I think that fleet replacement could go a little faster than 20 years. Also, the batteries in vehicles are not likely to be worn out when replaced, they'll just be degraded below transportation grade. So, I would guess that much storage will be in these used batteries. PG&E is already making contracts to purchase these.
A fleet replacement of established technology (ICE) with a non-commercial idea-stage technology for 20 years?
People, get some life. It took 20 years for ordinary hybrids to go from development to what? 2% of the market? How long will it take for plug-ins? For V2G? If I need to make a bet - no earlier than 2050.
"A total of 187,000 hybrid vehicles were sold in the United States in the first six months of 2007, according to J.D. Power. Sales of hybrid vehicles are expected to decline slightly in the second half of the year but, nevertheless, J.D. Power expects a total of 345,000 to be sold over the whole year. That would compare to 256,000 sold in 2006." CNN Money
PO will dramatically change the purchasing trends of automobiles. Toss old assumptions out the door.
It was PR. The commercial showed a bunch of kids hugging the hood. It was a commercial channel but I don't know which. But, it is interesting that GM would spend on a commercial for a product they won't produce yet for a while. Maybe they want to boost their credit rating by generating a wait list the way that the Prius had a wait list.
Plug-ins are an excellent idea because they fit in our current infrastructure. I expect them to become quite common within a decade or two. Maybe predominant - but just maybe - it is still to be seen what will be their cost and if they are able to scale fast enough.
V2G is another thing, because it will require additional infrastructure plus fitting the current one in it. Hell we don't even know if it will work - so far it's just a nicely sounding idea, demonstrated by noone.
Personally I think V2G won't work very well - and here is why: when I park my car at work in the morning I want to be able to start it any time I want to, and drive home or to my errands. Unfortunately after I park my car at work in the morning, this will be the most likely time millions of others will be doing the same and crank up those factories, A/Cs, computers etc. Hence we've got a certain negative correlation between parking and electricity demand. If this is the case when I leave work I will most likely find my car LESS charged than when I left it in the morning. If I use the ICE to recharge it on my way back, then effectively I am using my on-board ICE as a 15% efficient electrical generator that feeds the grid - burning (very expensive at that time) gasoline as a fuel. No good deal for me, thanks.
The bottom line is that V2G maybe sound like a good idea and may help a little but don't count on it yet. Let's see how plug-ins perform before we count those chickens.
LevinK, how about if we exclude you from V2G, because you seem to phrase criticisms as if you represent the world's style of life. Things will change and you won't have whatever power you want whenever you want it. Go with the flow.
But, lets not let go of, say, electric school buses. Known and regular hours on the road. But critically, an underused asset through most of the summer.
Just sitting there, 50kWh+ storage, providing ancillary services, up and down regulation, adding to spinning reserves.
Centrally located at a charging park. Oh, BTW, in emergencies, when locals seek shelter from the storm at a school, buses may provide local backup power.
Whether an idea will work or not has to be investigated prior to spending billions in implementing it. Or you don't agree? You guys are counting the chickens before they hatch and I bet none of you is even an electrical engineer. I am also not an engineer but at least I don't tell engineers what works or not.
EP's essay may be a good idea of what may possibly work. But it has to be proven in practice and this is the tough part. We can fantasize all we want.
I agree that this may possibly work. However, the potential payoff is enormous. Seeing if it does work should therefore be one of our highest priorities.
Edit: My degree says "BSEE" on it.
Unlike you, I have read (and more importantly, understood) the V2G papers at AC Propulsion's site. V2G is not a major method of storing energy for return to the grid. Its major uses are:
Schedulable/controllable demand, which is as good for the grid as schedulable generation (it is equivalent to spinning reserve);
Regulation, where power is taken or added on a time-scale of minutes to address immediate shortages or surpluses (and having little effect on battery SOC), and
Generating reactive power (which involves no energy transfer at all, but makes the grid more efficient).
I just saw your edit. Thank you very much for the clarification. And of course my non electrical engineering remark was not against you - you did not participate in this conversation at this point, so I still hold on my bet.
Now in the light of your clarification, would you go back to your original post and revise it? I'm sorry for this request, but the most critical point of your clarification is that V2G is NOT a significant source for grid energy storage. While in the original post you explicitly rely on it to store non-dispatchable solar energy. Considering the amounts of energy we're talking about you have to admit this is not a viable suggestion at all.
Of course I am at fault for not doing the research before posting, but to my credit the V2G faults I pointed out were leading to exactly the same conclusion - that V2G can not be a major source for grid storage. But it could be used to stabilize the grid - something which I did not think about and I thank you again for the fair clarification. I would be curious about the cost/benefit analysis for this, but for this one I promise I'll do my own research.
Now that we are back to lacking a viable way to store huge amounts of electric energy we are again in front of the classical problems what will provide the baseload power and what will be used as peaking power.
Clearly solar can be a part of both, but what exactly part remains to be seen. I see it doing relatively well combined with nuclear as a secure baseload and for the nights plus NG plus some DSM to handle the peaks. Like you pointed out V2G can be used both as DSM and to handle short-term shartfalls. 50% nuclear, 30% solar and 20% NG look pretty viable to me. With apologies to Alan, but if/when solar picks up I don't think wind will be referred to at all.
Yet again, Levin, You are wrong. See you don't bother to look up any references or educate yourself, so here is slide 15 of V2G Basics from a recent conference on plug-in vehicles sponsored by the IEEE:
Average car driven 1 hour/day --> time
parked is 23 hours/day
Daily average travel: 32 miles
Practical power draw from car: 10 - 20 kW
US power generation=811 GW; load=417 GW
US 191 million cars x 15 kW = 2,865 GW
Vehicle batteries in a converted nationwide fleet has some 7x the capacity of US load. That's non-negligible.
Let's go further. If each vehicle stores 10 kWh/day, that's 20% of US daily consumption of 10^10 kWh/day. With all-electric vehicles at 35 kWh, storage capacity is 70% of electricity consumed per day. So, for example, an electric fleet enables significantly more wind power to be productively used than without a storage method.
