, specialized equipment with correspondingly less manual labor. If
chainsaws and splitters are the primary material handling equipment, then a great deal of manual
labor is required to process the wood and feed the boilers. The up-side is that this equipment is
cheap, easy to fix (and quick to ship, if new is needed), and abundant in the villages. If
specialized harvesters and/or cutters are used, much of the manual labor is removed. However,
this equipment is expensive, cannot easily be replaced and there is likely only one of each per
village – so a failure means the operation is down until it is repaired. Each village in which a Garn
is installed will need to consider the associated material handling carefully in cooperation with
AWEA to determine the best solution for the village.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 21 - Final 20 Feb 2012
3.4 Supplemental Heat
There are conditions under which biomass alone or the combination of recovered heat and wood
heat may not be able to meet the heating load. It is possible to supply enough boiler capacity to
all but eliminate the need for oil (except in the case of biomass boiler failure, which is always
possible). However, it is not always the best answer financially.
In theory, one could pay someone to stand next to a Garn boiler and feed it continuously, which
would maximize its output. In reality, this is not practical; the study assumed a maximum of four
boiler “burns” a day. A burn is not necessarily a single loading and firing of the Garn; in this study
a burn means burning enough wood to raise the storage tank temperature from 120 deg F (min)
to 200 deg F (max) – in practice this might take two or more successive loads of wood in a row on
very cold days. Remember that heat is being extracted from the tank even as the heat from the
wood is charging it.
If the number of burns per day is limited, then it is possible (in fact common) that the expected
peak load of a building on the coldest days is slightly greater than the capacity of a given boiler,
or combination of boilers. In such a case, one could always either use the next larger boiler, or
add another to cover the peak loads. However, the economics of this are generally bad –
installing a very expensive boiler to run ten or so days a year makes no financial sense. It is
easier and less expensive to simply use the building’s existing oil boilers on such days to
supplement the biomass heat.
In the results summaries there is a value labeled “fraction of oil displaced”. A value of 1.000
means the biomass boiler has the potential to provide 100 percent of the building heat, with no
need for supplemental oil heat (in reality, there is maintenance downtime, so coverage will
probably never be 100 percent on an annual basis). In these cases, the peak load of the building
matched up well with the capacity of one or more of the stick-fired boilers. Often, however,
values such as 0.900 (90 percent) appear. In this case, the peak load is predicted to exceed the
capacity of the boilers for a very small number of hours each year – and thus supplemental heat
from the building oil boilers will be required.
3.5 Distribution
The heat generated by the external biomass boilers must be distributed to the various end-users.
Unless the biomass boiler is very, very close to the building (in which case the piping could be
above-ground), this is done by pumping hot water through buried distribution pipes to each
building. Traditionally, the piping used in this part of Alaska is a rigid system of pre-insulated
piping. A carrier pipe carries the fluid; this is standard steel piping. Rigid foam insulation
surrounds the carrier, and insulation in turn is protected by an outer spiral-wound metal jacket (or
casing). See Figure 3.3 below
Figure 3.3, Traditional “arctic pipe”
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 22 - Final 20 Feb 2012
This system provides superior heat loss characteristics (i.e. very low losses), but it is expensive,
and installation must be very well planned. It is expensive primarily because the whole piping
system is rigid. It must therefore be installed below the permafrost or frost heave will snap the
pipe – in many areas of interior Alaska this means 18 to 20 feet deep. The required trenching is
expensive, requires large equipment, and takes time.
Installation must be well planned out because the pipe is difficult to modify in the field. Cutting a
piece to length means cutting through all three layers; it is difficult to get a clean cut and
subsequent clean connection to the next piece. For that reason, the system is typically laid out in
great detail in the plans, and each piece and each fitting made for a specific spot in the system.
Thus any mistakes in fabrication or any damage to a piece in the field can take a long time to
repair.
