Loading...
The URL can be used to link to this page
Your browser does not support the video tag.
Home
My WebLink
About
Technical & Economic Feasibilty of Applying used EV Batteries in Stationary Applications 2003
SAND REPORT SAND2002-4084 Unlimited Release , 003 | -easibility of Applying onary Applications Storage Systems Program Erin Cready, John Lippert, Josh Pihl, Inwin Weinstock, Phillip Symons, and Rudolph G. Jungst Prepared by Sandia National Laboratories Albuquerque, New Mexico 87185 and Livermore, California 94550 Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under Contract DE-AC04-94-AL85000. Approved for public release; further dissemination unlimited. (fh) Sandia National Laboratories Issued by Sandia National Laboratories, operated for the United States Department of Energy by Sandia Corporation. NOTICE: This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government, nor any agency thereof, nor any of their employees, nor any of their contractors, subcontractors, or their employces, make any warranty, express or implied, or assume any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represent that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government, any agency thereof, or any of their contractors or subcontractors. The views and opinions expressed herein do not necessarily state or reflect those of the United States Government, any agency thereof, or any of their contractors. Printed in the United States of America. This report has been reproduced directly from the best available copy. Available to DOE and DOE contractors from U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831 Telephone: (865)576-8401 Facsimile: (865)576-5728 E-Mail: reports@adonis.osti.gov Online ordering: http://www.doe.gov/bridge Available to the public from U.S. Department of Commerce National Technical Information Service 5285 Port Royal Rd Springfield, VA 22161 Telephone: — (800)553-6847 Facsimile: (703 )605-6900 E-Mail: orders@ntis.fedworld.gov Online order: http://www.ntis.gov/help/ordermethods.asp?loc=7-4-0#onlinc SAND2002-4084 Unlimited Release Printed March 2003 Final Report Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications A Study for the DOE Energy Storage Systems Program Erin Cready, John Lippert, Josh Pihl, and Irwin Weinstock Sentech, Inc. 4733 Bethesda Avenue Bethesda, MD 20814 Phillip Symons Electrochemical Engineering Consultants, Inc. 1295 Kelly Park Circle Morgan Hill, CA 95037 Sentech Report Number: 068-006 Sandia Contract No. 20605 Abstract The technical and economic feasibility of applying used electric vehicle (EV) batteries in stationary applications was evaluated in this study. In addition to identifying possible barriers to EV battery reuse, steps needed to prepare the used EV batteries for a second application were also considered. Costs of acquiring, testing, and reconfiguring the used EV batteries were estimated. Eight potential stationary applications were identified and described in terms of power, energy, and duty cycle requirements. Costs for assembly and operation of battery energy storage systems to meet the requirements of these stationary applications were also estimated by extrapolating available data on existing systems. The calculated life cycle cost of a battery energy storage system designed for each application was then compared to the expected economic benefit to determine the economic feasibility. Four of the eight applications were found to be at least possible candidates for economically viable reuse of EV batteries. These were transmission support, light commercial load following, residential load following, and distributed node telecommunications backup power. There were no major technical barriers found, however further study is recommended to better characterize the performance and life of used EV batteries before design and testing of prototype battery systems. 3 Preface This study was funded by the U.S. Department of Energy through the Energy Storage Systems Program. The DOE Program Manager is Dr. Imre Gyuk. Sentech, Inc. carried out the work under Contract 20605 with Sandia National Laboratories. The Sandia technical contact is Rudolph G. Jungst, Long-Life Power Sources Department 2525. Sentech, Inc. contributors were Erin Cready, John Lippert, Josh Pihl, and Irwin Weinstock. Phillip Symons of Electrochemical Engineering Consultants, Inc. developed information regarding the potential stationary use applications. CONTENTS 1. EXECUTIVE SUMMARY .........cccssssssssssssssssessesssscssssssesessesecsesscsssesseneasessesesseeassseaseressessees ll 2. INTRODUCTION. .........cccsccssssscessccesscsesscsessssesesecsessesessseesessesscsesessesesseseesessseesesseseseseesesenses 15 D2 AODJSCH Ve Ob tis StUGY ct crecccctctctitecctsccticctacttadttadedcedsocdaceddcceatebuvsbucebuoseittetttnttittecetadstae 15 DIDTEDDSMMIGONS iti te sets c etseerette tbat ibegta tenia ttidgdad atdadsdoa tad dceuct coedccb cette eei tethers sigdatte 16 2.3. Status of EV Technologies and Markets............scsesscscscesecsececeseeeeeeeescescescesceasenseeeees 16 2.4. Present Markets for Used Batteries j2....sc2.o so ssccisccstcsccccsccascescasssusessssssossessccsosssesnesosseeses 17 25, ANI USABG 2) Uice Study ccc ere cee eee ect cet rede 20 DQISeL tility load FOU Owing aces tree e ct stcetecdccceseseceseveecsceceocceseees ooetecteceitiecere 21 2.5.2. Frequency regulation & spinning reserve ........ cece eseeseeeeeeeeeeeeeeeeeeseeeeenee 21 2.5.3. Uninterruptible power source and telecommunications applications ............. 23 2.5.4. Accelerated calendar life test. 24 2.5.5. Conclusions 24 3. A PROCESS FOR RE-APPLYING USED EV BATTERIES IN STATIONARY APPLICATIONS ........ccsscssscssssescesescccesecceseceececescececcceceaseceacescesesesseseeseaceceseacescacecesesceeesensees 25 OV ELVICW coerce ster tecenersesecetece ener ccrccc etc eece enters tee ee oa dae cae toe eaen ear coe doe cee geeeeee se neers 25 3 2) Acquiring Used EB ViBattetesiin tcretctistcticicdstesterddsdessisacssstecbisot osseeree cisbatstte 25 Pt CCR ee ee PC Rl icc 26 3.4. Reconfiguring Used EV Batteries for Stationary Applications .............c cece eee 27 3.5. Summary of Refurbishing Process............ccscsseesesesseseeseseeseseeseseeessenecseseeseseesesseeeeenees 29 4. APPLICATIONS 4.1. Transmission support 4.2. Area regulation & spinning reserve ............:ccceeeee 32 4.3. Load leveling/energy arbitrage/transmission deferral . 33 4.4. Renewables firming ocic.c.ccccccccecsccsecsocssssessosscsstsesrestors 4.5. Power reliability & peak shaving 4.6. Light commercial load follOWING........ eee cseeseesscseeeeessctseseeeesseseesecseeeseeseeseeeeeees 35 4.7. Distributed node telecom backup POWEL.........:.cccccscessesesseeesseeeseeeeseeseseseesessscescsseeeeeeees 36 4S EGResidential-load 1OMOWwine clic stedececeelcoscesressosecteoeteeccecetreit sttcedtatieacadscacedessysseuostccs- cose 36 5. ECONOMIC FEASIBILITY —- ASSUMPTIONS & METHODOLOGY........cccccsseseseee 37 5.1. Availability and cost of used EV batteries...........-.:....s.:ss;seseesousstceuescsersensssverensesesoesers 37 2-1 COMeCtON Gc transpOrtatlOns:eg-ssstases:coxuccescoseccs spore secre cite ree tag ie dedua aa scalatagaeseaes seoactees 38 Dio Ne@StHMG Ge TePACK AGING sttlacsee secede sicssccseccesscrecrssrotssccbutesecorsosteestecseececesscadescesccuse (rospeace 38 Side Capital COStS ta itecsscscesesassest losssseseosetosteces srcersecen crests teieeiesaeaetedteat deere descesssseduced 38 Final Report 5 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 5.3.2. Direct Costs.............ccsesccsscccsssssccesscescccsscccesseccsssncecssscceesssessseesesseeeesessscees 5.3.3. Indirect Costs 5.3.4. Earnings, taxes, and determination of reconfigured battery selling price....... 43 5.3.5. Cost breakdown ........cccccccesescesesseeseseseseesesceecesesessesacsesessesessesscsscsssaecsaeeseeeeaesenss 44 5.4. Energy storage system life Cycle COStS........cccccccesesceeeseeeseeseeeseseescsessetscesecneeeeeeeseeenes 45 5.4.1. Application requirements & system characteristics .......cccceseeeeseeeeeseesseeeeee 45 5.4.2. Duty cycle and projected life occ eeeeesecescesceeceecseeeeeeesaceseersaeeesaeeesseees 46 5.4.3. Battery COStS.......ccccceceessesessesesssescsesesesscseesesecsesessesessecsesscsesscseesesseseeaeeaeeaeeensens 46 5.4.4. Balance Of SysteM COStS ........cccccecsceseseeeeseseeseeesceecsecsesscseecseesesseseeseeeeeeseesesease 46 5.4.5. Operating and maintenance COStS..........cccceceseseeseeesessesesesescseseeeeseseesesesaeeeeseesees 47 5.4.6. Total System COStS ......cccsccesesesseseseseseeesesesesesesesssesesesecsesssesesescecsseseseeseaeeaeeasaes 48 5.5. Feasibility categories ...........ccccceceesesssseeesesesescseseseesesesssesecsesesesesseecscasseseecseseeaeeeeaeeaees 50 6. ECONOMIC FEASIBILITY — RESULTS OF ANALYSIS .........cccscssssssstecscsssessssesenees 50 6.1. Nickel/Metal Hydride 0.0... eee cscseseseescseeecseeecsesseseeseeaeseeaeseeaeeeaseeeeeeseseeeseeecees 51 6.1.1. Transmission SUpport ...........cceccsesesecesesescseseeescsesesesecsescseseseesescscseseraeeeeeteteeees 52 6.1.2. Area regulation & SpinMing FeSeTVE «0.0... cc ccecsceceeeeseseseesceeesceeeseseeaeeeeeeaeeaees 53 6.1.3. Load leveling/energy arbitrage/transmission deferral .............ccceeeeeeeeeeeeee 54 6.1.4. Renewables firming «0.0.0.0... ccc cece eseesceeeeceeesseesecsessecaeeseesecsesseeaeeeeeateneees 56 6.1.5. Power reliability & peak shaving... eee ececeeeeeeeeceeeeeeeeeeeeeeseneeseeseees 57 6.1.6. Light commercial load follOWIN...... eee eseeeeeeeseeseseeeeeeeeeeeeeeseseeseseestsaeaes 59 6.1.7. Distributed node telecommunications standby power ..........:e:csceeessesseseeseeeee 60 6.1.8. Residential load following 00... eee cece cseeeceecsececseeecseesesesetarseeeesesteaesees 62 6.2. Other Chemistries ...........c.cccececesesesesessseeeseecsescseseseesesescsesececsesesssesseseeeseseseseasacaeaeaseesees 63 7. ISSUES AND BARRIERS .......cccsssccsssssssssssessssssssssssssssscssssessssssssssssssssssssesesesssesesecesasacacers 67 7A. 7.2. 8. CONCLUSIONS........ 9. RECOMMENDATIONG.......ccccssssssssscerssesesncncsesesecesssescesssesesesceseseseeescesesssesesesesnsssssessseseses 71 9.1. Used EV battery module testing .........ccccecceceecscesececsecsesecsecscsecscseesseseseseneeeetetataees 71 9.2. Transmission support............... 9.3. Light commercial load following 9.4. Residential load following.............. 74 9.5. Distributed node telecOMmM ..........ccceescesssseeseeeseeeeeeeeeeseeeeseceseceseeneeeaeeeseeaeeeeseeatseescseees 76 Final Report 6 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications APPENDIX A uu... cecsssessssesecceseseseccscscesecseseccessneaseneasesecsaceseasencasecseseseneeseaseccaseneesecsasesseseceseneeseees A-l APPENDIX B uu... eccsssesesesesesesescsssesessescesesscesesesessssesesesesesecesesescaseneseseseseaeeneseseaeacaseseceseseacassesensets B-1 TABLES Table 1. Battery module characteristics for typical HEV and EV batteries. 00.0.0... 19 Table 2. Battery module characteristics for typical CEV and NEV batteries. .......... cece 20 Table 3. | Exide UPS warranty conditions and ANL test results ...........c:.cccsesseesceseseeseseeeeseeeeeees 23 DF ae 4 | Cap ta COS eee caer secu sebckdekebdetslssetdsdstecsuetabeseseactestesterusesbdenselaraeu 39 Table!) | HEMpIOYMIENE COs tS ccc ecceeeeena ses dstct detendetssuslersuoetersnssledstestctaeny 41 Table 6. Testing and repackaging COStS 0.0... ceceeeesesseseeseeseeseesseseesceseeseesecsecsecsecsecsecseeeseeeeeeeease 44 Table 7. BOS and O&M costs for analogous battery SysteMS.............:cceceeseeeeeeseeseeeeeseeseeeees 48 Table 8. Life cycle cost components for stationary energy storage systems based on used EV Table 9. Table 10. Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. Figure 9. Figure 10. Figure 11. Figure 13. Figure 14. Figure 15. Figure 16. Figure 17. Final Report Datterd es tee celle ata tatecseosatahdeseuderastseutstatesssbuesucderensattatsccedatesestessesderesterdcsabsssserasteed 49 Analysis of used Ni/MH EV batteries in utility applications «0.0.00... eee eeeseeeeeees 51 Analysis of used Ni/MH EV batteries in commercial & residential applications ...... 57 FIGURES Power profile for the load follow test conducted by PG&E for the PM250 power TO eer sedebeesatensatncdeccederesarsnduttsalatdabsassbaussbsnesstet seaue sorsesesucaesesdereedeneensasensssene 21 Utility Energy Storage (UES) cycle developed by SNL..u.... eee eseeseeseeseeseeseees 22 Frequency regulation subcycle of UES test profile ...........c cece cseeseeseeeeeeeseeeeeeees 23 Conceptual illustration of future life prediction... cece esscssesecsececeeeseseeeseeeee 27 Battery CONVErSION PLOCeSS .......eeeeeeeesesseseesseseeseeseeseeseesecseessesecseesscscsecsecsceseseeeeaees 29 Hypothetical used EV battery testing & packaging facility layout.......0... cee 42 Cost breakdown for reconfigured EV batteries ...........cccceceeseeseeeeseeseeseseeeseseeeeeeseesees 45 Life cycle cost as a function of battery life and buy-down for transmission support. 52 Primary cost drivers for transmission SUPPOMt ...........c ec eeseeeseesseseeeeseeseeseeseeeessetsetsees DD Life cycle cost for Ni/MH in area regulation and spinning reserve.............:eeeeeee 53 Primary cost drivers for area regulation & spinning reserve ............cceeceseeeseeeeeeeeees 54 Life cycle cost curves for load leveling/energy arbitrage/transmission deferral ........ 55 Life cycle cost curves for renewables firming ............ cc eeeeesceeeeeseeseeeeeeeeseeeeeeeeseeeees 56 Primary cost drivers for renewables firming ............cccceseeeeseeeeeseeeeseeeeseeesteetaeeeesees 56 Life cycle cost curves for power reliability and peak shaving...........cccccceseeeseeeeees 58 Primary cost drivers for power reliability and peak shaving ............:.:ccseseeseseeseseesees 58 7 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Figure 18. Figure 19. Figure 20. Figure 21. Figure 22. Figure 23. Figure 24. Figure 25. Final Report Life cycle curves for Ni/MH in light commercial applications «0.0.0.0... cecceeeeeseeeeseeee 59 Primary cost drivers for light comm. load follOWiNg............ccceseceeeeseeeeeeeeteeeeeeneee 59 Life cycle cost curves for distributed node telecom standby power compared to NRLA batteries lasting CWO Years eee anne tatastssacteedenetsetantendensotsedseastdelterdetdatd ona 61 Life cycle cost curves for distributed node telecom standby power compared to VRLA batteries lasting five years......... Life cycle cost of Ni/MH in residential application. .............cccccceeseseeseeeeeeeeeeeeeeeeeeee 62 Primary cost drivers for residential load following ............:.ccseeseeseeeseeeeeeeteeeeeeeeeeee 63 Life cycle cost curves for a generic 100Ah 12V EV battery that experiences a 10% capacity fade over its entire life... cecesescsseseseseseseseeeseeceesneeeseseseseseseeeessseseeeeeeees 65 Life cycle cost curves for a generic 100Ah 12V EV battery that experiences a 30% capacity fade over its entire Tife ott eee ee el detattactetletde 65 8 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Abbreviations and Acronyms Units of measure A amperes (amps) Ah amp-hours Vv volts WwW watts kw kilowatts MW megawatts Wh watt-hours KWh __ kilowatt-hours MWh __ megawatt-hours Hz hertz (cycles/second) kg kilogram ft. foot sec seconds yr year Vehicle and battery terminology EV electric vehicle HEV hybrid electric vehicle CEV _ city electric vehicle NEV neighborhood electric vehicle ZEV zero emission vehicle FLA flooded lead-acid VRLA _ valve-regulated lead-acid NiCd __ nickel/cadmium Ni/MH _ nickel/metal hydride Li-ion lithium-ion LMPB _ lithium metal-polymer OCV open-circuit voltage SOC _ state of charge DOD depth of discharge DST dynamic stress test AC alternating current DC direct current OEM _ original equipment manufacturer O&M operating and maintenance BOS balance of system HVAC heating, ventilation, and air conditioning UPS _ uninterruptible power system Organizations DOE __ Department of Energy SNL Sandia National Laboratories ANL Argonne National Laboratory USABC United States Advanced Battery Consortium CARB California Air Resources Board EPRI Electric Power Research Institute PG&E Pacific Gas & Electric OBC Ovonic Battery Company PREPA Puerto Rico Electric Power Authority Final Report 9 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications This page intentionally left blank. Final Report 10 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 1. EXECUTIVE SUMMARY This study evaluated the technical and economic feasibility of applying used electric vehicle (EV) batteries in stationary applications. Such a second use for a “spent” EV battery could possibly give EV owners an opportunity to reclaim a portion of the purchase price of the battery, effectively reducing its cost. This second use scenario could also make lower cost advanced batteries available to the stationary energy storage market and accelerate the establishment of a sustainable market for advanced EV battery technologies. Overall, the concept of EV battery reuse appears to be a viable one. The study team did not come across any insurmountable technical barriers to the implementation of a second use scheme. In fact, during the course of the study it was learned that there is considerable commercialization of used and reconditioned batteries. There are established markets for used, off-lease forklift batteries, reconditioned automotive starting, lighting, and ignition (SLI) batteries, and reconditioned lithium-ion batteries for laptop computers. Used nickel/metal hydride (Ni/MH) batteries from bench test and/or prototype EVs are being used in a solar program in Mexico. A study by Argonne National Laboratory examined the second use of Ni/MH EV batteries and showed that modules tested to end-of-life on the United States Advanced Battery Consortium (USABC) Dynamic Stress Test (DST) profile could provide performance competitive with new lead-acid batteries in stationary energy storage applications. While there are no technical “show stoppers,” there are some issues that will have to be dealt with before an EV battery second use scheme can be implemented. Several unresolved issues and potential barriers to the implementation of a second use scheme were identified during the development of the hypothetical reuse process. First, non-standardized battery modules and varying patterns of vehicle use could make assembly of matched strings of modules with similar capacities difficult. Second, the mechanism by which the value from the second use makes its way back to the EV buyer needs to be identified. Third, warranty terms and costs will be difficult to determine given the uncertainty in the performance and life of the used EV batteries. Finally, the perceived value of used batteries relative to new batteries in the consumer’s mind will have to be addressed to ensure widespread acceptance of used EV batteries. In addition to identifying potential barriers to EV battery reuse, this study considered the steps necessary to prepare used EV batteries for a second application. EV battery modules will have to be collected from vehicle dealerships or service centers, inspected to ensure physical and electrical integrity, tested to determine performance, and reconfigured into battery packs suitable for stationary applications. The study also considered stationary applications where used EV batteries could be applied. Eight potential stationary applications for used EV batteries were identified: transmission support, area regulation & spinning reserve, load leveling/energy arbitrage/transmission deferral, renewables firming, power reliability & peak shaving, light commercial load following, distributed node telecommunications backup power, and residential load following. The power, energy, and duty cycles required from batteries for these applications were identified. Two figures of merit were developed to quantify the economic benefits of battery energy storage in each application. The low value was estimated from the expected revenues or savings generated by a battery system operating in the application. The high value was typically extrapolated from Final Report il Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications what people have paid for similar systems, representing an upper limit to the allowable cost for the system. An economic analysis was performed to determine the feasibility of utilizing spent EV batteries in the eight potential stationary applications. The costs of acquiring, testing, and reconfiguring used EV batteries were estimated. For a typical 12V, 100 Ah Ni/MH battery module, the cost of testing and repackaging the module into a stationary battery pack came to about $64/kWh. Converting the used EV batteries to stationary applications is a labor-intensive process, with labor costs and overhead representing over half of the conversion costs. If the buy-down for a new EV battery is around $75/kWh based on the manufacturer’s rated capacity, the cost of the used battery to the second application comes to about $81/kWh at its reduced second-use rating. Adding this figure to the testing cost gives an estimated selling price for Ni/MH battery packs of $145/kWh. The results for other battery chemistries and module configurations would be similar. Costs for the assembly and operation of battery energy storage systems to meet the stationary applications were also estimated by extrapolating from data available on existing systems. The projected battery life, cost of the converted battery modules, balance of system costs, and system operating expenses were used to calculate the life cycle costs of a battery energy storage system designed to meet the requirements of each application. These life cycle costs were then compared to the economic benefits derived from the applications to determine the economic feasibility of applying used EV batteries for each application. The exception to this approach was distributed node telecommunications backup power, for which the costs of the used EV batteries alone, rather than a complete energy storage system, were compared directly to batteries currently used in the application. Based on the results of this analysis, the applications were categorized to indicate their economic feasibility. If the cost of the system was less than the low benefit estimate for the application, then the application was categorized as favorable. If the system cost was higher than the upper benefit estimate, the application was classified as unlikely. If the cost of the system fell between the two economic benefit estimates, the application was categorized as possible. It should be noted that for many of the systems, the battery costs were not the largest component of the life cycle costs. Battery costs over the life of the system ranged from 7% to 65% of the total system costs, and were the primary cost driver for only three of the applications considered in this study. Four of the eight applications were identified as being possible candidates for economically viable reuse of EV batteries. Transmission support was categorized as being a possible candidate for EV battery reuse, even at relatively high buy-down values. However, the difficulties involved in assembling a system large enough to meet the needs of this application may prove to be a major hurdle to its implementation. Light commercial load following was also identified as a possible application for reuse. Residential load following was the only application identified as a favorable candidate for reuse. Unfortunately, the market for this application is not well defined since it depends on the growth of distributed generation for residential use. Since the economic analysis for distributed node telecommunications backup power compared the cost of used EV batteries to that of competing battery technologies currently used in this application, the results are highly dependent on the relative lives of the batteries under consideration. Specifically, this application was classified as possible only when the used EV battery could provide a longer lifetime than competing technologies. It is unclear if this is likely Final Report 12 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications to be the case since how long batteries last in this application is primarily determined by calendar life and not cycle life. However, the size of the market for this application taken with the possibility for an economically feasible outlet for used batteries warrants further investigation into distributed node telecom as a candidate for reuse of EV batteries. The study team recommends that additional testing be performed on used EV batteries pulled from real-world vehicles to better characterize their performance in potential stationary applications. The team further recommends that prototype battery systems based on used EV battery modules be designed and tested for each of the four promising applications identified in the economic analysis. Final Report 13 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications This page intentionally left blank. Final Report 14 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 2. INTRODUCTION 2.1. Objective of this Study The high cost of batteries is often considered to be a major factor limiting the sale of electric vehicles (EVs). A considerable number of R&D dollars have been and are being spent by automakers and battery manufacturers around the world to reduce the cost of batteries in order to stimulate the sale of EVs. This study examined another opportunity: the potential for recovering some of the original value of a battery that has served its useful life in an electric or hybrid- electric vehicle, but which is still suitable for another application. The premise is that the battery could be refurbished, resold, and reused for another application, and some of the initial cost recovered through the sale of the battery into this second application. An EV battery is considered to have reached its end-of-life when it can no longer provide 80% of the energy (needed for vehicle range) or 80% of the peak power (needed for acceleration) of a new battery. The end-of-life criterion for HEV batteries is a 23% loss in power. A battery that cannot meet these performance criteria, however, may still be capable of satisfying the energy and power requirements of a less demanding application. For example, four alkaline AA cells can provide the power needed to operate a portable CD player for several hours and, when the player no longer spins the disk, the batteries can be removed and put into a clock or a TV/VCR remote where they may provide up to several more months of service. In this case, there was still energy remaining in the batteries when it could no longer spin the CD, but it could not be withdrawn at the relatively high power demanded by the CD spinner. It was, however, available at the lower power needed by the clock or TV/VCR remote. This technique of finding second uses for “spent” primary batteries may serve as a template for a way to extend the useful life of the advanced secondary batteries being developed for use in electric vehicles. For example, an EV battery that is no longer capable of delivering the power needed to accelerate up a freeway ramp may still provide many useful charge-discharge cycles in a less demanding, lower power application. Stationary battery applications often do not have the severe weight and volume constraints of the EV application and this can translate into lower energy and power requirements (on a unit weight or volume basis) for batteries. Since used EV batteries are still expected to be capable of storing and delivering substantial energy, it is possible that they might satisfy the requirements of these applications. Finding such a second use for a “spent” EV battery may give the EV owner an opportunity to reclaim a portion of the purchase price of the battery, effectively reducing its cost. This second use scenario would also make lower cost advanced batteries available to the stationary energy storage market and accelerate the establishment of a sustainable market for advanced EV battery technologies. The objective of this study was to determine the technical and economic feasibility of applying used EV batteries in stationary energy storage applications. The study identified several technical issues that need to be resolved before this reuse concept can become commonplace. The issues studied include: 1. What is the performance and projected remaining useful life of a battery removed from an EV? Final Report 15 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications N What are the potential stationary energy storage applications and what are their performance, life, and cost criteria? Ww Are used EV batteries capable of meeting these requirements? What needs to done to a used EV battery to prepare it for use in a stationary application? we Can used EV batteries be cost-competitive with other battery options for these stationary applications? 