You know in Texas they have installed so much wind power that now at night they can't use it all. Storage would significantly grow their use of wind. But then, you have a history here of arguing against significant amounts of wind.
Don't you think there's a reason people get so snippy with you?
You know in Texas they have installed so much wind power that now at night they can't use it all. Storage would significantly grow their use of wind.
Is nobody building pumped storage? At (IIRC) ~$100/MWh capacity, I would've expected selling dirt-cheap nighttime wind power at daytime peak rates would be economic. I could certainly be missing something, though.
Pitt, the comment was made last month in a public forum in front of audience of engineers, policy makers, business people, and utility people, and was made by Mark Kapner, PE, Senior Strategy Planner for Austin Energy.
My guess is the resources for pumped storage are just not there regionally.
They are, apparently, adding load balancing by putting windmills on the coast (where the winds blow at different times of day, even if the Load Factors are much lower than on the Texas Plains).
Warren Buffett is investing up to $4bn in Texas windmills.
If I am wrong, this mean EP is also wrong, and this paper here (PDF) is also wrong. Note this is a detailed technical paper not a power point presentation. I think you confuse grid regulation with energy storage. Grid regulation is the service of having a stand-by power to meet short-lived variations in demand. Utilities use hydro power, battery packs and capacitor banks to do it.
But it can not be used to smooth longer-term variation like the day/night cycle of solar. For these you need to have a significant storage in terms of GWh - like pumped hydro for example.
V2G does not have a huge capacity in terms of GWh. Your presentation and your calculation is bogus. The utility can not rely on all of the 10kwh stored in the car battery! Figure more like 2-3% of it. Why? Because if you jump in your car you need to know it is as fully recharged as possible - otherwise you are effectively feeding the grid with your ICE. Or the same thing I've been trying to point out all the time.
Overall I agree that V2G will be very useful. It will definitely enable more wind, because currently the V2G service is mostly performed by fossil plants operating in a spinning reserve mode. With V2G these can safely be retired.
BTW I'm expecting EP's response, at least he seems to know what he is talking about.
"...Because if you jump in your car you need to know it is as fully recharged as possible - otherwise you are effectively feeding the grid with your ICE."
There you go again with that ICE thing, and expecting whatever you want whenever you want it. You just don't get it. There are multiple services v2g enables. Check my first post on this thread.
By the way, it's not my presentation. It's from a guy who's been working on this for 10 years, and is currently riding around in a 35 kWh pure EV, and is working with a major utility and grid operator in a pilot program. The concept has been worked out. The concept has been demonstrated, and continues to be studied in order to get real-life experience with successively larger fleets.
I think it is obvious that V2G will provide some storage functions too. That may work nicely in the case with school buses which you pointed out.
This does not change that the 191mln.vehicles x 10kwh calculation was mildly said misleading. There will be real-world constraints on this number, with discharging limits depending on a number of factors. Even if we accept that the the on-demand user culture will have to change, I think the personal car will be the least useful part of V2G. Hopefully financial incentatives could address that somewhat.
I am not an expert in this, but the presentation quotes 10-20kw per vehicle. Is this viable? It seems to me typical households and neighborhood clusters are not calculated for charges like that... wouldn't it require some rewiring?
I am perfectly fine with all you've said, but I wasn't misleading, I simply pointed out order of magnitude values of certain quantities. Real-world constraints will of course limit what can be accomplished, but that is a discovery process that is underway.
The eBox in Willett's talk provides 120 kW. The U Delaware car gets taken out on 100+ mile trips around the Delaware valley with typically 30-40% of battery power left over before any recharging. Has cruised on I95 between Washington and Wilmington at 70+ mph without a problem.
As you know even better in local and stop and go traffic. EVs are amazing efficient due to additional regenerative braking for recharging the battery.
I just had a DOH! moment with all those things around V2G.
Tesla Roadster has a 53kWh battery pack. They claim it will last for 100,000 miles. They also assume energy efficiency of 110Wh/km or 177Wh/mile. So throughout the battery life they expect:
100,000 x 177Wh/mile = 17,700 kWh could be recycled though the battery.
Their battery pack consists of 6,800 18650 Li-Ion cells, currently selling for about $2.50/piece, wholesale. This is $17,000 just for the Li-Ion cells, and I'm assuming everything else could be reused after the cell is degraded (which is a very weak assumption).
So to cover only the degradation of my car battery, the utility will have to pay me:
17,000 / 17,700 = $0.96 / kWh!
Why in the world would they want to do it if the wholesale price of peaking power is more like $0.05 c/kWh? This is 19 times as expensive! And $0.96/kwth is just the beginning - we did not count the infrastructure and the original generation costs inside yet.
So, in order V2G to work we would need:
1) Batteries that don't degrade (ultracapacitors?) - the jury is still out on whether we'll see those
2) Either breakthrough on chemical battery life or on battery cost or a combination thereof. But what is the chance we could see breakthroughs that lower the cost 20 times!
The only thing I saw in the V2G papers about battery degradation is it will be a subject of later research. Isn't this way to convenient?
Please correct me if I'm missing something. If I don't then I'll consider this discussion to be over.
See my links and comments to Robert below, but in summary:
You're making money by just being plugged into the grid, providing spinning reserves. For regulation services, you can make $2500/yr.
On the battery lifetime, altairnano's specs claims a 15,000 deep-cycle lifetime, 41 years at one cycle per day. V2G will mostly be a small fraction of full cycle charge/discharge, so you might cut that lifetime by half.
Any reduced battery life is balanced against any revenue stream for providing services to the grid.
You're making money by just being plugged into the grid, providing spinning reserves.
"Spinning reserve" is the ability to produce energy on demand - and the bottom line is that you have to be able to produce it when requested at the market price.
Actually the way it works is at the time power is requested the loads are bid up with the lowest cost marginally produced power engaged first. In such environment V2G will never be used! The highest marginal cost peaking electricity is Natural Gas - at some $0.10/kwth it is 10 times less than batteries.