As noted in Section 2, therefore, AWEA proposes to use a flexible plastic piping system,
manufactured by Rehau, which uses PEX carrier pipes. Figure 3.4 is reproduced from Section
2.3:
Figure 3.4, PEX piping
Because the system is flexible, it can be installed in the active layer of the soil – in the Plant
proposed for Fort Yukon, the proposed depth in is 48 to 60 inches deep. Further investigation will
be needed to determine the appropriate depth in Venetie. Connections are simple, so the layout
does not need to be planned in great detail. Because it comes in rolls, hundred of feet of piping
can be laid out in very little time. Trenches are shallow and simpler to construct, generally using
equipment that may already exist in the villages.
The most significant negative aspect of the PEX system as opposed to the traditional system is
that the insulation is not as effective. Piping losses are greater with the PEX system, and piping
losses can have a significant effect on ongoing operating costs. In applications with free or low
cost recovered heat, these operational savings can be leveraged on an ongoing basis to
counteract the increased piping losses, allowing Venetie to realize the first cost savings
associated with the PEX system without the additional piping losses excessively damaging the
project financials.
This would not be the first installation of the PEX system in rural Alaska; thousands of feet of this
type of piping were recently installed in McGrath in less than one week.
3.6 Integration of Recovered Heat
Diagrammatically, the integration of recovered heat into a DH Plant is shown in Section 2.2. The
concept is very simple. Water in the cooling loop from the operating engine(s) producing power
in the power plant (or at the school) is routed first to the biomass boiler plant.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 23 - Final 20 Feb 2012
In the Plant, the hot cooling water flows through a heat exchanger. A three-way control valve on
the cooling loop side of the exchanger controls how much heat is rejected to the boiler loop. If
the valve is wide open, all the flow goes through the heat exchanger.
A temperature sensor in the boiler storage tank is used to modulate the heat exchanger 3-way
valve, extracting as much heat as possible from the engine cooling loop. If this is not enough
heat for to meet the load, additional wood or oil heat will be added to primary loop as required. If
the load is less than the available recovered heat, the three-way valve will modulate as required
to maintain the heating loop at setpoint, and the excess heat is sent to the existing radiators.
On the generator cooling loop side, the water leaving the Plant, having flowed either through the
heat exchanger or through the valve bypass, will return to the power plant cooler than it left. It will
then flow to the engine radiators. If it is already cooler than the radiator setpoint temperature,
then the radiator fans will not come on – the water continues back to the engine jacket to start the
cycle over. If the water from the DH Plant is warmer than the radiator setpoint, the fans will
comes on as needed to cool the water, and send it back to the jacket.
3.7 Design Considerations for the use of Stick Fired Boilers
Some of the material below has also been presented above. The intent is for this section to be a
stand-alone description of some of the implications of using stick-fired boilers without reference to
the remainder of the report.
As the name implies, stick-fired boilers burn round or split wood in relatively straight pieces. The
wood is minimally processed, being selected for a range of diameters and trimmed only for
length. If the diameter of the wood is too large, the wood may be split. Although the processing
is minimal (compared to chipping), it is generally all done manually (some splitting may be done
with a machine). Nevertheless, at the assumed unit costs, stick-wood is the cheapest energy
source available to the villages for generating thermal energy.
However, utilizing stick-wood means that much of the available biomass cannot be used. Wood
that is too large or too small, smaller tops and limbs that are bent and/or tangled, or tops which
contain leaves, cones, or needles are generally too difficult to handle in a stick-fired boiler. The
burn chamber of the boilers (see figure 3.5 below) is designed for straight stick-wood of a given
length. The wood used is generally air-dried, not mechanically dried; mechanical drying would be
very expensive on such small scales.
The Garn stick-fired boilers used as the basis of evaluation for this study are scheduled above in
this section and in Section 2.3.
