Felix Salmon

The enormous promise of vehicle-to-grid technology

By Felix Salmon
November 1, 2011

Dan Ferber’s 3,500-word article on Vehicle-to-Grid is far too long for you to read, especially when Greece is busy imploding, but it’s a very important idea. So let me give you the shorter version, starting with four facts about the energy industry.

  • The 146 million cars, SUVs, and pickup trucks in America, between them, produce seven times the power of all US power plants combined.
  • The supply of energy is volatile, and will get more so as we move to renewables like wind and solar. Those sources only produce energy some of the time.
  • The demand for energy is also volatile, going up during the day and when it’s hot outside.
  • Storing energy, by doing things like pumping water uphills into reservoirs, is expensive and cumbersome. And those energy sources can’t provide the small bumps in power needed to ensure that AC electricity is running at 60 hertz at all times.

All of which opens up an amazing opportunity for owners of electric vehicles — be they electric, hybrid, or fuel cell. Those vehicle owners can basically become baby energy traders, fueling up their cars at night, when electricity is cheap, or at the pump. And then plugging their cars into the grid, where they can sell energy back to the grid for much more than they paid for it.

Willett Kempton of the University of Delaware has already set up his electric Scion to do just that; it’s been earning him $300 a month since 2009.

This is a fantastic idea, and it’s a no-brainer, really, that all electric cars should have the ability to power the grid, rather than just drawing power from it. The number and size of power plants is a function of peak electricity demand; if electric-car owners collectively can help meet peak demand, then that means we need fewer power plants. And, the revenue from selling that electricity would help offset the extra cost of buying an electric car in the first place.

The batteries in electric cars are expensive and valuable pieces of technology which go unused for most of the time. Let’s put those things to use, and make money doing so! The only real question is why this isn’t happening already.

20 comments so far | RSS Comments RSS

One of the reasons is that it can significantly reduce the life span of the car’s (very expensive) battery pack. The other is that our electricity industry is run by very conservative, minimally competent regulators who often just do what incumbent firms want. And the incumbents don’t want distributed energy of any type, as it’s just more competition for them.

Posted by SohoD | Report as abusive

The main reason people can make money buying power at night and selling it back during the day is because there is not as much demand for electricity at night. So if everybody has an electric car or giant battery and starts buying electricity at night, the price will go up and there will be no reason to sell it back. Plus I’m guessing some of those electric cars will be driven during the day, so they won’t be able to sell the electricity back to the utility.

I have an electric car, and even if I could connect it to the grid during the day, I wouldn’t, as it would reduce the range and defeat the whole purpose of having the car in the first place (to take me places). Rather than buying a $10,000 battery with an expensive car ($20-100K) attached to it (that you can’t use if it’s being used for energy arbitrage), people can just buy the $10,000 batteries and do the same thing. If you can get $300/mo for storing electricity in a $10K system, it would be a great investment, but my guess is the battery wouldn’t last long enough to earn a significant profit.

But it would be great to have all those electric cars on the road.

Posted by KenG_CA | Report as abusive

Batteries, and energy storage generally, have been the laggard of the industrial revolution. They seem to be catching up slowly, and at some point could be truly revolutionary. As electric cars become more common, charging will need to be channeled to off-peak hours, and that alone will make the grid more efficient. Good things will happen if we can some day park battery packs in homes — or cars — that can stockpile electricity more effectively than is feasible today.

Posted by kenjd | Report as abusive

SohoD is right. Batteries wear out. It wouldn’t surprise me to find out it is uneconomic to use batteries to power the grid even if could charge them for free. You can pay $300 for a 100Ahr 12V deep-cycle lead-acid battery. That stores around $1 of electricity and you get about 300 charge-cycles.

Posted by Mr.Do | Report as abusive

In most states, regulations prevent utilities from charging residential customers different prices for electricity at different times of day, so the financial benefits or charging off-peak wouldn’t be be captured by users.

Posted by PandR | Report as abusive

There will be a cottage industry in this, just not today. In ten years time, enterprising businesses will buy up used car batteries (still holding perhaps half of their original charge, but not enough to power a car) and will eke out a few more years of life in this scheme.

But burning expensive oil to produce cheap electricity? Even the power companies don’t do that any more, and their equipment is more efficient than yours.