2.2. Definitions Before continuing with the details of the study, it is necessary to define several terms used throughout this report to refer to the various components that make up a battery to ensure that the reader is clear on what items are being described. A battery cell is a group of electrodes in a single container exhibiting the fundamental voltage of the battery chemistry. Battery cells are typically 1-4 volts A battery module is a group of cells mechanically attached to each other and electrically connected in series/parallel arrangements to form the building block of the battery pack. In EVs, battery modules are typically 10-30 volts. A battery pack consists of a number of battery modules in a single container connected in series/parallel arrangements to achieve the desired voltage and capacity. The battery pack also contains the electronics and thermal management system required to operate the battery. In EVs, the battery pack typically consists of 10-40 battery modules and usually operates in the 100-350 volt range. The battery or battery system may be a single battery pack (as in EVs), or it may be a number of battery packs electrically connected together to meet the requirements of an application. 2.3. Status of EV Technologies and Markets For the purposes of this study, electric vehicles sold in the U.S. (including those that have gone out of production), as well as some demonstration and prototype vehicles, were broken into four broad categories. Their descriptions are as follows: e EV-Electric Vehicle: full size passenger automobile with a battery-based electric drive train. e CEV - City Electric Vehicle: small (short wheelbase) passenger automobile with a battery-based electric drive train designed for urban commuting; top speed approximately 60 mph. e NEV - Neighborhood Electric Vehicle: small, low speed (<25 mph) passenger automobile with a battery-based electric drive train designed for short trips. e HEV — Hybrid Electric Vehicle: full size passenger vehicle with an internal combustion engine coupled with an electric motor and battery. Over the past few years, sales of full-sized EVs in the U.S. amounted to about 4,500 vehicles. Several automakers have demonstration models of CEVs and Ford offers the TH!NK City vehicle for sale in Europe and plans to offer it for sale in the U.S. in the fall of 2002. Several car makers have also demonstrated prototype NEVs, and DaimlerChrysler has sold about 6,200 of its Final Report 16 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications GEM. In addition, more than 25,000 Honda Insight/Toyota Prius HEVs were sold in the United States through the end of 2001. Each of these classes of vehicles has different power and energy requirements for the battery. For example, EVs typically have about 30 kWh of energy storage, CEVs and NEVs require about 10 kWh, and HEVs typically have a battery with around | kWh of storage. In addition to considering different types of vehicles, this summary also covers systems based on each of the major battery chemistries used in electric vehicles: ¢ Ni/MH — Nickel/Metal Hydride e Li-ion — Lithium-ion ¢ Ni/Cd — Nickel/Cadmium ¢ LMPB - Lithium Metal Polymer ¢ VRLA - Valve Regulated Lead-Acid e FLA - Flooded Lead-Acid. Table | presents broad descriptions of typical batteries for EVs and HEVs, organized by chemistry and vehicle type; similar data for CEVs and NEVs is shown in Table 2. These tables provide a rough idea of the range of batteries that will be available for potential use in an EV battery second use program. While it is not an exhaustive list, it does contain the specifications of the battery systems installed in most EV’s that have been sold in the U.S. (including those that have gone out of production), as well as some demonstration and prototype vehicles. The list does not cover most of the conversion vehicles currently available. The information used here was obtained from the Electric Vehicle Association of the Americas website, the Department of Energy's Office of Transportation Technologies Alternative Fuels Database, vehicle manufacturers, battery manufacturers, and the “Advanced Batteries for Electric Vehicles: An Assessment of Performance, Cost and Availability” report submitted by the Year 2000 Battery Technology Advisory Panel to the California Air Resources Board. More detailed tables are available in Appendix B. 2.4. Present Markets for Used Batteries A literature search related to current markets for used batteries was conducted as part of this study. Originally, the search was to focus solely on EV and HEV batteries, but since few EVs and/or HEVs have been sold to-date, and far fewer have had their batteries reach end of life, the scope of the search was expanded to include the reuse of batteries taken from non-EV/HEV applications. It was assumed that knowledge gained on existing secondary markets for other batteries would be useful to determine the feasibility of establishing a second market for spent EV and HEV batteries. The results of the literature search demonstrate that there is considerable commercialization of used and reconditioned batteries, including batteries from EVs: ¢ Amateur radio emergency operators acquire used gel cell lead-acid batteries removed by hospitals from medical instruments on a scheduled basis. e There is established commerce in used off-lease forklift batteries, reconditioned automotive starting, lighting, and ignition (SLI) batteries, and reconditioned lithium-ion Final Report 17 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications batteries for laptop computers, as well as batteries with various chemistries for energy storage in small renewable energy systems. e AC Propulsion, a small-volume electric vehicle manufacturer, is implementing a successful secondary market for spent Optima® deep-cycle spiral-wound recombinant lead-acid batteries. e Energy Conversion Devices, a manufacturer of both solar photovoltaic modules and nickel/metal hydride (Ni/MH) batteries, is participating in a solar program in Mexico, which incorporates used Ni/MH EV batteries taken from EV bench tests and prototype EVs. Details of the literature search and on the used battery markets may be found in the Interim Task 1 Report, issued by SENTECH, Inc., included as Appendix A in this report. Final Report 18 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 1. Battery module characteristics for typical HEV and EV batteries. Vehicle Type HEV Ev Chemistry Ni/MH Li-ion? LMPB? Ni/MH Li-ion? VRLA LMPB® Battery Voltage (V) 144-274 346 288 - 343 360 312 260 Battery Capacity (Ah) 6.5 3.6 77-95 90 60 - 85 119 Battery Capacity (kWh) 0.94 - 1.8 1.2 26 - 32 32 19-27 31 Cells/battery 120 — 228 96 240 - 286 96 - - Modules/Battery 20 - 38 2 24 - 28 12 26 - 39 13 Module Voltage (V) a 173 50 12-13.2 30 8-12 20 Module Capacity (Ah) 6.5 3.6 14 77-95 90 60 - 85 119 Module Capacity (kWh) 0.047 0.62 0.7 150-11 2.7 0.68 - 0.72 2.38 Module Output Power (kw)! ~0.9 16 3.2-4 12.5 4-5 4.9 Motor Output (kW) 10 - 33 17 49 - 102 62 67 - 102 - Battery Manufacturers PEVE' Shin Kobe Avestor |PEVE Shin Kobe —_|Panasonic Avestor SAFT Texaco Ovonic |SAFT East Penn Vehicles Honda Insight |Nissan Tino None Chevy S-10 Nissan Altra |Baker/Ford USPS |None Toyota Prius |Dodge Durango Ford Ranger Ford Ranger Dodge ESX3 GM EV-1 Chevy S-10 Honda EV Plus GM EV-1 Toyota Rav4 Solectria Force Solectria Force Notes: ' PEVE - Panasonic EV Energy, a Matsushita company. > The lithium ion characteristics listed here are for Shin Kobe EV and HEV modules. Module level data were not available for SAFT Li-ion batteries since none of the vehicles using SAFT Li-ion batteries are in production yet. > The LMPB characteristics listed here are for prototype modules developed by Avestor, and some of the numbers are extrapolations from cell-level performance. Avestor has not yet mass-produced any LMPB battery packs, and there are no automakers with plans at the current time to use LMPB systems in a production vehicle, although they have been demonstrated in the Ford TH!NK City and the GM Precept. * Module Output Power was estimated using the specific power for the module and an estimated module mass calculated from the total battery pack mass and the number of modules 19 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Final Report Table 2. Battery module characteristics for typical CEV and NEV batteries. Vehicle Type CEV NEV Chemistry Ni/MH Li-ion Ni/Cd FLA Battery Voltage (V) 288 120 114 72 Battery Capacity (Ah) 28 90 100 130 Battery Capacity (kWh) 8.1 1 1 9.4 Cells/battery 240 32 - 7 Modules/Battery 24 4 19 6 Module Voltage (V) 12 30 6 12 Module Capacity (Ah) 28 90 100 130 Module Capacity (kWh) 0.34 2.7 0.6 1.6 Module Output Power (kW) 11 12.5 1.6 Motor Output (kW) 19 24 27 5-25 Battery Manufacturers PEVE Shin Kobe SAFT Trojan ISAFT Vehicles Honda City Pal |Nissan Hypermini [THINK City Dynasty IT Toyota ecom Solectria Force |GEM E825 Solectria Flash [TH!NK Neighbor 2.5. ANL USABC 2” Use Study Between September 1996 and August 1997 Argonne National Laboratory conducted a study for the United States Advanced Battery Consortium (USABC) that examined the second use of Ni/MH EV batteries’. Eight Ovonic Battery Company (OBC) Ni/MH modules that had completed more than 500 Dynamic Stress Test (DST) cycles simulating EV lifetime usage were tested to determine their performance in non-EV applications. Test results on the used EV batteries were compared to lead-acid battery test data or warranties supplied on lead-acid batteries by the manufacturer for the specific applications on which they were tested. The approach and results of this work are summarized below. After completing DST testing, each of the eight modules was characterized according to standard USABC procedures to determine remaining capacity and power capability. Based on these tests, the Ni/MH batteries were derated to either 30 Wh/kg or 45 Wh/kg, representing standard production lead-acid units and advanced technology lead-acid batteries, respectively. The Ni/MH modules were characterized again at their new capacity rating and then placed on life test under one of four test regimes: 1) utility load following; 2) utility frequency regulation and spinning reserve; 3) uninterruptible power sources (UPS) for stand-by power and telecommunications applications; and 4) accelerated life testing. ''N. Pinsky (USABC Program Manager) et al., Electric Vehicle Battery 2" Use Study, Argonne, IL: Argonne National Laboratory, Electrochemical Technology Program, (May 21, 1998), 80 pages plus Appendix. Final Report 20 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 2.5.1. Utility load following Two modules, one at each rating, were tested for utility load following based a the test profile developed by Pacific Gas & Electric (PG&E) to characterize the PM250, a 250 kW, 167 kWh modular power unit containing 384 lead-acid batteries. The PG&E test applied repeated 3 hour sine shaped discharges with an 86.1 kW peak designed to remove the maximum storage capacity of the battery (167 kWh). Each discharge was followed by a charge to 100% SOC. Every ten discharges, the maximum storage capacity was removed in | hour at a 167 kW constant power rate, and the battery was recharged to 100% SOC. This 11 discharge cycle was repeated until the battery string voltage dropped below the defined 0% SOC point before the maximum capacity had been discharged. ANL scaled the power levels for the PG&E tests down to a single Ni/MH module to perform their tests. A plot illustrating the power profile used in testing the PM250 is shown in Figure 1. PM250 Power (kW) Time (h) Figure 1. Power profile for the load follow test conducted by PG&E for the PM250 power module The PM250 system was only able to complete 33 of the sine wave discharges, falling well short of the expected 150 cycles. If other discharges performed by the PM250 battery during testing are included in the total, the battery completed 72 deep discharge cycles before reaching end-of- life. The goal for the Ni/MH modules was to achieve 250 sinewave discharges before reaching the minimum voltage cutoff signifying end-of-life. The 30 Wh/kg derated module completed all 250 planned sinewave discharges and a total of 294 deep discharges before testing was terminated. It still had not reached end-of-life at this point. The 45 Wh/kg module completed 133 sinewave discharges and a total of 162 deep discharge cycles before reaching end-of-life. 2.5.2. Frequency regulation & spinning reserve Two modules were placed on a test profile designed to simulate frequency regulation and spinning reserve operations. This Utility Energy Storage (UES) profile, developed by Sandia National Laboratories (SNL), is based on the actual loads encountered by the Puerto Rico Electric Power Authority (PREPA) battery energy storage facility. Each UES cycle contains Final Report 21 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications three 160 minute long frequency regulation sessions consisting of 32 asymmetrically arranged five minute segments at three different power levels. Each of these sessions is followed by a charge period. At the end of the third charge period, a spinning reserve discharge is performed. This discharge consists of a 15-minute constant power discharge followed by a 15-minute ramp down to zero power. Finally, the battery is recharged to 100% SOC before starting the next UES cycle. As with the PM250 tests, the power levels for this test profile were scaled down to the level of a Ni/MH module. The UES test cycles were performed on one Ni/MH module rated at 30 Wh/kg and one at 45 Wh/kg. The UES test cycle load profiles are illustrated in Figure 2 and Figure 3. D I s . c Spinning H F Reserve requency Frequency A rene! Regulation Regulation G . \ ; rule _ uly — elle — H A R G E ONE WEEK Figure 2. Utility Energy Storage (UES) cycle developed by SNL Final Report 22 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 12 Power (MW) == | _ ao | FS a | Co Te “12 0 20 40 60 80 100 120 140 160 Time (min) Figure 3. Frequency regulation subcycle of UES test profile Both OBC modules completed all 16 of the planned UES cycles without reaching the minimum voltage limit. The 45 Wh/kg derated module showed a capacity loss of about 1%, and the capacity loss in the 30 Wh/kg derated module was negligible. This compared favorably with the twelve C&D Technologies lead-acid batteries subjected to similar testing at SNL, which showed 8.5% capacity loss after 13 UES cycles. 2.5.3. Uninterruptible power source and telecommunications applications Three of the Ni/MH modules, all rated at 45 Wh/kg, were tested under a series of constant current discharge regimes to prove that they could meet or exceed the warranty conditions for an Exide flooded lead-acid battery. Exide’s warranty specified certain cycle lives under eight different discharge conditions. Each module was expected to perform at least two of these warranty conditions in succession. The tests and their results are summarized below. Table 3. Exide UPS warranty conditions and ANL test results Warranty’ Module Discharge Discharge Warranted Cycles condition rate time es completed 1 A Cc thr 80 63 (80) 2 A 4c 1.5 min 210 100 3 A C/8 8 hr 40 25 4 B 4c 4-15 min 120 120 5 B C/4 4hr 60 54 6 Cc 4c 30 sec 2700 2700 7 Cc 2C 30 min 100 67 8 : 4C 45 sec 960 - The tests on these modules were plagued by uncertain charge procedures, erroneous equipment settings, and unexpected failures. There were a number of changes made to the original test plan, particularly regarding which module would perform what tests and the order of the tests performed, to try to get around these difficulties. Module A reached end of life at cycle 63 of the C rate, 100% discharges, but due to an improper setting the test was allowed to continue to the Final Report 23 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications full 80 cycles at an end of discharge voltage of less than the limit selected for the test. Ignoring this deviation, each of the modules completed one of the Exide warranty conditions before reaching end-of-life. Of the other five warranty conditions, four were attempted but not met, and the fifth was never attempted. The study authors also note that the warranties for VRLAs, obtained after test protocols had already been developed, warrant somewhat higher cycle lives than those for flooded batteries. 2.5.4. Accelerated calendar life test One Ni/MH module was placed on a 14 V float charge in an oven at 40°C to accelerate corrosion reactions and possible failure mechanisms. The battery was left at elevated temperature for a total of 144 days, with reference performance tests performed at 96 days and at the end of the test. Over the test period, the C/3 capacity decayed by 13.2% and the peak power degraded by 20% from the start of the test. There was no comparison made between the performance of this module and results from tests of other batteries. 2.5.5. Conclusions Overall, the test results showed that used OBC Ni/MH EV modules perform at least as well as new lead-acid batteries in stationary energy storage applications. In some instances, the used EV modules appeared capable of performing the same functions as new lead-acid batteries over longer lifetimes. The study demonstrated that reusing Ni/MH EV modules in secondary applications could be technically feasible from a battery performance perspective. Final Report 24 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 3. A PROCESS FOR RE-APPLYING USED EV BATTERIES IN STATIONARY APPLICATIONS 3.1. Overview This section discusses the process by which used batteries taken from EVs and/or HEVs might be selected, tested, and refurbished for a second use in a stationary application. The scenario assumes that EV dealers/service centers would dispose of used EV batteries at a refurbishing facility, where the used batteries would be screened, sorted, tested, and finally reconfigured for their second use in stationary applications. 3.2. Acquiring Used EV Batteries Electric and/or hybrid electric vehicles will most likely be serviced by the same dealerships that sell them. A battery may be removed from an EV for a number of reasons — the vehicle owner complains of inadequate performance, onboard diagnostics indicate that the battery is not performing to specifications, or a predefined mileage or time limit (e.g., 6 years or 60,000 miles) has been reached. If vehicle is brought to the dealership because it is performing below owner expectations, the service department will probably perform some diagnostic tests to determine whether the fault is caused by a limited number of defective modules or is an overall battery problem. In those cases where there are only a limited number of below-par modules, these may be repaired or replaced and the balance of the battery would be reinstalled in the vehicle. There are a number of defects that could reduce the performance from a particular module, and the nature of the defect will determine whether a module is repaired or replaced. A module exhibiting high resistance due to corroded or loose module and cell interconnections could probably be repaired and reinstalled. Modules exhibiting leaks, high internal impedance, or an internal short circuit would most likely be replaced. Such modules would probably not be good candidates for reuse. For an EV battery pack that is approaching or has exceeded its expected life in the vehicle, the dealer service department will probably replace the entire pack. Automobile OEM representatives have indicated that, in such a case, the battery modules would be removed and replaced while the remaining components of the battery pack (interconnects, cooling systems, electronics, and the tray itself) would be reused in the vehicle. The dealerships would thus have a number of used battery modules for disposal — either as scrap or as candidates for installation in a stationary application. The batteries in HEVs, on the other hand, will be part of an integrated system made up of the electrochemical cells, the thermal management system, and the other components needed to manage the battery in the vehicle environment. When an HEV battery fails prematurely or achieves its expected lifetime in the vehicle, this entire integrated unit will be replaced. It was determined during the course of this analysis that the labor required to disassemble this integrated unit, sort and cull the cells, and reassemble them into modules of reasonable size would make the cost of the used batteries prohibitively expensive. As a consequence, batteries Final Report 25 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary A pplications that fail prematurely due to the failure of individual cells or other hardware problems will probably not be viable candidates for second use. Likewise, HEV batteries that have met their expected lifetime of 15 years will probably not be good candidates for use in a second application. The perceived value of a 15-year-old battery will probably limit the amount of money that people will be willing to pay for it. Based on our estimates, it is unlikely that such batteries would command a high enough price to accommodate the costs of the refurbishing process. Thus, HEV batteries were not included in the detailed economic analysis performed for this report. 3.3. Testing Used EV Batteries Used EV battery modules, after arrival at a refurbishing facility, will be visually examined to separate out modules with obvious physical damage, leaks, or other signs of abuse. The initial inspection will also include determination of the module’s manufacturer specifications and age from labels or bar codes. Voltage and resistance measurements will be performed to identify modules that have failed (due to short circuits or dried-out separators, for example). Physically sound modules that have not surpassed their calendar lives will then be subjected to limited cycle testing to determine their capacity and their power capability, and to make some prediction of their expected life in the second application. The test regime envisioned is based on the type of characterization tests performed on newly : 2 : manufactured EV modules, using the USABC Reference Performance Test” sequence, and in conversations with EV battery manufacturers. The sequence includes 1. Establish the module capacity via four charge-discharge cycles, charging per the manufacturer’s recommended profile and discharging at C/3 (based on manufacturer’s original rating) to 100% of capacity. to Establish the power capability by recharging, discharging at C/3 to 50% DOD, and determining the sustained (30 sec) power capability at 2/3 of the module’s OCV. 3. Sort modules by capacity, power capability, and calendar age for later assembly into packs for use in stationary applications. The stabilized capacity after the four C/3 cycles may be used to predict the expected life in the second application. Figure 4 shows a hypothetical plot of capacity vs. cycle number data that might be gathered by a manufacturer of EV battery modules as being representative of the expected cycle life of modules discharged at a standard rate, say C/3. By superimposing the capacity data collected on the charge-discharge cycles after EV end-of-life, it is possible to determine where on this standard life curve the used EV modules lie and, therefore, what the expected remaining life is. > USABC Electric Vehicle Batteries Test Procedures Manual, Revision 2, January 1996. Final Report 26 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Capacity vs Cycles 110 100 TTT MITTIN NNN rarer rrrieseerrer pees eel 80 Capacity measured on four 70 consecutive cycles after EV EOL and superimposed on he manufacturer's life data C/3 Capacity (% of original) 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Cycle Number Figure 4. Conceptual illustration of future life prediction 3.4. Reconfiguring Used EV Batteries for Stationary Applications A canvass of the stationary applications considered in this study shows a wide range in power and energy requirements. Small commercial systems, for example, are likely to be in the 25 kW, 100 kWh range, while utility systems may require up to 100 MW of power and tens of megawatt-hours of storage. Moreover, many of the applications are likely to be unique, requiring a battery configuration unlike any other. The larger systems will be equivalent to more than 100 full EV packs, complicating shipping and handling if fully assembled at the refurbishing facility. On the other hand, assembling thousands of individual battery modules together at the application site could result in high installation costs. As a consequence, it is likely that the used EV modules will be assembled at the refurbishing facility into battery packs that are small enough for convenient handling, but large enough to reduce installation costs on larger systems. A survey of the applications considered for this study indicates that 25 kWh is a convenient building block to meet the smaller applications (except for small household systems, which could use individual modules). It is possible that used EV battery modules could be assembled into much larger units (for example, skid mounted shipping containers), particularly for the larger applications. However, it is likely that the pack assembly costs on a per module basis would be of similar magnitude for a wide range of pack sizes. The analysis performed here focused on 25 kWh units that could meet the smaller commercial applications as sort of a “placeholder” to estimate pack assembly costs. This size is similar to EV battery packs, allowing for estimation of materials and assembly costs by analogy. The hypothetical stationary battery pack, or “StatPack,” designed for this analysis would contain 21 12V, 100Ah EV modules. These could be arranged into three rows of seven side-by-side modules to create a battery pack with a convenient size. Taking into account space requirements and additional weight for cooling systems and electronics, each StatPack would be roughly 4” Final Report 27 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications wide x 4’ long x |” high and weigh around 1000 kg. This size is about the same as a standard shipping pallet, and the weight is small enough to allow handling with forklifts or other commonly used equipment. The modules would probably be connected in series, generating 252 V. Various battery chemistries or module characteristics could lead to slightly different StatPack configurations. StatPacks would be fabricated from modules that have been fully characterized at the refurbishing facility. The modules in a StatPack would be matched on the basis of cycle tests so that they all have similar capacity and power capability. The StatPacks would contain all the necessary components for thermal and electrical management of the batteries, including fans or coolant channels, module interconnects, sensors, and electronics. After fabrication, the StatPacks would be stored at the refurbishing facility and eventually sold to whomever would assemble and integrate the battery energy storage systems. Final Report 28 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 3.5. Summary of Refurbishing Process The complete process for converting used EV batteries to stationary applications is illustrated in Figure 5. ¥ Collect used EV modules from dealerships and transport to refurbishing facility. v Visually inspect modules for damage or signs of abuse. v Determine manufacturer ratings and age from label or barcode. v Measure woltage and resistance to identify failed modules. v Load modules onto test equipment. v Determine module capacity through 4 charge/discharge cycles: Charge according to manufacturer profile Discharge at C/3 to 100% rated capacity v Determine module power capability: Charge according to manufacturer profile Discharge at C/3 to 50% rated capacity Discharge at 2/3 OCV for 30 sec. v Disconnect modules from test equipment. ¥v Sort modules by capacity, power capability, and age. v Assemble into new battery packs v Ship to assembler of complete energy storage system. & Figure 5. Battery conversion process Final Report 29 v Recycle damaged, old, or failed modules 6 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications This page intentionally left blank. Final Report 30 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 4. APPLICATIONS A total of eight stationary applications of batteries were analyzed for this study and considered in the economic feasibility analysis. This chapter describes the requirements, benefits, and economic values of each of the applications considered. Most of the application requirements, in particular those listed for the utility scale applications and for power reliability/peak shaving, are based on updates or projections from the “Opportunities Analysis” report developed by Sandia National Laboratories.