Study Finds PHEV Li-ion Iron Phosphate Cells Show Little Capacity Fade Under Combined Driving and V2G Usage; Economic Model Suggests Incentives Will Be Required for Vehicle Owners to Participate in V2G
20 October 2009
[This is a revision of an earlier post, which had been pulled due to the status of the referenced papers as working papers. Both have now been revised and accepted by the Journal of Power Sources and are in press.]
| Degradation as a function of (a) capacity (Ah) processed by cell or (b) energy (Wh) at different DoD. Different DoD did not have a large impact on capacity fade. Source: Peterson et al. (CEIC-09-02) Click to enlarge. |
Researchers at Carnegie Mellon Electricity Industry Center have concluded that a PHEV pack comprising lithium iron phosphate cells would incur little capacity loss from combining vehicle-to-grid (V2G) activities with regular driving. Statistical analyses indicated that rapid battery cycling incurred when driving degraded the cells more than slower, vehicle-to-grid galvanostatic cycling.
Scott Peterson, Jay Apt, and Jay Whitacre also found that the percent capacity lost in the cells (they used A123Systems 26650 M1 cells, which are used in the Hymotion PHEV conversion packs) per normalized Wh or Ah processed is quite low even based on just use in a dynamic driving cycle—more than 95% of the original cell capacity remained after thousands of driving days worth of use. However, in a companion paper assessing the economics of V2G for consumers, they also concluded that the maximum annual profit for a PHEV owner to engage in V2G (~$10-$120) would likely prove insufficient to encourage use of the battery pack for grid electricity storage and later off-vehicle use.
| Test current profile used to simulate driving day for cells showing all trips. The times after trips 3 and 4 when V2G discharge was simulated are indicated. Source: Peterson et al. (CEIC-09-02) Click to enlarge. |
Performance. The quantify the capacity degradation of the pack, the authors derived nominal urban driving and driving/V2G power profiles and correlated battery test regimes by combing several common data sets. To determine the quantity and rate of energy transferred to and from a battery during driving conditions, they used a simple physics model that computed the energy needed to propel a typical vehicle through the trip profile.
To calculate the power vs. time battery duty cycle needed to achieve this velocity/acceleration profile, the vehicle was assumed to have the physical characteristics of a 2008 Toyota Camry. The efficiency of power transfer from regenerative braking to batteries was assumed to be 40%, the efficiency from battery to wheels was assumed to be 80%. The battery pack energy capacity was assumed to be 16 kWh. An 800 watt constant load was added to account for the power needed for all activities unrelated to movement such as heater, air conditioner, radio, lights and other accessories.
Among their findings were that the cell depth of discharge (DoD) does not does not have nearly as great an effect on lifetime as previously reported values for other battery chemistries. In cells discharged to 95% DoD per cycle, their measurements predicted that 5,300 cycles will be needed before reaching 80% of initial capacity instead of around 1,500 cycles as indicated by other data. Daily cycles with shallower DoD values do not appear to increase cycle life.
They also found that there is a difference between driving energy withdrawn and constant discharge—i.e., for V2G. The low rate constant discharge for V2G resulted in roughly half the degradation of that of the dynamic discharge for driving: -2.7x10-3 percent capacity lost per normalized Wh or Ah processed for V2G support vs.-6.0x10-3 percent for dynamic driving support. These values show that several thousand driving/V2G driving days incur substantially less than 10% capacity loss regardless of the amount of V2G support used, they concluded.
This result implies that a LiFePO4/graphite–based PHEV battery pack with properly matched cells can be cycled though a very broad state of charge range without incurring any significant increase in capacity loss as a function of Ah or Wh processed. In principle, a PHEV can utilize a smaller battery and use a greater proportion of the battery, however doing so might make discharge rate and associated ohmic heating more of an issue.
...the cycle DoD and relative fraction of low-rate galvanostatic cycling vs. acceleration/regenerative braking current pulses are not important even over thousands of driving days. Rather, it is the integrated number of lithium ions that have been intercalated/de-intercalated into the electrodes, regardless of the DoD at which these events occur.
