Battelle Ventures and Affiliate Invest $1.9M in Traffic Management Startup
Global NGV Fleet at Almost 7.3 Million

Xcel Energy Announces Six-Month Test of V2G and Plug-In Hybrid Electric Vehicles

Six plug-in hybrid electric vehicles (PHEVs) will be on the road by the end of 2007 as part of a demonstration test of vehicle-to-grid (V2G) technology by Xcel Energy. The project, which will convert six Ford Escape Hybrids to PHEVs equipped with V2G technology so each can charge and discharge power to and from the grid, is one of the nation’s first real-world demonstrations of the V2G technology.

With operations in eight states, Xcel Energy will study how the vehicles perform in varied geographic regions and climates over a six-month period. Three company employees will serve as test drivers, using three of the PHEVs in typical home settings. The remaining three PHEVs will be used in the company’s fleet.

With every US home connected to the electricity grid, vehicle-to-grid technology could be key to meeting our growing energy needs. This project will allow us to explore how PHEVs can become an integrated part of a ‘smart house’ and our vision of the smart grid energy system of the future—one that allows customers and utilities to work together to balance the power grid, lower greenhouse gas emissions and improve our nation’s energy security.

—Michael Lamb, executive director of Xcel Energy Utility Innovations

Xcel Energy’s demonstration will build on its previous PHEV impact study by examining how drivers—and their vehicles—will react and perform in real-world settings. The project will explore the potential benefits of widespread PHEV use including: reducing petroleum-related emissions and greenhouse gases, enhancing energy security by reducing dependence on foreign oil, improving the reliability and cost-effectiveness of the electricity grid, exploring ways to make PHEVs more accessible and lowering vehicle fuel costs.

The project is a joint collaboration between Xcel Energy; Hybrids Plus Inc. in Boulder, Colo.; V2Green Inc. in Seattle, Wash.; and the US Department of Energy’s National Renewable Energy Laboratory in Golden, Colo.

Hybrids Plus (earlier post) will replace the cars’ nickel-metal hydride batteries with a 12 kWh lithium-ion phosphate battery pack using 26650 cells from A123Systems for the base conversion to a plug-in hybrid.

To make the cars V2G capable, each will be equipped with a V2Green Connectivity Module that controls vehicle charging, collects data and communicates via a cellular modem (earlier post); and an Inverger (a 6-kilowatt inverter and charger in a single unit) from Hybrids Plus. V2Green will also supply server software enabling remote control of smart charging and V2G functions.

By outfitting the vehicles with these components, Xcel Energy can remotely control the battery cycles in each vehicle by requesting that each postpones charging or begins discharging energy back to the electricity grid.




I agree with plug in hybrids - I can see the advantage of shifting some of the power generation towards a point source emission that can be monitored, regulated, and controlled. I would caution people to double check their "Fuel Efficiency" numbers as these can be misleading.


Is there any reason to do Vehicle to Grid technology with a normal IC engine? It only makes sense to me if the engine in the car is inherently more efficient and pollutes less than the power plant (or weighted average of the power plants) that provides power to the Grid.

Am I missing something here?

Bob Bastard

TR, the majority of energy supplied to the grid by the vehicle's batteries would probably not have come from the vehicle's ICE, but rather from the grid itself. The idea is to use plugged in vehicle batteries to allow for more capacity during peak demand, rather than bringing extra power plants online.
The analogy I would use is the idea of adding capacitors to a circuit to cover transient power demands, instead of simply beefing up the power supply to cover the maximum demand that could ever occur.
Evidently, investing the up front capital in order to purchase batteries or other energy storage devices for this purpose on a large scale has not been economically viable for utilities in the past. However, if large numbers of electric cars were already plugged into the grid, it only makes sense for utilities to try exploit such a large energy source for brief periods of peak demand, as an alternative to firing up an idle power plant.


Spot on Bob. I actually see the mass acceptance of PHEVs as being a huge boon to utility companies. Their idle capacity at night could charge a vast fleet of PHEVs. While these vehicles are parked around mid-day, some of the energy used to charge them can be drained to provide peak load.

I would like to see some sort of cost-benefit assessment from the utilities end. Perhaps they should foot part of the cost of the battery, maybe as much as 1/4?