* The requirements for the remaining applications were estimated by the study team. Two values are listed for the economic benefits of each application. For most of the applications, the lower value is a projection of the typical savings or added revenue derived by the owner of the battery system from its Operation. This value is an estimation of what the battery system might be worth to potential users, and hence represents the most somebody might be willing to pay to purchase such a system if they expect to obtain a profit. The second, higher value reflects what people have paid for similar systems currently or formerly in operation. This value gives an idea of the allowable system cost for installation in particularly high value situations or for early adopters of the technology. Three of the applications are exceptions to these values: light commercial and residential load following and distributed node telecommunications standby power. For the two load following applications, the high and low values were estimated by the study team based on projected benefits derived from operation of such systems. The analysis for distributed node telecom standby power was different from the other applications in that only the batteries were considered, not an entire energy storage system. The high and low benefit values were determined from the price range for lead-acid batteries, the technology currently used to meet the needs of this market. * P. Butler, Battery Energy Storage for Utility Applications: Phase I — Opportunities Analysis, Sandia National Laboratories Report SAND94-2605 (10/1994). Final Report 31 Technical and Economic Feasibility of. Applying Used EV Batteries in Stationary A pplications 4.1. Transmission support In this application, the battery system provides pulses of real and reactive power to stabilize transmission lines. The battery must be of sufficient size to Application Req Typical power rating 10 sec (up to 5 pulses in Discharge time sequence during discharge support transmission assets, which time) implies 10s to 100s of megawatts. Since | Energy delivery Up to 100 MWs per pulse this is a pulse power application, and the | Frequency of use 1/month pulses are somewhat infrequent, not Low benefit estimate * $50/KW/yr much storage capacity is required, and High benefit estimate * $150/kW/yr the life of the batteries would primarily be determined by calendar life limitations. The first pulse may be discharge or charge, depending on the cause and nature of the particular de-stabilizing event, so the battery must be maintained at an intermediate state-of-charge. The electricity storage system allows transmission lines in a constricted network to be more heavily loaded during periods of peak demand by customers. This allows utilities to defer investments in transmission assets. The economic benefits listed above are based on estimates of the value of such deferred investments. 4.2. Area regulation & spinning reserve These ancillary services are typically provided by generating assets Area regulation 20 MW Spinning reserve 20 MW Typical peak power rating operating at zero or Average power rating +10MW 20 MW partial loading. A battery | d Cycled 15 min full power, system can provide load AAT Mie dant 15 min ul down following (real and Energy delivery +2 MWh 7.5 MWh possibly reactive power) Frequency of use Continuous 1/month for area regulation Low benefit estimate ° $700/kW (frequency regulation for High benefit estimate i $1500/kW an island system) and provide an alternative method for short term, fast response spinning reserve. As in transmission support, the first use for area regulation may be discharge or charge, so the battery must be maintained at a partial state of charge. i Updated and inflated from Hurwitch et. al., EPRI Energy Storage Workshop materials (1991). i Study team projection. ° B. Louks, EPRI Journal 1988, based on Dynastor projections and inflated (1988). 7 Inflated cost of PREPA BESS, from A. Akhil, S. Swaminathan, and R. Sen, Cost Analysis of Energy Storage Systems for Electric Utility Applications, Sandia National Laboratories report SAND97-0443 (1997). Final Report 32 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications This set of applications requires a fairly strenuous duty cycle. Spinning reserve requires 15 minutes at full power and 15 minutes of ramp down from full power to zero. These events only happen about once a month, but they would require a complete discharge of the battery. Area regulation requires zero net unscheduled power flow between control areas in each 15 minute period. The energy transferred to meet this application is only about 25% of that for spinning reserve, but the battery is cycled continuously. The benefit of the electricity storage system in both spinning reserve and area regulation derives from reducing or eliminating the fuel and maintenance costs that are normally associated with underutilized generating assets. The benefit derived from area regulation is probably inadequate to justify a battery, but once an electricity storage system has been installed, a battery could be the least cost alternative for this service. 4.3. Load leveling/energy arbitrage/transmission deferral This is the classic utility application for icati SH LUiKo iit energy storage: store cheap electricity Average power rating 10 MW/15 MVA generated off peak and sell it on-peak Discharge time At least 5 hours when more expensive generators are Energy delivery 50 MWh required. Alternatively, the use of Frequency of use 100-200 days/yr nighttime electricity on-peak can allow Low benefit estimate ° $50/kW/yr deferral of transmission expansions. High benefit estimate ” $150/kW/yr This application requires large storage capacities, with discharges of five hours or more favored by most utilities, particularly for transmission deferral and arbitrage. Each discharge removes most of the capacity of the battery, and discharges would occur every weekday when power use is high, i.e., 100 to 200 days per year. The benefit of the electricity storage system in this application is the difference between the cost for supplying electricity close to the loads from on-peak generation and transmission assets and the cost for supplying electricity from off-peak assets. 8 Updated and inflated from Hurwitch et. al., EPRI Energy Storage Workshop materials (1991). ° H. Zaininger, Analysis of the Value of Battery Storage with Wind and Photovoltaic Generation to the Sacramento Municipal Utility District, Sandia National Laboratories report SAND98-1904 (1998). Final Report 33 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 4.4. Renewables firming Most renewable energy resources, such as wind and solar energy, are intermittent in nature — they do not provide a reliable, continuous source of power. This limitation prevents system operators from having the same type of control over renewable generating assets that they have over other generating assets. For this reason, prices paid for electricity PW) NICHI cating tts Typical peak power rating SMW Average power rating 1 MW Discharge time 1-10 hours Energy delivery 1-10 MWh Frequency of use 10-20 days/month Low benefit estimate '° $1000/kW High benefit estimate '° $1500/kW generated by renewables (unfirm power) are typically lower than what is paid for firm power. An energy storage system can follow the renewable generation (and to a lesser extent the system load) and allow the renewable generator to be counted a firm resource. This application requires a wide range of storage capacities, depending on the nature of the renewable resource and the presence or absence of other generators that fill in the gaps. The duty cycle for this application depends on the nature of the renewable resource, but would probably be similar to that found in the other load following applications, with many shallow DOD cycles superimposed on daily deep discharges. The first event after a period of inactivity may be discharge or charge, depending on the needs of the electric system and the renewable resource. The benefit of the electricity storage system for renewables is the extra revenue for firm electricity as compared to electricity from a non-firm resource. Additionally, variations in the power from renewables can cause problems with transmission, since wind and solar farms are often placed remote from loads and are often connected through weak lines. The benefit estimates used here are derived from avoided transmission upgrades. 4.5. Power reliability & peak shaving Application Requirements An energy storage system can provide electricity during extended outages and Typical peak power rating 2MW reduce the purchase cost for electricity Average power rating 1 MW (demand charges, time-of-day prices) by Discharge time 3-4 hours shaving peaks. The second use of EV Energy delivery 3-4 MWh batteries for Uninterruptible Power Frequency of use 6/yr up to daily Source (UPS) applications alone appears —_| Low benefit estimate |! $120/kW/yr very unlikely, given the low cost of lead- High benefit estimate MN $250/kW/yr acid batteries for these applications and the fact that they are widely used and have well-defined warranties. Thus, peak shaving must be used together with the power reliability function. In this case, the customer will have to decide '© Based on H. Zaininger, Analysis of the Value of Battery Storage with Wind and Photovoltaic Generation to the Sacramento Municipal Utility District, Sandia National Laboratories report SAND98-1904 (1998). "| Proprietary source. Final Report 34 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications on the value of the system for each application and then decide how much capacity to hold back for power reliability. Battery systems designed to meet this application could be as large as 2 MW in rated power output, but will most likely consist of 100 kW modules. Three to four hours of storage will be required to provide blackout ride-through and significant peak shaving benefits. Blackouts may only occur a few times per year, but peak shaving could be used almost every workday depending on the electricity tariff for the site. The benefit of the electricity storage system in this application is mostly in the power reliability function, with peak shaving being used to offset the total costs of the system. 4.6. Light commercial load following A battery will likely be used in tandem Typical peak power rating 200 kW with most distributed generation technologies (including renewables) to Average power rating 25 kW allow more efficient and more reliable Discharge time 3 hours operation. The battery system would be Energy delivery 75-100 kWh used for load-following, thereby allowing Frequency of use Daily a generator to run at relatively constant Low benefit estimate !” $10/kW/month power delivery or a renewable resource to High benefit estimate ! $20/kW/month better match the load. This mode of operation would require the battery to be in use (charge or discharge) most of the time, and it would be at a partial state-of-charge for much of the time. The benefit of the electricity storage system in these applications is in allowing more efficient and more dispatchable local generation. Battery systems would only be practical for these applications if a utility connection were not economically viable, if a battery system owner could arrange to receive a high price for any excess electricity that could be sold back to the utility, or if the battery reduces the cost of the distributed generating system by avoiding the need for an oversized generator to meet peak loads. The benefit estimates shown here were based on avoided utility demand charges. 0 Study team projections. Final Report 35 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 4.7. Distributed node telecom backup power Lead-acid batteries already provide power eta Sg for distributed nodes (fiber nodes) of the Typical peak power rating 5 kW telecomm system during electric utility Average power rating <5 kW outages. The replacement of lead-acid Discharge time 5-10 hours batteries for telecom “switches” is Energy delivery 25-50 kWh deemed very unlikely, but lithium-ion Frequency of use 2/yr batteries are already being supplied in test | Low benefit estimate !° $120/kWh quantities for the distributed telecom node | High benefit estimate 13 $200/kWh application. Very high reliability, i.e., the ability to deliver the stated capacity and power, is a must for this application (in order to minimize costly service calls). Since the batteries are used for backup power, the duty cycle in this application is fairly benign. However, VRLAs used in this application have shown lifetimes as short as one year due to the acceleration of aging processes by the high temperatures frequently encountered in telecom equipment boxes. Advanced battery technologies may show less performance degradation during high temperature float or standby compared to lead acid batteries, resulting in longer battery lifetimes. The benefit of an alternative to lead-acid batteries in this application is in lower life cycle costs due to longer time between replacements. The benefit estimates listed above are based on the current price for VRLAs. 4.8. Residential load following This application is very similar to light i Sg commercial load following, just on a Typical peak power rating 10 kW smaller scale and operating under Average power rating 1kW different load profiles. Distributed Discharge time 3 hours generation technologies for residential use | Energy delivery 3-4 kWh will likely be paired with a battery system | Frequency of use Daily to improve their efficiency and reliability. | Low benefit estimate '* $5/kW/month The benefit of the electricity storage High benefit estimate $10/kW/month system in these applications is in allowing more efficient and more dispatchable local generation. Battery systems would only be practical for these applications if a utility connection were not economically viable, if a battery system owner can arrange to receive a high price for any excess electricity that can be sold back to the utility, or if the battery reduces the cost of the distributed generating system by avoiding the need for an oversized generator to meet peak loads. The benefit estimates listed above were based on the avoided demand charges calculated for light commercial load following. '3 Estimated current cost of batteries for telecommunications. '4 Study team projections. Final Report 36 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 5. ECONOMIC FEASIBILITY —- ASSUMPTIONS & METHODOLOGY The first part of this assessment focused on three separate tasks: identifying the steps necessary to prepare used EV batteries for a second application, estimating the requirements and values for a range of potential stationary applications, and projecting the remaining life and performance of a used EV battery module. Once these tasks were completed, an economic analysis that combined the results of all three efforts was performed to determine the economic feasibility of generating a second revenue stream for EV batteries from stationary applications. A straightforward approach to this analysis was adopted. It began with estimating the cost of each step in the process to convert a module from vehicle to stationary use. This included the design of a hypothetical battery testing and pack assembly facility. The costs from this facility and the other steps in the conversion process were used to generate an estimate of the selling price of a stationary application battery pack comprised of used EV battery modules. The performance projections for the batteries were then used to design a battery system that would meet the requirements for each of the stationary applications. A life cycle cost estimate based on the cost of the stationary battery packs, the number of packs required to build the system, the number of battery replacements over the system life, and the balance of system (BOS) costs was performed for each stationary battery system. This life cycle cost was then compared to the estimated value of the application to identify promising stationary applications. The assumptions used for each of these steps are identified in the following sections. 5.1. Availability and cost of used EV batteries The goal of the economic analysis was to determine the feasibility of EV battery reuse when electric vehicles are common. For the purposes of this analysis, it was assumed that the only significant sales of electric vehicles in the U.S. in the near-to-mid term will occur in the state of California as a consequence of the California Air Resources Board’s (CARB) zero emission vehicle (ZEV) mandate. The impact of the CARB mandate on vehicle sales is fairly difficult to interpret due to partial credits for vehicles such as HEVs and NEVs. The mandate has also become a bit of a moving target in recent years as the regulations have been rewritten and revised. At the time of this writing, the mandate does require a certain level of full-function ZEV (either battery electric or fuel cell vehicle) sales as the mandate comes into force. CARB estimates that by 2006 sales of full function ZEVs will be roughly 10,000 vehicles per year. This study assumes that 10,000 all-battery electric vehicles will be sold each year, and the batteries from these vehicles will be available several years later when the vehicles need battery replacements. At 25-30 battery modules per pack, this corresponds to 250,000-300,000 EV battery modules available for reuse each year. The value of the used EV batteries for a second application determines the amount of buy-down or rebate that can be applied to the electric vehicle. Conversely, a certain buy-down may be required to ensure commercial viability of EVs. High volume battery production cost estimates for the two leading technology contenders (Ni/MH and Li-ion) range from $225/kWh to $300/kWh. The United States Advanced Battery Consortium (USABC) mid-term commercialization goal for battery cost is $150/kWh. Thus, the cost of the battery to the EV user will have to be reduced by up to $150/kWh to meet the USABC’s goals for a marketable Final Report 37 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications electric vehicle. Since there is uncertainty regarding both the projected production costs of EV batteries and the allowable price of the battery in a commercially viable vehicle, this analysis treats buy-down cost as a variable and considers a range of values (from $0/kWh to $150/kWh). 5.2. Collection & transportation The first cost item in the reuse process is transporting the pallets of used modules from dealerships to a facility for testing, sorting, and reassembly into battery packs suitable for stationary applications. It is likely that the dealerships that sell and service EVs will be spread across the major metropolitan areas of California (Los Angeles, Sacramento, San Diego and the San Francisco Bay Area). These areas are somewhat far apart, so transporting the used battery modules to a centrally located facility could result in significant costs. To avoid these costs, this analysis assumes a total of four facilities will be built, one in each metropolitan area. The cost estimate for each facility includes the price of a medium duty truck equipped to carry batteries and a full time driver to make regular collection stops at local vehicle dealerships 5.3. Testing & repackaging Assuming EVs are spread evenly across California’s high population density areas, each of the four testing and repackaging facilities will have to handle about 62,500 modules each year (the facility was actually designed for an input of 67,200 modules per year to make test equipment design easier). To determine the costs associated with testing, sorting, and repackaging the used EV modules, a hypothetical facility with all the equipment necessary to handle this level of throughput was designed. The facility is assumed to operate for 350 days per year to maximize the use of the capital equipment. This requires an average daily input of 192 modules. A description of the cost components associated with the facility follows. 5.3.1. Capital Costs Testing Equipment: The first step in designing the facility was to specify the type and cost of testing equipment required to perform the test procedures outlined in Section 3. Most EV modules have a rated capacity of around 100 Ah, so the test stand would have to handle around 33 A during the C/3 discharge. EV battery modules vary in voltage, but most are between 10 and 30 volts. Quotes were obtained from Bitrode Corporation and Maccor, Inc. for a test station that could handle +/-40 A and 40 V. Bitrode designed a system consisting of 16 channels that is expected to cost $43,725. The cycle tests outlined previously (Section 3.3) require around 40 hours to complete. If time is added for connecting and disconnecting the modules, each module will be on test for approximately 48 hours. The facility must therefore have 318 modules on test at any one time to meet the 192 module input requirement. At 16 channels each, 24 test stations are required. Each day, 12 of these stations will have all their battery modules disconnected and replaced with new modules while the other 12 stations will continue cycling the batteries that were hooked up to them on the previous day. Since battery test equipment costs scale with rated power, and since only a single pulse will be performed on each module to get an idea of its power capabilities, the test stations were not designed to handle the pulse power characterization. Instead, an estimate was obtained for a separate single channel unit rated at 200 A and 30 V, which would cost about $8,000. One pulse test unit will be required for every two test stations that are switched each day, for a total of six Final Report 38 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications pulse test units. Six computers will also be needed to operate all of the test equipment and collect data. Bitrode estimated the cost of each computer along with its associated peripherals, software, and cables at $7,236. Materials handling equipment: A conveyor system will move the used EV modules from the loading dock to the test stations and from the test stations to the pack assembly area. Equipment will be required to lift the modules from the conveyor belts and place them into the StatPack containers. The assembled StatPacks will be stored on racks within the facility until they are shipped to the end user. Finally, a forklift will handle unloading of the pallets of modules as well as moving and loading the stationary battery packs. Table 4 contains a listing of all the capital costs for the facility, including the materials handling equipment. Office and other equipment: The office area of the facility will be equipped with computers, furniture, a photocopier, a printer, a fax machine, and furniture. In addition, the facility will contain a workshop for calibration, diagnosis, and repair of the test equipment. This workshop will be equipped with various tools and devices (such as multimeters, power sources, a soldering iron, etc.). These items are included in the capital cost estimate for the facility and are shown in Table 4. Table 4. Capital costs : jerry elite oF Be ee ou es eee Test Equipment $1,140,816 Bitrode Model MCN16-40-40 battery cycler 24 $43,725 $1,049,400 Bitrode Model RCN1-200-24 pulse test unit 6 $8,000 $48,000 Computers 6 $7,236 $43,416 Materials Handling $138,462 Conveyors $28,871 Module lifting equipment 8 $1,000 $8,000 Storage Racks $5,182 Nissan PE30YSC forklift 1 $18,595 $18,595 Medium duty truck 1 $50,000 $50,000 Office and Other Equipment $20,500 Total Capital Costs $1,299,778 5.3.2. Direct Costs Materials: The primary material cost for the testing facility comes from the used EV modules. This cost was determined by multiplying the buy-down cost input by the rated energy capacity of the module when it was new to determine the cost per module to the testing facility. Packaging components for the stationary battery pack are the only other significant material cost. Each battery pack will consist of 21 modules arranged in series, giving an overall rated voltage of 252 V and an energy storage capacity of about 25 kWh. Each battery pack will require interconnects, cooling components (such as fans or coolant tubing), electronics, and the package in which the modules are assembled. These components were estimated to cost about $250/battery pack. Labor: The number of full time employees required to operate the facility was determined from an estimate of the effort required to carry out the testing protocols listed above and generate an output of 192 modules per day. An analysis of the tasks required to disconnect a set of modules from a test station, reconnect a new set of modules, and set up a series of tests indicated that it will take roughly four hours for a technician to completely switch out the modules on a test Final Report 39 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications station. In a single day, a technician could therefore switch out two test stations. This implies the need for six technicians on duty for eight hours a day, seven days a week. The facility would have to employ about eight test technicians. Eight StatPacks will be produced each day. Assuming roughly one person-day for pack assembly, eight employees will be required to assemble packs eight hours a day, seven days a week, for a total of 12 full time pack assemblers. There will be two supervisors to ensure that one is always on duty during operating hours. The facility will also employ a plant manager, an electrical engineer to maintain the test equipment, a sales and logistics manager who will handle battery shipments and contracts, an office administrator to handle bookkeeping and correspondence, three full time security guards for round-the-clock surveillance, a forklift driver, a truck driver, and a part time janitor. Estimates for the hourly wages for each of these employees were obtained from the Department of Labor’s Bureau of Labor Statistics (BLS) website and converted to annual salaries. '° These salaries were then increased by a factor of 27.4%, the national average for non-wage compensation of civilian workers.'° The salaries and resulting employment costs are summarized below. 'S U.S. Department of Labor Bureau of Labor Statistics, National Compensation Survey, http://www.bls.gov/ncs/home.htm#data (12/2001) US. Department of Labor Bureau of Labor Statistics, Employer Costs for Employee Compensation for March 2001, http://www.bls.gov/news.release/ecec.toc.htm (6/2001) Final Report 40 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 5. Employment costs : Description Number AnnualSalary Annual Total Test technicians 8 $41,210 $370,890 Pack assemblers 12 $23,490 $281,880 Forklift driver 1 $26,940 $26,940 Truck driver 1 $26,940 $26,940 Supervisors 2 $46,560 $93,120 Plant manager 1 $85,450 $85,450 Electrical engineer 1 $66,320 $66,320 Sales/logistics rep 1 $58,630 $58,630 Office administrator 1 $28,220 $28,220 Security guards 3 $19,470 $58,410 Janitor Ya $19,880 $4,970 Total wages $1,101,770 Non-wage compensation factor 27.4% Non-wage compensation $301,895 Total Employment Costs $1,403,665 Rent: After determining the equipment and personnel necessary to run the testing and repackaging facility, a sample floor plan was drawn up to estimate the amount of space required for the operation (see Figure 6. Hypothetical used EV battery testing & packaging facility layout). Based on this floor plan, roughly 10,173 ft’ of space will be required for the facility. The average rent for R&D/flex facilities from the fourth quarter of 2001 for the four areas where facilities might be built (Los Angeles, Oakland, Sacramento, and San Diego) was $11.27 per square foot per year (triple net), corresponding to annual rent of about $1 15,000." " Grubb & Ellis, Asking rental rates (R&D-Flex, triple net) for second quarter 2001 from "Industrial Market Trends: A Survey of the Nation's Industrial Markets" (Fall 2001) Final Report 41 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications rai 94.50 ft Supemisor | Supervisor Scale: 1” = 20° Office Offi Pad r a 8 Break room a Gravity Feed Conveyor 3 J) Powered Conveyor 7 4! _Test Equiornent Cabinet Battery Test Rack jp Computer Workstation rm GOROOLBEOCCED | =I crim met CY - imi a qa 7 % a H 5 3 o Ly = =~ ABB A 3 £ > TOO TOTAL 7 al mm a re | ya Oo G8 ya ppoocg Pop cot HA AB 3 = Ld Joo L Pea . | DSSS TO See Pack Assembly Pallet of Battery Modules Werkstation 7 1 | | | Stqrage Rbcks | | L | - — 8 25 Ht — Figure 6. Hypothetical used EV battery testing & packaging facility layout Electricity: Electricity costs were estimated for battery testing as well as general lighting, heating, and cooling. Testing for each module will require about 9 kWh of electricity for a total of 604,800 kWh of electricity per year. According to data obtained from the DOE Energy Information Administration’s website, this electricity would cost $43,300/yr at the average commercial electricity rate of 10.4¢/kWh for the utilities serving the four areas mentioned above.'® Based on an average annual energy expenditure of $1.58 per square foot of commercial floor space obtained from the Department of Energy’s Office of Buildings Technology, HVAC and lighting for the refurbishing facility would cost around $7,200/yr.'? Transportation: Assuming the truck for collection of modules from dealerships travels 50,000 miles per year, and maintenance and fuel cost $0.325/mile, the truck would cost $16,250/year to operate. Other direct costs: This catchall, estimated at 2% of labor costs, includes phones, postage, office supplies, and the other costs of doing business. 5.3.3. Indirect Costs Insurance, general & administrative costs, and warranty costs are all estimated as a percentage of direct costs: 3%, 16%, and 4%, respectively. 