Posted by TFF | Report as abusive

I’ll pose this as a question for someone who knows more about the electrical grid than me, but it would seem like there’d be issues managing grid stability if electric cars were a meaningful amount of electricity supply. For example, I’d think a high percentage of electric cars would unplug from the grid at the same time – morning rush hour, lunch, and evening rush hour. At a minimum, there’d still be the need to maintain standby capacity for a high percentage of the supply provided by electric vehicles, minimizing the savings of this idea.

Posted by realist50 | Report as abusive

What KenG said- the more arbitrageurs playing this game the less profit they will receive- they will increase demand at night and increase supply during the day- which will correspondingly shrink their margins.

Posted by AdamJ23 | Report as abusive

I think I’m missing something here.

I don’t understand the relevance of the first point. That’s supplied overwhelmingly from petrol so surely just shows how inadequate the power stations would be if large numbers of cars connected to the grid. By the time people charged the car for what they actually needed (which would surely be used in the day, when the energy grid most needs the energy people would be least likely to provide it) it would take significant amounts to then store some to sell back.

Also, it strikes me as fairly inefficient to use gas/coal plants to charge a car battery in the night just so the car can sell it to the grid in the day. I don’t know how much, but there must be a decent amount of lost energy there. Considering how rarely power stations are flat out it seems easier to let them just provide all the energy in the first place?

Posted by ABT | Report as abusive

How on earth is he making $300/month doing this? Let’s make two assumptions:

1) This car battery is capable of delivering 30,000 watts of power to the grid.
2) This car battery can do so for one hour per day before needing to be recharged.

You now have a 30 kw mini power plant. If peak electric prices for the hour in question are $300/MWh (a very high price), then our 30 kwh of power will sell for $9 per day ($300/1000*30). If you do this 30 days per month, you will gross $270. Unfortunately, the real-time price of electricity for PJM Eastern Hub exceeded $300/hour a mere 11 times in July and August, and on only 37 different days did the real-time price of electricity exceed $100/hour.

This, of course, says nothing about the $0.10/kwh you will pay for charging the vehicle. You may notice that $0.10/kwh at the retail level is equivalent to $100/mwh at the wholesale level.

Posted by MitchW | Report as abusive

SohoD and Mr.Do make the crucial points. Current EV battery manufactures are struggling to achieve a 10 year useful life on these $10k batteries. Giving power back to the grid is going to cut that life significantly, which makes achieving anything close to payback even harder.

Posted by MickieB | Report as abusive

Ken G@ 2:52pm wrote:

“But it would be great to have all those electric cars on the road”

CORRECTION: coal powered cars. So called ‘electric cars’ are a myth.

Posted by crocodilechuck | Report as abusive

Felix, thanks for the shout-out to this story of mine in Miller-McCune magazine. I thought it was a cool idea, too. And it’s beyond the idea stage now: I describe in the story how it’s been piloted at the University of Delaware, complete with payments from the region’s grid operator, PJM Interconnection, and how at least one business is already investing significantly in vehicle-to-grid.

Good comments, too. I wanted to respond to a couple of them. To SohoD,Mr Do and MickieB: The effects on a car’s battery depend on how the battery will be used. If the battery were to be drawn down significantly on a regular basis, that would cut into the life of an expensive battery, and you would be right. But that’s not how V2G is being implemented right now. The folks commercializing V2G are plugging into the frequency regulation market, not the bulk power market. Frequency regulation requires only very slight amounts of electricity drawn from (or added to) the battery–not enough to alter the life of the battery, according to experts I spoke with. However, it has to be done within seconds, and automatically, when the car receives an electronic signal from the grid operator. This service supplies the power needed to keep the electricity in our outlets at 60.0 Hz (no more, no less), which is necessary to keep all our clocks running on time and our electrical equipment from malfunctioning, and is required by law. The hardware and software to use car batteries to do this have already been developed at the University of Delaware, as I described in the story.

Because frequency regulation is technically difficult to pull off, it’s worth much more than bulk power. So, MitchW, you’re looking at a different, and less lucrative, market than the one they’re actually tapping into. PJM is not paying the University of Delaware for bulk electricity. They’re paying them to have their electric cars ready to provide frequency regulation. This is different from what the average consumer or business customer deals with, and pays a much higher price–enough for a single car to pull in $10 a day ($300 a month), as the U Delaware cars did.