The only way what you suggest to work is to mandate utilities to NOT maintain enough lower cost spinning reserve thus creating artificial shortage in the market! Aren't you stretching this a little bit? How do you expect consumers will react to a $1/kwth price on their bills? Utilities will not pay for standby they will NEVER EVER use. They will build up the lower cost peaking generation until there is a glut of it - which is the case everywhere in the developed world. They will simply choose to ignore anybody who tries to sell them 20 times more expensive electricity. Or do you suggest the government mandates them to accept the bitter pill?
I agree that if it delivers, Altairnano's battery may address the cost issue to some extent. Assuming it reaches the same cost as Li-Ion and it has 10x times the battery life, the cost of the power (from degradation only) would be $0.10/kwth. Add infrastructure and premium costs and it would go to $0.15/kwth - closer to competitiveness but still remains to be seen. Just like with Eestor the question remains open.
Even then the utilities will prefer buying Altarirnano/Eestor batteries themselves. Why all the trouble of building V2G infrastructure and paying premiums if they can be in a full control and take all the benefits for themselves? Are you going to force them not to do it? Moreover the goal of accommodating renewables will be more easily reached this way.
I think you should abandon the idea at this point of time, it's getting way too funky.
Tesla uses conventional li-ion batteries. At about $400/kwh, and perhaps 500 cycles, the cost per discharge is about $.80, far too high for utility storage. This is generally understood - no one would suggest using conventional li-ion batteries for utility storage.
First, you have to realize that V2G isn't the most important use of vehicle batteries for utility load leveling. Instead, the the most important use of vehicle batteries for utility load leveling is dynamically scheduled charging, which will make a dramatic difference.
V2G won't be needed for large-scale utility load-leveling until wind & solar reach more than 20% each of market share - that won't be for at least 10 years, and that would be under a crisis mode installation program.
2nd, 500 cycles is conservative for conventional li-ion in a vehicle with sophisticated charge & temp management.
3rd newer li-ion batteries, such as A123systems, or Altair, have much more than 10x the cycle life of conventional li-ion.
4th, li-ion costs will continue to drop by 7-10% per year - it's pure economy of scale & manufacturing experience.
By the time V2G is needed, it will be cost-effective.
Guys, this is getting surrealistic. V2G will NEVER be competitive, because there is an inherent much better deal - do it yourself battery to grid. The utilities have a century long experience in doing it and so far I don't see them complaining.
No matter how battery technology evolves, it will always be more profitable for the utility to buy it's own batteries and do it itself instead of building and enormous infrastructure for effectively renting mobile batteries.
In addition utility scale batteries will always have different requirements than vehicle batteries. Utility scale batteries would also benefit from economies of scale. BTW why is nobody suggesting renting our cell phone batteries? If you add them up they will form enormous unutilized capacity.
If the utility pays you for your battery it will have to cover the following:
- battery degradation
- V2G infrastructure, maintenance and profit margin
- premium to the car owners to cover his cost of frequently buying new batteries and making him interested in the whole schema
If the utility byus its own batteries it will have to pay for:
- battery degradation (depreciation)
- nothing else
The first option will ALWAYS be much more expensive. Something plus something is always more than something plus zero. I would expect V2G to put at least a $0.10/kwth charge over the battery degradation cost. It is ridiculous to think they would ever consider such a deal.
There is a caveat: if ultra capacitors come to life and degradation costs get close to zero then the game becomes different. In this case the V2G infrastructure + car owner premium will have to compete with the cost of capital for buying the ultracapacitor over its lifetime. I would not hold my breath though - complex schemas like V2G are never cheap. It will be many years of investing billions upfront before the costs are brought down to competitive level... in a competitive market nobody would do it if they have an easy, quickly deployable and scalable alternative at hand.
Have a good evening and sorry for spoiling this party.
I think E-P has answered most of your concerns, but I'm not sure you really absorbed what I or he was saying, so I'll try again:
1) Tesla's batteries are far more expensive than other chemistries, because Tesla wanted the maximum energy density, which the older, conventional li-ion's provide. Firefly or A123systems batteries would be far, far cheaper than the $1/KWH that you're using, more on the order of 10 cents per KWH.
2) The most important thing is not V2G, which is energy flowing from the car. The most important thing is utility managed charging, which will buffer wind & solar. That will suffice for at least 10 years. By that point the infrastructure for utility managed charging will have been in place for years - the utilities (PG&E, etc), car companies (Tesla, GM) and software companies are already planning for this to be in place when the cars are sold. That infrastructure will seamlessly handle V2G.
3) A123systems batteries, which appear likely to win the Volt contract, have sufficiently long cycle life that effectively there is no degradation cost to the car owner to reselling energy back to the utility.
4) vehicle owners will pay for batteries for their transportation utility, and utilities won't have to pay the full cost of the battery.
By the time V2G is needed, if it ever is, the batteries & infrastructure will be ready.
OTOH, V2G may not be needed. Geographical diversity, long distance transmission, PHEV dynamic charging, pumped storage, flow batteries, Firefly lead-acid in utility scale installations....all of these may do the job. I suspect it will have an important role at least for the small-scale services that have been discussed, but we'll see.
"Spinning reserve" is the ability to produce energy on demand - and the bottom line is that you have to be able to produce it when requested at the market price.
Actually the way it works is at the time power is requested the loads are bid up with the lowest cost marginally produced power engaged first. In such environment V2G will never be used! The highest marginal cost peaking electricity is Natural Gas - at some $0.10/kwth it is 10 times less than batteries.
Actually you are wrong. We're not addressing peak demand. Spinning reserves is a reserve when power is required quickly, within minutes, from 'spinning' generators and ready to go at a moments notice. It is paid for by (1)the amount of time it's reserve is ready to go and available, and (2) the $/kWh of actual energy delivered. Ignore (2) as insignificant for now, let's focus on (1) revenue for being 'available'. A 100 kW battery plugged in for an hour provides 0.1 MW-hour of reserve. Agregated in a fleet of ten cars provides 1 MW - hour. No power need be exchanged in this service unless the ISO requests it. PJM (the ISO or grid operator in the mid-atlantic region) pays for this reserve $14 per MW - hr. Stay plugged in 10 hrs, that's $14/car per day. Multiply by days available per year for annual revenue. How often is spinning reserves called on? In the PJM service area, all of 21 hours for all of 2005 (pg 36 from Ref).