'S Department of Energy Energy Information Administration, Form EIA-861, "Annual Electric Utility Report" http://www.eia.doe.gov/cneaf/electricity/esr/esrt1 5p2.html#cal (4/2002) 7 Department of Energy Office of Buildings Technology State & Local Programs, BTS Core Databook http://btscoredatabook.eren.doe.gov/tableview.asp?TableID=108 &t=pdf (10/2001) Final Report 42 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 5.3.4. Earnings, taxes, and determination of reconfigured battery selling price The testing and packaging facility is assumed to have an after-tax internal rate of return (IRR) of 15%. This figure is typical of what many firms require before making an investment in a particular venture. The analysis also assumes that the facility costs will be recovered in five years. While this is somewhat short, potential investors would expect, and perhaps demand, a rapid return from a unique and potentially high-risk venture. In addition, obsolescence may require replacement of some of the testing equipment after five years, particularly if EV battery technologies change significantly. Finally, since testing and packaging the batteries is primarily a labor-intensive effort, this short lifetime has little effect on the overall cost of the stationary battery packs. The cost of the stationary packs on a per module basis was calculated by determining the revenues necessary to achieve a 15% IRR after state and federal taxes of 8.84% and 34%, respectively. IRR is defined as the discount rate required to make the net present value of a particular venture equal to zero, where the net present value is the sum of all the cash flows over the life of a venture discounted back to the present. Expressed as an equation, “.— EXP—TAX + REV _ 20 (1+ 7RR)! NPV =-CC + 0 where: NPV = net present value CC = capital costs EXP = expenses TAX= taxes REV = revenues IRR = internal rate of return i= year n= lifetime of facility Taxes are equal to the tax rate times the adjusted gross income for each year: TAX =TR(REV — EXP — DEP) where: TR = tax rate DEP = depreciation = CC/n (linear depreciation) Substituting these values, the expression for NPV becomes: *, — EXP —TR(REV ~ EXP-“°)+ REV NPV =-CC+)) " =0 i (1+ /RR)' Solving this expression for the annual revenues yields: Final Report 43 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications n EXP+TR(EXP— © n i=0 (1+ JRR)' “, 1-TR >~——— (1+ IRR)! Thus, the revenues could be determined from the capital costs, the annual expenses, the tax rate, the internal rate of return, and the projected lifetime of the facility. The selling price of the converted batteries on a per module basis was calculated by dividing the total annual revenues by the total number of modules that pass through the facility each year: REV MODULES PRICE = A summary of the costs associated with a refurbishing facility assuming battery modules are obtained at a buy-down of $75/kWh is shown in Table 5. This table also shows the projected selling price of the converted battery packs on a per-module and per-kWh basis. Table 6. Testing and repackaging costs Description Us ees Total Cost ($/yr) Direct Costs Batteries (buy-down) 60,480 modules $98.10/module $5,933,088 ($75/kWh) Pack materials 2,880 packs/yr $250/pack $720,000 Labor $1,403,655 Rent 10,173 ft.? $11.27/ft.7/yr. $114,679 Electricity $79,245 Testing 604,800 kWh 10.4¢/kWh HVAC & lighting 10,173 ft? $1.58/ ft? Transportation 50,000 miles $0.365/mile $18,250 Other direct costs 2% of labor $22,035 Indirect Costs Insurance 3% of Direct Costs $250,051 G&A 16% of Direct Costs $1,333,064 Warranty 4% of Direct Costs $333,401 Capital recovery, earnings, and taxes $218,403 Required annual revenues $10,724,437 Battery module throughput 60,480 Battery pack selling price ($/module) $176.43 Battery pack selling price ($/kWh) $147.02 5.3.5. Cost breakdown The cost for collecting, testing, and repackaging used EV battery modules into battery packs for stationary applications is broken down into its various components in Figure 7. This figure was generated by taking the cost components listed in Table 6 and dividing them by the total capacity of all the battery packs that would be produced by the refurbishing facility in a year. Final Report 44 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications The primary cost driver for these battery packs is the purchase price of the used EV batteries, which is determined by the buy-down given to the EV buyer. The battery procurement costs account for over half of the stationary battery pack costs. Testing and conversion of the used EV batteries is primarily a labor intensive process, with employment and overhead costs each accounting for around 12-13% of the total battery pack costs. The other major cost component is the materials for repackaging the used EV modules into a battery pack, which makes up about 7% of the total pack cost. Reconfigured EV Battery Pack Cost Breakdown ($/kWh) $81.75 BBatteries $9 92 @ Packaging Materials $3.14 BTesting equiprnent $3.43 Glnsurance $18.28 OG8A _-— $4.57 OWarranty XY $3.01 SCapital Recovery, Earnings, and Taxes — $2.00 GAll Others Figure 7. Cost breakdown for reconfigured EV batteries 5.4. Energy storage system life cycle costs The price of the batteries represents only a single (and often small) component of the cost of a complete energy storage system. There are a number of other costs that are incurred during assembly, installation, and operation of such systems that need to be considered to determine life cycle costs. Racks and wiring are required for connecting the battery modules and holding them in place. Nearly all energy storage systems include a power conditioning system (PCS) to ensure that the output of energy storage device matches the requirements of the application (for example, 240V, 60Hz AC power). Sensors must be installed to monitor the performance and condition of the battery. Also, both regular and unscheduled maintenance will be required over the life of the system. Furthermore, the batteries themselves will most likely need replacement over the expected life of the energy storage system. This section describes the methodology for estimating these and other costs. It also explains how all of the costs were brought together to determine the life cycle costs of an energy storage system designed to meet each of the applications described in Section 4. 5.4.1. Application requirements & system characteristics The first step was to design an energy storage system for each application that could meet that application’s power and energy requirements. The number of battery modules required to meet Final Report 45 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications the needs of each application was calculated based on the modules’ projected power and energy capabilities. 5.4.2. Duty cycle and projected life To determine the replacement costs of the battery, the battery life had to be estimated. Since both the amount of time that the battery will remain in the vehicle and the number of discharge cycles it will see before being removed are uncertain and will vary from one vehicle to the next, the life of the battery in the stationary application is difficult to estimate. For this analysis, battery life in the second application was treated as a variable. To generate the plots shown in Section 6, battery life was arbitrarily varied from | to 9 years. In addition to these cost vs. life curves, a best guess value was established for the life of the battery in each application by comparing the projected cycle life and the expected calendar life. Cycle lives were determined by dividing the annual energy throughput required by the application duty cycle by the expected throughput remaining in the battery after use in an EV (based on cycle lives projected by battery manufacturers, where available). For a somewhat simplified example, if an EV battery module had a projected life of 1500 cycles, and it were removed from the vehicle after 1000 cycles, it would be expected to have 500 cycles remaining in the second application. If this same module had a capacity of about | kWh, it should have an energy throughput of around 500 kWh remaining for the second application. If the duty cycle for the application required 100 kWh of energy throughput each year, the battery module would be expected to last about 5 years in that application. The calendar life in the stationary application was estimated to be 5 years. If the battery is designed for a 10-year life, and it is removed from the vehicle after around five years, that would leave around five years remaining for the second application. Please note that this is a somewhat arbitrary guess since it is unclear how long the battery may actually last in the vehicle. There is also some uncertainty on the impact of the vehicle environment (particularly temperature extremes) on battery calendar life. Furthermore, the life in the second application may be considerably longer since the environment for stationary applications is typically much more benign than that found in vehicles. For the best guess estimates, the life of the battery was set as the lesser of the cycle and calendar lives. 5.4.3. Battery costs The cost of a single set of batteries was calculated by multiplying the number of modules required to meet the application by the selling price of the refurbished module (see Section 5.3). The number of battery replacements was then determined by dividing the system design life (typically 20 years) by the expected life of the battery. The present value of all the battery purchases over the system life was calculated by discounting the costs of all the replacements back to present day dollars and adding them to the initial purchase cost. 5.4.4. Balance of system costs Typically, the procurement costs of all the non-battery equipment that goes into an energy storage system, as well as the cost of assembling and starting up the system, are lumped together as balance of system (BOS) costs. These include (but are not limited to) the facilities for housing the batteries, equipment to connect the battery modules together and monitor their performance, equipment to connect the battery to the local electricity supply, electronics to control the Final Report 46 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications operation of the battery system and condition its output appropriately, and transportation for all the components to the application site. For this analysis, the BOS costs were broken down into three categories: accessories, facilities, and transportation; PCS, interface equipment, and controls; and installation and startup. For each of the applications, an existing analogous system was chosen as a starting point for estimating BOS costs. For example, GNB has installed a backup power/peak shaving system at its lead smelting facility in Vernon, California, to allow orderly plant shutdown during blackouts and avoid demand charges by providing peak power. The costs of this system have been detailed in several reports.” This system was used as the template for several applications, including power reliability/peak shaving. The cost components for this battery system were separated into the groups listed above. The cost of each group was then divided by an appropriate size metric to generate a scaled cost. For example, the battery accessories & facilities costs were divided by the storage capacity of the Vernon system to generate a $/kWh unit value for those components, while the PCS, interface equipment, and controls costs were divided by the rated power output to give a $/kW figure. See Table 7 for all of the unit BOS costs for each of the applications. These unit costs were then multiplied by the appropriate system parameters for each of the stationary applications to give a scaled estimate of the various BOS costs. The resulting projected costs are summarized in Table 8. Note that distributed node telecom has no BOS because this study looked only at the batteries for this application, not the entire system. There are no similar systems available for residential load following, so the costs for these systems were estimated by the study team based on expected PCS and maintenance costs. 5.4.5. Operating and maintenance costs Operating and maintenance (O&M) costs must also be considered if an accurate life cycle cost is to be determined. These items include routine and unscheduled maintenance, repair or replacement of failed or damaged battery modules or packs, periodic monitoring to ensure nominal system operation, and the cost of the electricity used in operating the system (powering sensors, controls, cooling devices, etc.) as well as that consumed by the system due to losses during normal operation. Annual O&M cost estimates were determined in much the same way BOS costs were estimated. The O&M costs for an analogous system were divided by the power rating of that system to give a $/kWyr value (see Table 7). These unit values were then multiplied by the rated power of the hypothetical system designed to meet the stationary application to give a scaled O&M cost estimate. The results of this calculation are summarized in Table 8. ?0 A. Akhil, S. Swaminathan, and R. Sen, Cost Analysis of Energy Storage Systems for Electric Utility Applications, Sandia National Laboratories report SAND97-0443 (1997) and S. Swaminathan, N. Miller, and R. Sen, Battery Energy Storage Systems Life Cycle Costs Case Studies, Sandia National Laboratories report SAND98-1905 (1998). Final Report 47 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 7. BOS and O&M costs for analogous battery systems PCS. Accessories, eee Installation Similar facilities, and L t Application cue ane Ce eae controls Units $/kWh $/kKW $/kWh $/kWyr Transmission Support PREPA $525 $442 $360 $38 Area Regulation & Spinning Reserve PREPA $525 $442 $360 $18 Load Leveling /Arbitrage/ Trans. Deferral | Chino $117 $336 $52 $29 Renewables Firming Vernon $482 $319 $90 $58 Power Reliability & Peak Shaving Vernon $482 $319 $90 $58 Light Commercial Load Following Vernon $482 $319 $90 $58 Distributed-Node Telecomm Standby _ Batteries only N/A N/A N/A N/A Residential Load Following N/A Included $100 Included — $102/yr One of the costs associated with installing and operating a battery system is proper disposal or recycling of the batteries when they reach the end of their useful life. At present, the battery manufacturer is responsible for the battery from cradle to grave, and the cost of handling the batteries at end of life is included in the price of the new EV battery. It is assumed here that recycling/disposal costs will continue to be paid by the primary user of the battery (the EV owner). The money set aside for recycling/disposal will somehow be passed along with the battery through its various applications until it reaches end of life, or the responsibility for proper disposal will remain with the battery manufacturer. Either way, the fees will not have to be directly borne by the user of the battery in the second application. Therefore, disposal/recycling fees were not included in the system costs for this analysis. Salvage value for the balance of system equipment was also not included in this analysis. If the energy storage systems last for as long as planned here (20 years for most applications), most of the balance of system components will have lived out their useful lives. Some components may be obsolete, and others may not be sufficiently portable to warrant use in another system or for another application. Therefore, it was assumed that there would be very little salvage value left in the system when it reaches end of life, and whatever is remaining is expected to offset any decommissioning expenses. 5.4.6. Total system costs The total capital cost for installing the system was calculated by adding together the battery costs and the BOS costs. The total capital cost was then amortized over the life of the system to give an annualized capital cost. This annualized capital cost was added to the annual O&M costs to determine the annualized cost of installing and operating the system, which this study refers to as the life cycle cost. This value was then divided by the size of the system to give a per unit figure of merit in $/kW/yr or $/kWh/yr. Table 8 summarizes the results of all of these calculations for a *! Costs for analogous systems inflated from A. Akhil, S. Swaminathan, and R. Sen, Cost Analysis of Energy Storage Systems for Electric Utility Applications, Sandia National Laboratories report SAND97-0443 (1997) and S. Swaminathan, N. Miller, and R. Sen, Battery Energy Storage Systems Life Cycle Costs Case Studies, Sandia National Laboratories report SAND98-1905 (1998). Final Report 48 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Application Requirements & System Characteristics Table 8. Life cycle cost components for stationary energ Trans. re) enon Cae stora: Area Reg. & Spinning e systems based on used EV batteries PC Caer to minis ces dul ieee Ce ET Etat) art Pie uisabcabelill® asm Lale Node Following _ &Peak Following Telecom Peak power required KW 100,000 20,000 15,000 5,000 2,000 200 5 10, Required capacity KWh 140 40,000 50,000 10,000 8,000 100 50 4 Modules required mod 16,667 33,334 41,667 8,334 6,667 84 42 5 Peak power available KW 100,002 80,002 100,001 20,002 16,001 202 101 12) Capacity available kWh 20,000 40,001 50,000 10,001 8,000 101 50 6§ Duty Cycle & Projected Life Typical energy delivery kWh 140 2000/7500 100,000 10,000 4,000 100 50 4 Frequency of use Uses/yr 12 17520/12 200 240 256 365 2 365) Annual energy throughput kWh 1,680 35,130,000 20,000,000 2,400,000 1,024,000 36,500 100 1,460) Throughput per module KWh/mod 0.10 1,054 480 288 154 435 2 292! Projected battery life Yr 5.00 1.09 2.40 4.00 5.00 2.65 5.00 3.94) Projected system life Yr 20 20 20 20 10 10 20 20) Battery Costs Cost of one set of batteries $ $2,900,000 $5,880,000 $7,400,000 $1,470,000 $1,180,000 $14,800 $7,400 $880, Battery purchases 4 19 9 6 2 4 4 6 Present value of battery purchases $ $8,000,000 $62,800,000 $38,700,000 $5,220,000 $2,050,000 $46,900 $20,100 $3,150} Balance of System Costs Similar system PREPA PREPA Chino Vernon Vernon Vernon _ batt. only N/A Accessories, facilities, transportation $ $10,500,000 $21,000,000 $5,840,000 $4,820,000 $3,850,000 $48,600 N/A Included PCS, interface equipment, controls $ $44,200,000 $8,840,000 $5,040,000 $1,590,000 $638,000 $63,800 N/A $1,001 Installation & startup $ $7,210,000 $14,400,000 $2,610,000 $902,000 $721,000 $9,090 N/A ned | Total BOS first cost $ $61,900,000 $44,300,000 $13,500,000 $7,310,000 $5,210,000 $121,000 N/A $1,001 Operations & Maintenance Costs Total O&M costs S/yr $3,830,000 $367,000 $434,000 $290,000 $116,000 $11,600 N/A $102) [Total System Costs Capital costs $ $69,900,000 $107,000,000 $52,200,000 $12,500,000 $7,260,000 $168,000 $20,100 $4,150} Annualized capital costs $/yr $6,550,000 $10,000,000 $4,890,000 $1,170,000 $1,030,000 $23,900 $1,880 $389, Total annual expenses $/yr $10,400,000 $10,400,000 $5,330,000 $1,460,000 $1,150,000 $35,500 $1,880 $491 System unit annual cost $/kW/yr $104 $520 $355 $293 $573 $177 $38 $49) Final Report 49 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications used EV battery with a $75/kWh buy-down, roughly 1000 cycles remaining in its useful life, and an expected calendar life of 5 years. 5.5. Feasibility categories After all of the calculations were completed, the system life cycle cost was compared to the application values listed in Section 4 to determine the feasibility of using the reconfigured EV batteries in each application. For the purposes of this report, three categories of feasibility were developed. If the system life cycle cost is more than the high value estimate, the application is deemed “unlikely” for that battery. If the life cycle cost is less than the high value but more than the low value, the application is described as “possible” for the battery. Finally, if the life cycle cost is less than the low value estimate, the application/battery combination is termed “favorable.” The results of the calculations and the feasibility assessments are summarized in the following section. Final Report 50 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 6. ECONOMIC FEASIBILITY — RESULTS OF ANALYSIS 6.1. Nickel/Metal Hydride Ni/MH batteries are the most mature advanced battery chemistry used in electric vehicle applications in the U.S. As such, obtaining data on life and performance was relatively easy compared to other battery chemistries considered. Based on conversations with battery manufacturers, it was assumed that a used Ni/MH EV module could provide a cycle life roughly equal to that of the vehicle application at a slightly de-rated capacity. Further, we assumed that the used EV battery would have a calendar life of 5 years in the second application. The battery replacement schedule was determined using the lesser of the predicted cycle and calendar lives. Half of the applications considered in the study involve systems that would be installed by a utility to provide benefits or services to the grid as a whole. The results of the analysis for these four utility applications are summarized in Table 9. The lifetimes shown represent the best estimate of how long the battery would last in the application. The term in parentheses after the life estimate indicates whether the life in the application was limited by duty cycle or calendar life. Table 9. Analysis of used Ni/MH EV batteries in utility applications PS Cli) Buy-down cycle cost Low value High value EN Life (yr) ($/kWh) — ($/kW/yr) — ($/kW/yr)_ ($/kWi/yr) _ Analysis [Transmission stabilization 5 (calendar) $50 $150 $0 $116 Possible $50 $119 Possible $100 $121 Possible $150 $125 Possible Area regulation & 1.1 (cycle) $35 $75 lspinning reserve $0 $319 Unlikely $50 $453 Unlikely $100 $587 Unlikely $150 $721 Unlikely _| Load leveling 4.8 (cycle) $50 $150 $0 $158 Unlikely $50 $222 Unlikely $100 $286 Unlikely $150 $350 Unlikely Renewables firming 4.0 (cycle) $50 $75 $0 $226 Unlikely $50 $270 Unlikely $100 $315 Unlikely $150 $356 Unlikely Final Report 51 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 6.1.1. Transmission support Of these four applications, only transmission support appears to be a possible candidate for a second use of EV battery modules. For the entire range of buy-down values, the life cycle system cost is between the two value estimates, categorizing the application as “possible” for used Ni/MH EV modules. This is graphically represented in Figure 8, which shows the variation in life cycle cost with both battery life and buy-down. Ni/MH - Transmission Support $200.00 $$$ ee > = 18000 4 2 = = = $1600.00 Fc ee ee > Buy-down ($/kWh): S : a $140.00 4 $0 2 $120.00 4 —s— $50 wo @ a Bion oe = —+— $100 5 | é —— $150 8 $80.00 4 — —Low Value © yon fe ee CK. --- High Value 3. LS | o @ $000 4 $ — $2000 J 2 we $0.00 + + ' + 0 2 4 6 8 10 Battery life (yrs) Figure 8. Life cycle cost as a function of battery life and buy-down for transmission support Note that since this application requires only pulse discharges, the total energy that passes through the battery is quite small. Thus, there is no predicted cycle life for this application. The life of the battery will be limited by calendar aging processes. The horizontal dashed lines on the graph in Figure 8 indicate the estimated values of the application and serve as boundaries between the three viability categories. As expected, the life cycle cost increases with increasing buy-down and decreasing life, but only to a small degree. This relative insensitivity to battery costs is due to the nature of the application, which requires short duration, high power pulses. These requirements result in the balance of system components being the major cost driver for the battery system, as illustrated in Figure 9. The PCS, interface, monitors and controls account for 40% of the total life cycle costs, O&M costs account for an estimated 37%, and battery accessories make up about 9%. The battery and its replacements account for only 7% of the total life cycle costs. Final Report 52 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Due to this insensitivity to battery costs, transmission support can tolerate buy- downs as high as $150/kWh at battery calendar lifetimes as short as two years. The only point where the cost curves look unfavorable is at high buy-downs and a one-year life. Since the battery is expected to be calendar life limited in this application, it is unlikely that the life would be this short. Complete Energy Storage System Life Cycle Cost Components 7% a] (Transmission Support) 7% ut Present value of battery 9% purchases fe ‘a gAccessories, facilities, . transportation OPCS, interface equipment, contrals Glnstallation & startup %e ne Present value of OAM costs 6.1.2. Area regulation & spinning reserve Figure 9. Primary cost drivers for transmission support The life cycle cost curves for area regulation & spinning reserve are shown in Figure 10, below. $700.00 7——— + $500.00 + $400.00 + $300.00 4 $200.00 + Life cycle cost of system ($/kVWiyr} $100.00 4 $0.00 \ ge00.00 4 4 j—— Life projected 5 from duty cycle NUMH - Area Regulation & Spinning Reserve Buy-down ($/kVWh): —o— $0 —s— $50 —+— $100 —— $150 — — Low Value ----- High Value Battery life (yrs) Figure 10. Life cycle cost for Ni/MH in area regulation and spinning reserve The vertical white line on this plot indicates the projected battery life based on the expected energy throughput remaining in the battery and the duty cycle for the application. Area Final Report 53 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications regulation is a fairly strenuous application, requiring continuous cycling of the battery, and the effect on life is apparent in this plot. However, even at high lifetimes, and even if the batteries could be obtained for free, this application would not be viable. This is due to both the relatively low value of the estimated economic benefits derived from this application and the high cost of installing and operating the system. As Figure 11 illustrates, battery costs dominate the life cycle costs for a system designed to provide area regulation & spinning reserve. There are therefore two potential pathways to reduced costs: lower conversion costs or significantly improved battery performance. It is unlikely that a quantum leap in battery life Complete Energy Storage System Life Cycle Cost Components (representing a doubling (Area Regulation & Spinning Reserve) of the projected cycle life, 4% @ Present value of battery for example) is just purchases around the corner, so the most likely path to cost reduction is in reduced battery testing and reconfiguring costs. While it is possible that these costs could be somewhat lower than what was estimated for BAccessories, facilities, transportation OPCS, interface equipment, controls Glnstallation & startup Present value of O2M costs this study, it is unlikely that the cost reductions ji | ; necessary to make this Figure 11. Primary cost drivers for area regulation & spinning application viable are reserve achievable. 6.1.3. Load leveling/energy arbitrage/transmission deferral Load leveling/energy arbitrage/transmission deferral looks somewhat better than area regulation & spinning reserve, but it is still classified as unlikely. The life cycle cost curves are shown in Figure 12. Final Report 54 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications NiMH - Load Leveling/Energy Arbitrage/Transmission Deferral $1 000 oo - - eoreurneremmmurrnce: — <> $900.00 4 Life projected = ss0a.00 | Z from duty cycle = Buy-down ($/kVVh): = 700.00 4 $0 5 z|| a $600.00 4 3 $50 o $500.00 5 #100 6 | —— $150 8 $400.00 4 — —Low Value ° ay 2 g90000 42 High Value o 2 $20.00 4 ee = sio0.00 4 Possible L | Favorabl $0.00 ; | Favorable 0 2 4 6 3 10 Battery life (yrs) Figure 12. Life cycle cost curves for load leveling/energy arbitrage/transmission deferral This application appears to be just on the border of possible when the used EV batteries are obtained for free. However, any reasonable buy-down that could lower the purchase price of EVs puts the life cycle cost well above the expected economic benefit from the application. This Complete Energy Storage System Life Cycle Cost Components (Load Leveling/Arbitrage/Transmission Deferral) B% Present value of battery purchases gAccessories, facilities, transportation OPCS, interface equipment, controls Olnstallation & startup Present value of O&M costs is one of the higher value applications, but it also has one of the highest life cycle costs of the systems studied here. The vast majority of this cost (68%) comes from the batteries themselves, as shown in Figure 13. Much as with area regulation & spinning reserve, the only way to significantly reduce the life cycle costs of these systems would be reduced Figure 13. Primary cost drivers for load leveling/energy arbitrage/transmission deferral Final Report 5S testing & reconfiguring costs. Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 6.1.4. Renewables firming The fourth utility scale application, renewables firming, is similar to area regulation and spinning reserve in that it is estimated to have a fairly low economic benefit, but high system life cycle costs. The life cycle costs as a function of life and buy-down are shown below in Figure 14. Ni/MH - Renewables Firming $9800.00 | epee pre S = $500.00 4 x . Life projected | Buy-down ($/kWWh) & = 7 from duty cycle | = $400.00 + 2 2 2 | —s— $50 & 5500.00 s|| —— #10 ‘—_ 4 t + i 3 i } |) —+— $150 8 i eee | — — Low Value $200.00 4 | ----- High Value | oO 2 $100.00 4 4 [ro EES Re ena eee eee ~ | Possible $0.00 | Favorable 0 2 4 6 8 10 Battery life (yrs) Figure 14. Life cycle cost curves for renewables firming Complete Energy Storage System Life Cycle Cost Components (Renewables Firming) @& Present value of battery purchases gAccessories, facilities, transportation 6% OPCS, interface equipment, controls 10% Blnstallation & startup Present value of O&M costs 31% Figure 15. Primary cost drivers for renewables firming Final Report 56 The life cycle costs for a system designed to meet this application are split fairly evenly between the various cost components. As Figure 15 illustrates, the life cycle costs derive mostly from the batteries, the battery accessories and facilities, and O&M costs. It is doubtful that sufficient cost reductions in all of these areas to make this application possible could be achieved. Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications The remaining applications considered in this study would most likely be installed on the customer side of the meter to provide benefits to a particular customer or facility. The results for these commercial and residential applications with Ni/MH batteries are shown in Table 10. Table 10. Analysis of used Ni/MH EV batteries in commercial & residential applications System life Buy-down cycle cost Low value High value Application Life (yr) CU Ls) CUA CUA) Sue ess Power reliability & peak shaving 5 (calendar) $120 $250 $0 $473 Unlikely $50 $540 Unlikely $100 $606 Unlikely $150 $672 Unlikely Light commercial load following 2.6 (cycle) $120 $240 $0 $155 Possible $50 $169 Possible $100 $184 Possible $150 $200 Possible Distributed node telecom 5 (calendar) $32 $53 $0 $12 Favorable $50 $29 Favorable $100 $46 Possible $150 $63 Unlikely | Residential load following 3.9 (cycle) $60 $120 $0 $29 Favorable $50 $42 Favorable $100 $56 Favorable $150 $69 Possible 6.1.5. Power reliability & peak shaving The life cycle cost curves for power reliability & peak shaving are shown below in Figure 16. This application is expected to generate a fairly high economic benefit, but the system designed to meet the application requirements is fairly expensive, putting it well into the unlikely range even with free batteries that last a long time. Final Report 57 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Ni/MH - Power Reliability & Peak Shaving $1,600.00 7 ee sieieemniianono ee eerie © $1,400.00 4 \ es = fe projecte | g $1,200.00 4 \ from duty cycle : | Buy-down ($/KVWVh): E > —+— $0 g $1,000.00 + 3 $50 a 2 a s $100 +5 $800.00 ‘ial 3B S $600.00 4 —_ a alte DBD | fe em High Value S $400.00 + oO OO ee a $200.00 + Possible | $0.00 ; : : Favorable 0 2 4 6 8 10 Battery life (yrs) Figure 16. Life cycle cost curves for power reliability and peak shaving Power reliability and peak shaving is primarily an energy intensive application, and this is illustrated by the cost breakdown shown in Figure 17. Nearly half of the cost of the system for comes from battery accessories and facilities, while about a quarter of the cost comes from the batteries themselves. It is Complete Energy Storage System Life Cycle Cost Components possible that facilities costs (Power Reliability & Peak Shaving) for many potential 10% installations could be Present value of battery purchases reduced through creative siting of the battery system, Accessories, facilities, but this reduction would aati probably not be enough to MPCS, interface equipment, | make this application viable controls in and of itself. It is unlikely that the costs for racking, interconnects, and Present value of O&M a scr esodries-could be Its significantly reduced. 8% Glnstallation & startup Figure 17. Primary cost drivers for power reliability and peak shaving Final Report 58 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 6.1.6. Light commercial load following In contrast, the other two commercial applications look much more promising. Light commercial load following is categorized as “possible” for nearly the entire range of buy-downs and battery lifetimes considered here. The life cycle cost curves are shown in Figure 18 below. NiMH - Light Commercial Load Following Battery life (yrs) $300 oo nv = =] Life projected 2 _ ‘ Fie = S \ from duty cycle = S 8250.0 4 ee ad z NN Buy-down ($/kWh) & CNG —+— 90 = $200.00 4 z @ & . ee : 3 —s— $50 & ssen00 oe ee eS —+— $100 6 a a —— $150 8 p— ae — — Low Value Oo @ 3100.00 4 ee ee ee | C----- High Value S = a 3 2 $50.00 4 s 4 aw $0.00 t T T T 0 2 4 6 8 10 Figure 18. Life cycle curves for Ni/MH in light commercial applications Complete Energy Storage System Life Cycle Cost Components (Light Commercial Load Following) 19% Present value of battery purchases gAccessaories, facilities, transportation OPCS, interface equipment, controls 19% Glnstallation & startup Present value of O&M costs 25% Figure 19. Primary cost drivers for light comm. load following Load following is a fairly strenuous application, requiring nearly continuous low depth of discharge cycles superimposed on daily deep discharges. At higher buy-downs, battery life becomes increasingly important. Based on the projections made here, light commercial load following could support moderate buy-downs even at the somewhat short lifetimes expected given the application duty cycle. However, shorter than Final Report 59 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications predicted battery lives combined with high buy-downs could push this application into the unlikely range. 6.1.7. Distributed node telecommunications standby power The other commercial application that may show promise for the second use of EV batteries is distributed node telecommunications standby power. This application is a bit of a special case in this study in that the economic analysis considered only the batteries rather than a fully integrated energy storage system. This was done because it is likely that the used EV batteries will be used as a direct replacement for the VRLAs currently used for this application and will probably utilize existing electronics, wiring, containers, and other components. Another difference between distributed node telecom standby power and the other applications considered in this study lies in how the economic benefit of the batteries was derived. For this application, the high and low estimates for the economic benefit derived from the used EV batteries were based on the cost range for VRLAs. Since the economic analysis for this study was conducted on a life cycle cost basis, the expected life of the VRLAs was needed to calculate the economic benefit of the batteries over the system life. Unfortunately, definitively determining the life of VRLAs in distributed node telecom applications is impossible. Since they are being used for backup power, the batteries are not cycled frequently. Calendar life is typically the limiting factor that leads to battery replacement in this application. Variations between sites (temperature extremes, exposure to the elements, and other factors) and different batteries lead to a range of lifetimes. To account for these variations, two cases were run in the analysis performed for this study in an attempt to bound expected battery lifetimes. The first case assumed a VRLA life of two years, and the second considered a five year life. These two cases lead to very different results, as shown below in Figure 20 and Figure 21. Final Report 60 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications NUMH - Distributed Node Telecom Standby Power (compared to 2 yr VRLAs) $100.00 4 + 7 = s000 | = saoo0 | Buy-down ($/kWh) tel & $70.00 4 —+— $0 & —s— $50 ® $60.00 | ne > © 350.00 eh 78150 2 2 — — Low Value sue Bl} ----- High Value @ $30.00 = 3 = o g2000 | = 3 $1000 4 = $0.00 . : 5 0 2 4 6 8 Battery life (yrs) Figure 20. Life cycle cost curves for distributed node telecom standby power compared to VRLA batteries lasting two years Ni/MH - Distributed Node Telecom Standby Power (compared to 5 yr VRLAs) $100.00 5 = qo000 J = Buy-down ($/kVVh = $8000 | 7 GaWn) & >|| —+—$0 a $70.00 4 3 : 550 2 $60.00 5 5 —+— $100 & ss000 | —— 1150 oS — —Low Value % $40.00 4 3 4000 NN Cit... High Value @ $3000 4 = $2000 4 | @ $. Possible = $10.00 4 | | Favrabe $0.00 + r r 1 i 0 2 4 6 8 10 Battery life (yrs) Figure 21. Life cycle cost curves for distributed node telecom standby power compared to VRLA batteries lasting five years Final Report 61 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications When compared to VRLAs that last five years, used Ni/MH EV batteries do not appear to be a good candidate for this application. The feasibility is only categorized as possible at low buy- down values and fairly long (four years or more) lifetimes. However, if the competing VRLAs last only two years, then under certain conditions the economic feasibility is classified as possible or even favorable. Since only the batteries are considered in the analysis for this application, there is a strong dependence on both buy-down cost and battery life. In general, if the used EV battery lasts significantly longer than the VRLAs (at least 50% longer), the application can support modest buy-downs. As the lifetime of the EV battery increases, the feasibility of the application improves and higher level buy-downs can be achieved. 6.1.8. Residential load following The residential load following application appears promising. In fact, this is one of only two applications studied that generates a life cycle cost less than the low value estimate, classifying it as “favorable” (see Figure 22). NiMH - Residential Load Following $180 oo 7 7 eae a . es 1 © $160.00 4 \ = \ = sranoo | \ S ‘ © $120.00 +-----\- Ween neces teeny —+— 90 2 \ —a— $50 & 100007 $100 = $380.00 4 x —— $150 8 ‘ — —Low Value 3 $60.00 + 4 ee es a |ti«C} ---- High Value Vv S $40.00 4 2 $20.00 4 $0.00 r t T 1 a 2 4 6 8 10 Battery life (yrs) Figure 22. Life cycle cost of Ni/MH in residential application Final Report 62 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications The favorable economic predictions for this application are due in part to the low balance of system costs, illustrated in Figure 23. Since a residential load following battery will most likely be teamed with a distributed generator (such as a photovoltaic array or fuel cell), most of the necessary power electronics will already be in place. The only additional equipment required is a charge controller and Complete Energy Storage System Life Cycle Cost Components possibly some (Residential Load Following) switchgear. Furthermore, maintenance on these @ Present value of battery types of systems is purchases expected to be very low. Due to these low BOS costs, the system cost for @PCS, interface equipment, 7 a ee 7 controls this application is very sensitive to the cost of the battery. At the projected Present value of O2M life of four years, this costs application can accept buy-downs of up to $150/kWh and still be in the favorable range. However, as battery life decreases, the viability of the application drops off rapidly, particularly for high buy-downs. Still, this application probably represents the best opportunity for the second use of EV batteries. The biggest drawback to this application is that only a handful of modules are required to meet the application requirements, so a large number of customers would be required to provide a large enough market to handle the hundreds of thousands of EV battery modules that may reach the market. 19% Figure 23. Primary cost drivers for residential load following There are also some questions regarding the maturity of the residential distributed generation market, and the widespread market penetration necessary to ensure a home for each used battery may not occur for a number of years. 6.2. Other chemistries A number of battery chemistries have been considered in the past or are currently under development for electric vehicle use. Before the rise of nickel/metal hydride, most full function EVs were made with either lead-acid or nickel/cadmium batteries. Lead-acid batteries are still some of the least expensive and most widely used secondary batteries, and that will not likely change any time soon. In recent years, there has been a trend toward using advanced batteries in full-function EVs, although nearly all of the low-speed NEVs on the market utilize lead-acid batteries. Even so, it is unlikely that lead-acid batteries will find application in a second use scheme since very little benefit will be derived from using refurbished lead-acid EV batteries rather than new lead-acid batteries. In fact, new lead-acid batteries are low enough in cost that a refurbished EV battery could end up being more expensive after testing and repackaging than its new counterpart. Furthermore, limited life is one of the drawbacks of new lead-acid batteries, and this problem will only be compounded by exposing a battery to the environment found inside an EV for an extended period of time before it is placed in the stationary application. Final Report 63 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Nickel/cadmium found extensive use in EVs in Europe, and has been mentioned as a candidate EV battery by at least one U.S. automaker. Even so, there is currently a big push by environmentalists in both Europe and the U.S. to ban the use of cadmium due to concerns regarding the final fate of the heavy metal after disposal. The European Union has already made moves to phase out the use of cadmium, and similar efforts are being pursued in the U.S. Thus, it is unlikely that Ni/Cd batteries will see much use in EVs in the future. There are several other battery technologies under consideration for use in EVs. Lithium-ion (Li-ion) batteries in particular are receiving a great deal of interest due to their potentially high specific energy and specific power. To date, they have not yet proven capable of meeting the cycle/calendar life demands of electric vehicle applications. Lithium-polymer batteries are also under development for EVs, but these are further still from achieving cycle life performance goals. Since these batteries are still developmental in nature, the study team was unable to obtain much data from which to project the performance and life of these batteries in second applications. Much of the analysis described in the previous sections of this report is independent of battery chemistry. In fact, all of the transportation, testing, and balance of system costs should be independent of the type of battery. Packaging cost is probably the only item that may change significantly from one technology to another. Li-ion modules, in particular, require electronics at the module and, possibly, even the cell level to prevent overcharge and cell reversal, both of which can lead to thermal runaway and possible catastrophic disassembly. Incorporation of these electronics into the stationary battery pack may add to the packaging costs. It is unclear at this point what types of electronics will be installed on Li-ion EV modules and how those will be integrated with the battery pack and the vehicle as a whole. It is possible that some of the electronics could be re-used in the stationary battery pack, which could reduce or eliminate the aforementioned increase in packaging costs. Regardless, packaging materials represent a fairly small part of the overall cost of the stationary battery pack, so it is unlikely that these changes will significantly affect the results of the analysis. The only other inputs to the analysis that may change with battery chemistry are the battery capacity and the projected cycle and/or calendar life. Most EV modules currently under development have about the same rated capacity at the start of their lives, so the only major differences in performance over the 2" life will be due to capacity fade and remaining calendar life. The two sets of life cycle cost curves in the next two figures are based on “generic” 12-V 100-Ah EV modules. The modules are assumed to go through a vehicle life that is equivalent to 1000 DST cycles. The first module (Figure 24) is assumed to have a 10% fade in capacity over both lifetimes, leaving it at 90 Ah at the end of the stationary application. The second module (Figure 25) is assumed to fade 30% to 70 Ah under the same conditions. The four vertical white lines indicate cycle lives relative to the life in the vehicle. For example, the 0.25 line represents a module that operates for only 25% of the vehicle life, or about 250 cycles. Similarly, the 2.0 line identifies a module that operates for twice the vehicle life, or 2000 cycles. Final Report 64 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Residential Load Following - 10% Capacity Fade 0. 25 045 10° 2 0 2nd life relative to EY life $1901.000) |-teemetetetemenser veeiereteensel dees certaeeeereernemeieeeneieealeeain! S $160.00 + = S 3] XS $140.00 4 S| Buy down ($/kWh) FS & $12000 +----4--)-------------------------------------5: —+— 30 2 ‘ % —a— $50 Zz $100.00 + . 2 $100 S = ane = $80.00 | & $150 3° — — Low Value °° o $60.00 aoa =e 7 o----- High Value Vy ~~ > SL o.oo | naa SS 2 © i Se ae eS = “0 -—4——4—___¢- 44 5 $20.00 4 : : w $0.00 i T : r ; a 2 4 6 8 10 Battery life (yrs) Figure 24. Life cycle cost curves for a generic 100Ah 12V EV battery that experiences a 10% capacity fade over its entire life Residential Load Following - 30% Capacity Fade $180.00 a 25 0. 5. 1.0 2 20 2nd life relative to EY life S $160.00 = $140.00 | Buy down ($/k¥Vh) Ss © $120.00 —— 40 2 —s— $60 = $100.00 $100 = $80.00 —— $150 3 — — Low Value 9 ® 00.00 (a2 Se Oe eee eC; ---- High Value 2 o 2 a $40.00 $20.00 | $0.00 1 + + “ 0 2 4 6 8 10 Battery life (yrs) Figure 25. Life cycle cost curves for a generic 100Ah 12V EV battery that experiences a 30% capacity fade over its entire life Final Report 65 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Comparing these two plots reveals very little difference between the economic feasibility of the two modules. In fact, the additional 20% change in capacity fade had only a marginal impact on the life cycle cost of the battery system. It also slightly shifted the projected lifetime in years for the various cycle life projections. It would appear that the analysis performed above for Ni/MH should hold for any battery chemistry that can provide a similar cycle life in the second application. Furthermore, the economic feasibility of the various applications is only affected when the life of the used EV module is very short (a year or less). Unfortunately, the study team was unable to obtain cycle life projections for the more developmental battery technologies. Should such data be obtainable in the future, plotting the projected life on the figures developed for this report should give a good first order estimate of the economic feasibility of a second use for the battery module in question. Final Report 66 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 7. ISSUES AND BARRIERS In the process of determining the technical and economic feasibility of using EV batteries in second applications, the study team identified several issues that may prove to be hurdles to battery reuse if they are not addressed. Some of these issues are technical in nature, while others are more economic or market oriented. 7.1. 7.2. Technical Lack of available battery life & performance data for modules coming out of EVs. It is difficult to determine whether or not using EV batteries in second applications will work without more test data on the performance of modules after they are removed from vehicles. Life projections in the second application are compounded by the variability that can be expected in vehicle duty cycles, which will affect the cycle life, as well as climate extremes, which could impact calendar life. Definitive data on the spread of EV module capacities and power capabilities, as well as cycle life tests in second applications, will be required before second use of EV batteries can be considered feasible. Difficulty in obtaining modules with similar capacities for matched strings. The modules in a battery string need to have similar capacities in order to ensure maximized utilization of battery capacity and prevent possible damage to individual modules. This is particularly true for deep discharge applications, although it is less important for pulse power or low depth of discharge cycling. If the variability of the EV application described above results in a wide range of capacities in used EV modules, it may be difficult to obtain sufficient modules to assemble a matched string, particularly for larger applications. Non-standardized battery modules. Both of the aforementioned issues are compounded by the potentially large number of battery suppliers involved in the market. If every automaker uses a different module configuration, locating matched modules for assembly into packs, and matched packs for assembly into larger systems could prove difficult. Integration of power electronics for Li-ion modules. Li-ion cells are susceptible to thermal runaway and catastrophic failure during overcharge or cell reversal. Li-ion battery packs will likely have special electronics to prevent these conditions from occurring. These electronics will at least reside at the module level and may even be present at the cell level. When the used EV battery modules are reconfigured into stationary battery packs, these cell and module level electronics will either have to be integrated into the pack level electronics or replaced with similar components that can be interconnected with the rest of the battery management system. Since it is currently unclear exactly what a Li-ion EV battery module will look like, the implications of integrating these electronics on both the economic and technical feasibility of a battery reuse process were not determined in this study. Market Unclear chain of custody for an EV battery reuse process. If the second use is to have any impact on the commercial viability of electric vehicles, the buy-down from the second use must find its way back to the vehicle purchaser. While this could take the form of a rebate Final Report 67 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications at the time of battery trade-in, it is unlikely that most people would be willing to pay more upfront for the car based on the assumption that they will get their money back at a later date. The most effective approach would be to give the vehicle purchaser the discount up front. This could be accomplished in several ways. 1. The automaker could purchase the battery from a supplier at full price and not charge the vehicle owner for the entire cost of the battery based on the assumption that they will recoup the loss when they sell the battery into the second application. N Alternatively, the battery manufacturer could maintain custody of the battery for the entire vehicle life, and essentially lease it to the EV owner. When it is removed from the vehicle, the battery could then be reconfigured by the battery manufacturer and sold to a stationary application. However, one of the battery manufacturers consulted during this study expressed displeasure with the idea of leasing the battery or dealing with used EV battery modules. 3. The reconfiguring could also be performed by a third party who would then sell it to the stationary application user or a system integrator. The point of all this is that there are a number of ways a battery reuse process could be carried out. While this does not necessarily create a barrier to the concept, a chain of custody will have to be established that results in reduced cost for the EV buyer and is acceptable to all parties involved. Chain of custody may also be important for determining warranties — who will warrant what to whom. e §=Warranty terms and costs. Given the uncertainty in the used battery performance and life together with the lack of experience with some of the systems considered here, it will be difficult to establish warranties and budget for them appropriately. However, warranties will be absolutely essential to the success of any second use process. Simply put: no warranties, no sales. e = Perceived value. The utility industry places a premium on reliability and is notoriously skeptical of new technologies and unproven approaches. It is unlikely that utilities will readily buy into the concept of used batteries. Other potential customers may be more open-minded, but the perceived value of used goods may play an important role in the success of EV battery reuse. e = EV production uncertainty. Given the flexible nature of the mandates that currently appear to be driving EV production, the potential for other vehicle technologies (primarily fuel cells), and the whims of the consuming public, the future of battery electric vehicles is far from certain. As long as there is not a guaranteed stream of EV batteries, it will be difficult to convince anybody to invest in the facilities necessary to process EV batteries for a second use. Final Report 68 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 8. CONCLUSIONS Overall, the concept of EV battery reuse appears to be a viable one. The study team did not come across any insurmountable technical barriers to the implementation of a second use scheme. In fact, during the course of the study it was learned that there is already considerable commercialization of used and reconditioned batteries. Furthermore, a study by Argonne National Laboratory examined the second use of Ni/MH EV batteries and showed that modules tested to end-of-life on the United States Advanced Battery Consortium (USABC) Dynamic Stress Test (DST) profile could provide performance competitive with new lead-acid batteries in stationary energy storage applications. While there are no technical “show stoppers,” there are some issues that will have to be dealt with before an EV battery second use scheme can be implemented. First, non-standardized battery modules and varying patterns of vehicle use could make assembly of matched strings of modules with similar capacities difficult. Second, the mechanism by which the value from the second use makes its way back to the EV buyer needs to be identified. Third, warranty terms and costs will be difficult to determine given the uncertainty in the performance and life of the used EV batteries. This is an important issue since warranties will be absolutely essential in achieving market acceptance of the batteries. Finally, the perceived value of used batteries relative to new batteries in the consumer's mind will have to be addressed to ensure widespread acceptance of used EV batteries. Used EV batteries will most likely be available as individual modules rather than entire EV battery packs. These modules will have to be collected from vehicle dealerships or service centers, inspected to ensure physical and electrical integrity, tested to determine performance, and reconfigured into battery packs suitable for stationary applications. The testing and reconfiguration process considered here is relatively simple, but essential for the success of the EV battery reuse concept. Testing is required to establish the capacity and power capabilities of the used EV modules. This information will be needed in designing the stationary battery system. It will also aid in predicting a module’s remaining life, which will be necessary in establishing warranties. The labor and equipment required for EV battery testing and reconfiguration will represent a significant portion (roughly half) of the final selling price of the used EV modules into the second application. In spite of the costs involved in the refurbishing process, EV battery reuse looks like it could be an economically viable concept. The economic analysis conducted for this study looked at the feasibility of applying used EV batteries in eight different stationary applications. Of these eight, the feasibility of applying used EV batteries in four of the applications (transmission support, light commercial load following, distributed node telecommunications backup power, and residential load following) was classified as either favorable or possible. It should be noted that, for many of the applications considered, the battery was not the primary cost driver in the life cycle costs. Often the most expensive component of the system was the power conversion electronics or operating and maintenance costs. The cost of the battery alone is insufficient to determine the economic feasibility of a stationary energy storage system. Surprisingly, the four applications classified as economically feasible do not have a lot in common. In fact, they include both the largest and the smallest of the battery sizes considered Final Report 69 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications here. They also do not show any similarity in how much of the system cost is represented by the battery. It is difficult to draw any generalizations on application requirements, system sizes, or battery characteristics from the results shown here. It would appear that used EV battery/stationary application combinations have to be considered on a case-by-case basis. Finally, there are sufficient uncertainties in the values used in this analysis (particularly in the performance of the batteries and the economic benefits derived from the applications) to warrant testing and demonstration of the concepts presented here. Potential testing and demonstration programs are detailed in the following section. Final Report 70 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 9. RECOMMENDATIONS Based on the analyses performed for this study, using EV batteries in a second stationary application appears to be a viable concept. However, there are sufficient uncertainties and limitations in the scope of the present study to warrant additional work in this area. The largest unknown in any hypothetical second use process is the performance and life remaining in the battery when it is removed from the vehicle. While conversations with battery manufacturers indicate that some of this testing has been performed, at present there is no publicly available data regarding used EV battery characteristics. Any future programs seeking to further the reuse concept should include significant testing of modules taken from vehicles to characterize their performance and evaluate their capabilities under various stationary application duty cycles. Implementation of second uses of EV batteries will also require proof of concept and early deployment programs to show potential investors that the idea is a valid one and to identify implementation difficulties. The analysis performed for this study indicates that future demonstrations should focus on four applications: transmission support, light commercial load following, residential load following, and distributed node telecommunications backup power. A description of possible programs and a cost estimate for each follows. 