—Peterson et al. (CEIC-09-02)
Economic model. In the companion working paper, the authors examined the potential economic implications of using plug-in vehicle batteries to store grid electricity generated at off-peak hours for off-vehicle use during peak hours. They used hourly electricity prices in three US cities to arrive at daily profit values, while the economic losses associated with battery degradation were calculated based on data from the first study.
For a 16 kWh vehicle battery pack, the maximum annual profit with perfect market information and no battery degradation cost ranged from ~$140 to $250 in the three cities. If the measured battery degradation is applied, however, the maximum annual profit (if battery pack replacement costs fall to $5,000 for a 16 kWh battery) decreases to ~$10-$120. It appears unlikely that these profits alone will provide sufficient incentive to the vehicle owner to use the battery pack for electricity storage and later off-vehicle use.
—Peterson et al. (CEIC-09-03)
They also estimated grid net social welfare benefits from avoiding the construction and use of peaking generators that may accrue to the owner, and found that these are similar in magnitude to the energy arbitrage profit.
...the vehicle owner might be able to avoid ~$200 of peaking costs. In states with traditional regulated electricity, the public utility commission might elect to avoid paying the utility to install and run a peaker, instead giving some of the avoided cost to V2G owners. In restructured states, the ISO/RTO may pay an aggregator to provide V2G power instead of paying a generator a capacity payment; the aggregator would then pay some of their revenue to the vehicle owner. In the absence of such incentives, it is unlikely that large-scale grid energy storage in PHEVs will be attractive to a large number of vehicle owners.
—Peterson et al. (CEIC-09-03)
Resources
Scott Peterson, Jay Apt, and Jay Whitacre (2009) Lithium-Ion Battery Cell Degradation Resulting from Realistic Vehicle and Vehicle-to-Grid Utilization (CEIC-09-02)
Scott Peterson, Jay Whitacre, and Jay Apt (2009) The Economics of Using PHEV Battery Packs for Grid Storage (CEIC-09-03)
I wouldn't let the utilities at my $5K battery pack for $120 / year - sorry, no way.
I might use it to reduce my own peaking use, but I wouldn't let anyone else at it.
I think it would be a better idea to have separate B2G batteries where you could have a few KWh of heavy batteries in your house to do a little load shifting. If you had some solar cells, this would work even better.
If peak power is a problem for vehicle batteries, add in some supercaps.
Also, in western Europe, peak demand is about 5.30- 6 pm, when many people are in their cars.
Posted by: mahonj | 22 August 2009 at 09:19 AM
I'm also skeptical about V2G.
The big problem with EVs so far is insufficient battery capacity to allow an EV to provide similar functionality to a gas car. That anyone would want to dump their EV's charge into the grid seems really farfetched for the forseeable future.
Furthermore, it would stand to reason that EV batteries will always be much more expensive than fixed energy storage devices. Stationary batteries don't have the constraint of having to be low weight like EV batteries do. And, lower-cost mass storage approaches other than batteries are likely going to be the mainstay of a renewables-based energy economy, e.g. molten salt storage systems accompanying "power tower" solar thermal plants.
Posted by: bot_feeder | 22 August 2009 at 10:22 AM
Very relevant findings! Very good data to help reduce cost for the GM Volt and similar PHEV's. The Volt can have smaller battery, but the engine will have to kick in occasionally to supplement power in order to reduce the rate of discharge of the battery. The engine may have to be insulated to prevent heat loss when not in used, in order to prevent emission problem. Or, Ultracapacitor can provide for the extra current needed during acceleration.
Posted by: Roger Pham | 22 August 2009 at 11:06 AM
So the govmnt could "incentivize" EV owners to allow V2G.
This would be a small penalty – umm wait; penalize EV owners ? – a really bad idea.
Looks like V2G is another pipe dream that falls apart in the face of common sense.
OK then. Looks like the Volt can double it AER soon.
Ultracapacitirs would allow smaller batteries for the same range but if cap cost and weight are not significantly less, just keep it simple with the 16kWh of batteries (to keep the power) and get increased AER.