And then there are regulation services, the revenue of which was discussed by me elsewhere in this thread and comes from this work.
I think you should abandon the idea at this point of time, it's getting way too funky.
You insist on looking at it from the POV of V2G vehicle owner. You imagine getting a fixed stream of income basically for having a battery.
Guess what? It won't happen. Anyone with cache on hand can buy a battery and in the case of utilities they have to be crazy to pay all those premiums to you, not keep them for themselves.
Like I explained - you can not make them accept V2G if its cost per kwth is higher than what they already can get from existing spinning reserve. Since installed spinning reserve is more than enough, you are basically suggesting that they will be paying for an insurance for event that will never happen. It's like buying a life insurance for dead person.
The utility can't justify the capital cost just to have spinning reserve. But if you're paying for the battery via the difference between the cost of energy from electricity and the cost of energy from gasoline, the extra revenue from services such as regulation and spinning reserve is worth the expense in control systems and the minor impact on durability.
(If that impact exists at all; recall that AC Propulsion's regulation test caused the measured capacity of the Panasonic lead-acid battery pack to increase.)
When you fill your posts with multiple egregious errors of fact, you are not going to get the kind of answer you want. You are going to have to straighten both your facts and your reasoning out first, THEN you will get the kind of response that you desire.
After some digging into it I admit you two are right and I don't/didn't quite understand the nature of the peak load, spinning reserve and regulation services. I'm sorry.
Now I'm back to square one: what would be the cost/benefit analysis of V2G? It looks the primary service V2G could provide is regulation, it should be less competitive as spinning reserve or peak load. For these two I'd also think discharge limitations will limit the actual resource base - I am at loss what happens after the storage is drained if peak load/spinning reserve mode is engaged - aren't those contracts supposed to be for continuous power? I need to do some more digging but I would suspect that the V2G contracts would be depending both on MWs and the GWh-s available.
The market for grid frequency regulation services open to new service providers in the United States in 2004 was valued at nearly US$360 million--and is expected to increase in 2005
$360mln. or even $500mln. yearly is too thin of a market IMO. I think it will all depend on the cost of implementing the V2G system itself vs the expected revenue. Solar and especially wind may increase this market, but this would increase their overall costs too.
I don't really think more wind and solar plus V2G will be enough to retire significant number of existing units. Obviously spinning reserve is avoided for coal power plants as much as possible, so I don't think many of those are or will be kept running just because of that. If I understand correctly ancillary services don't reduce the need for base load power.
P.S. Somehow you forget that the $0.03-0.05/kwth electricity spinning reserve generators provide ALREADY includes its capital costs inside. And a generator may work 60 years. How long on-grid battery would last? 2 years?
"Spinning reserve" is the ability to produce energy on demand
Not quite. It's the ability to make up the difference between immediate supply and immediate demand on command (response time of a few seconds). This has historically been done with generators on-line but idling, but it can be done just as easily and far more efficiently by varying a large controllable load. In the case of (PH)EVs, the available spinning reserve is equal to the full charging load (and that's without making demands on the batteries to feed back to the grid).
and the bottom line is that you have to be able to produce it when requested at the market price.
Spinning reserve has its own market price, quite different from e.g. a MW of base load generation. If you have 200 million vehicles connected to the grid at a minimum of 6.6 kW peak charging load apiece (220 V 30 A connection; some will have more), that's 1.32 TW of load available to manage the grid. If it's averaging 1 kW per vehicle, you've got 200 GW (roughly 45% of today's average generation in the US) to play with as spinning reserve.
You can do considerably more than that, of course. Spinning reserve is there to make up for a large generator going off-line. V2G would allow back-feeding the grid for a few critical minutes while slower-reacting plants were brought up to take up the slack (or other demand was taken off-line). It has to be there, but it gets used very infrequently. In this way, V2G would supply "spinning reserve" at zero cost in fuel, minimal cost in equipment and a very low cost in battery life.
Historically the generators you find obsolate now were providing spinning reserve at the cost of $0.02-0.05/kwth.
Your genial innovation is going to provide them with spinning reserve for $1/kwth. OK lets allow innovations etc. to bring this down to say $0.20/kwth. Just perfect!
Just to mix in the conversation a little is the Flow battery. I dont think I've seen anyone mention it in this long thread. http://www.vrbpower.com/
We are putting one of these in for UPS of a Telco site I am in the middle of engineering. It will be 100Kw.
I think these may be more what we will end up seeing at the power utility level as more wind and solar projects come on-line. They are already using them with ff electric plants now at the end of long transmission line runs. Charging at night and using them for peaking during the day.
I dont have the numbers in front of me, but I recall it being 2x cheaper than lead-acid for mass storage and battery life. High up-front costs but last longer than Lead acid that we use in DC plants of telcos.
Thoughts on NaS batteries? AEP recently prchased some MWs, and IIRC, price was $4-4.5 per Watt.
Snipped press release from AEP below. Sounds like they're also interested in flow.
What's most interesting is they want to get 1 GW of storage.
“We’re first movers on advanced storage among U.S. utilities, a position we’ve held on a wide number of technologies in our century of existence,” Morris said. “Our near-term goal is to have at least 25 megawatts of NAS battery capacity in place by the end of this decade. But this is just a start. Our longer-term goal is to add another 1,000 megawatts of advanced storage technology to our system in the next decade. We will look at the full spectrum of technologies – flow batteries, pumped hydro, plug-in hybrid vehicles and various other technologies in early stages of development today – to determine their feasibility and potential for commercial application.”