GreenPlease, there are a couple of papers (including an economic analysis) that you may be interested in here:


Personally, I think the economic viability of V2G ranks right up there
with a perpetual motion machine. By far, the most expensive component
of an EV or PHEV will be the battery. This is particularly the case
when you consider that batteries wear out and will probably need to be
replaced at least once in the life cycle of an EV or PHEV, both of
which involve regular deep cycling of the battery. Deep cycling reduces battery life. V2G by definition
results in additional deep cycling of a very expensive vehicle battery
that has a limited life, thus reducing that limited life.

If the economics of storing off-peak energy in a battery made sense,
the utilities would all be doing so themselves. They aren't.
Instead, they would rather have their customers pay for this
uneconomic means of energy storage by wearing out their expensive HEV or PHEV
batteries that much sooner.


V2G by definition results in additional deep cycling of a very expensive vehicle battery that has a limited life, thus reducing that limited life.

Why would it need to deep cycle?


Greener: In order to avoid putting any real wear and tear on your battery or leaving you without sufficient charge, V2G will only use a very small percentage of each battery. Using a small percentage has virtually no effect on the life of the battery. The main purpose of V2G is to "smooth out the grid" rather than be used for massive storage.

Roger Pham

Following Greener's concern, let's consider two scenarios:

The use of BEV with A123 Lithium at 2000 discharge cycles, at a low projected future price at $500 USD/kwh (current price $1000-2000/kwh) . Over 2000 cycles, the cost per kwh is 500/2000= ~$0.25.

Whereas if the Prius is used as a plug-out generator at 20 kw capacity, with its engine/generator capable of 10,000 hrs of continuous use as a generator. This is not unreasonable, considering other dedicated piston generators capable of 50,000 hrs of use. A "barebone" or stripped Prius now retail for $21,000, so we can estimate that the engine + generator would not cost over $10,000. So, in 10,000 hrs x 20kw = 200,000 kwh for $10,000 worth, or $0.05 /kwh depreciating cost. Adding 1-2 more cents/hr for oil changes and maintenance periodically, and it is still far cheaper than BEV as V2G.

If adapted for NG at 1/2 the cost of gasoline, this HEV plug-out V2G is competitive with NG gas turbine power plants at 40% efficiency, since the Prius itself is capable of 37% tank-to-wheel efficiency.

The cost of turbine power plant is ~$500-700 /kw. The Prius power plant at under $10,000 for 20 kw is also under $500. BUT, the Prius is paid for by its owner, who will very unlikely in its life span, ever drive the car near its maximum mileage capacity. By V2G using HEV like the Prius, it will be a win-win situation for Prius owner to recoup some of the investment in the car by being able to sell peak-demand electricity back to the grid, AND to the utility company, thereby saving the utility company the upfront investment of a gas turbine peak-capacity plant, while being able to purchase power at competitive price ONLY when peak demand is needed.

So, in 10,000 hrs x 20kw = 200,000 kwh for $10,000 worth, or $0.05 /kwh depreciating cost. Adding 1-2 more cents/hr for oil changes and maintenance periodically, and it is still far cheaper than BEV as V2G.

How much oil need 20 kW generator per hour ?

I'm sure that oil will be much more than 1-2 cents/kwh, close to 100 cents/kwh.


Actually, it's probably not 100c/kwh. My wild guesstimate is around 25c/kwh with fuel.

A Prius will get about 45-50 mpg on the highway. If it cruises at 45-50 mph, then it is consuming about one gallon of fuel per hour. When running at an easy cruise, the engine is obviously not putting out all its rated power, but we can assume it is putting out up to 27 horsepower or so in order to overcome drag and rolling resistance. Perhaps a bit less, as the Prius has low drag.

(On the subject of engine output at highway cruise, see the fifth paragraph in:

If a cruising Prius is actually putting out 27 hp -- which is 20 kw, by the way -- and consumes one gallon per hour, and a gallon of gas costs $3, then you are getting 20 kwh for 300 cents, or 15c / kwh. If a cruising Prius actually puts out less power than 27 hp, then we would need to run the engine faster to sustain 20 kw output. But even if requirements were 33% higher, that would only be 20c / kwh in fuel. Add that to Roger's figures for the cost of buying and maintaining the engine, and you have no more than 25c/kwh or so in costs.