3.7.1 Sizing, Boiler Control, and Utilization Rate.
A primary feature of the Garn boiler is the built-in thermal storage. Physically, this is a large hot
water tank that surrounds the combustion chamber. Functionally, the tank “decouples” the burn
rate of the boiler from the actual heat load requirements. In essence, the process of combustion
heats the tank and the tank serves the load (through pumps and a piping system), but not at the
same rate. This is illustrated in figure 3.2 above. In the WHS 3200, for instance, the process of
combustion generates up to 925 KBTU/h. The storage tank can hold 2,064 KBTU. So, if the
“burn” lasts a little over two hours, it will completely charge the tank. If at the same time the
building heating load is 500 KBTU/h, it will take a little over four hours to deplete the tank – thus
the rate of charging the tank is decoupled from the rate of extraction (serving the heating load) by
the storage tank.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 24 - Final 20 Feb 2012
Heating a building is a continuous process, heating the tank in a Garn is a “batch” process. The
thermal storage tank bridges the gap between the continuous process of heating and the batch
process of burning. A batch process is one in which an event takes place at intervals – for
instance, every eight hours, one fills the Garn with a “batch” of wood, and burns it.
This decoupling effect eliminates the need for sophisticated combustion controls that would allow
the boiler to track the load; that is, to match the burn rate to the heating load. The boiler is
manually fed, and manual started. This results in a very simple boiler, which holds down first
cost. The primary control function of the Garn is combustion control – simply ensuring that the
combustion air is controlled such that the wood burns hot and clean.
The decoupling effect also means that sizing is less of an issue than it would be with a chip-fired
boiler. If more capacity is needed to meet load, the operator can simply conduct more “burns”
per day (within the limits described above). When less capacity is needed, fewer burns are
performed.
There are limits to this, of course. An operator would not want to have to feed the boiler once
every three hours round the clock, especially in the -50 deg F temperatures that can occur in the
interior of Alaska. In this study, the assumption was that if more than four burns per day were
required to meet peak heating load, a larger boiler would be used, or another boiler would be
added to the installation. Four burns per day imply a minimum of six hours between burns.
Adding another boiler increases the time between burns, but it adds significant cost as well.
In addition, the number of stick-fired boilers per installation is generally limited to three. Beyond
this limit, it is felt, the installations would get too large and too expensive. (None of the
combinations of boilers in Venetie would require three boilers.) Because of the thermal storage,
the boilers are quite large, and they require at a minimum a covered roof and flat slab floor;
ideally they would be completely enclosed. Equally important, the utilization rate of the
equipment drops as the number of boilers increases. If one boiler is adequate in “warm” weather,
two required for “cool” weather, and all three for “cold” weather, then the overall utilization rate of
the plant is probably no more than about one half (50 percent). Installing equipment in the interior
of Alaska is expensive; the higher the utilization rate, the more cost-effective the installation.
In the summer, heat loss from the tank may become a significant factor. The seasonal range of
heating loads in the interior of Alaska is the highest in the country. The heat load at -60 deg F is
20 times higher (or more) than the load at 80 deg F (when the load is probably only domestic hot
water). So a burn that only lasts six hours at peak theoretically lasts 120 hours in the summer.
Obviously, in 120 hours, more heat is going to be lost through the tank insulation than is going to
be actually used. It might therefore be more practical to run the existing oil-fired boilers when the
load drops too low. However, for the purposes of this study, although insulation losses were
accounted for, it was assumed that the Garn boilers were not turned off in summer.
3.7.2 End‐user Issues.
All of the facilities included in this study already exist (a possible exception is the new clinic, but
this study assumes it will get constructed before a biomass plant is constructed), thus any
installation of a wood-fired boiler would by necessity be a retrofit to an existing heating system.
The intent is that the boilers be installed in such a way as to be transparent to the end-user. That
is, the occupants cannot tell whether the heat is coming from the existing oil-fired equipment or
from the proposed wood-fired equipment. Moreover, the mechanical heating system must
operate the same way regardless of heat source; switching from one source to the other must be
as simple as opening and closing valves. Finally, the intent is that the systems will be installed in
such a way that a failure of a wood-fired boiler automatically starts the oil-fired back-up, and
ideally, notifies the operator of the failure.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 25 - Final 20 Feb 2012
The Garn boilers do have one major limitation in terms of end-user transparency; they cannot
control the hot water supply throughout the burn cycle. When a burn finishes, the storage tank is
at design temperature (200 deg F is the design temperature for Garn). However, as the hot water
is pumped through the heating system, it gives up heat to the space. As a result, when it gets
back to the tank it is colder than when it left – the difference between supply and return
temperature, called the delta T (or change in T) depends on the type of heating equipment (air
handling unit, baseboard heat, radiator, etc) and the heating load.