9.1. Used EV battery module testing Due to the uncertainties in used EV battery performance, any demonstration program must be preceded by, or run concurrently with, a focused testing program designed to characterize the used modules and quantify their capabilities and lifetimes under stationary application load profiles. The testing should include a fairly large number of used modules to provide a statistically significant sample that could identify differences between chemistries, vehicle use patterns, and climate extremes. 9.2. Transmission support While transmission support proved to be the most PPeUWSsiiy sens economically viable utility scale application in the | Number of systems 1 analysis for this study, it will also be one of the Length of demonstration 3 months most difficult to demonstrate. Based on the EV battery modules 600 assumptions used here, a 100 MW transmission Equivalent EV battery packs 24 support system would require over 17,000 used System storage capacity 650 kWh EV modules, which would require the batteries System rated power 3.2 MW from nearly 700 vehicles. In fact, no automaker has sold more than 1,000 full function EVs with the same battery. Thus, it is unlikely that more than 10,000 or so of the same battery module would be available for demonstration programs in the near term. However, a bench-scale test of used EV modules under a simulated transmission support load profile would be insufficient to prove the validity of the concept. One of the biggest challenges in deploying used batteries for this application is showing that large numbers of modules can be strung together, and large numbers of strings can be interconnected in a reliable fashion to generate high voltages and meet the high current demands for the application. Thus, any Final Report 71 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications demonstration for this application should involve a battery system with a voltage over a thousand volts, comprised of several strings containing hundreds of modules each. A reduced scale system would be unable to meet the needs of a real world transmission support application. Furthermore, it is unlikely that any utility would be willing to potentially jeopardize the integrity of a transmission asset to demonstrate an unproven concept utilizing used components. Therefore, any demonstration of this concept will likely rely on testing in a laboratory setting with sophisticated simulation equipment. There are a number of facilities located in the U.S. and Europe that perform testing and validation of transmission equipment. It is likely that any of these could handle the types of tests required to put a battery system for transmission support through its paces. The study team recommends a program including the design, assembly, and testing of a reduced scale battery system for transmission support. The system should be comprised of roughly 600 battery modules, divided into three strings of 200 modules each. Such a design would generate 2400V, store about 650 kWh, and be capable of a pulse power output of around 3.2 MW. The battery should be integrated with the necessary electronics, monitors, and controls and installed at a transmission equipment test facility for several months of testing under simulated transmission support conditions. The first step in this program should involve identifying partners and locations for both obtaining used EV batteries and performing the testing. It is likely that the modules from these batteries could be obtained free of charge from automakers or battery manufacturers through some sort of cost sharing partnership. After it is clear that sufficient EV battery modules are available, system design and component procurement can proceed. As with all of the demonstration programs, the used EV battery modules should be characterized before they are assembled into the energy storage system. Assembly and integration of the system will probably occur onsite at the testing facility. Several months of testing will be required to characterize the capabilities of the battery system and identify potential problems with actual installations. Since transmission support is primarily a pulse power application, the life of the batteries will be limited by calendar life issues, not cycle life limitations. Therefore, parallel accelerated calendar life testing may be of interest. Once testing is completed, the results of the program should be analyzed and reported. A summary of these steps and the costs associated with them is shown below. Activity Cost 1. Identify partners and site for testing $50,000 2. Obtain used EV modules of the same type Cost share 3. Design system $200,000 4. Obtain the balance of system components PCS, interface equipment, controls and monitors $2,870,000 Interconnects, racking, cooling $175,000 5. Perform baseline performance tests to characterize the modules $600,000 7. Assemble and install system $100,000 8. Perform system testing $85,000 9. Analyze success of system; identify barriers to implementation and quantify value $50,000 TOTAL $4,130,000 Final Report 72 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 9.3. Light commercial load following In a light commercial load following application, ey the battery system is utilized to allow a distributed ht Commercial Load Following : Number of systems 5 generator to operate at or near constant loading. Length of demonstrations 2 years This is typically a more efficient mode of EV battery modules/system 100 operation for the generator, and it can Equivalent EV battery packs 4 significantly reduce the size (and cost) of the System storage capacity 110 kWh distributed generator since it does not have to System rated power 200 kW meet the entire peak load. When operating in a grid-connected mode, a battery can be used to avoid demand charges. The demonstration program for this application should include several installations to test the use of the battery with an array of different distributed generation technologies, such as microturbines, fuel cells, and fossil-fueled gensets. The study team recommends a total of five installations, the demonstration at each site lasting for two years. Each system will utilize roughly 100 used EV battery modules, contain about 100 kWh of storage capacity, and generate around 200 kW of peak power. Exact values for these system parameters will depend on the needs of the site where the system is installed. The first step in mounting this demonstration program will be to identify sites and partners for installing the battery systems. Facilities operating under rate structures with high on-peak electricity charges or high demand charges should be the focus of this effort. This application offers an interesting opportunity for leveraging R&D funds with other DOE programs seeking to demonstrate distributed generation (DG) in a light commercial setting. Sites already selected for DG projects could be retrofitted with refurbished EV batteries. Alternatively, a system could be designed from the ground up in partnership with a DG demonstration program. Once sites and partners have been selected, the energy battery systems will need to be designed. It is likely that each site will require a different system, depending upon the DG technology in use and the application load profile. After the designs are completed, the system components will have to be procured. It is likely that the EV battery modules could be obtained through some sort of cost-sharing arrangement with an EV battery manufacturer or automaker. Furthermore, the distributed generator could be purchased by another DOE program. Depending on the distributed generator technology, it may be possible to integrate the battery system with the electronics built into the generator to minimize the cost of the power electronics. The costs of the facility to house the energy storage system (which could be as simple as a space in the basement of the building or a trailer parked outside) would probably be provided as a cost share by the owner/operator of the property where the system is installed. Once the EV batteries have been characterized, the system can be assembled. While the batteries could be connected together and placed in some sort of shipping container elsewhere, integration of the generator into the system will probably have to occur onsite. After the site has been prepared, system installation, startup, and shakedown can proceed. This will probably take a couple of weeks of labor from an engineer and a technician. During operation, system performance can be monitored autonomously, but occasional visits for data collection and analysis will be required over the two year operating period. Regular and unscheduled maintenance visits will also be necessary to keep the system operating in an optimum manner. The study team figures that a one day visit to each site once a week by an Final Report 73 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications engineer familiar with the system would be sufficient to monitor system operation and handle maintenance issues. At the end of the two years, a report analyzing the success of all five systems and identifying pitfalls and unsuspected problems should be generated. The estimated costs for all of these steps are summarized below. Activity Cost/system Total cost 1. Identify partners and site for demonstration —— $50,000 2. Design system $50,000 $250,000 3. Obtain used EV modules of the same type cost share 4. Obtain the balance of system components Generator cost share PCS, interface equipment, controls and monitors $130,000 $650,000 Interconnects, racking, cooling $21,000 $105,000 Facility cost share 5. Perform baseline performance tests to characterize the modules $100,000 $500,000 6. Assemble system $100,000 $500,000 7. Prepare site, install and test system $14,000 $70,000 8. Monitor and maintain system $120,000 $600,000 9. Analyze success of system; identify problems and quantify value ne $50,000 TOTAL $485,000 $2,775,000 9.4. Residential load following This application is much like light commercial load following, only on a smaller scale. In Residential Load Following addition to the microturbines, fuel cells, and ; a eae ations 2 aan fossil-fuel gensets mentioned above, residential EV battery modules/system 5 load following could also involve renewable Equivalent EV battery packs 02 generators such as PV or possibly even a small System storage capacity 5 kWh wind turbine. We would recommend at least four System rated power 10 kw types of systems for demonstration: 1. grid-connected household with a roof-top PV array and net metering 2. off-grid household with PV 3. off-grid household with a propane genset (and possibly wind or PV) 4. off-grid household with a fuel cell In all, a total of ten systems would be desirable. Each system would consist of 5 or so EV battery modules, with a total capacity of about 5 kWh and a rated power of around 10 kW. At this size, ten systems would be feasible. Two year tests would be desirable to show the operation of the system over varying seasons and prove its long term durability. As with all the demonstrations suggested here, the first step will be to identify partners and locations for potential installations. Much as with light commercial load following, opportunities for synergies with other government and commercial programs exist for this application. Several fuel cell companies are in the process of testing and demonstrating prototype fuel cell systems for residential power generation, and they may be willing to utilize used EV battery modules in their systems. The Department of Defense is ramping up a residential fuel cell demonstration Final Report 74 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications program at bases across the country. And Texaco Ovonic Battery Co. is already installing used EV battery modules with PV arrays in remote Mexican villages. Leveraging funds and facilities with these programs could significantly reduce the costs of installing a demonstration system. Unlike the other demonstration programs described here, the battery systems designed for each of the residential sites would probably be quite similar. To reduce costs, the study team recommends that a single battery system be designed for integration into a number of residential distributed generation systems. This system could be assembled into a box or two at a central facility and shipped to the application site for integration into the generating system. Some residential generators may already have power electronics capable of handling the battery system, although others will require additional components. It is possible that cost of the electronics, controls, and monitors will have to be borne by the demonstration program, but the distributed generator and any facility costs will likely be covered through cost-sharing partnerships with other DOE programs or with the end user of the system. The used EV batteries will probably also be obtained free of charge through a cost share from a battery or vehicle manufacturer. After the system design is complete and the components are obtained, the batteries can be tested and assembled into a self contained system at some central facility. It is assumed here that assembly of the battery systems and initial shakedown will take about a week for an engineer and a technician. After this initial checkout, the system will be ready for transport to the application site. Another week of labor will be required to integrate the battery system with the generator and test the entire power system. Monitoring, maintenance, and repair will be required once the system is up and running. Most of the monitoring for these systems could be done offsite, and an allowance for telemetry was included in the cost estimate for monitors and controls given below. Each site should only take a half day or so of an engineer's time to analyze the performance of the system once a week. An additional visit to the site once every two months should cover any regular maintenance or unscheduled maintenance for each site. All of these functions should be continued for the entire two years of the demonstration. When the two years are completed, a report summarizing the performance of all the systems and analyzing their economic benefits should be produced. All of these steps and their approximate costs are summarized below. Activity Cost/system Total cost 1. Identify partners and site for demonstration -—- $50,000 2. Design system —- $50,000 3. Obtain used EV modules of the same type cost share 4. Obtain the balance of system components Generator cost share PCS, interface equipment, controls and monitors, interconnects, racking $20,000 $200,000 Facility cost share 5. Perform baseline performance tests to characterize the modules $5,000 $50,000 6. Assemble system $5,000 $50,000 7. Prepare site, install and test system $7,000 $70,000 8. Monitor and maintain system $48,000 $480,000 9. Analyze success of system; identify problems and quantify value ooo $50,000 TOTAL $83,000 $1,000,000 Final Report 75 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications 9.5. Distributed node telecomm Unlike the other three applications mentioned in ee De eG this section, distributed node telecommunications backup power has a well-established market Number of systems currently being met by batteries. This application a - ; offers a potential early entry market for used EV Length of demonstrations batteries. It also presents the opportunity to work Phase I 1 year alongside with the potential end users of the Phase II 2 years battery system to accurately specify the EV battery modules/system 48 application requirements, design a system that Equivalent EV battery packs 2 meets their needs, and test it in real-world System storage capacity 52 kWh situations alongside competing technologies. System rated power SkW However, distributed node telecomm standby power requires an extremely reliable battery system. Before telecommunications companies are willing to accept used EV batteries as a replacement for the new VRLAs they current use in this application, they will have to be convinced of the reliability and durability of these batteries. The study team recommends a two phase demonstration program designed to show that used EV batteries can meet the needs of this application in a reliable fashion. Furthermore, this program should be run in such a way as to minimize both the risk and the cost to the telecommunications company that is involved. The first phase of the demonstration program will consist of prototype design, assembly, and testing. Since it is unlikely that telecommunications companies will be willing to install an unproven system in the field, this prototype will be put under extensive testing at a telecom company test facility for about a year. During this testing period, the prototype system will be operated under a wide range of conditions to monitor its performance and ensure it will function as designed under all the conditions it would be expected to encounter in actual installations. If Phase I is successful, the program will proceed with Phase II, in which 9 battery systems will be placed in the field for two years and carefully monitored. As with all of the programs recommended here, the first step will be to identify partners and sites for the demonstration. The telecommunications company will be a particularly key player, and will need to be identified early in the process to ensure involvement in all facets of the program. Once the players have been identified, the engineering team can work closely with the telecommunications company to design a system that meets their needs. This will be followed by procurement of the various components for one system. As with the other programs, the EV batteries could be obtained free of charge through a cost sharing agreement with either a battery manufacturer or automaker. It is assumed that most of the rest of the system components will be provided as a cost share from the telecommunications company since the system will have to be integrated into their equipment. There is an allowance for monitors and telemetry included in the budget estimate for this program since these systems will likely be heavily instrumented to monitor their behavior. Once the first set of battery modules has been benchmarked, it will be ready for integration into the prototype. It is assumed here that assembly will take approximately two weeks of labor from an engineer and a technician, and shakedown will take another two weeks. Once the system has been checked out, it will be ready for testing by whatever protocols the telecommunications Final Report 76 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications representatives recommend. This is expected to take about one day per week of time from an engineer working onsite at the test facility to monitor the system performance, set up tests, move the system around if necessary, and perform maintenance or repairs. After Phase I has been successfully completed, the prototype system design can be evaluated to identify necessary improvements. The materials for the nine Phase IT systems can be obtained from the same sources used for the prototype system. After battery benchmarking, the second set of systems can be assembled and tested, which is predicted to take a total of two weeks for each system. Installation in the field should take a day or so. Monitoring will most likely be performed at a central facility via telemetry installed in the battery system and should only require a couple of days a week to analyze the data from all the systems. One maintenance or repair visit is scheduled each month to meet the needs of all the systems. Finally, once the second phase of the project reaches completion, a report analyzing the performance and economics of the used EV batteries in this application should be generated. The steps involved in both Phases and their estimated costs are summarized below. Activity Cost/system Total cost 1. Identify partners and site for demonstration —— $50,000 2. Design system + $50,000 3. Obtain used EV modules of the same type cost share 4. Obtain the balance of system components PCS, interface equipment, controls and monitors $1,000 $10,000 Interconnects, racking, cooling cost share from telecom Facility 5. Perform baseline performance tests to characterize the modules $48,000 $480,000 6. Assemble & test prototype a $61,000 7. Assemble remaining systems $9,900 $99,000 8. Prepare site, install and test systems $2,100 $21,000 9. Monitor and maintain systems —— $182,000 10. Analyze success of systems; identify problems and quantify value ooo $50,000 TOTAL $37,000 $1,218,000 Final Report 77 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications This page intentionally left blank. Final Report 78 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications APPENDIX A: INTERIM REPORT ON BATTERY REUSE INFORMATION SEARCH Introduction The current high cost of batteries is a major barrier hindering the more widespread adoption of electric vehicles (EVs) and hybrid-electric vehicles (HEVs). Most efforts addressing this obstacle consist of research and development aimed at reducing the cost of producing EV and HEV batteries. With support from Sandia National Laboratories, this study is examining another often overlooked opportunity: the potential for recovering some of the original value from a battery that has served its useful life in an electric or hybrid-electric vehicle, but which is still suitable for another application. The premise is that the battery could be refurbished, resold, and reused for another application, and some of the initial cost recovered through the sale of the battery in a secondary market. The objective of this study is to determine the technical and economic feasibility of reusing spent EV and HEV batteries, and to identify potential stationary applications. Methodology A literature search was conducted as part of this study. Originally, it was to focus solely on EV and HEV batteries. The scope of the search was expanded, however, to also include the reuse of batteries taken from non-EV/HEV applications during their first lifetime. Very few electric vehicles (EVs) and hybrid-electric vehicles (HEVs) have been commercialized to date, and far fewer have had their batteries reach end of life. Consequently, few EV and HEV batteries have been removed from vehicles and sold or used in a secondary market. It was assumed that knowledge gained on existing secondary markets for other batteries would be useful to determine the feasibility of establishing a secondary market for spent EV and HEV batteries. The literature search involved making personal contacts by phone and/or e-mail, and conducting literature and Internet searches. Contacts were made with organizations and companies dealing with electric and hybrid-electric vehicles, or which install battery renewable energy systems, to determine if they knew of any secondary battery markets. Searches on the Internet as well as through non-peer-reviewed publications were conducted on the secondary use of batteries. In addition, publicly available online databases (such as SciFinder© and INSPEC) were used to search peer-reviewed publications for information on EV/HEV battery reuse as well as issues surrounding cycle life and performance degradation in EV and HEV batteries. Final Report A-1l Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Summary of results The study findings demonstrate that there is considerable commercialization of used and reconditioned batteries, including batteries from EVs: e Amateur radio emergency operators acquire used gel cell lead-acid batteries from hospitals. e There is established commerce in used off-lease forklift batteries, reconditioned automotive starting, lighting, and ignition (SLI) batteries, and reconditioned lithium-ion batteries for laptop computers, as well as batteries with various chemistries for energy storage in small renewable energy systems. e AC Propulsion, a small-volume electric vehicle manufacturer, is implementing a successful secondary market for spent Optima® deep-cycle spiral-wound recombinant lead-acid batteries. e Energy Conversion Devices, a manufacturer of both solar photovoltaic modules and nickel/metal hydride (Ni/MH) batteries, is participating in a solar program in Mexico, which incorporates used Ni/MH EV batteries taken from EV bench tests and prototype EVs. e Argonne National Laboratory conducted a study in 1996-1997 for the United States Advanced Battery Consortium (USABC), examining the second use of Ni/MH EV batteries. The study demonstrated that there is merit in considering the reuse of EV batteries for other applications. Second Use of Electric and Hybrid-Electric Vehicle Batteries A small secondary market for various battery chemistries exists for a range of applications. The most extensive secondary market is for lead-acid batteries, several types of which are being reused in a number of applications. Comprehensive Automotive Reclamation Services (CARS) of Maryland, a Baltimore company that disassembles used and damaged cars and light trucks, added reclaiming and reselling lead- acid batteries after purchasing a Pulsetech digital battery analyzer to determine which batteries had bad cells and which could be reused. CARS of Maryland shifted from recycling all the batteries, earning about $1.50 each, to reselling reconditioned batteries for about $25 each. “In just one month, CARS of Maryland has recovered about 500 batteries using the Pulse Recovery System™, The company can now resell these batteries to new markets such as cab and other fleet customers.” [1] International Business Club, a Canadian company, acts as a broker for business opportunities, including reconditioning automotive lead-acid batteries. [2] Its website provides information on services the company provides to set up a battery reconditioning business. Few technical details on how to recondition batteries or on their Batteries Tech 2000 reconditioning system are provided other than the statement that Batteries Tech 2000 can recondition 10 batteries simultaneously in a period of 20 hours. They have reconditioned and sold more than 25,000 batteries over the past three years. All batteries were sold with 6- to 12-month warranties, and less than 5% of the reconditioned batteries were returned. Defective batteries were replaced with reconditioned batteries. Battery recyclers were one source where they received their batteries to Final Report A-2 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications recondition. Their customers have included used automobile dealers, new and used auto parts dealers, fleet owners, retailers of trucks and heavy equipment, and to a lesser extent golf courses. Hospitals regularly replace sealed lead-acid (SLA) batteries (gel cells) used to power medical diagnostic instruments, alarm systems, and uninterruptible power sources (UPS) on a fixed schedule before they are worn out. These batteries often can be obtained for free to power emergency communications activity. An Amateur Radio Emergency Service (ARES) member gives advice to other members on how to inspect, recharge, and test donated gel cells. [3] Inspection: Check out open circuit voltage to expedite distribution by sorting out batteries, which may be load tested immediately. Any 12-volt batteries having an open circuit voltage (Voc) of 12.8 V or greater are ready for load testing. Those with Voc of less than 12.8 V are charged by connecting in parallel across a regulated 13.8-V power supply. Any that are not accepting charge after 4 hours are discarded. Total charging should not exceed 140% of capacity. Testing: Batteries which accept charge to Voc greater than or equal to 12.8 V must still be load tested after recharging. An easy quick check is to apply a load in amperes which approximates battery capacity in amp-hours, for 10 seconds, monitoring