Now, 20c - 25c / kwh for electricity, delivered to your door, is not a particularly cheap price -- although it happens to be in the same ballpark as electricity in the part of Boston where I live. But consumer rates tend to reflect the average cost of service (cheap base-load coal plus expensive peaking natural gas), and the important things in V2G are the marginal cost of power at peak demand moments, and/or the marginal cost of dedicated smoothing and stabilization equipment if the grid gets choppy, as it does from time to time.

Concentrating only on the former, it is the case that in many parts of the country, peak power will cost more than 20c - 25c per kwh to generate and transmit to the local distributer's hub. Part of that cost is simply running an expensive peak-demand power plant, which might be less efficient or consume expensive fuel. Another part of that is recovering the capital cost of having a peak-demand power plant sit around idle most of the time. Yet another part of the cost is transmission, which grows more expensive as congestion goes up. Building more high-voltage power lines just to meet peak demand ties up capital just like building an extra power plant.

Point is, getting all your electricity from the generator on your Prius is probably not very cost effective, but using V2G to help cover peak demand can be.


I see the arguement but am not totally convinced.

Is this over simplifying the Grid? How is an ISO supposed to manage the signals to all the households? How is the ISO supposed to know how much capacity (in terms of cars or kW-hr) it has to use?

Wouldn't something like Beacon Power's Fly Wheels be better suited (in the near-term) for this type of Grid-smoothing regulation?


You are all on drugs or simply very ignorant. This V2G concept will never work.

First is the energy stored in a vehicle battery. Maybe it is from grid during off-peak, maybe residual from the last driving cycle. Anyway, as already pointed out, batteries are very expensive. Grid putting in and drawing out energy adds more battery cycles thus accelerating battery degradation. Check out discussion about high battery cost and how it is impacting that vehicle's design. The last thing any owenr wants to do with his expensive battery pack is put additional wear while not driving. At the very least I would expect the power company to pay a certain portion of the price for a new battery pack when I need to replace it.

Next, most cars are parked in a parking lot of an office building, not plugged in. During the peak hours, such as a hot summer day my car is not plugged in. By the time I return home and plug it in, say between 6 and 7 pm the peak demand is about to end. Plus I just drained the battery during my commute home from work. In fact, that battery will be full some time near morning when demand as the lowest. So odds are very good that my car is not plugged in and/or battery is empty when grid wants to pull electricty from it.

How do I tell the grid when it can pull electricity out? Say I did plug it in while at work because office garage does have outlets. It is a hot summer day, A/Cs are full blast on, so grid pull energy from my battery. Next I finish work, get into the car to drive home ... ooops battery is empty!!!!! If it is a hybrid it can still turn on ICE, sure. But WTF, I bought a hybrid to drive more on electric drive and less on ICE. Otherwise I wouldn't bother with a hybrid drive. So if my battery werel always drained because grid keeps emptying it at the time when I need battery energy then I would be very angry.

Thus grid pulling any sort of energy from my car's battery during day time hours is just not going to happen. I need that energy myself.

Then you talked about ICE acting as a generator. Again this is crazy. I don't want the grid starting my ICE while car is parked in the garage and filling garage with fumes. I certainly don't want to discover an empty fuel tank in the morning.

Heck, even if vechilce were parked outside say office building parking lot, still we don't want a bunch of cars generating CO2 in the middle of a city. The point of generating stations that are far aways from cities is to move pollution away from cities. There is plenty of pollution just from driving. We don't want more pollution in cities from generating electricity. On the contrary, we should be using solar panels, wind turbines, etc.

So again, we are not going to use any car ICE for generating energy. It is insane. In fact, the entire concept of many small distributed power sources like these cars sending energy into the grid only to come back to us is insane. Instead, we shall use our cars to power our own homes when grid fails. There is no need to send power away only to return back. We generate our own power for our own needs at minimum travel distance.

That may be solar panels and wind turbines for normal operation, as well as a car for emergencies. In any case, the grid serves only one purpose well. It takes gobs of energy from large sources (generators) to smaller consumers (office buildings, houses, etc.) It only makes sense to send energy backwards from homes when it is very cheap and plentiful. Such us from solar panels which would be wasted otherwise.