The cooler return water immediately begins to dilute the 200 deg F water, cooling it. Once the
burn is done, no more heat is being added to the tank, but heat is continuously being removed to
heat the space – thus the tank temperature falls throughout the tank’s “draw-down” cycle. Garn
considers the tank to be “depleted” when it reaches 120 deg F. The basis of the heat storage
capacities listed in figure 3.2 is the assumption that the tank is heated to 200 deg F, and then
heat is extracted until it reaches 120 deg F, at which time, another burn is initiated.
However, in a retrofit situation, 120 deg F hot water may not be suitable. Many hot water heating
systems are designed to use hot water at 180 deg F or more when at peak load. For instance,
the heating coil in an air handling unit may have been sized to provide the required peak heating
output using 180 deg F supply water (180 deg F is a very common coil temperature). In such a
case, with the heating load at or near peak, the Garn boiler will be able to meet load as long as
the storage tank temperature equals or exceeds 180 deg F, but as it falls below that value, the air
handling unit may no longer be able to meet the load. By the time the supply temperature falls to
120 deg F, the air handling unit will be operating significantly below design capacity.
As a specific example, assume an air handling unit that requires 180 deg F supply water at peak
load (i.e., there was no spare capacity in the coil at peak load), then in effect the storage capacity
of the Garn boiler would be reduced by 3/4: 1 - [(200 – 180) / (200 – 120)] = 0.75. The WHS
3200 that has been used as an example above would have a storage capacity of only 516 kBTU,
rather than 2,064 kBTU. At the same time, the time between burns would also be cut to 1/4 of
the calculated time, although each “burn” would be much shorter, since the burn only had to raise
the temperature of the tank by 20 deg F.
Practically speaking, most heating systems use hot water in the 140 deg F – 200 deg F range.
Only radiant floor systems typically use hot water as low as 120 deg F. Thus, in almost all cases,
it would appear that the storage capacity of the Garn units would have to be de-rated. This study
did not de-rate capacity of the Garn for three reasons: 1) there was not sufficient time to survey
all the existing equipment, and related drawings and specs, to determine the design supply
temperatures, and 2) In all cases, at load conditions not at or close to peak, 120 deg F water may
suffice – thus the number of hours per year when the de-rate would be applied may be quite
small (however, this is when the weather is coldest, and the most labor is required to maintain the
fuel supply and burn rate), and 3) direct observations of systems in Alaska have shown that many
are so over-sized that they operate even with low hot water supply temperatures. It was therefore
assumed that the specified storage capacity could be used in full; before any installation design is
finalized, however, this assumption should be confirmed.
In Venetie, there is one system of specific concern – the clothes dryers in the washeteria. The
inlet air for these dryers is heated by hot water coils served by the existing oil-fired boilers.
Drying clothes is a function of not just BTUs, but also of air temperature. During our site visit, the
boiler was set for 160 deg F supply water. If this the temperature required for the clothes dryers,
then a Garn boiler would not be a good application to serve the dryers. It is likely that a small
dedicated boiler would be required to serve very hot water to the dryers as needed. This would
alter the economics of serving this building with heat from wood, but the effect is not likely to be
large enough to make it infeasible – the dryers are not large and do not run continuously.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 26 - Final 20 Feb 2012
3.7.3 Material Handling.
As noted above, the Garn boilers are manually fed. For each burn, the operator must clean out
any ash as needed, load the combustion chamber with new stick-wood, and manually start the
fire. Once the fire is lit, the chamber door is shut, and the fire burns until all the fuel is consumed.
However, as noted above, it would take a little over two hours of burn to fully heat the storage
tank (using the WS 3200 as an example again). A single load of wood will not burn for two hours,
meaning that each burn must consist of more than one load of wood. In addition, during that time
that the burn is taking place, heat is being extracted from tank to meet the heating load. So
although a “burn” is treated as a single event in this study, it is important to note that at or near
peak load, a burn could take as long as three hours to complete, and require two to three
“reloads” of the combustion chamber. (A complete burn is defined herein as burning enough fuel
to raise the storage tank from 120 deg F to 200 deg F, even as heat is being extracted from the
tank for ongoing heating.) Thus although the number of burns is limited (in this study) to no more
than four a day, this could still imply roughly 9 - 10 hours a day of loading and cleaning the
combustion chamber in cold weather.