voltage drop. In a "good" battery the voltage drops, but quickly stabilizes after a few seconds, does not continue to fall and recovers within a few seconds after the load is removed. The author has gotten reliable service from 12-V batteries that don't drop below 11.7 V at "C" load for 10 seconds. If time is available, and there is only a small number of batteries, a better test is to approximate a continuous work load for at least a full minute. The author tests 12-V batteries up to 2 Ah with an 8-W fluorescent light at 0.6-A load. Larger ones up to 10 Ah can use a 12-V, 50-W incandescent lamp at 4-A load. For larger batteries, connect the test battery to the intended transmitter and the transmitter to a nonradiating dummy load, monitoring voltage drop for a minute of full-power key-down. Accept for reissue batteries which do not exhibit more than 0.5-V voltage drop at normal working load and duty cycle. From the author’s personal experience, one in ten donated batteries is rejected and recycled. When subjecting a battery to a current load which exceeds C/5, or 1/5, of its amp-hour capacity, expect a 25-30% reduction in its delivered capacity. At lower temperatures available capacity is further reduced. Lead-acid batteries typically lose 50% of their capacity at 32° F! A ham radio operator and member of the Electric Vehicle Association of Washington, DC confirmed that ham radio operators often use used 12-volt gel cell batteries. [4] Deep-cycle lead-acid batteries are commonly found in secondary markets. A member of the Electric Vehicle Association of Washington, DC, who is starting to design his own three-wheel EV, said that a store, Battery Warehouse on Gude Drive in Alexandria, Virginia, sells used batteries, such as forklift batteries. [4] SURPLUS TRADERS, a major buyer and importer of manufacturers’ excess inventories, maintains a business-to-business website aimed at redistributing these materials to manufacturers, exporters and large quantity users. [5] On 9/18/01 there was the following posting: “FORK LIFT BATTERIES WANTED: We have buyers for off-lease and surplus forklift batteries in usable condition and forklift battery chargers in any condition.” Final Report A-3 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications AC Propulsion, developer of the “tzero” high-performance electric sports car about to enter small-volume production reported that it has implemented a very successful secondary market for spent Optima® yellow top batteries. The used EV batteries are “sold for SLI (starting, lighting, and ignition) application at $25 each—about 25% of the original cost—with a two-year warranty.” [6] pp 9-10. One typical application for used EV and HEV batteries is a small renewable energy system. According to Richard Perez [7] of Home Power magazine, a publication devoted to helping individuals set up and operate small renewable energy systems, dozens of articles on used batteries have appeared in the do-it-yourself renewable energy magazine over the years. One homeowner, for example, recently reported on his successful use of used C&D lead-acid batteries in his stand-alone solar-electric photovoltaic (PV) home system. [8] He had acquired these batteries as surplus from telephone company backup service in 1983 when they were five years old. After having been told frequently that his batteries must be about to crash or were somehow unfit to rely on, he decided to conduct a drawdown and load test run. The homeowner contacted the battery manufacturer in Pennsylvania and identified his battery bank as consisting of C&D KCT-720 cells rated at 720 Ah at the 8-hour rate, or 882 Ah at the 20-hour rate. In April 1999, he disconnected the PV array and turned on many of the household loads to discharge the battery to a 20 percent state of charge, which he achieved after 4 days. He then turned on all the lights in the house, and loaded the inverter with the freezer and the deep-well pump to test whether the voltage would hold up to keep the Trace SW4024 inverter on for all essential services. He then added a 1-kW hotplate to the load. The voltage held at 21.4 volts under a 104-amp load on the severely discharged batteries. The test was successful, with the inverter operating without failure. He then shut down all the loads, reconnected the PV system, and began charging the battery at a 10-amp rate. After four variably sunny days, the battery was fully charged. The doomsayers who said the homeowner’s system was about to crash were proven wrong, since he indicated that the batteries performed normally during the two years since the test. Oasis Montana Inc. Alternative Energy Supply and Design maintains a website that sells surplus and used renewable energy components, including used batteries. [9] On 9/18/01 there was a posting for “16 Marathon 12V 125Ah Batteries. Approx. 12 years old.” The origin of the batteries was not given except to indicate that they are located in Saskatchewan, Canada. Paul Hess at the Energy Efficiency and Renewable Energy Clearinghouse, the U.S. Department of Energy’s clearinghouse for information on energy efficiency and renewable energy, indicated that he had met someone who was at Solar Energy International in Carbondale, CO about 10 years ago who reconditioned batteries and resold them for renewable energy systems. [10] Solar Energy International (SEI), which installs small renewable energy systems and conducts training courses on teaching others how to do installation, reported that the battery distributor formerly located in Colorado closed his marginal operation several years ago. [11] Peter Lowenthal, of the Solar Energy Industries Association, the trade association of U.S. solar energy companies, reported that many of the batteries in small renewable energy systems in developing countries are lead-acid, either automotive starting, lighting, and ignition (SLI) or deep-cycle lead-acid. Often they are poorly maintained or abused, are discharged extensively, and are not adequately charged before the next usage. [12] Final Report A-4 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications In addition to deep-cycle lead-acid batteries, commercial and concept electric vehicles employ nickel/metal hydride and lithium batteries. In Europe, nickel/cadmium batteries are commonly used as well. Hugh Marrow of the International Cadmium Association reported by phone that used nickel/cadmium batteries develop high impedance, but can be used for low drain applications for as long as 10 more years after no longer being of use in EVs. [13] Jade Mountain, a company specializing in the retail of renewable energy products and systems, several years ago had reconditioned batteries for sale: nickel/cadmium, nickel/iron, and Ni/MH batteries. [14] They stopped selling them because of low demand and difficulty in obtaining supply. “Problem was that we couldn't afford to stock expensive, slow-selling products such as this, and the supply wasn't too reliable, so when orders finally came in, we'd typically find that there were no batteries to be had,” they reported by e-mail. Both the editor of EV World®, a website specializing in electric and hybrid electric vehicles, and the President of the Electric Vehicle Association of Washington, DC, indicated that Energy Conversion Devices was exploring a secondary use for their used Ni/MH batteries. [15] [16] Energy Conversion Devices, a manufacturer of both solar photovoltaic modules and nickel/metal hydride (Ni/MH) batteries, reported by telephone that they are participating in a solar program in Mexico which incorporates Ni/MH EV batteries that have completed lab testing or useful life in prototype EVs. [17] Dell Computer, a computer retailer, sells refurbished lithium-ion batteries for laptop computers on its website: <http://www.dell.com/us/en/gen/default.htm>. [18] A search conducted on September 26, 2001 on “Reconditioned batteries” at their Home and Home Office use site, <http://accessories.us.dell.com/sna/index.asp?customer_id=19>, produced seven hits on refurbished lithium-ion batteries for sale. In order to get an idea of how these refurbished batteries differ in price and warranty from new batteries, the prices for these refurbished lithium-ion batteries were compared to prices for new, comparable lithium-ion batteries listed for the same Dell laptop computers. In some cases, the watt-hour listing on the Dell refurbished batteries differed slightly from the amp-hour and voltage ratings listed by the other two retailers. Final Report A-S Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 1. Price Comparison for New vs. Refurbished Computer Batteries Dell Laptop Computer Model Dell Computer’ Batteries Direct” Raymond Sarrio® | (refurbished) (new) (new) Inspiron 3000 (~40Wh) $89.95 $189.00 Inspiron 3500 (~45 Wh) $119.00 $189.00 Inspiron 5000 (~52 Wh) $84.95 $129.00 Latitude CS (~50 Wh) $109.00 $189.00 Latitude LM (~42 Wh) $129.00 $194.00 $179.00 Notes to Table 1: 'Dell provides free 3-5 Day Ground shipping. No warranty information is provided on the refurbished batteries, or explanations of how they “refurbish” the batteries. *Batteries Direct (<http://batteriesdirect.com>) provides free Standard Delivery (4-7 days) and charges $3.95 for 2-3 Days Priority Delivery. Laptop computer batteries are covered by a PRORATED one-year warranty from the date of purchase. *The Raymond Sarrio Company (<www.sarrio.com>) provides a 30-day “no questions asked return policy” and a 1- year warranty against defects on all batteries sold. They also provide free UPS 3-Day Select shipping anywhere in the continental United States. Of major interest is a study conducted by Argonne National Laboratory between September 1996 and August 1997 for the United States Advanced Battery Consortium (USABC) that examined the second use of Ni/MH EV batteries. [19] Tests were conducted on eight Ovonic Battery Company (OBC) Ni/MH batteries that had completed more than 500 Dynamic Stress Test life cycles, simulating EV lifetime usage. Modules were tested to determine their performance relative to that of lead-acid modules in five non-EV applications: 1) utility load following; 2) utility frequency regulation and spinning reserve; 3) commercial and industrial off-road vehicles; 4) uninterruptible power sources (UPS) for stand-by power; and 5) accelerated life testing. Test results on the reused EV batteries were compared to lead-acid battery test data or warranties supplied on lead-acid batteries by the manufacturer for the specific applications on which they were tested. The OBC batteries were initially tested for performance, their module-rated capacities were reduced, and they were selected for specific application tests. The specific energy range to be examined in the study was determined by reviewing previous EPRI lead-acid battery test data. Standard production lead-acid units had a specific energy of approximately 30 Wh/kg (Hawker and Optima® batteries), and advanced technology lead-acid units were approximately 45 Wh/kg, which was the best measured for the Electrosource Horizon® battery. Two of the OBC Ni/MH modules were derated to 30 Wh/kg, and the remaining six were derated to 45 Wh/kg. 1) The utility load follow tests were based on tests used by Pacific Gas and Electric Company (PG&E) to characterize the PM250 Power Management System. The OBC modules were put through repeated 3 hour sine-wave discharges to 80% depth of discharge followed by charges to 100% SOC. They were also subjected to periodic (every 10 cycles) 1-hour constant power discharges at the I-hour rate. One each of the standard and advanced technology derated modules were put through 250 load follow cycles. The standard technology derated module completed all 250 load follow cycles and a total of 294 deep discharge cycles; the advanced technology derated module only Final Report A-6 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications completed 133 load follow cycles before failing on a constant power discharge. However, this still accounted for a total of 162 deep discharge cycles to end of life (EOL), which was far better than the 72 cycles performed by the Delco-Remy lead-acid modules tested by PG&E. 2) Two more OBC modules (one each of the advanced technology and standard technology derated modules) were tested under the Sandia National Laboratories (SNL) Utility Energy Storage (UES) test cycle. Developed to model the expected duty cycle of the 20-MW Battery Energy Storage system installed by the Puerto Rico Electric Power Authority (PREPA), the UES cycle contained three 50-hour-long frequency regulation sessions separated by intermediate charges. These sessions were followed by a spinning reserve discharge (15-minute constant power discharge followed by a 15-minute ramp down) and a final charge to 100% SOC. The test plan called for reference performance tests to be conducted every four UES cycles to establish a performance baseline. Both OBC modules completed all 16 of the planned UES cycles. The advanced technology derated module showed a capacity loss of about 1%, and the capacity loss in the standard derated module was negligible. This compared favorably with the lead-acid batteries subjected to similar testing at SNL, which showed 8.5% capacity loss after 13 UES cycles. 3) After the stationary application tests were conducted on the four utility modules, the modules were placed on off-road vehicle life test using the Battery Council International (BCI) cycle life test procedure for deep-cycle batteries (e.g., marine, golf, RV). The BCI procedure defines a constant-current (CI) cycle life test with 2-h rate discharges to 100% DOD discharge cutoff voltage (11 V) until the battery capacity declines to less than 50% of its 2-h rating. Three of the four Ni/MH modules were still performing BCI cycles when testing was halted due to dwindling funds. No comparison was made with the performance of lead-acid batteries for this application. 4) Three modules were tested for UPS application. Test conditions for the UPS test were based on the Exide lead-acid battery (flooded cell type) warranty. Exide indicates a warranted number of cycles for eight discharge conditions (combinations of eight discharge rates and eight discharge times). Three OBC modules (derated to 45 Wh/Kg) underwent tests in an attempt to satisfy all eight Exide warranted conditions. Each of the three modules successfully completed at least one of the Exide warranted conditions, but they fell well short of the eight warranted conditions that were planned. The report authors felt that the tests were performed under accelerated conditions with uncertain charge procedures, and the modules could have performed better under different circumstances. However, they also acknowledged that the warranted conditions tested were for flooded cells, and advanced technology lead-acid batteries are warranted for significantly higher numbers of cycles. 5) One battery underwent float-charge life testing at an elevated temperature (40°C) to accelerate possible corrosion reactions and failure. Reference cycles were conducted at ambient temperature (approximately 23°C) periodically during this life test to determine the effect of the elevated temperature on battery deep-discharge capacity, resistance, and power. The peak power of the module was declining at a rate of 0.11 W/kg per day at 40°C, which is close to the decline of 0.087 W/kg per day seen in a new OBC module Final Report A-7 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications subjected to similar tests. This led the authors to conclude that the impact of temperature on performance degradation in the OBC modules is similar throughout their cycle lives. Overall, the test results showed that used OBC Ni/MH EV modules perform at least as well as, if not better than, new lead-acid batteries in stationary energy storage applications. In fact, the derated EV modules appeared capable of performing the same functions as lead-acid batteries over longer lifetimes. The study demonstrates that reusing Ni/MH EV modules in secondary applications could be technically feasible from a battery performance perspective. Peer-reviewed journal search Searches performed utilizing online databases turned up no peer-reviewed articles on battery reuse. There were a number of articles on cell or electrode-level cycle life for both nickel metal hydride and lithium-based chemistries, but very few of these dealt with entire battery packs or even mass-produced cells. Most of the published information on cycle life and performance degradation for entire batteries identified during this search dealt with lead acid systems. The most informative of these papers is listed below: Aurbach, D. et al. “Factors Which Limit the Cycle Life of Rechargeable Lithium (Metal) Batteries.” Journal of The Electrochemical Society 147, no.4 (2000): 1274-1279. Crow, J. et al. “Summary of Electrical Test Results for Valve-Regulated Lead-Acid (VRLA) Batteries.” Journal of Power Sources 95 (2001): 241-247. Hollenkamp, A.F. “When is Capacity Loss in Lead/Acid Batteries *Premature’?” Journal of Power Sources 59 (1996): 87-98. Nakamura, K. et al. “Failure Modes of Valve-Regulated Lead/Acid Batteries” Journal of Power Sources 59 (1996): 153-157. Peters, K. “Review of Factors That Affect the Deep Cycling Performance of Valve- Regulated Lead/Acid Batteries.” Journal of Power Sources 59 (1996): 9-13. Zhang, Lu. “AC Impedance Studies on Sealed Nickel Metal Hydride Batteries over Cycle Life in Analog and Digital Operations.” Electrochimica Acta 43, nos. 21-22 (1998): 3333-3342. Conclusion Entrepreneurial individuals and companies have set up secondary markets for a variety of types of batteries. These batteries, once used in one type of operation, have been taken and applied to a new application where they have been given a new, second life. Spent nickel/metal hydride batteries, at least those used in electric vehicles, appear to have sufficient capacity and life to be used for a number of stationary applications. Additional tests will need to confirm this, as well as determine if this is true for HEV batteries, and for batteries with other chemistries, especially lithium-ion batteries. Setting up a successful secondary market for spent EV and HEV batteries will also require favorable economics. Final Report A-8 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications References [1] “CARS of Maryland Uses PRS to Recover & Resell Batteries.” Originally presented in PulseTalk, The Official Newsletter of PulseTech Products Corporation, Fall 1998. Available online at <www.pulsetech.com/about/pulsetalk/fall98/cars.html>. (2] “The International Business Club has Developed a Unique and Very Efficient Battery Reconditioning System.” Available online at <http://www.clubclub.com/english/index.html>. [3] E. Harris, “Inspection and Test of Donated Used Batteries for ARES,” Fairfax County, VA, (December |, 1998), 2 pages. Available online at <www.va- ares.org/library/free_batteries.html>. This article also appeared in the SARINFO reference library (<sarinfo@minklink.net>), the MVARC and NVFMA newsletters, and on the ARRL Members Only website (<http://www.arrl.org>). 4] Approximately two dozen members of the Electric Vehicle Association of Washington, DC, were queried during a monthly meeting held on August 14, 2001 in Bethesda, Maryland. 5] SURPLUS TRADERS. Available online at <http://www.surplustraders.net> 6] A.N. Brooks and T.B. Gage, The tzero Electric Sports Car—How Electric Vehicles Can Achieve Both High Performance and High Efficiency, (October, 2000), 21 pages. Paper presented at the 17th Electric Vehicle Conference (EVS17), in Montreal, Canada. 7] E-mail communication with Richard Perez of Home Power magazine. The additional articles he mentioned were not purchased because they did not appear to deal with EV/HEV batteries, nor with advanced, non-lead-acid batteries. 8] L. Barker, “PV Household Storage Battery Test,” Home Power, 83 (June/July 2001): 50-53. 9] Oasis Montana Inc. Alternative Energy Supply and Design. Available online at <www.oasismontana.com>. 10] Telephone conversation with Paul Hess, Energy Efficiency and Renewable Energy Information Clearinghouse. 11] E-mail communication with Johnny Weiss, Solar Energy International. 12] Personal communication with Peter Lowenthal, Solar Energy Industries Association. 13] Telephone conversation with Hugh Marrow, International Cadmium Association. 14] E-mail communication with Jade Mountain. Jade Mountain can be reached on the Internet at <http://www.jademountain.com/> and via e-mail at <info@jademountain.com>. 15] E-mail communication with Bill Moore, editor of the EV World® website: <www.evworld.com>. [16] Personal communication with David Goldstein, President, Electric Vehicle Association of Washington, DC, who stated that he had heard that ECD had a proposal to use 50% capacity Ni/MH batteries in UPS units. [17] Telephone communication with Homero del Basque, Energy Conversion Devices. Final Report A-9 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications [18] Dell Computer Corporation. Available online at <www.dell.com>. [19] N. Pinsky (USABC Program Manager) et al., Electric Vehicle Battery 2™ Use Study, Argonne, IL: Argonne National Laboratory, Electrochemical Technology Program, (May 21, 1998), 80 pages plus Appendix. Final Report A-10 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications APPENDIX B: EV BATTERIES AVAILABLE ON THE U.S. MARKET This appendix contains a summary of the battery systems found in electric vehicles sold or demonstrated in the U.S. The purpose of this summary is to provide a rough idea of the range of batteries available for potential use in an EV battery secondary use program. As such, it is not an exhaustive list. It does, however, contain the specifications of the battery systems installed in most EV’s that have been sold in the U.S. (including those that have gone out of production), as well as some demonstration and prototype vehicles. It does not contain most of the conversion vehicles currently available. The information used here was obtained from the Electric Vehicle Association of the Americas website, the Department of Energy’s Office of Transportation Technologies Alternative Fuels Database, vehicle manufacturers, battery manufacturers, and the “Advanced Batteries for Electric Vehicles: An Assessment of Performance, Cost and Availability” report submitted by the Year 2000 Battery Technology Advisory Panel to the California Air Resources Board. For the purposes of this study, electric vehicles were broken into four broad categories to allow for comparisons and simplification of the data. Their descriptions follow: EV -Electric Vehicle: full size passenger automobile with a battery-based electric drive train. HEV — Hybrid Electric Vehicle: full size passenger vehicle with an internal combustion engine coupled with an electric motor and battery. CEV -— City Electric Vehicle: small (short wheelbase) passenger automobile with a battery-based electric drive train designed for urban commuting; top speed approximately 60 mph. NEV — Neighborhood Electric Vehicle: small, low speed (< 25 mph) passenger automobile with a battery-based electric drive train designed for short trips. Each of these classes of vehicles has different power and energy requirements for the battery. For example, EV’s typically have about 30 kWh of energy storage, CEV’s and NEV’s require about 10 kWh, and HEV’s typically have a battery with around | kWh of storage. In addition to considering different types of vehicles, this summary also covers systems based on each of the major battery chemistries used in electric vehicles: Ni/MH — Nickel/Metal Hydride Li-ion — Lithium-ion NiCd — Nickel/Cadmium LMPB - Lithium Metal-Polymer VRLA — Valve-Regulated Lead-Acid FLA — Flooded Lead-Acid. Final Report B-1 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Tables | and 2 below contain broad descriptions of typical batteries organized by chemistry and vehicle type. The remaining tables contain more detailed specifications for the battery systems found in most electric vehicles available on the U.S. market today, as well as a number of prototype and demonstration vehicles expected to reach production in the next few years. Final Report B-2 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 1. Battery Module Characteristics for typical HEV and EV batteries. Vehicle Type HEV EV Chemistry Ni/MH Li-ion LMPB* Ni/MH Li-ion® VRLA LMPB* Battery Voltage (V) 144 — 274 346 288 - 343 360 312 260 Battery Capacity (Ah) 6.5 3.6 77-95 90 60 - 85 119 Battery Capacity (kWh) 0.94 - 1.8 1.2 26 - 32 32 19-27 31 Cells/battery 120 — 228 96 240 - 286 96 = - Modules/Battery 20 - 38 2 24 - 28 12 26 - 39 13 Module Voltage (V) 7.2 173 50 12-13.2 30 8-12 20 Module Capacity (Ah) 6.5 3.6 14 77-95 90 60 - 85 119 Module Capacity (kWh) 0.047 0.62 0.7 1.0-1.1 2.7 0.68 - 0.72 2.38 Module Output Power (kW)* ~0.9 16 3.2-4 12.5 4-5 4.9 Motor Output (kW) 10 - 33 17 49 - 102 62 67 - 102 : Battery Manufacturers PEVE' Shin Kobe Avestor |PEVE Shin Kobe — |Panasonic Avestor SAFT [Texaco Ovonic |SAFT East Penn Vehicles Honda Insight |Nissan Tino None Chevy S-10 Nissan Altra |Baker/Ford USPS |None Toyota Prius |Dodge Durango Ford Ranger Ford Ranger Dodge ESX3 GM EV-1 Chevy S-10 Honda EV Plus GM EV-1 Toyota Rav4 Solectria Force Solectria Force Notes: 'PEVE — Panasonic EV Energy, a Matsushita company. *The lithium-ion characteristics listed here are for Shin Kobe EV and HEV modules. Module level data were not available for SAFT Li-ion batteries since none of the vehicles using SAFT Li-ion batteries are in production yet. *The LMPB characteristics listed here are for prototype modules developed by Avestor, and some of the numbers are extrapolations from cell-level performance. Avestor has not yet mass-produced any LMPB battery packs, and there are no automakers with plans at the current time to use LMPB systems in a production vehicle, although they have been demonstrated in the Ford TH!NK City and the GM Precept. “Module Output Power was estimated using the specific power for the module and an estimated module mass calculated from the total battery pack mass and the number of modules. Final Report B-3 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 2. Battery Module Characteristics for typical CEV and NEV batteries. Vehicle Type CEV NEV Chemistry Ni/MH Li-ion NiCd FLA Battery Voltage (V) 288 120 114 72 Battery Capacity (Ah) 28 90 100 130 Battery Capacity (kWh) 8.1 1 1 9.4 iCells/battery 240 32 - - Modules/Battery 24 4 19 6 Module Voltage (V) 12 30 6 12 Module Capacity (Ah) 28 90 100 130 Module Capacity (kWh) 0.34 2.7 0.6 1.6 Module Output Power (kW) 1.41 12.5 1.6 Motor Output (kW) 19 24 27 5-25 Battery Manufacturers PEVE Shin Kobe ISAFT Trojan SAFT Vehicles Honda City Pal |Nissan Hypermini |TH!NK City Dynasty IT Toyota ecom Solectria Force |GEM E825 Solectria Flash [THINK Neighbor For additional details on the individual vehicles, see Tables 3-8. Final Report B-4 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 3. HEV’s by Manufacturer — Honda, Nissan, Toyota, DaimlerChrysler Manufacturer Honda Nissan Toyota DaimlerChrysler | DaimlerChrysler Model Insight Tino Prius ESX3 Durango Chemistry Ni/MH Li-ion Ni/MH Li-ion Li-ion Battery Manufacturer PEVE Shin Kobe PEVE SAFT SAFT Battery Model HEV Mn type HEV HEV Hi Power Hi Power Battery Voltage (V) 144 345.6 273.6 165 259.2 Battery Capacity (Ah) 6.5 3.6 6.5 8 16 Battery Capacity (kWh) 0.936 1.24416 1.7784 1.32 4.1472 Modules/Battery 20 2 38 Module Voltage (V) 7.2 172.8 7.2 Module Capacity (Ah) 6.5 3.6 6.5 Module Capacity (kWh) 0.0468 0.62208 0.0468 Module Output Power (kW) ~0.9 ~0.9 Cells/Module 6 48 6 Connection series series series Cell Voltage (V) 1.2 3.6 1.2 3.6 3.6 Cell Capacity (Ah) 6.5 3.6 6.5 8 16 Cells/battery 120 96 228 72 Cell Design cylindrical Dcell| cylindrical |prismatic, plastic| cylindrical cylindrical Motor Output (kW) 10 17 33 Sales Figures: 1996 1997 1998 1999 (1st qtr) 1999 (2nd qtr) 1999 (3rd qtr) 1999 (4th qtr) 17 2000 (1st qtr) 397 2000 (2nd qtr) 1149 2000 (3rd qtr) 1290 2610 2000 (4th qtr) 952 2952 2001 (1st qtr) 597 2490 2001 (1%*- 4" qtr) 3901 12116 Total 7706 17678 Availability Retail Japan Only Retail Prototype Prototype Final Report B-5S Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 4. EV’s by Manufacturer — DaimlerChrysler, Ford Manufacturer DaimlerChrysler Baker/Ford Ford Model EPIC Minivan _|ECRV Postal Delivery Veh. Ranger EV' Chemistry Ni/MH VRLA VRLA Ni/MH Battery Manufacturer SAFT East Penn East Penn PEVE Battery Model NH 12.4 (old) UX 168 UX 168 EV 95 Battery Voltage (V) 336 312 312 300 Battery Capacity (Ah) 95 85 85 95 Battery Capacity (kWh) 31.92 26.52 26.52 28.5 Modules/Battery 28 39 39 25 Module Voltage (V) 12 8 8 12 Module Capacity (Ah) 95 85 85 95 Module Capacity (kWh) 1.14 0.68 0.68 1.14 Module Power Output (kW) 3.2 4 4 3.8 Cells/Module - - - 10 Connection - - - series Cell Voltage (V) - - - 1:2 Cell Capacity (Ah) - - - 95 Cells/battery - - - 250 Cell Design monoblock monoblock monoblock prismatic, plastic Motor Output (kW) 75 67 67 67 Sales Figures: 1996 1997 17 27 1998 0 310 1999 (1st qtr) 10 28 4999 (2nd qtr) 41 119 1999 (3rd qtr) 23 179 1999 (4th qtr) 55 207 2000 (1st qtr) 51 46 2000 (2nd qtr) 9 130 2000 (3rd qtr) 0 73 2000 (4th atr) 0 140 2001 (1st qtr) 0 500 2 Total 206 500 1261 Availability Out of Production| _500 CA & DC USPS? Fleet and Retail Notes: 1 The Ford Ranger EV was initially offered with VRLA batteries only. In recent model years, customers could select Ni/MH batteries as an option. The sales figures are for both options together. 