Roger Pham

NG-powered Prius will bring the fuel cost/kwh to half that of using gasoline.
Note that since the Prius' efficiency is comparable to that of a gas turbine, if NG can be purchased at the same bulk price as the utility company can, there will be hardly any fuel cost differential as the result of V2G via NG-powered HEV.

If this is seriously contemplated, perhaps a means of communication to the utility network via a radio transceiver would be necessary. Or, perhaps an internet network signal going over the power circuitry line may do it for communication.


I once attended a lecture given by Prof. Andy Frank (bottom right) who is considered the "father" of the PHEV. At no time in the lecture did he ever suggest that V2G be used for electrical generation. His whole point was to allow the battery storage of the V2G EVs to allow for the use of less oil (and NG). Using the cars as generators would be counter productive to the goal of independence from petroleum. The goal is to take advantage of a small portion of a great many batteries to even out the demand side of the equation and thus make better use of the supply capacity.

Roger: on several occasions I've seen you amortize the cost of an EV battery based on deep cycle lifespan. This would be by far a worst case scenario. The wear and tear on a battery is geometrically proportional to the depth of discharge. You can get millions of 1% discharges out of the same battery that you could only deep discharge 2000 times.

Roger Pham

Welcome to the 21st century, when micro-computer and power electronics seems to be able to do the hitherto impossible. It can allow for very good synchronization with the grid frequency. A dedicated V2G car will be connected to the grid during predicted peak power demand somehow. A dedicated parking spot with electrical socket will have to be provided.

Each Prius can generate 20 kw, so there must be more than enough V2G generating capacity than the highest peak demand in order for the scheme to work.

Admittedly, the scheme is harebrained. But, look at the summer black out in California some years ago, and don't you just wonder if there is a quick way to generate quick peak power capacity without major investment by the utility company?

Roger Pham

Do you have data on the relationship between what percentage of SOC vs frequency in order to max out a battery kwh lifetime potential?

My cell phone instruction stated that since the battery can be recharged up to~300 times, so not to recharge it until the battery is low in order to max out the battery's wh potential in its lifespan.


Technically it may be possible and even feasable. I didn't dispute technical feasability. I am a programmer, so know a few things about computers.

Of course, utilities would love to save money by using consumers' cars. You still didn't explain why would consumers want to offer their cars for utilities to use.

If there is a blackout such as in California or the big one couple of year ago in the north east, do you think that I would offer my Prius to generate electricity for the grid while I sit in the dark?????

I read an article about some man who plugged in Prius into his house during a hurricane when electricity was out. Of course, everyone will use their own Prius for their own power needs BEFORE offering energy to grid.

The point is that all/most of the benefit is for utilities and grid, while consumers see little in return. Thus consumer WILL NOT WANT to offer their Prius cars for this purpose.


Prius uses 40% to 80% SOC range as maximizing NiMH chemistry lifespan. This is the common knowledge that NiMH don't like deep discharge nor full charge. Hence Prius' miniscule battery pack capacity is further reduced due to this limitation of being able to use only half of that capacity.

Lithium chemistry might tolerate deeper discharge, so probably handles a wider range. However they generally handle well over 300 cycles.

Also keep in mind that car batteries are designed to be much more robust then phone batteries.


Roger: When I attended Prof. Franks lecture I spent much of my time thinking that the last thing I wanted was someone else using up my precious battery cycles. So after the lecture I got a chance to talk to him directly. When I queried him about this concern he answered me to the effect that they wouldn't be cycling the battery deep enough to effect its lifespan. I'll look for a more formal link but for now I found the following quote from the battery university (the site looks a little out of date)

"A battery often receives many short discharges with subsequent recharges. With the smart battery, these cycles do not count because they stress the battery very little. On satellites, the depth-of-discharge is only about 10%. Such minute discharge cycles put the least amount of stress on the batteries in space. With shallow discharges, however, nickel-based batteries require a periodic deep discharge to eliminate memory.

Lithium and lead-based batteries do not require a periodic full discharge. In fact, it is better not to discharge them too deeply but charge them more often. Using a larger battery is one way to reduce the stress on a battery. "

Roger Pham

Let's imagine in the future, HEV market penetration will be 1/10 or higher, and imagine further that HEV's will be offered with the V2G option, paid for by the utility company, not all at once, but monthly, based on regular participation. The V2G car owner will be give credit or be paid monthly for each kwh supplied to the grid during peak demand.
Over the life span of the car, this payment may be as much as 5,000 to 10,000 dollars, thus more than the hybrid option for the car. The V2G option is already free.