As with any stick-fired appliance, the fuel should be kept dry, and should be located close to the
point of use. Therefore, any building or structure constructed to house the boiler should have
sufficient space to stack cord wood. The amount of wood to be stored within the building (as
opposed to in a wood yard) depends on the site conditions. In harsh conditions, it may be
desirable to store several days’ worth of cord wood (at peak load consumption rate) in the boiler
building, in case weather keeps the operator from being able to re-stock the building from the
wood yard. On the other hand, in all cases the existing oil-fired system is assumed to be left in
place as back-up, so this may limit how much wood the operator chooses to store in the boiler
building.
Regardless of how much wood is stored in the boiler building, considerable manual labor would
be required to get the sticks from the wood yard to the building; labor to load, unload, and stack
the wood. Because no equipment is required once the stick-wood reaches the yard, the material
handling, though labor intensive, is not subject to equipment breakdowns.
There are a number of options for discarding the ash. It is likely the ash would be collected in a
small bin or dumpster, and emptied only as this gets full.
3.7.4 Emissions Controls/Efficiency.
Garn boilers have no active emissions controls. The boiler uses an induced draft (ID) fan to
ensure that enough air is present to provide complete combustion. This alone helps eliminate or
mitigate many emissions. It helps prevents the formation of carbon monoxide (CO), which forms
as a result of incomplete combustion. It minimizes smoke and particulates (but cannot extract
any particulate formed and emitted), by burning clean and hot thus leaving very little behind but
incombustible ash.
Using more air than is strictly necessary simplifies the control, and makes for a clean burn, but it
also reduces efficiency. Excess air cools the boiler down as it enters and brings additional
moisture with it, both of which require excess heat to bring them up to combustion temperature.
The Garn does provide good transfer from the stack gas to the hot water storage tank. The stack
gas essentially passes through the tank five times (a five-pass heat exchanger); see Figure 3.5
below. There are four horizontal passes and one vertical pass. Overall, the efficiency of the
Garn is good – the study assumed 75 percent of the net useable heat content of the wood was
transferred into the tank.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 27 - Final 20 Feb 2012
3.7.5 Maintenance.
There is very little maintenance required on a Garn boiler, and in fact, there is not much that an
operator could do. Figure 3.5 below shows a cross section of a Garn boiler. The wood is burned
in the primary combustion chamber, “E”. In the secondary combustion chamber, “F”, only hot
gases are burned. As long is the ash is removed from “E” before each burn, there is not much to
maintain. The ID fan (“H”) must be repaired or replaced if it fails.
Figure 3.5 also shows the “tubes” that transfer heat to the storage tank. The tubes (from the end
of “F” through the end of “I”), must be cleaned; if not, then any scaling or fouling of the tubes is
not removed, and these will gradually erode the efficiency of the boiler (or even cause the tubes
to fail). The tubes can generally be cleaned by running a wire brush through them. The
frequency of cleaning depends in part on how clean the wood is; clean forest wood should have
no inclusions, while scrap and construction debris often do. If these inclusions (adhesives,
preservatives, etc) do not burn completely, they often plate out on the tubes, degrading
performance.
Between cleanings, efficiency will slowly degrade as deposits accumulate, until the next cleaning.
Figure 3.5 also shows that the boiler has two vertical tubes sections – these are more difficult to
clean. All feeding, de-ashing, and cleaning are manual. The timing of the feeding is manual,
although that could be partly automated (i.e., a control system could alert the operator when the
tank is nearly depleted).