Posted by: ToppaTom | 22 August 2009 at 08:09 PM
Too many variables.
Surely the 'incentive' for using V2G is the local utility prices and the difference between on and off peak. (prices not mentioned in the stidy?) Peak is often double the cost per kWh of off peak so simply using your own off peak charged EV battery to power your house during the day is incentive enough!
It's also likley that someone tuned in enough to own an EV and who is considering V2G will be odds on the have installed microgeneration of some kind on their property. In this case the profitability of V2G could be enhanced by Feed in tariffs.
Posted by: Paul | 22 August 2009 at 11:12 PM
"Peak shaving" (sending power back to the grid when demand is high) is NOT the only service that could provide profit to a BEV or PHEV owner. V2G could give utilities new ways to provide regulation services (keeping voltage and frequency stable) and provide spinning reserves (meet sudden demands for power). In future development, it has been proposed that such use of electric vehicles could buffer renewable power sources such as wind power, for example, by storing excess energy produced during windy periods and providing it back to the grid during high load periods, thus effectively stabilizing the intermittency of wind power.
Some have claimed the total profit from all of these could be between $1000 - $5000 per year.
http://www.udel.edu/V2G/
Posted by: ai_vin | 23 August 2009 at 07:50 AM
Great paperwork - really of no use in the real world.
I wonder if they considered the voiding of the warranty for the $8,000 battery pack as soon as it is used for V2G.
It seems like the customer would need a very large incentive for that.
The Mellon academics live in a world of simulation and do-gooder naivety. They need to understand some basic fundamentals like electric power sufficiency, durability margins, and the nature of new car ownership.
Posted by: frankbank | 23 August 2009 at 07:55 AM
Why is this of no use? Has there been a similar study? Is it common knowledge that V2G has little effect on the capacity of the battery? It seems that if no one asks these question and applies solid analysis, then we would all be in the dark on the topic.
In the GGC review for the second paper they state clearly that there is little economic incentive to use V2G. If was preparing a business plan and trying to account for V2G then this type of analysis is quite useful.
Posted by: ssintay | 23 August 2009 at 07:28 PM
mahonj
I wouldn't let the utilities at my $5K battery pack for $120 / year - sorry, no way.
Can you provide some more insight in why you think that the utilities will ruin your battery? Do you really think that your battery is in the hands of the utilities and they can do with it as they please? That they can drain it to 0% if they like? How naive.
Everybody knows that shallow cycles do very little damage to a battery. Your car will limit the minimum charge & discharge levels to safe values so your battery will not wear out.
The idea of having batteries in your house is very bad. You will never recoup those costs. The additional use of the car battery that you need anyway does not cost any extra investment and does no measurable damage to your battery. That can be profitable. Simple economics.
I'd happily do V2G if that reduces the operation cost of my EV.
Posted by: Arne | 24 August 2009 at 04:03 AM
bot feeder:
Furthermore, it would stand to reason that EV batteries will always be much more expensive than fixed energy storage devices. Stationary batteries don't have the constraint of having to be low weight like EV batteries do.
You make the same mistake as mahonj. A dedicated load leveling battery will never give a return on your investment at current battery costs. The additional use of a car battery that you need to buy anyway can be profitable.
Posted by: Arne | 24 August 2009 at 04:05 AM
Paul
Peak is often double the cost per kWh of off peak so simply using your own off peak charged EV battery to power your house during the day is incentive enough!
Where are you during the day? I am usually at work, as 80% of the population. Not a practical idea.
Posted by: Arne | 24 August 2009 at 04:06 AM
frankbank
I wonder if they considered the voiding of the warranty for the $8,000 battery pack as soon as it is used for V2G.
You do not understand how this will work in practice. The car will always be in control of the battery. So it will allow the utility to so much draw power as to not do harm to the battery. Who says the warranty is void? Or did you make that up?
Posted by: Arne | 24 August 2009 at 04:10 AM
If no lasting harm is done the vehicle battery pack (and availability) and you can make $$$/year buying power at low rate and reselling at much higher rate while increasing the overall power grid security, why NOT.