7500 ECRV Postal Delivery Vehicles were ordered by the US Postal Service and will be deployed in California and the District of Columbia. The USPS has the option of purchasing 5000 more of the vehicles after evaluation. Final Report B-6 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 5. EV’s by Manufacturer - GM Manufacturer GM Chevy Model EV1' S-10 Electric’ Chemistry VRLA Ni/MH VRLA Ni/MH Battery Manufacturer Panasonic Texaco Ovonic Panasonic Texaco Ovonic Battery Model EV 1260 EV 1260 Battery Voltage (V) 312 343 312 343 Battery Capacity (Ah) 60 77 60 77 Battery Capacity (kWh) 18.72 26.4264 18.72 26.4264 Modules/Battery 26 26 26 26 Module Voltage (V) 12 13.2 12 13.2 Module Capacity (Ah) 60 77 60 77 Module Capacity (kWh) 0.72 1.0164 0.72 1.0164 Module Output Power (kW) 5 4 5 4 Cells/Module - 1 - 11 Connection : series - series Cell Voltage (V) - 1.2 - 1.2 Cell Capacity (Ah) - 77 - 77 Cells/battery - 286 - 286 Cell Design monoblock prismatic, metal monoblock prismatic, metal Motor Output (kW) 102 102 85 85 Sales Figures: 1996 39 1997 264 278 1998 258 99 1999 (1st qtr) 93 45 1999 (2nd qtr) 0 1 1999 (3rd qtr) 0 57 1999 (4th qtr) 45 10 2000 (1st qtr) 89 2000 (2nd qtr) 65 2000 (3rd qtr) 2000 (4th atr) 2001 (1st qtr) Total 853 500 Availability Out of Production Out of Production Notes: ' The GM EV-I and Chevy S-10 Electric were both initially offered with VRLA batteries only. In recent model years, customers could select Ni/MH batteries as an option. The sales figures are for both options together. Final Report B-7 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 6. EV’s by Manufacturer — Honda, Nissan, Toyota, Solectria Manufacturer Honda Nissan Toyota Solectria Model EV Plus Altra EV RAV4 Force Chemistry Ni/MH Li-ion Ni/MH VRLA, NiCd, or Ni/MH Battery Manufacturer PEVE Shin Kobe PEVE Battery Model EV 95 Mn type EV MHB-100 Battery Voltage (V) 288 360 288 156 Battery Capacity (Ah) 95 90 95 Battery Capacity (kWh) 27.36 32.4 27.36 Modules/Battery 24 12 24 13 Module Voltage (V) 12 30 12 12 Module Capacity (Ah) 95 90 95 Module Capacity (kWh) 1.14 2.7 1.14 Module Output Power (kW) 3.8 12.5 3.8 Cells/module 10 8 10 Connection series series series Cell Voltage (V) 1.2 3.75 1.2 Cell Capacity (Ah) 95 90 95 Cells/battery 240 96 240 Cell Design prismatic, plastic cylindrical prismatic, plastic Motor Output (kW) 49 62 50 42 Sales Figures: 1996 1997 105 69 1998 133 30 359 1999 (1° qtr) 29 26 79 1999 (2° qtr) 21 4 78 1999 (3 qtr) 7 0 51 1999 (4" qtr) 5 0 47 2000 (1° qtr) 0 11 1 2000 (2 qtr) 0 4 18 2000 (3 qtr) 0 5 77 2000 (4" qtr) 0 30 10 2001 (1° qtr) 0 1 27 2001 (2™ atr) 0 19 26 2001 (3 atr) 0 0 77 2001 (4'" qtr) 0 0 26 ‘Total 300 130 945 Availability Out Of Production| Fleet in CA Fleet Retail Final Report B-8 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Table 7. CEV’s by Manufacturer — Honda, Nissan, Toyota, Ford TH!NK Manufacturer Honda Nissan Toyota Ford THINK Model City Pal Hypermini e-com (62 mph) | City (56 mph) Chemistry Ni/MH Li-ion Ni/MH NiCd Battery Manufacturer Panasonic Shin Kobe Panasonic SAFT Battery Model EV 28 Mn type EV EV 28 STM 5-100 MRE Battery Voltage (V) 288 120 288 114 Battery Capacity (Ah) 28 90 28 100 Battery Capacity (kWh) 8.064 10.8 8.064 11.4 Modules/Battery 24 4 24 19 Module Voltage (V) 12 30 12 6 Module Capacity (Ah) 28 90 28 100 Module Capacity (kWh) 0.336 2.7 0.336 0.6 Module Output Power (kW) act 12.5 11 1.6 Cells/Module 10 4 10 - Connection series series series - Cell Voltage (V) 1.2 3.75 1.2 - Cell Capacity (Ah) 28 90 28 - Cells/battery 240 16 240 - Cell Design prismatic, plastic cylindrical prismatic, plastic monoblock Motor Output (kW) 24 19 27 Sales Figures: 1996 1997 1998 1999 (1st qtr) 1999 (2nd qtr) 1999 (3rd qtr) 1999 (4th qtr) 2000 (1st qtr) 2000 (2nd qtr) 2000 (3rd qtr) 2000 (4th qtr) 2001 (1st qtr) Total Availability Demo Demo Demo Demo (retail in EU) B-9 Technical and Economic Feasibility of Applying Final Report Used EV Batteries in Stationary Applications Table 8. NEV’s by Manufacturer — Dynasty, GEM, Solectria, Ford TH!NK GEM Manufacturer Dynasty (DaimlerChrysler) Solectria Ford TH!INK Model IT NEV E825 Flash Neighbor Chemistry FLA FLA FLA FLA? |Battery Manufacturer Trojan Trojan Battery Model 30 XHS Battery Voltage (V) 72 72 144 72 Battery Capacity (Ah) 130 Battery Capacity (kWh) 9.4 Modules/Battery 6 6 12 6 Module Voltage (V) 12 12 12 12 Module Capacity (Ah) 130 Module Capacity (kWh) 1.6 Cells/Module - - - - Connection - - - - Cell Voltage (V) - - - - Cell Capacity (Ah) - - - - Cells/battery - - - - Cell Design : - - : Motor Output (kW) 25 2.5 34 5 Sales Figures: 1996 1997 1998 1999 (1st qtr) 1999 (2nd qtr) 1999 (3rd qtr) 1999 (4th qtr) 2000 (1st qtr) 2000 (2nd qtr) 2000 (3rd qtr) 2000 (4th qtr) 2001 (1st qtr) 6261 [Total 6261 Availability Retail Retail Retail Retail (Fall 2001) Final Report B-10 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications References Electric Vehicle Association of the Americas website: http://www.evaa.org DOE Office of Transportation Technologies Alternative Fuels Data Center: http://www.afdc.doe.gov/afvehicles.html M. Anderman, F. Kalhammer, and D. MacArthur, Advanced Batteries for Electric Vehicles: An Assessment of Performance, Cost and Availability, Prepared for the State of California Air Resources Board (June 22, 2000). Final Report B-\1 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications This page intentionally left blank. Final Report B-12 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Distribution Dutch Achenbach Gerald W. Braun 2111 Buffalo 14801 Braemar Crescent Way Casper, WY 82604 Darnestown, MD 20878 Per Danfors Bill Erdman c/o 1029 W. Zermatt Dr. #8 850 Greenstone Ct. Midway, UT 84049 Brentwood, CA 94513 Brad Johnson Nick Magnani 2122 Natahoa Court 2494 Windmill Point Rd. Falls Church, VA 22043 White Stone, VA 22578 Hans Weinrich Bob Weaver ABB Power T&D Co., Inc. 777 Wildwood Lane 1460 Livingston Ave. Palo Alto, CA 94303 P.O. Box 6005 North Brunswick, NJ 08902-6005 Final Report D-1 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Bryan Plater Active Power, Inc. Corporate Headquarters 11525 Stonehollow Drive, #135 Austin, TX 78758 Joe Thomas Active Power, Inc. 65 Glen Road, Suite 4-287 Garner, NC 27529 Robert O. Bartlett AFS Trinity Power Corporation 7617 Irongate Lane Frederick, MD 21702 Tim Mack Airak, Inc. 9058 Euclid Ave. Manassas, VA 20110 Final Report Jim Balthazar Active Power, Inc. Corporate Headquarters 11525 Stonehollow Drive, #110 Austin, TX 78758 William V. Hassenzahl Advanced Energy Analysis 1020 Rose Avenue Piedmont, CA 94611 Paul Grems Duncan Airak, Inc. 9058 Euclid Avenue Manassas, VA 20110 Dennis Meiners Alaska Energy Authority 813 West Northern Lights Boulevard Anchorage, AK 99503 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications David Lockard Alaska Energy Authority/AIDEA 813 West Northern Lights Blvd. Anchorage, AK 99503 Phillip R. Beaver Albemarle Corporation 8000 GSRI Avenue P.O. Box 14799 Baton Rouge, LA 70820 Christian Sasse ALSTOM Research & Technology Centre PO Box 30; Lichfield Road Stafford, ST17 4LN UNITED KINGDOM Dave Nichols American Electric Power 1 Riverside Plaza Columbus, OH 43215 Final Report D- Peter Crimp Alaska Energy Authority/AIDEA 813 West Northern Lights Blvd. Anchorage, AK 99503 Burkhard R6mhild ALSTOM Power, Inc. 2800 Waterford Lake Drive Midlothian, VA 23112-3981 Ali Nourai American Electric Power 1 Riverside Plaza, 20th Floor Columbus, OH 43215 Meera Kohler Anchorage Municipal Light & Pwr 1200 East Ist Avenue Anchorage, AK 99501 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Ira Bloom Argonne National Laboratories 9700 South Cass Avenue CTD, Bldg. 205 Lemont, IL 60439-4837 Herb Hayden Arizona Public Service 400 North Fifth Street P.O. Box 53999, MS8931 Phoenix, AZ 85072-3999 Robert Hammond Arizona Public Service Co. (APS) Solar Test and Research Center 1500 E. University Drive Tempe, AZ 85281 Rick Lawrence AVO International P.O. Box 9007 Valley Forge, PA 19485-1007 Final Report Christian St-Pierre ARGO-TECH Productions, Inc. Subsidiary of Hydro-Quebec 1580 de Coulomb Boucherville, QC J4B 7Z7 CANADA Ray Hobbs Arizona Public Service 400 North Fifth Street P.O. Box 5399, MS8931 Phoenix, AZ 85072-3999 Violaine Dorval Avestor 1600 De Coulomb Boucherville, QC J4B7Z7 CANADA Matthew L. Lazarewicz Beacon Power Corp. 234 Ballardvale Street Wilmington, MA 01887 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications William Capp Beacon Power Corp. 234 Ballardvale Street Wilmington, MA 01887 Chris Marnay Berkley National Laboratory MS 90-4000 1Cyclotron Rd. Berkeley, CA 94720 Larry Stoddard Black and Veatch 11401 Lamar Ave. Overland Park, KS 66211 Art Day Boeing Phantom Works P.O. Box 3999 MS 73-09 Seattle, WA 98124-2499 Final Report Michael L. Bergey Bergey Windpower 2001 Priestley Avenue Norman, OK 73069 Klaus Kraemer Berliner Kraft und Licht (BEWAG) Bewag AG Energy Planning & Trading Berlin, D-12432 GERMANY Michael Strasik, Ph.D. Boeing P.O. Box 3999 MS 73-09 Seattle, WA 98124-2499 Mike Hoffman Bonneville Power Administration P.O. Box 3621 905 N.E. 11th Avenue Portland, OR 97208-3621 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Roy Forey British Embassy 3100 Massachusetts Avenue, NW Washington, DC 20008-3600 Dr. Sudhan S. Misra C&D Technologies 1400 Union Meeting Road Blue Bell, PA 19422-0858 Terry Surles, Ph.D. California Energy Commission 1516 9th Street MS43 Sacramento, CA 95814-5504 Alex Q. Huang CPES Virginia Polytechnic Institute 674 Whittemore Hall (0179) Blacksburg, VA 24061-0179 Final Report Septimus van der Linden Brulin Associates, L.L.C. 14418 Old Bond Street Chesterfield, VA 23832 Pramod P. Kulkarni California Energy Commission Research & Dev. Office 1516 9th Street, MS42 Sacramento, CA 95814-5512 John Cooley Chugach Elec. Association, Inc. P.O. Box 196300 Anchorage, AK 99519-6300 Michael A. Lankford Detroit Edison 15600 19 Mile Road Clinton Township, MI 48038 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Haukur Asgeirsson : . Joe Iannucci Detroit Edison Company 2000 2nd Ave. 435 SB Detroit, MI 48226-1279 Distributed Utility Associates 1062 Concannon Blvd. Livermore, CA 94550 Murray W. Davis Steven G. Whisenant DTE Energy Duke Power Company 37849 Interchange Dr. P.O. Box 1006 Farmington Hills, MI 48335 Charlotte, NC 28201 Alan Collinson Jim DeGruson EA Technology Limited Eagle-Picher Industries. Inc. Capenhurst, Chester, England CH1 6ES C & Porter Street UNITED KINGDOM Joplin, MO 64802 Alfred Rufer Ecole Polytechnique Federale de Lausanne K.M. Abraham (EPFL) E-KEM Sciences LEI-STI-EPFL P.O. Box 920401 Lausanne, CH1015 Needham, MA 2492 SWITZERLAND Final Report D-7 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Final Report Gail Hendrickson (15) Electric Drive Transporation Association 701 Pennsylvania Avenue NW 3rd Floor - East Building Washington, DC 20004 Steve Chapel Electric Power Research Institute P.O. Box 10412 Palo Alto, CA 94303-0813 Robert B. Schainker Electric Power Research Institute (EPRI) 3412 Hillview Avenue P.O. Box 10412 Palo Alto, CA 94304 Dave Feder Electrochemical Energy Storage Systems, Inc. 35 Ridgedale Avenue Madison, NJ 07940 -8 Mark Skowronski Electric Power Group 201 South Lake Avenue, Suite 400 Pasadena, CA 91101 Steve Eckroad Electric Power Research Institute P.O. Box 10412 Palo Alto, CA 94303-0813 Anthony Price Electricity Storage Association C/O Innogy, Windmill Hill Business Park Whitehill Way Swindon, SN5 6PB UNITED KINGDOM Jack Brown ElectroEnergy, Inc. 30 Shelter Rock Road Danbury, CT 06810 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Michael Eskra ElectroEnergy, Inc. 30 Shelter Rock Road Danbury, CT 06810 Brian Marchionini Energetics, Inc. 901 D St SW Suite 100 Washington, DC 20024 Edward Skolnik Energetics, Inc. 901 D St SW Suite 100 Washington, DC 20024 Lauren Giles Energetics, Inc. 7164 Columbia Gateway Drive Columbia, MD 21046 Final Report D-9 Chuck Whitaker Endecon Engineering 347 Norris Ct. San Ramon, CA 94583 Christina Termaath Energetics, Inc. 901 D St. SW Suite 100 Washington, DC 20024 Howard Lowitt Energetics, Inc. 7164 Gateway Drive Columbia, MD 21046 Ndeye K. Fall Energetics, Inc. 901 D St SW Suite 100 Washington, DC 20024 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Rich Scheer Scheer Energetics, Inc. 901 D St. SW Suite 100 Washington, DC 20024 Philip C. Symons Energy Storage Association/EECI 1295 Kelly Park Circle Morgan Hill, CA 95037 Tom Key EPRI PEAC Corp. 860 Corridor Park Blvd. Knoxville, TN 37932 Erik Henning EUS GmbH Munscheidst 14 Gelsenkirchen, D-45886 GERMANY Final Report Amber Gray-Fenner Energy Communications Consulting 7204 Marigot Rd. NW Albuquerque, NM 87120 R.B. Sloan Energy United P.O. Box 1831 Statesville, NC 28687 Haresh Kamath EPRI PEAC Corporation 942 Corrider Park Boulevard Knoxville, TN 37932 James P. Dunlop Florida Solar Energy Center 1679 Clearlake Road Cocoa, FL 32922-5703 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Tom Lovas Four Dam Pool Power Agency 703 West Tudor Road, Suite 102 Anchorage, AK 99503-6650 Nick Miller General Electric Company 1 River Road Building 2, Room 605 Schenectady, NY 12345 Raj M. Bharadwaj General Electric Global Research One Research Circle, ES 601 Niskayuna, NY 12309 Les Fairchild Georgia Power Co. 4404 N. Shallowford Rd. Atlanta, GA 30308 Final Report Bob Zrebiec GE Industrial & Pwr. Services 640 Freedom Business Center King of Prussia, PA 19046 Declan Daly General Electric Drive Systems 1501 Roanoke Blvd. Salem, VA 24153 Daniel P. Smith General Electric R&D P.O. Box 3, ES-142 Schenectady, NY 12301 Gerry Woolf Gerry Woolf Associates 17 Westmeston Avenue Rottingdean, East Sussex, BN2 8AL UNITED KINGDOM Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Christopher John GNB Industrial Power 829 Parkview Blvd. Lombard, IL 60148 George Hunt GNB Technologies Woodlake Corporate Park 829 Parkview Blvd. Lombard, IL 60148-3249 Sanjay Deshpande’ GNB Technologies Woodlake Corporate Park 829 Parkview Blvd. Lombard, IL 60148-3249 Ben Norris Gridwise Engineering Company 121 Starlight Place Danville, CA 94526 Final Report D-12 Robert Schmitt GNB Industrial Power, Division of Exide Technologies 829 Parkview Boulevard Lombard, IL 60148 Joe Szymborski GNB Technologies Woodlake Corporate Park 829 Parkview Blvd. Lombard, IL 60148-3249 Steven Haagensen Golden Valley Elec. Assoc., Inc. 758 Illinois Street P.O. Box 71249 Fairbanks, AK 99701 Philip Kunsberg Haviland Holdings, Inc. 134 East Gate Drive, Suite B Los Alamos, NM 87544 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Clyde Nagata Hawaii Electric Light Co. P.O. Box 1027 Hilo, HI 96720 Larry Meisner HOPPECKE Batteries, Inc. 1960 Old Cuthbert Road Suite 130 Cherry Hill, NJ 08034 Mohammad Shahidehpour Illinois Institute of Technology ECE Department, 3301 S. Dearborn Chicago, IL 60616 Jerome F. Cole ILZRO 2525 Meridian Parkway PO Box 12036 Research Triangle Park, NC 27709 Final Report Jacqui L. Hoover Hawaii Ocean Science & Technology Park 73-4460 Queen Kaahumanu Hwy., #101 Kailua-Kona, HI 96740-2632 Andre Belanger Hydro-Quebec Research Institute 1800 Boul. Lionel Boulet Varennes, Quebec J3X 1S1 CANADA Carl D. Parker ILZRO 2525 Meridian Parkway P.O. Box 12036 Research Triangle Park, NC 27709 Patrick Moseley ILZRO 2525 Meridian Parkway P.O. Box 12036 Research Triangle Park, NC 27709 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Final Report Ron Myers Imperial Oil Research Centre 3535 Research Road NW Room 2E-123 Calgary, Alberta, T2L 2K8 CANADA Robert Wills Intergrid Consulting 39 Union Street Peterborough, NH 03458 Rosalind Volpe International Lead Zinc Research Organization, Inc. (ILZRO) PO Box 12036 Research Triangle Park, NC 27709-2036 Douglas Freund Invensys Energy Management Division 2023 North Atlantic Avenue, #276 Cocoa Beach, FL 32931 D- Ken Belfer Innovative Power Sources 1419 Via Jon Jose Road Alamo, CA 94507 A. Kamal Kalafala Intermagnetics General Corp. 450 Old Niskayuna Road P.O. Box 461 Latham, NY 12110-0461 Albert R. Landgrebe Int'l Electrochemical Sys & Technology B14 Sussex Lane Long Neck, DE 19966 Willie Ozbirn Invensys Energy Solutions 1034 East Ash Street (62703) P.O. Box 19424 Springfield, IL, IL 62794-9424 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications David Bertagnolli ISO New England One Sullivan Road Holyoak,, MA 01040-2841 Gerard H. C. M. Thijssen KEMA T&D Consulting Utrechtseweg 310 P.O. Box 9035 Armhem, 6800ET NETHERLANDS Frank McLarnon Lawrence Berkeley National Lab University of California One Cyclotron Road Berkeley, CA 94720 Grayson Heffner Lawrence Berkeley Nat'l Lab University of California One Cyclotron Road M/S Wash DC Berkeley, CA 94720 Final Report Imelda G. Francis Just Do IT 21658 W. Ravine Rd. Lake Zurich, IL 60047 Joseph M. Derby Kerr-McGee Chemical LLC PO Box 25861 Oklahoma City, OK 73125 Elton Crains Lawrence Berkeley Nat'l Lab University of California One Cyclotron Road Berkeley, CA 94720 Joe Eto Lawrence Berkeley Nat'l Lab University of California One Cyclotron Road, MS 90-4000 Berkeley, CA 94720 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications J. Ray Smith Lawrence Livermore Nat'l Lab University of California P.O. Box 808, L-644 Livermore, CA 94551 Patrick J. McGinnis, P.E. Machinery Research & Engineering Directorate, Naval Surface Warfare Center - Carderock Division NSWCCD-SSES, Building 77L 5001 South Broad Street Philadelphia, PA 19112-5083 John Sawarin McKenzie Bay International Ltd. 143 Windson Ave. London, Ontario, N6C2A1 Rene Pardo Mega-C Power Corporation 100 Caster Avenue Vaughan, ON L4L 5Y9 CANADA Final Report D-16 Susan M. Schoenung Longitude 122 West, Inc. 1010 Doyle Street Suite 10 Menlo Park, CA 94025 Jim Fiske Magtube, Inc. 5735-B Hollister Ave. Goleta, CA 93117 Lewis Taylor Mega-C Power Corporation 100 Caster Avenue Vaughan, ON L4L 5Y9 CANADA Floyd Russell Metlakatla Power & Light P.O. Box 359 3.5 Mile Airport Road Metlakatla, AK 99926 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Peter Donalek Montgomery Watson Harza (MWH) Global 175 W Jackson Blvd, Flr 19 Chicago, IL 60604 Dr. Amit Singal NanoPowder Enterprises, Inc. 120 Centennial Avenue Suite 106 Piscataway, NJ 08854-3908 Tony Schaffhauser National Renewable Energy Laboratory 1617 Cole Blvd. Golden, CO 80401-3393 Kae Pohe Nevarest 443 Nevada Highway Boulder City, NV 89005 Final Report D- Jagoron Mukherjee MPR Associates, Inc. 320 King Street Alexandria, VA 22311 Dr. Norman Holcombe National Energy Technology Laboratory 626 Cochrans Mill Road P.O. Box 10940 Pittsburgh, PA 15236-0940 Steven P. Lindenberg National Rural Elec Cooperative Assoc. 4301 Wilson Blvd. SSER9-207 Arlington, VA 22203-1860 Andrew L. Rosenthal New Mexico State University Southwest Tech. Dev. Institute Box 30001/Dept. 3SOL Las Cruces, NM 88003-8001 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Satish J. Ranade New Mexico State University Klish School of ECE, Box 3001, Dept. 3-O Las Cruces, NM 88003 Toyoo Ted Takayama NGK Insulators, LTD/NAS Battery Division 2-56, Suda-cho, Mizuho Nagoya, Aichi 467-8530 JAPAN Edward Torrero NRECA Cooperative Research Network 4301 Wilson Boulevard, SS9-204 Arlington, VA 22203-1860 Brendan Kirby Oak Ridge National Laboratory P.O. Box 2008 Bldg. 3147, MS-6070 Oak Ridge, TN 37831-6070 Final Report D-18 L. Daniel Mears NGK Insulators, LTD/NAS Battery Division 6540 Lusk Boulevard, Suite C-102 San Diego, CA 92121 Eddy Ozawa NGK-Locke, Inc. 2525 Insulator Drive Baltimore, MD 21230 Clinton Winchester NSWC Carderock Division, Code 644 9500 MacArthur Boulevard West Bethesda, MD 20817-5700 John Stoval Oak Ridge National Laboratory P.O. Box 2008 Bldg. 3147, MS-6070 Oak Ridge, TN 37831-6070 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Michael A. Karnitz Oak Ridge National Laboratory P.O. Box 2008 MS 6186 Oak Ridge, TN 37831-6186 Douglas R. Danley Orion Energy Corporation 4580A Mack Avenue Frederick, MD 21703 Stan Sostrom POWER Engineers, Inc. P.O. Box 777 3870 US Hwy 16 Newcastle, WY 82701 Todd Jaffe Power Asset Brokers, LLC 3420 Keyser Road Baltimore, MD 21208 Rick Winter Jerry Neal Private Consultant 246 Canyon Woods Way Suite C San Ramon, CA 94583 Public Service Co. of New Mexico Alvarado Square MS-BA52 Albuquerque, NM 87158 Roger Flynn Wenceslao Torres Public Service Co. of New Mexico Puerto Rico Elec. Pwr. Authority Alvarado Square MS-2838 P.O. Box 364267 Albuquerque, NM 87158 San Juan, PR 00936-4267 Final Report D-19 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Mark Kuntz Scott Hassell RAND Corporation 1200 South Hayes Street Arlington, VA 22202-5050 Regenesys Technologies Ltd. 303 East Walker Drive Suite 1200 Chicago, IL 60601 James T. Reilly eee eee Reliable Power Inc. 2300 Clarendon Blvd. Suite 401 Arlington, VA 22201 Reilly Associates P.O. Box 838 Red Bank, NJ 07701 John P. Venners Malcolm T. Jacques Reliable Power Inc. 2300 Clarendon Blvd. Suite 400 Arlington, VA 22201 Reliable Power, Inc. 2300 Clarendon Blvd., Suite 401 Arlington, VA 22201 Matthew Karcher Markus Becker Reliable Power, Inc. RWE North America 2300 Clarendon Blvd., Suite 401 555 12th Street, Suite 630 Arlington, VA 22201 Washington, DC 20004 Final Report D-20 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Brad Roberts S&C Electric Company 2010 Energy Drive P.O. Box 879 East Troy, WI 53120 Salah Oweis Saft America, Inc. Advanced Battery Systems Division 107 Beaver Court Cockeysville, MD 21030 Michael C. Saft SAFT Research & Dev. Ctr. 107 Beaver Court Cockeysville, MD 21030 G. E. "Ernie" Palomino Salt River Project P.O. Box 52025 MS PAB 357 Phoenix, AZ 85072-2025 Final Report James McDowall Saft America, Inc. 3 Powdered Metal Drive North Haven, CT 06492 Guy Chagnon SAFT Research & Dev. Ctr. 107 Beaver Court Cockeysville, MD 21030 Mindi J. Farber-DeAnda SAIC 901 D Street, SW, Suite 201 Washington, DC 20024 Fred Schwartz San Francisco Public Utilities Commission 1155 Market Street, 4th Floor San Francisco, CA 94103 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Leonard Lansing Satcon Power Systems 835 Harrington Court Burlington, Ontario L7N 3P3 CANADA Jonathan Hurwitch Sentech, Inc. 4733 Bethesda Avenue Suite 608 Bethesda, MD 20814 Kurt W. Klunder Sentech, Inc. 4733 Bethesda Avenue Suite 608 Bethesda, MD 20814 Srikesh Sridharan Sentech, Inc. 4733 Bethesda Avenue, Suite 608 Bethesda, MD 20814 Final Report Irwin Weinstock Sentech, Inc. 4733 Bethesda Avenue Suite 608 Bethesda, MD 20814 Krista Long Sentech, Inc. 4733 Bethesda Avenue, Suite 608 Bethesda, MD 20814 Rajat K. Sen Sentech, Inc. 4733 Bethesda Avenue Suite 608 Bethesda, MD 20814 Richard N. Schweinberg Southern California Edison 6070 N. Irwindale Avenue Suite I Irwindale, CA 91702 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications James T. Collinson Southern Electric Exchange 3379 Peachtree Road NE Suite 250 Atlanta, GA 30326 Harold Gotschall Technology Insights 6540 Lusk Blvd. Suite C-102 San Diego, CA 92121 Tom Anyos The Technology Group, Inc. 63 Linden Avenue Atherton, CA 94027-2161 Donald A. Bender Trinity Flywheel Power 6724D Preston Avenue Livermore, CA 94550 Final Report Katsuji Emura Sumitomo Electric Industries, Ltd. 1-1-3 Shimaya Konohana Osaka, 554-0024 JAPAN Charles E. Bakis The Pennsylvania State University Dept. of Engineering Science & Mechanics 212 Earth-Engineering Science Bldg. University Park, PA 16802 Bill Roppenecker Trace Engineering Division 5916 195th Northeast Arlington, WA 98223 Melissa Reading Trinity Flywheel Power 6724D Preston Avenue Livermore, CA 94550 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Dr. Joe Hoagland TVA/Pubic Power Institute P.O. Box 1010 Reservation Rd., Muscle Shoals, AL 35662 (FedEx) Muscle Shoals, AL 35662-1010 Jim Drizos Trojan Battery Company 12380 Clark Street Santa Fe Springs, CA 90670 Dr. Gerald P. Ceasar Dan Sanchez U.S. Department of Commerce U.S. Department of Energy NIST/ATP Albuquerque Operations Office Bldg 101, Room 623 P.O. Box 5400 Gaithersburg, MD 20899 Albuquerque, NM 87185 Russ Eaton Samuel F. Baldwin U.S. Department of Energy U.S. Department of Energy Golden Field Office Office of Energy Efficiency and Renewable Energy 1617 Cole Blvd. 1000 Independence Avenue SW Bldg. 17 Washington, DC 20585-0121 Golden, CO 80401 Alex O. Bulawka Gil Bindewald U.S. Department of Energy U.S. Department of Energy 1000 Independence Avenue SW 1000 Independence Avenue SW. EE-11 FORSTL EE-2D FORSTL Washington, DC 20585 Washington, DC 20585 Final Report D-24 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Gary D. Burch U.S. Department of Energy 1000 Independence Avenue SW EE-11 FORSTL Washington, DC 20585 James Daley U.S. Department of Energy 1000 Independence Ave SW EE-2D FORSTL Washington, DC 20585 Peter R. Goldman U.S. Department of Energy 1000 Independence Avenue SW EE-12 FORSTL Washington, DC 20585 Dr. Imre Gyuk U.S. Department of Energy 1000 Independence Avenue SW EE-2D FORSTL Washington, DC 20585 Final Report Jack Cadogan U.S. Department of Energy 1000 Independence Avenue SW EE-11 FORSTL Washington, DC 20585 Tien Q. Duong U.S. Department of Energy Office of FreedomCAR & Vehicle Technologies 1000 Independence Avenue SW EE-32, FORSTL, 5G-030 Washington, DC 20585 Sigmund Gronich U.S. Department of Energy 1000 Independence Avenue SW EE-15 FORSTL Washington, DC 20585 Debbie Haught U.S. Department of Energy 1000 Independence Avenue SW Washington, DC 20585 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Patricia A. Hoffman U.S. Department of Energy 1000 Independence Avenue Sw EE-2D FORSTL Washington, DC 20585 Eric M. Lightner U.S. Department of Energy Communications and Control Program 1000 Independence Avenue SW EE-16 FORSTL Washington, DC 20585 William P. Parks U.S. Department of Energy 1000 Independence Avenue SW EE-10 FORSTL Washington, DC 20585 Neal Rossmeissl U.S. Department of Energy 1000 Independence Avenue SW EE-13 FORSTL Washington, DC 20585 Final Report Richard J. King U.S. Department of Energy 1000 Independence Avenue SW. EE-11 FORSTL, 5H-095 Washington, DC 20585 Philip N. Overholt U.S. Department of Energy Transmission Reliability 1000 Independence Avenue SW. EE-2D FORSTL Washington, DC 20585 James E. Rannels U.S. Department of Energy 1000 Independence Avenue SW. EE-11 FORSTL Washington, DC 20585-0121 Dan T. Ton U.S. Department of Energy 1000 Independence Avenue SW. EE-11 FORSTL Washington, DC 20585 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Henry Chase U.S. Flywheel Systems 21339 Nordhoff St. Chatsworth, CA 91311-5819 Edward Beardsworth UFTO 951 Lincoln Avenue Palo Alto, CA 94301-3041 Bor Yann Liaw University of Hawaii 1680 East West Road, Post 112A Honolulu, HI 96822 Max Anderson University of Missouri - Rolla 112 Electrical Eng. Bldg. Rolla, MO 65401-0249 Final Report Wayne Taylor U.S. Navy Code 83B000D, NAWS China Lake, CA 93555 Richard Lemmons United States Air Force OO-ALC/LIPC 6040 Gum Lane Hill Air Force Base, UT 84056-5825 Mariesa L. Crow University of Missouri - Rolla 1870 Miner Circle, 233 EECH Rolla, MO 65401 John Herbst University of Texas - CEM 10100 Burnet Road, Building 133 Austin, TX 78712 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Bob Lasseter University of Wisconsin 1415 Engineering Dr. Home- 2913 Walnut Wood Ct. 53711 Madison, WI 53706 Colin Davies Urenco Power Technologies Virginia Avenue, NW Suite 610 Washington, DC 20037 David St. Angelo Valence Technology, Inc. 6504 Bridge Point Pkwy. Austin, TX 78730 Rusi Patel Xenergy/KEMA consulting 3 Burlington Woods Burlington, MA 01803 Final Report Alan Palin Urenco (Capenhurst) Ltd. Capenhurst, Chester Cheshire, CH1 6ER ENGLAND John Woodthorpe Urenco Power Technologies Virginia Avenue, NW Suite 610 Washington, DC 20037 Michael Benke Xantrex Technology, Inc. 161-G South Vasco Road Livermore, CA 94550 Yoshiyasu Yamamoto YC Consulting 6102 Shelby Street Indianapolis, IN 46227 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications R. Kristiansen Yuasa-Exide, Inc. 35 Loch Lomond Lane Middleton, NY 10941-1421 Greg H. Nelson ZBB Energy Corporation W93 N14475 Whittaker Way Menomonee Falls, WI 53051 Peter Lex ZBB Technologies, Inc. N93 W14475 Whittaker Way Menomonee Falls, WI 53051 MS-0889/01832 MS-0512/02500 MS-0521/02501 MS-0613/02521 MS-0614/02522 MS-0614/02522 MS-0614/02522 MS-0614/02522 MS-0614/02523 MS-0613/02525 MS-0741/06200 Final Report Henry W. Zaininger Zaininger Engineering Co., Inc. 9959 Granite Crest Court Granite Bay, CA 95746 Mike Hughes ZBB Technologies Inc. N93 W14475 Whittaker Way Menomonee Falls, WI 53051 Robert J. Parry ZBB Technologies, Inc. N93 W14475 Whittaker Way Menomonee Falls, WI 53051 Jeff W. Braithwaite James K. Rice J. Thomas Cutchen Daniel H. Doughty Nancy H. Clark Paul C. Butler Stanley Atcitty Thomas D. Hund Robert W. Bickes, Jr. Rudolph G. Jungst (5) Marjorie L. Tatro D-29 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications Final Report MS-0708/06202 MS-0708/06214 MS-0753/06218 MS-0753/06200 MS-0710/06251 MS-0710/06251 MS-0710/06251 MS-0710/06251 MS-0557/09133 MS-9018/08945 MS-0612/09612 MS-0899/09616 Christopher P. Cameron Henry M. Dodd Ward I. Bower Paul C. Klimas Abbas A. Akhil Garth P. Corey John D. Boyes (10) John W. Stevens Angel Urbina Central Technical Files Review & Approval For DOE/OSTI (1) Technical Library (2) D-30 Technical and Economic Feasibility of Applying Used EV Batteries in Stationary Applications