Figure 3.5, Cross Section Through a Garn Boiler
3.8 Site Issue.
As noted above, the Garn boilers are quite large as a consequence of the storage tank. The
WHS 3200, the largest Garn boiler considered in this study, is 14’ – 3” long, 7’ -2” wide, and 7’ –
9” high. Each unit (full of water) weighs 34,500 lb. The largest Garn plant “allowed” in this study
would include three WHS 3200 units (we arbitrarily limited the number of boilers per Plant to
three, but no Plant at Venetie required three boilers). Not including interior wood storage (but
including clearance around each unit), this would require a minimum of 678 square feet (20’ – 3”
long by 33’ – 6” wide) with an average floor loading of 153 lb/sf. This floor loading will likely
require a relatively thick reinforced concrete slab to prevent differential settlement. However, the
issue of buildings and concrete slabs are not necessary if a “Garn in a Box” method of installation
is used. This is the recommended alterative.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 28 - Final 20 Feb 2012
The recommended site is shown in Appendix C with all of the buildings to be heated labeled. The
Venetie Village Council who also operates the power plant owns the area behind and to the west
of the current power plant. This area is a large enough site to house both the dual Garn boiler
plant and a wood yard large enough to store over half of the annual wood supply needed to
operate the plant. This site makes heat recovery into the DH plant and distribution
straightforward with minimal distances to each building.
SECTION 4
4.1 Limits
As with any performance evaluations, the quality and validity of the outputs and subsequent
conclusions depends on the quality of the inputs and the methodology. Methodology is discussed
in Section 4.2 below. The input data gathered for used in the analyses performed as a part of this
study include:
• Building specific data
• Historical village PCE (electrical consumption) data
• Proposed equipment data
• Annual oil consumption, by building
• Annualized weather data (bin data) from the Fort Yukon airport
• Village maps and plans
• Interviews with Civil Engineers, contractors, and consultants with significant experience in
the interior of Alaska
• Pricing data from boiler manufacturers, piping suppliers and other AK vendors
• Performance data from Garn
• Input from other Alaska Wood Energy Associate team members
What was not performed a part of this analysis was detailed measurements of building heat loads
and existing equipment performance. A building heating load profile is central to predicting
annual fuel consumption. Ideally, this would be generated by directly measuring heating load
throughout the year. At the same time, the actual operating efficiency of the existing boilers and
distribution system would be measured. This would provide a highly detailed profile of heating
load and the energy required to meet that load, for any condition throughout the year.
In practical terms, however, the required measurements are difficult to perform, and not cost-
effective. The equipment needed to make these measurements is not present at any of the
installations in the village, and would have to be flown in and installed. The measurements would
need to continue from winter to summer, to generate a complete load profile. The resulting
incremental increase in the accuracy of the load profile cannot justify that level of cost. As
Section 4.2 explains, even without the measurements, the data that were collected limit the load
profiles to within a narrow range of values.
Also missing from this analysis is a detailed analysis of electrical load profiles (and thus available
recovered heat). The only data available was monthly. Ideally, at least some hourly or daily data
would be included in any Level 3 analysis.
4.2 Methodology
4.2.1 Energy Savings.
The performance of the existing and proposed heating systems was modeled using
spreadsheets; the type of model used is known as a “bin model”. In this case, the bins are
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 29 - Final 20 Feb 2012
ranges of outside air temperatures (OATs). Temperature bins are used because heating load is
very closely correlated to OAT. Each “bin” of OAT is 2 deg F wide – for instance, 40 – 42 deg F
is a bin, with the midpoint temperature of 41 deg F. For each OAT bin, the heating load profile
assigns a heating load to that temperature bin. The actual “bin data” is the number of hours per
year that the outside air temperature falls into each specific bin.
Bin weather data is published for numerous sites, including many in Alaska. However, no such
data are published for Venetie. Therefore, actual hourly temperature data from the Fort Yukon
airport was used to construct a bin table for this site. The weather data came from calendar
years 2007, 2008, 2009 and 2010. The data were combined to come up the average number of
hours per year that the OAT in the Yukon Flats area falls into each bin. This was done a on a
monthly basis – for instance, the data in Figure 4.1 shows the spread of temperatures in the first
thirteen bins; from the 85 deg F bin to the 61 deg F bin (each 2 deg F bin is labeled by the
midpoint temperature).