I think that fleet replacement could go a little faster than 20 years. Also, the batteries in vehicles are not likely to be worn out when replaced, they'll just be degraded below transportation grade. So, I would guess that much storage will be in these used batteries. PG&E is already making contracts to purchase these.
Chris
A fleet replacement of established technology (ICE) with a non-commercial idea-stage technology for 20 years?
People, get some life. It took 20 years for ordinary hybrids to go from development to what? 2% of the market? How long will it take for plug-ins? For V2G? If I need to make a bet - no earlier than 2050.
"A total of 187,000 hybrid vehicles were sold in the United States in the first six months of 2007, according to J.D. Power. Sales of hybrid vehicles are expected to decline slightly in the second half of the year but, nevertheless, J.D. Power expects a total of 345,000 to be sold over the whole year. That would compare to 256,000 sold in 2006."
CNN Money
PO will dramatically change the purchasing trends of automobiles. Toss old assumptions out the door.
Interesting that I saw a television add for the Volt already.
Chris
Did they mention the price? Which channel?
It was PR. The commercial showed a bunch of kids hugging the hood. It was a commercial channel but I don't know which. But, it is interesting that GM would spend on a commercial for a product they won't produce yet for a while. Maybe they want to boost their credit rating by generating a wait list the way that the Prius had a wait list.
Chris
Plug-ins are an excellent idea because they fit in our current infrastructure. I expect them to become quite common within a decade or two. Maybe predominant - but just maybe - it is still to be seen what will be their cost and if they are able to scale fast enough.
V2G is another thing, because it will require additional infrastructure plus fitting the current one in it. Hell we don't even know if it will work - so far it's just a nicely sounding idea, demonstrated by noone.
Personally I think V2G won't work very well - and here is why: when I park my car at work in the morning I want to be able to start it any time I want to, and drive home or to my errands. Unfortunately after I park my car at work in the morning, this will be the most likely time millions of others will be doing the same and crank up those factories, A/Cs, computers etc. Hence we've got a certain negative correlation between parking and electricity demand. If this is the case when I leave work I will most likely find my car LESS charged than when I left it in the morning. If I use the ICE to recharge it on my way back, then effectively I am using my on-board ICE as a 15% efficient electrical generator that feeds the grid - burning (very expensive at that time) gasoline as a fuel. No good deal for me, thanks.
The bottom line is that V2G maybe sound like a good idea and may help a little but don't count on it yet. Let's see how plug-ins perform before we count those chickens.
LevinK, how about if we exclude you from V2G, because you seem to phrase criticisms as if you represent the world's style of life. Things will change and you won't have whatever power you want whenever you want it. Go with the flow.
But, lets not let go of, say, electric school buses. Known and regular hours on the road. But critically, an underused asset through most of the summer.
Just sitting there, 50kWh+ storage, providing ancillary services, up and down regulation, adding to spinning reserves.
Centrally located at a charging park. Oh, BTW, in emergencies, when locals seek shelter from the storm at a school, buses may provide local backup power.
And for some new EV performance results: click me
I don't see the point of your snippy comment.
Whether an idea will work or not has to be investigated prior to spending billions in implementing it. Or you don't agree? You guys are counting the chickens before they hatch and I bet none of you is even an electrical engineer. I am also not an engineer but at least I don't tell engineers what works or not.
EP's essay may be a good idea of what may possibly work. But it has to be proven in practice and this is the tough part. We can fantasize all we want.
I agree that this may possibly work. However, the potential payoff is enormous. Seeing if it does work should therefore be one of our highest priorities.
Edit: My degree says "BSEE" on it.
Unlike you, I have read (and more importantly, understood) the V2G papers at AC Propulsion's site. V2G is not a major method of storing energy for return to the grid. Its major uses are:
I just saw your edit. Thank you very much for the clarification. And of course my non electrical engineering remark was not against you - you did not participate in this conversation at this point, so I still hold on my bet.
Now in the light of your clarification, would you go back to your original post and revise it? I'm sorry for this request, but the most critical point of your clarification is that V2G is NOT a significant source for grid energy storage. While in the original post you explicitly rely on it to store non-dispatchable solar energy. Considering the amounts of energy we're talking about you have to admit this is not a viable suggestion at all.
Of course I am at fault for not doing the research before posting, but to my credit the V2G faults I pointed out were leading to exactly the same conclusion - that V2G can not be a major source for grid storage. But it could be used to stabilize the grid - something which I did not think about and I thank you again for the fair clarification. I would be curious about the cost/benefit analysis for this, but for this one I promise I'll do my own research.
Now that we are back to lacking a viable way to store huge amounts of electric energy we are again in front of the classical problems what will provide the baseload power and what will be used as peaking power.
Clearly solar can be a part of both, but what exactly part remains to be seen. I see it doing relatively well combined with nuclear as a secure baseload and for the nights plus NG plus some DSM to handle the peaks. Like you pointed out V2G can be used both as DSM and to handle short-term shartfalls. 50% nuclear, 30% solar and 20% NG look pretty viable to me. With apologies to Alan, but if/when solar picks up I don't think wind will be referred to at all.
Yet again, Levin, You are wrong. See you don't bother to look up any references or educate yourself, so here is slide 15 of V2G Basics from a recent conference on plug-in vehicles sponsored by the IEEE:
Average car driven 1 hour/day --> time
parked is 23 hours/day
Daily average travel: 32 miles
Practical power draw from car: 10 - 20 kW
US power generation=811 GW; load=417 GW
US 191 million cars x 15 kW = 2,865 GW
Vehicle batteries in a converted nationwide fleet has some 7x the capacity of US load. That's non-negligible.
Let's go further. If each vehicle stores 10 kWh/day, that's 20% of US daily consumption of 10^10 kWh/day. With all-electric vehicles at 35 kWh, storage capacity is 70% of electricity consumed per day. So, for example, an electric fleet enables significantly more wind power to be productively used than without a storage method.
You know in Texas they have installed so much wind power that now at night they can't use it all. Storage would significantly grow their use of wind. But then, you have a history here of arguing against significant amounts of wind.