Every workplace will, by local regulation, offer a certain number of V2G sockets on certain dedicated parking slot, similar to the requirement for handicap parking spots, and people with V2G-capable car will sign up for these preferential parking slot, which is located closer and has a roof over it for economic reason, thus protecting your car from damaging UV light, rain and hail. It's kinda like a VIP parking. Look at all the incentives for choosing a V2G option.

If each HEV can put out on average 10 kw power, it will take only 100 cars to generate 1 Mw of capacity. A few thousand cars, which is only 1-2% of cars in a small to medium size municipality, and you will way more than cover peak demand for that city.


So where is the business case for utilities paying HEV owners $5K to $10K over 10-20 years? That would be great as it would pay for several battery pack replacements. However, there is no information pointing to utilties wanting to pay so much.

Again, where is the business case for cities to build the infrastructure (sockets and such)? Often cities stuggle with existing expenditures and don't have much extra money for such projects. Also why would cities want HEVs generating MORE pollution in city cores???? Makes no sense.

"If each HEV can put out on average 10 kw power ...."
IT CANNOT put out 10 kw power. Neither city nor HEV owners want the car engines running right under their office windows. Nobody wants HEVs adding yet more pollution to cities. On the contrary, they are trying to reduce pollution from car travel.

Just picture a bunch of HEVs generating electricity during one of those hot summer days and all the smog getting yet worse. It would be a disaster. Why would I want my car running all day long during a hot day and engine getting very hot, etc. Plus I walk out of office directly into all these fumes?!?!? It is nuts.

Again, consider other factors besides technical feasability. Consider pollution. You cannot just run bunch of ICEs any place any time.

In fact, it is far easier to force people to reduce A/C usage than run a bunch of HEVs.


G, they aren't going to be remotely starting cars and running their engines to generate electricity. You're making an empty objection.


10 kW out of PHEV...??
10 kW / 200 V (Gen III Prius Voltage) = 50 Amps.

At 6.5 Amp-Hr capacity and cycling it at only 1% of capacity (i.e. 0.65 available A-Hr) you get 47 seconds worth of smoothing capacity per vehicle. And that's assuming 100% inverter efficiency.

There is definitely potential for V2G - no doubt it makes some sense. But IFF the Generator of the car is cleaner and more efficient than that of the grid (see my first comment). When we have an engine that runs off fairy-dust/hopes/dreams/the cynicism of Americans than I'll sign up.


I am not a V2G expert or anything, but I do not think that there is any effort going to be made to start the engines without drivers. The capacitor analogy is, I think, the most appropriate. Instead of numbers like 10 kW from 100 cars, I think that with population and penetration of this technology, that they are thinking more along the lines of 0.01 kW from 10,000 cars.

q, your example of a hot day is appropriate, but instead of the PHEV generating at that time, the electricity provider wants to be able to draw a little off the battery (knowing that you have an ICE to get you home) or at least to stop your car from drawing juice from the grid (if it does not have at least ~60% charge in the batteries).

Roger, it sure would be nice if these roofed parking structures were covered with PVs


If utility wants to draw a tiny portion of a battery capacity and considering relatively small battery packs with SOC limitations, it seems to me that they would not be able to get much energy at all. Not enough to make any difference.

Let's look at some real numbers:

Latest model has 1.3 KWh battery pack
Next gen should double the pack, so say it has 2.6 KWh

SOC range is 40% 1.04 KWh to 80% 2.08 KWh = 1 KWh

You can only get about 1 KWh from the next gen Prius. Say utility only draws 10%, so 0.1 KWh per car.

For a utility anything below a 1 MWh is a pocket change.
1,000 / 0.1 = 10,000 cars

You need at least 10,000 cars to have any sort of impact!!! Of course, some cars may not be plugged in (in use), some may be empty, etc. so you may only count on half the cars actually being available to use. Now you need 20,000 cars to be able to get 1 MWh.

How much infrastructure is that? How much cost? Utility will decide it is far too much hassle. Much easier to build a coal plant and start it as needed. Much less problems.

The comments to this entry are closed.