Figure 4.1. Fort Yukon Airport Bin Data
In the calculations performed within the model, individual calculations are completed in a series of
tables that have the same format at the original bin temperature data (see Appendix A for sample
calculations). Figure 4.2 below shows a portion of a calculation used in this study (in this case,
the calculations involved with the School in Venetie).
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 30 - Final 20 Feb 2012
Figure 4.2. Partial Bin Model, Venetie School
The basis of all calculations is the heating load profiles for each building included in the study.
These load profiles reflect all the information available about the building, such as heating
equipment capacity, operating data, historical oil consumption data, size and type of building, etc.
As noted above, it is difficult to measure heat load directly, but simple to measure oil
consumption. So the first profiles generated are oil consumption profiles. These profiles assign a
specific oil consumption rate to each OAT bin. Using the calculation format above, the model
calculates the amount of oil required to heat each building, and then compares that to known
consumption – obviously, the model must be able to back-predict known consumption if it is to be
used to predict future consumption.
Subsequent calculations are done to determine how much oil/wood/recovered heat is required to
meet the predicted heating load – one table for each energy source (complex rules determine
which source is the primary, secondary, or tertiary source in each load condition). Once the
required oil (for instance) is calculated for each spot in the table, the SUMPRODUCT function
sums up all the oil required for each month. This is the basic format of the calculations.
Once the oil consumption profiles are verified, the oil consumption profile is converted to a space
heating load profile, by multiplying BTU of input heat (oil) times the efficiency of the boiler/furnace
to arrive at the actual heat to the space. Once these space load profiles are generated, they are
fixed. A visual example of oil consumption versus OAT load profile is shown in Figure 4.3 below.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 31 - Final 20 Feb 2012
Figure 4.3. Oil Consumption versus OAT for Fort Yukon School
No matter what heat source is used, or how great any parasitic or piping losses are, any
proposed mini plant or biomass boiler must at the end of the day deliver that same amount of
BTUs to the space as the current oil-fired appliance does. Once the space load profiles are
established, the spreadsheet models the various mini plants and boilers to determine how much
energy they would consume to provide this required space heat.
As noted above, in addition to producing a set amount of BTUs for space heat, a mini plant must
produce enough additional BTUs to heat the plant itself (parasitic loss) and to overcome the heat
lost from the distribution piping into the ground (piping losses). Finally, the model must calculate
how much pumping (electrical) energy must be used to get the heat to the buildings. Once inside
the buildings, the electrical energy used for pumping is the same for the existing systems as it
would be with a mini Plant in place, so this energy is not calculated or accounted for in the model.
Additional key load profile assumptions:
• Space heating load varies linearly with OAT (a 10 deg F drop in OAT results in twice the
increase in load that a 5 deg F drop causes)
• There is an OAT at which space heating stops – the OAT combined with the internal loads in
the building (people, lights, equipment) are such that no additional heat is required; above
this temperature, the only load is DHW. This value can be set individually for each building.
• However, in many buildings, there is heating required above this point, for heating domestic
hot water. This DHW heating value can also be set individually for each building.
Figure 4.4 below shows the table where the user inputs there values:
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 32 - Final 20 Feb 2012
Figure 4.4. Building Characteristics
The column “min as a % of max” has to do with the space heating load. A value of 0.020 for
instance, means that at the temperature at which space heat begins (the last column), the space
heating load at that temperature is 2.0 percent of the maximum space heating load. If the heat
goes to a process (such as heating village water to keep it from freezing), this value can be set to
1.000 – in other words, the load is the same at all temperatures. The column “% of peak to DHW”
is the percent of peak heating load assigned to heating domestic hot water. Some buildings use
their boilers to heat DHW, and some have separate heaters. If the boiler heats the DHW, this
column is an attempt to estimate how much of the boiler heat goes to DHW.
4.2.2 Recovered Heat.
Just as with heating loads, a “recovered heat” profile is generated. This profile assigns a specific
village power requirement to each temperature bin. This is less straightforward than assigning
heat loads versus OAT, because the correlation between village power and OAT is not as strong
– there is also a time-of-day component to power output. However, bin models predict long term
average performance, not hour-by-hour performance. A bin model that predicts consumption
accurately on a monthly basis is generally as specific as is required – most utilities bill and/or
report consumption on a monthly basis
A preliminary temperature versus power curve was developed for Venetie which accurately
predicts monthly consumption. However, in a Level 3 analysis, this relationship would need to be
better understood and defined.