Don't you think there's a reason people get so snippy with you?
Is nobody building pumped storage? At (IIRC) ~$100/MWh capacity, I would've expected selling dirt-cheap nighttime wind power at daytime peak rates would be economic. I could certainly be missing something, though.
Pitt, the comment was made last month in a public forum in front of audience of engineers, policy makers, business people, and utility people, and was made by Mark Kapner, PE, Senior Strategy Planner for Austin Energy.
My guess is the resources for pumped storage are just not there regionally.
They are, apparently, adding load balancing by putting windmills on the coast (where the winds blow at different times of day, even if the Load Factors are much lower than on the Texas Plains).
Warren Buffett is investing up to $4bn in Texas windmills.
If I am wrong, this mean EP is also wrong, and this paper here (PDF) is also wrong. Note this is a detailed technical paper not a power point presentation. I think you confuse grid regulation with energy storage. Grid regulation is the service of having a stand-by power to meet short-lived variations in demand. Utilities use hydro power, battery packs and capacitor banks to do it.
But it can not be used to smooth longer-term variation like the day/night cycle of solar. For these you need to have a significant storage in terms of GWh - like pumped hydro for example.
V2G does not have a huge capacity in terms of GWh. Your presentation and your calculation is bogus. The utility can not rely on all of the 10kwh stored in the car battery! Figure more like 2-3% of it. Why? Because if you jump in your car you need to know it is as fully recharged as possible - otherwise you are effectively feeding the grid with your ICE. Or the same thing I've been trying to point out all the time.
Overall I agree that V2G will be very useful. It will definitely enable more wind, because currently the V2G service is mostly performed by fossil plants operating in a spinning reserve mode. With V2G these can safely be retired.
BTW I'm expecting EP's response, at least he seems to know what he is talking about.
"...Because if you jump in your car you need to know it is as fully recharged as possible - otherwise you are effectively feeding the grid with your ICE."
There you go again with that ICE thing, and expecting whatever you want whenever you want it. You just don't get it. There are multiple services v2g enables. Check my first post on this thread.
By the way, it's not my presentation. It's from a guy who's been working on this for 10 years, and is currently riding around in a 35 kWh pure EV, and is working with a major utility and grid operator in a pilot program. The concept has been worked out. The concept has been demonstrated, and continues to be studied in order to get real-life experience with successively larger fleets.
But as I have said, don't sign up.
As for your bet, I don't want your money.
I think it is obvious that V2G will provide some storage functions too. That may work nicely in the case with school buses which you pointed out.
This does not change that the 191mln.vehicles x 10kwh calculation was mildly said misleading. There will be real-world constraints on this number, with discharging limits depending on a number of factors. Even if we accept that the the on-demand user culture will have to change, I think the personal car will be the least useful part of V2G. Hopefully financial incentatives could address that somewhat.
I am not an expert in this, but the presentation quotes 10-20kw per vehicle. Is this viable? It seems to me typical households and neighborhood clusters are not calculated for charges like that... wouldn't it require some rewiring?
I am perfectly fine with all you've said, but I wasn't misleading, I simply pointed out order of magnitude values of certain quantities. Real-world constraints will of course limit what can be accomplished, but that is a discovery process that is underway.
The eBox in Willett's talk provides 120 kW. The U Delaware car gets taken out on 100+ mile trips around the Delaware valley with typically 30-40% of battery power left over before any recharging. Has cruised on I95 between Washington and Wilmington at 70+ mph without a problem.
As you know even better in local and stop and go traffic. EVs are amazing efficient due to additional regenerative braking for recharging the battery.
I just had a DOH! moment with all those things around V2G.
Tesla Roadster has a 53kWh battery pack. They claim it will last for 100,000 miles. They also assume energy efficiency of 110Wh/km or 177Wh/mile. So throughout the battery life they expect:
100,000 x 177Wh/mile = 17,700 kWh could be recycled though the battery.
Their battery pack consists of 6,800 18650 Li-Ion cells, currently selling for about $2.50/piece, wholesale. This is $17,000 just for the Li-Ion cells, and I'm assuming everything else could be reused after the cell is degraded (which is a very weak assumption).
So to cover only the degradation of my car battery, the utility will have to pay me:
17,000 / 17,700 = $0.96 / kWh!
Why in the world would they want to do it if the wholesale price of peaking power is more like $0.05 c/kWh? This is 19 times as expensive! And $0.96/kwth is just the beginning - we did not count the infrastructure and the original generation costs inside yet.
So, in order V2G to work we would need:
1) Batteries that don't degrade (ultracapacitors?) - the jury is still out on whether we'll see those
2) Either breakthrough on chemical battery life or on battery cost or a combination thereof. But what is the chance we could see breakthroughs that lower the cost 20 times!
The only thing I saw in the V2G papers about battery degradation is it will be a subject of later research. Isn't this way to convenient?
Please correct me if I'm missing something. If I don't then I'll consider this discussion to be over.
$0.96 cents / kWh!
I wouldn't settle for less than a $1/kWh!
See my links and comments to Robert below, but in summary:
You're making money by just being plugged into the grid, providing spinning reserves. For regulation services, you can make $2500/yr.
On the battery lifetime, altairnano's specs claims a 15,000 deep-cycle lifetime, 41 years at one cycle per day. V2G will mostly be a small fraction of full cycle charge/discharge, so you might cut that lifetime by half.
Any reduced battery life is balanced against any revenue stream for providing services to the grid.
You're making money by just being plugged into the grid, providing spinning reserves.
"Spinning reserve" is the ability to produce energy on demand - and the bottom line is that you have to be able to produce it when requested at the market price.
Actually the way it works is at the time power is requested the loads are bid up with the lowest cost marginally produced power engaged first. In such environment V2G will never be used! The highest marginal cost peaking electricity is Natural Gas - at some $0.10/kwth it is 10 times less than batteries.