Most people use their car about 1/24 of the time. The other 23/24 is available for recharge or discharge. Why not put the battery pack to work a bit more if no lasting harm is done? Controlling recharge and discharge depth + level of charge needed at any given times should not be an impossible task to do.
Posted by: HarveyD | 24 August 2009 at 08:23 AM
Anne,
I hate to have to burst the dream bubble of V2G. But if it really didn't cost more in battery wear-out than it saved in peak load shaving, then power companies would buy their own batteries, and locate them right next to the plant. It would be even more cost effective and efficient for them to operate that way...if it were true.
But of course, all batteries wear out. Have you suggested that they will not?
The party holding the bill for battery replacement (car mfgr, under warranty, leaseholder if leased, or car owner if owned outright) is rightly concerned to protect their pricey car battery investment.
V2G is an academic dream exercise that electric power companies have encouraged to look for a free lunch at someone else's expense.
Posted by: frankbank | 24 August 2009 at 10:26 AM
frankbank: I love that you brought that up. You posed a fine conditional, so please check this out:
http://www.powermag.com/issues/departments/focus_on_o_and_m/Advanced-batteries-supply-ancillary-services_1589.html
There are dozens of these trailers in service around the country, and more being built. Of course, if the power companies can get you (the vehicle owner) to subsidize the cost of the batteries, then the economics work out even better for them. In other words, if those trailers were off-line (being driven or not plugged in) 9 hours a day, they would provide 15/24-ths as much regulation service, so the utility could to pay for up to 15/24-ths the cost of the battery and come out ahead.
Right?
Posted by: Nat Pearre | 20 October 2009 at 06:50 AM
Home photovoltaic systems have intertie inverters that can direct excess energy back to the grid. State laws require payback at market price from the power company. We could do the same with batteries. Charge them at off peak rates at night and sell it back during the day. This would work best with a dedicated set of batteries in the basement for example.
The problem with car batteries is you need to use the battery in the morning to drive to work. When you get home in the evening, you need to charge it right away to go the the supermarket. However if you have a Li-Air battery with a 500 mile range you only need to charge it twice a month. So you could sell it's energy to the power company in the early evening and charge it later at night.
Most people don't need to carry a 500 mile battery in the car all the time. I think batteries will be modular in the near future. Keep a 50 mile range battery installed at all times. Charge it every night at home. Keep more batteries in the home which can be used to buy low and sell high for load leveling with the power company. When you need to take a long trip, transfer some batteries from the house to the car.
Posted by: Zhukova | 20 October 2009 at 06:55 AM
But if it really didn't cost more in battery wear-out than it saved in peak load shaving, then power companies would buy their own batteries, and locate them right next to the plant.
frankbank,
You continue to not understand it.
V2G is ADDITIONAL use of an investment that has to be made anyway. What you are proposing are DEDICATED batteries. That is an entirely different case with different economics. It doesn't in any way prove that V2G can't be economical.
Think deeper.
Posted by: Arne | 20 October 2009 at 07:19 AM
"That anyone would want to dump their EV's charge into the grid seems really farfetched for the forseeable future."
The research here suggests that people aren't using much of their battery capacity:
http://www.greencarcongress.com/2009/10/ukccc-20091014.html
V2G makes a lot of sense initially for load balancing if the impact on battery cycle life is minimal. In the future as costs come down then I'll be happy to give up charge. Both services for a suitable fee of course.
Posted by: Scatter | 20 October 2009 at 07:57 AM
The average BEV, if recharged at the lowest night rate, for about 8 Kwh/day, will not use much more than $150/year from the power grid.
A V2G configuration with proper current limiter + disconnet features to stop battery discharge from going under 30% should not damage the battery pack and could make a BEV almost energy cost free if you can negotiate a V2G power credit of at least $150/year or more. Something like $200 to $250 a year would represent a minor profit at no cost to BEV owners.
On-board ultra caps would certainly make most battery packs last much longer. It should be considered for all heavy users like taxis, delivery vehicles, city buses and for people doing lots of city driving.