4.2.3 Cost Estimates.
The other component required to calculate the payback of any given scenario is the project cost.
One estimate was prepared for each proposed mini plant and individual boiler installation. The
current cost estimates contains the best knowledge of the team members regarding construction
in bush Alaska. As the study level proceeds, these costs estimates will be constantly updated to
reflect the current design. Ultimately, actual vendor quotes and contractor’s estimates or bids will
be used for the final cost estimates.
4.3 Business Structure
A sustainable business structure is critical for a successful project. This report recommends that
the business structure be based on a for-profit model. That is the District Heating Utility takes in
more funds than it expends and it sets up a fund that will cover the costs of equipment
maintenance and replacement.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 33 - Final 20 Feb 2012
The NSP calculation in the feasibility model of demonstrates the potential amount of savings
($208,413 annually) that are available at a cost of $300 per cord fed into the boiler displacing
$8.50 per gallon diesel fuel. We estimate approximately $50 per cord of labor to cut the wood
into appropriate sized sticks and to feed the boiler. So that leaves a potential to pay $250 for a
delivered cord of wood. The model also assumes that the boilers are fed up to four times per day
on extremely cold days and they are fed seven days a week when needed. That is the only way
the maximum amount of fuel oil can be displaced. Any time the optimum amount of wood based
on the demand of the system is not fed into the boiler, then a less than optimum amount of fuel is
displaced and the business revenues are less than optimum. However, a realistic expectation
would be to capture 90% of the optimum values as there will be times when each buildings back
up heat will be used.
The feasibility analysis demonstrates that a for profit heat utility is fully feasible. The basic
benefits include almost doubling the efficiency in the use of diesel generators for making
electricity and heating buildings, keeping $208,413 in the community for jobs and infrastructure
maintenance and stabilizing the cost of heating fuel for primary public buildings.
The most practical business structure, since the VCC owns the power plant and the heat
recovered from that facility is a key part of the of the District Heating System, is for the VCC to
own the heat system as well. There are some business synergies that make a lot of sense. The
powerhouse operator could easily pick up the responsibility of managing the boiler plant. This
does not mean that person will be the only one to feed the boiler; but that the person is
responsible to make sure that the boiler is fed. Feeding the boiler could be part of a wood
gathering contract or be contracted separately. Another way of operating the plant itself would be
to contract that as well.
Billing of customers will be based on BTU meters in each connected building just as the electric
meters are used. This can be done in a prepay manner or as you go. Both ways the meters
must be read and billing must occur. The same office that does the electric billing could
accomplish the heat billing and bookkeeping.
Wood delivery can be accomplished in several ways. Probably the most effective would be for
the VCC to contract for wood delivery with key wood gathers as well as be willing to buy wood
from independent wood haulers on an as needed basis. A prime contractor would lease any
procured harvest equipment from the VCC at a minimal cost, but with key issues of insurance and
maintenance spelled out in the contract and lease.
The discussion above is meant to start the process of creating a business structure. Once a
decision is made to move forward with a biomass project and the VCC has worked out a general
business structure a full-blown business plan will need to be developed with the final design
process, costs and structures.
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 34 - Final 20 Feb 2012
Appendix A. Wildland Fire History Near Venetie
Fire history from 1950-2004 around Venetie
!
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 35 - Final 20 Feb 2012
Appendix B. Photos of Harvest Equipment
Tractor pulling a log trailer with a grapple arm for loading
Self-loading log trailer
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 36 - Final 20 Feb 2012
Tractor with a harvesting head for cutting, limbing and cut to length
Forcat with self-loading trailer for transporting wood to the village
Example of an armored articulated tractor being used in Fort Yukon
BIOMASS HEATING FEASIBILITY Level 2 Study
Venetie, Alaska Final
Alaska Wood Energy Associates - 37 - Final 20 Feb 2012
Appendix C. Boiler Site and Proposed Heated Buildings