Which brings me to KenG_CA’s point that “I have an electric car, and even if I could connect it to the grid during the day, I wouldn’t, as it would reduce the range and defeat the whole purpose of having the car in the first place (to take me places).” Luckily, even if everyone felt like KenG (and most people do) it wouldn’t make a lick of difference for the viability of vehicle-to-grid. That’s because even at the busiest time of day, evening rush hour, research shows (counterintuitively) that about 90 percent of cars still sit parked. So if you had 1000 cars ready for vehicle to grid, every driver could go where they want, when they want, and V2G would still work. All this and more is explained in the Miller-McCune story itself.

Posted by DanFerber | Report as abusive

Chuck, not all electricity is generated by coal. That’s only true mostly in the states that are committed to supporting incumbent parasitic power companies, and there’s no reason it can’t change (other than people like tearing the tops off mountains and dumping it in their rivers, lakes, and streams).

Posted by KenG_CA | Report as abusive

@DanFerber said “This service supplies the power needed to keep the electricity in our outlets at 60.0 Hz (no more, no less), which is necessary to keep all our clocks running on time” This quoted statement is a bit extreme (almost fear mongering) and opens up this important topic to criticism. Direct synchronous clocks are rare and only becoming more rare. most clocks use a quartz crystal oscillator to keep time. More and more systems are using internet time servers, cell networks, or gps satellites for time syncing. there was some talk a while back of dropping the requirement of averaging 60Hz over a 24 hour period, which was intended to keep grid synced clocks in sync. I’m not sure if this proposed deregulation ever occurred, but if it did, then grid synced clocks have already been affected.

Posted by patentguy | Report as abusive

@patentguy. Fair enough. I was speaking loosely, and there are other kinds of clocks besides electrical ones. But the reason why the frequency regulation market exists is because straying even a little away from 60 Hz can mess up all sorts of electric-powered equipment, including industrial equipment. The folks at PJM told me that grid operators are required by FERC to maintain frequency very close to 60 Hz. Maybe someone from the utility industry could provide more detail. See:

http://green.blogs.nytimes.com/tag/frequ ency-regulation/

http://bit.ly/v4lgYf (This link also has some background about the frequency regulation market.)

And here’s a piece on proposed changes in FERC rules so that regulation sources that respond more quickly, like flywheels or batteries, get paid more on the frequency regulation market. (Frequency regulation is also sometimes just called “regulation.”)


Posted by DanFerber | Report as abusive

DanFerber and patentguy, the frequency of electricity in the US must be within 0.1 HZ of 60 HZ in the US, but that isn’t a problem. There’s no reason cars or batteries that are connected to the grid cannot maintain that frequency; solar systems and wind turbines for the home have no trouble doing so.

Dan, the problem with connecting cars to the grid is not that they will always need to be driven (“even at the busiest time of day, evening rush hour, research shows (counterintuitively) that about 90 percent of cars still sit parked”), but rather people are going to be reluctant to give their stored energy back to the grid while they’re parked at work. Maybe when the range of most electric vehicles gets extended to > 300 miles, people will feel more confident giving up some of their charge during the day, but that will also require more complex charging stations in parking lots – if I’m going to pay for electricity at night to charge my car, I’ll want to get credit for what I give back at work during the day. It’s not a simple thing to implement, and until there are enough cars with enough range, it won’t be worth creating and deploying the technology to do that.

Posted by KenG_CA | Report as abusive

@DanFerber, if the main vehicle-to-grid interaction is frequency regulation, rather than bulk power transfer, then it’s unlikely to lead to the type of load-shifting and peak-shaving that Felix suggests, correct?

Also, I’m not sure how large the frequency regulation market is – would prices crash if supply (i.e. vehicles available to regulate frequency) multiplied by orders of magnitude?

Posted by PandR | Report as abusive

For those of you that buy into the “EVs just shift pollution from the road to a power plant” argument, consider the beauty of solar photovoltaics (PV) coupled with EVs. I’ve got 5kW of PV on my residence, and it has paid for itself already in the 10 years I’ve had it. I now generate enough power for two EVs and my house, and drive *truly* clean and free of fuel costs.

I’d like to see you make transportation fuel on YOUR roof.

Posted by gregbrew56 | Report as abusive

Selling back the electricity seems like a really good idea, but there isn’t enough electric car drivers to capitalise on that. However, when more and more electric car owners use this method to sell back electricity, electricity costs at night would rise throughout the board as well. In the future, costs at night would probably be as expensive as in the day, and this diminishes the profit one can earn.
Peter – http://www.pmwltd.co.uk/

Posted by Peter_Mould | Report as abusive

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/