The only way what you suggest to work is to mandate utilities to NOT maintain enough lower cost spinning reserve thus creating artificial shortage in the market! Aren't you stretching this a little bit? How do you expect consumers will react to a $1/kwth price on their bills? Utilities will not pay for standby they will NEVER EVER use. They will build up the lower cost peaking generation until there is a glut of it - which is the case everywhere in the developed world. They will simply choose to ignore anybody who tries to sell them 20 times more expensive electricity. Or do you suggest the government mandates them to accept the bitter pill?
I agree that if it delivers, Altairnano's battery may address the cost issue to some extent. Assuming it reaches the same cost as Li-Ion and it has 10x times the battery life, the cost of the power (from degradation only) would be $0.10/kwth. Add infrastructure and premium costs and it would go to $0.15/kwth - closer to competitiveness but still remains to be seen. Just like with Eestor the question remains open.
Even then the utilities will prefer buying Altarirnano/Eestor batteries themselves. Why all the trouble of building V2G infrastructure and paying premiums if they can be in a full control and take all the benefits for themselves? Are you going to force them not to do it? Moreover the goal of accommodating renewables will be more easily reached this way.
I think you should abandon the idea at this point of time, it's getting way too funky.
Tesla uses conventional li-ion batteries. At about $400/kwh, and perhaps 500 cycles, the cost per discharge is about $.80, far too high for utility storage. This is generally understood - no one would suggest using conventional li-ion batteries for utility storage.
First, you have to realize that V2G isn't the most important use of vehicle batteries for utility load leveling. Instead, the the most important use of vehicle batteries for utility load leveling is dynamically scheduled charging, which will make a dramatic difference.
V2G won't be needed for large-scale utility load-leveling until wind & solar reach more than 20% each of market share - that won't be for at least 10 years, and that would be under a crisis mode installation program.
2nd, 500 cycles is conservative for conventional li-ion in a vehicle with sophisticated charge & temp management.
3rd newer li-ion batteries, such as A123systems, or Altair, have much more than 10x the cycle life of conventional li-ion.
4th, li-ion costs will continue to drop by 7-10% per year - it's pure economy of scale & manufacturing experience.
By the time V2G is needed, it will be cost-effective.
Guys, this is getting surrealistic. V2G will NEVER be competitive, because there is an inherent much better deal - do it yourself battery to grid. The utilities have a century long experience in doing it and so far I don't see them complaining.
No matter how battery technology evolves, it will always be more profitable for the utility to buy it's own batteries and do it itself instead of building and enormous infrastructure for effectively renting mobile batteries.
In addition utility scale batteries will always have different requirements than vehicle batteries. Utility scale batteries would also benefit from economies of scale. BTW why is nobody suggesting renting our cell phone batteries? If you add them up they will form enormous unutilized capacity.
If the utility pays you for your battery it will have to cover the following:
- battery degradation
- V2G infrastructure, maintenance and profit margin
- premium to the car owners to cover his cost of frequently buying new batteries and making him interested in the whole schema
If the utility byus its own batteries it will have to pay for:
- battery degradation (depreciation)
- nothing else
The first option will ALWAYS be much more expensive. Something plus something is always more than something plus zero. I would expect V2G to put at least a $0.10/kwth charge over the battery degradation cost. It is ridiculous to think they would ever consider such a deal.
There is a caveat: if ultra capacitors come to life and degradation costs get close to zero then the game becomes different. In this case the V2G infrastructure + car owner premium will have to compete with the cost of capital for buying the ultracapacitor over its lifetime. I would not hold my breath though - complex schemas like V2G are never cheap. It will be many years of investing billions upfront before the costs are brought down to competitive level... in a competitive market nobody would do it if they have an easy, quickly deployable and scalable alternative at hand.
Have a good evening and sorry for spoiling this party.
I think E-P has answered most of your concerns, but I'm not sure you really absorbed what I or he was saying, so I'll try again:
1) Tesla's batteries are far more expensive than other chemistries, because Tesla wanted the maximum energy density, which the older, conventional li-ion's provide. Firefly or A123systems batteries would be far, far cheaper than the $1/KWH that you're using, more on the order of 10 cents per KWH.
2) The most important thing is not V2G, which is energy flowing from the car. The most important thing is utility managed charging, which will buffer wind & solar. That will suffice for at least 10 years. By that point the infrastructure for utility managed charging will have been in place for years - the utilities (PG&E, etc), car companies (Tesla, GM) and software companies are already planning for this to be in place when the cars are sold. That infrastructure will seamlessly handle V2G.
3) A123systems batteries, which appear likely to win the Volt contract, have sufficiently long cycle life that effectively there is no degradation cost to the car owner to reselling energy back to the utility.
4) vehicle owners will pay for batteries for their transportation utility, and utilities won't have to pay the full cost of the battery.
By the time V2G is needed, if it ever is, the batteries & infrastructure will be ready.
OTOH, V2G may not be needed. Geographical diversity, long distance transmission, PHEV dynamic charging, pumped storage, flow batteries, Firefly lead-acid in utility scale installations....all of these may do the job. I suspect it will have an important role at least for the small-scale services that have been discussed, but we'll see.
"Spinning reserve" is the ability to produce energy on demand - and the bottom line is that you have to be able to produce it when requested at the market price.
Actually the way it works is at the time power is requested the loads are bid up with the lowest cost marginally produced power engaged first. In such environment V2G will never be used! The highest marginal cost peaking electricity is Natural Gas - at some $0.10/kwth it is 10 times less than batteries.
Actually you are wrong. We're not addressing peak demand. Spinning reserves is a reserve when power is required quickly, within minutes, from 'spinning' generators and ready to go at a moments notice. It is paid for by (1)the amount of time it's reserve is ready to go and available, and (2) the $/kWh of actual energy delivered. Ignore (2) as insignificant for now, let's focus on (1) revenue for being 'available'. A 100 kW battery plugged in for an hour provides 0.1 MW-hour of reserve. Agregated in a fleet of ten cars provides 1 MW - hour. No power need be exchanged in this service unless the ISO requests it. PJM (the ISO or grid operator in the mid-atlantic region) pays for this reserve $14 per MW - hr. Stay plugged in 10 hrs, that's $14/car per day. Multiply by days available per year for annual revenue. How often is spinning reserves called on? In the PJM service area, all of 21 hours for all of 2005 (pg 36 from Ref).