Posted by: HarveyD | 20 October 2009 at 09:32 AM
Most people don't need to carry a 500 mile battery in the car all the time. I think batteries will be modular in the near future. Keep a 50 mile range battery installed at all times. Charge it every night at home. Keep more batteries in the home which can be used to buy low and sell high for load leveling with the power company. When you need to take a long trip, transfer some batteries from the house to the car.
You wouldn't need to tranfer the batteries if you kept them on a trailer; it's call the 'car and 1/2' idea.
Posted by: ai_vin | 20 October 2009 at 11:42 AM
V2G does not sound far fetched or whacked out. Once the battery tech is robust and commonplace you won't be skitish about 'lending' back some power, for $.
This would provide grid stability and benefit all.
Posted by: danm | 20 October 2009 at 01:08 PM
The same argument could be run around load shedding where you ask (program)the meter to switch off certain appliance air con, fridge, wash machines, pool crawler and the vacuuming robot maid until the lower tariff returns.
New appliances would inevitably be designed to smooth over these off times by IE status lights to discourage use, stable deeper and or up sizing insulation redesigning the shelf layout with more stable long storage and less stable areas for fast moving foods.
Numerous if not most appliances can be made more efficient.
Of course the same naysayers will say "not my fridge you won't."
regardless the fact that they will be the very same people complaining that their power bills escalate.
Many simple consumer items use batteries and capacitors.
That economically function in a similar way to smart grid (in all its manifestations) But in a way we 'Are familiar with and accept as necessary and doesn't challenge the status quo.
Why would one even try to understand when being dragged (as if anyone would bother)
kicking and screaming is so dramatic and then get to complain about the power bill. Or lack of supply.
Demanding the status quo seems like much more fun
Larger battery packs may be very useful in a car anytime soon, the salt towers may see service one day but they wont make the car go further.
I think the main objection is that the report is done by "academics collaborating with power companies for a free lunch at someone elses expense."
Except this is a favourable report by academics in collaboration with power companies to integrate renewable energy, Allow the roll out of electric vehicles and provide consumers with an environmentally efficient money saving way to enjoy the maximum benefit from this huge sector.
And when scarce fuel resources and rationing become reality?
And vehicle owners need to access the low carbon renewables?
Posted by: arnold | 20 October 2009 at 03:24 PM
Anne,
Here's what I get, batteries wear out.
When you say "ADDITIONAL use of an investment that has to be made anyway", you must think of a battery like its a monument that remains the same value, and does wear out, no matter what you do with it. Its as if the battery has infinite life.
So thinking deeper, a battery has a finite cycle life, can you agree to that? Consider a particular EV battery pack has a cycle life of say, 20 megawatt hours of total discharge. So the "investment" of say, $10,000 for that pack is depreciated with its use. So each kWhr of discharge depreciates the pack by 50 cents. Whether the pack is consumed by driving a car or by V2G dreams is not relevant when accounting for a pack with finite life.
The power company wants to use the pack for peak shaving. Will they pay $0.50 per kWhr for it? No, but they will encourage people to think about it to them for free. Get your own battery.
Posted by: frankbank | 20 October 2009 at 04:30 PM
I would participate in this if the power company provided free power and the charger to charge my EV during the day.. they would need to install many chargers all over the city and in my garage and all free.
Posted by: Herm | 20 October 2009 at 05:11 PM
If you use a 16 kwh battery, and discharge it by 5% for peak shaving, that's .8 kwh of energy. If you buy this energy at night for $.06/kwh and sell it back in the day for $.10 you get a $.03 payback. The article says you can do this 5,000 times before getting to 80% capacity. So that's only about $150.
But you're deep discharging the same battery every day while driving the car. That limits the lifetime to about 1,500 charges, so the payback for peak shaving is only about $35 over about four years. You also would need something like a Sunny Boy intertie inverter to send power to the grid. The 700 watt model cost about $1,200. I don't see a realistic payback with current battery technology.
Posted by: Zhukova | 20 October 2009 at 07:26 PM