And then there are regulation services, the revenue of which was discussed by me elsewhere in this thread and comes from this work.
I think you should abandon the idea at this point of time, it's getting way too funky.
Not for me.
You insist on looking at it from the POV of V2G vehicle owner. You imagine getting a fixed stream of income basically for having a battery.
Guess what? It won't happen. Anyone with cache on hand can buy a battery and in the case of utilities they have to be crazy to pay all those premiums to you, not keep them for themselves.
Like I explained - you can not make them accept V2G if its cost per kwth is higher than what they already can get from existing spinning reserve. Since installed spinning reserve is more than enough, you are basically suggesting that they will be paying for an insurance for event that will never happen. It's like buying a life insurance for dead person.
The utility can't justify the capital cost just to have spinning reserve. But if you're paying for the battery via the difference between the cost of energy from electricity and the cost of energy from gasoline, the extra revenue from services such as regulation and spinning reserve is worth the expense in control systems and the minor impact on durability.
(If that impact exists at all; recall that AC Propulsion's regulation test caused the measured capacity of the Panasonic lead-acid battery pack to increase.)
I don't think you even bother to read my posts.
Go for it man, I'm with you.
It's probably because you don't seem to get the whole spinning reserves, regulation, peak load thing.
If you do, you're not conveying it in a way that this argument/debate can move forward.
If you think it's not economic for utilities to invest in V2G, great. Repetition on your part doesn't improve your message.
And don't worry 'bout your tax dollars going for v2g. They're being well-spent wasting American and Iraqi lives as we quibble.
When you fill your posts with multiple egregious errors of fact, you are not going to get the kind of answer you want. You are going to have to straighten both your facts and your reasoning out first, THEN you will get the kind of response that you desire.
After some digging into it I admit you two are right and I don't/didn't quite understand the nature of the peak load, spinning reserve and regulation services. I'm sorry.
Now I'm back to square one: what would be the cost/benefit analysis of V2G? It looks the primary service V2G could provide is regulation, it should be less competitive as spinning reserve or peak load. For these two I'd also think discharge limitations will limit the actual resource base - I am at loss what happens after the storage is drained if peak load/spinning reserve mode is engaged - aren't those contracts supposed to be for continuous power? I need to do some more digging but I would suspect that the V2G contracts would be depending both on MWs and the GWh-s available.
http://findarticles.com/p/articles/mi_hb5050/is_200504/ai_n18342790
$360mln. or even $500mln. yearly is too thin of a market IMO. I think it will all depend on the cost of implementing the V2G system itself vs the expected revenue. Solar and especially wind may increase this market, but this would increase their overall costs too.
I don't really think more wind and solar plus V2G will be enough to retire significant number of existing units. Obviously spinning reserve is avoided for coal power plants as much as possible, so I don't think many of those are or will be kept running just because of that. If I understand correctly ancillary services don't reduce the need for base load power.
P.S. Somehow you forget that the $0.03-0.05/kwth electricity spinning reserve generators provide ALREADY includes its capital costs inside. And a generator may work 60 years. How long on-grid battery would last? 2 years?
Not quite. It's the ability to make up the difference between immediate supply and immediate demand on command (response time of a few seconds). This has historically been done with generators on-line but idling, but it can be done just as easily and far more efficiently by varying a large controllable load. In the case of (PH)EVs, the available spinning reserve is equal to the full charging load (and that's without making demands on the batteries to feed back to the grid).
Spinning reserve has its own market price, quite different from e.g. a MW of base load generation. If you have 200 million vehicles connected to the grid at a minimum of 6.6 kW peak charging load apiece (220 V 30 A connection; some will have more), that's 1.32 TW of load available to manage the grid. If it's averaging 1 kW per vehicle, you've got 200 GW (roughly 45% of today's average generation in the US) to play with as spinning reserve.
You can do considerably more than that, of course. Spinning reserve is there to make up for a large generator going off-line. V2G would allow back-feeding the grid for a few critical minutes while slower-reacting plants were brought up to take up the slack (or other demand was taken off-line). It has to be there, but it gets used very infrequently. In this way, V2G would supply "spinning reserve" at zero cost in fuel, minimal cost in equipment and a very low cost in battery life.
Historically the generators you find obsolate now were providing spinning reserve at the cost of $0.02-0.05/kwth.
Your genial innovation is going to provide them with spinning reserve for $1/kwth. OK lets allow innovations etc. to bring this down to say $0.20/kwth. Just perfect!
Just to mix in the conversation a little is the Flow battery. I dont think I've seen anyone mention it in this long thread. http://www.vrbpower.com/
We are putting one of these in for UPS of a Telco site I am in the middle of engineering. It will be 100Kw.
I think these may be more what we will end up seeing at the power utility level as more wind and solar projects come on-line. They are already using them with ff electric plants now at the end of long transmission line runs. Charging at night and using them for peaking during the day.
I dont have the numbers in front of me, but I recall it being 2x cheaper than lead-acid for mass storage and battery life. High up-front costs but last longer than Lead acid that we use in DC plants of telcos.
Thoughts on NaS batteries? AEP recently prchased some MWs, and IIRC, price was $4-4.5 per Watt.
Snipped press release from AEP below. Sounds like they're also interested in flow.
What's most interesting is they want to get 1 GW of storage.
“We’re first movers on advanced storage among U.S. utilities, a position we’ve held on a wide number of technologies in our century of existence,” Morris said. “Our near-term goal is to have at least 25 megawatts of NAS battery capacity in place by the end of this decade. But this is just a start. Our longer-term goal is to add another 1,000 megawatts of advanced storage technology to our system in the next decade. We will look at the full spectrum of technologies – flow batteries, pumped hydro, plug-in hybrid vehicles and various other technologies in early stages of development today – to determine their feasibility and potential for commercial application.”