November 07, 2009
Chevy Volt Batteries In Cold Weather

A Fortune Magazine article by Alex Taylor points to qualifiers on claims of the performance of the forthcoming Chevy Volt pluggable hybrid electric vehicle (PHEV). Read the whole article. But I found the part about batteries most interesting.

That's not all. Also under scrutiny is GM's oft-repeated assertion that the Volt will have an all-electric range of 40 miles. Critics point out that the car needs ideal conditions to do that.

For one thing, the 40-mile range depends on ambient temperatures of 60 degrees to 65 degrees. When the temperature drops below 60, the batteries become less efficient. And if it gets hotter than 65, the air conditioner can impose an additional load on the Volt's batteries. Either way, the range diminishes.

But no quantitative information about how range goes down with dropping temperature. A question for any reader who knows a lot about battery performance: How much does battery performance decline with temperature? What sort of range can a guy in Wisconsin or upstate Minnesota expect in January with a Volt that can go 40 miles in late spring?

Also, how does battery performance fare in the Mojave Desert in July?

Also, does the extent of battery performance decline with temperature vary by battery technology? How do lead acid, NiMH, and Lithium ion batteries compare? Do some lithium chemistries to better than others in cold weather?

If battery range drops only to, say, 30 miles then that doesn't seem like a show-stopper. The majority of the American commuting public goes less than 30 miles round trip to work every day.

Update: In the comments "bbm" reminds me that GM has built the Volt to have a substantial margin its battery capacity. Of the 16 kwh of battery capacity the Volt needs only 8 kwh to go 40 miles. GM does not want to discharge more than 70%. So effectively there's a 2.4 kwh reserve available for colder weather. So the Volt might start out with 10.4 kwh available in mild weather.

Also in the comments agesilaus points to a research paper about Li ion battery performance in cold weather. At least with the Li ion battery used as an example the battery lost 15% going down to 0 C (32 F) from a mild temperature. Well, a 15% loss off of 10.4 kwh would still leave the volt above 8 kwh. at -10 C (14 F) the Volt would lose another 10% and might be slightly below 40 mile range. Though I wonder how much heater power would be needed to stay warm with an outside air temperature that cold.

Share |      Randall Parker, 2009 November 07 07:29 PM  Energy Electric Cars

Dan said at November 8, 2009 8:03 AM:

I am by no means an expert, but I can offer one datapoint:

The Tesla roadster actively maintains the battery temperature within a window to maximize battery lifetime. The energy required to do this does not appear to affect the overall range to an important degree. i.e. the load from the AC/heater is sufficiently less than what is required for the motor that it is not a primary influence on the overall range. Because the temperature is held constant (more-or-less), ambient temperature also does not affect capacity. Note that the roadster controls/maintains battery temperature even when the car is OFF.

I expect the Volt will also employ active control of battery temperature. Note that even though the range (40miles vs 250miles) is different between the two vehicles, the battery packs are somewhat more similar (16kW vs 53kW).


Kieth said at November 8, 2009 10:08 AM:

It would make a big difference if the battery heater was used while the car is "plugged in" as opposed to when it's
sitting on the street in St. Paul. I'm a little bit skeptical that the latter case would be negligible. I am also curious
to know if the battery is encased in insulation? or whether the insulation would be counterproductive during warm

Abaddon said at November 8, 2009 10:13 AM:

Why it almost seems as if GM put the car into development hoping that battery tech would mature before production began.

Insufficiently Sensitive said at November 8, 2009 10:19 AM:

Note that the roadster controls/maintains battery temperature even when the car is OFF.

In other words, this roadster consumes energy 24/7, and when "we" are all driving our oh-so-correct electrigizmo hybrids, the sum of their carbon footprints will be more enormous than the supposed saving from junking fluorescent lightbulbs.

Yes, practical considerations are subordinated to the green religion, and Al Gore is more overfed and overserved with energy with each passing day.

Chuck said at November 8, 2009 10:21 AM:

40 Miles? I do 36 round trip daily. This vehicle would be useless if anything less than perfect (only 10% reserve) would strand me for however long it would take to get a tow and then charge 50% to get home.

Chuck said at November 8, 2009 10:22 AM:

40 Miles? I do 36 round trip daily. This vehicle would be useless if anything less than perfect (only 10% reserve) would strand me for however long it would take to get a tow and then charge 50% to get home.

ptleahy said at November 8, 2009 10:27 AM:

Why not just wind it up with a giant rubber band?

The Other Hand said at November 8, 2009 10:40 AM:

A giant rubber band's performance varies widely depending on the temperature.

Steve said at November 8, 2009 10:41 AM:

Chuck, the Volt has a gasoline powered engine to drive the generator. Is the car really useless if you have to extend the range by running the engine a few miles?

Stephen Osder said at November 8, 2009 10:46 AM:

Comments regarding use of battery power to raise the ambient temperature make no sense. In conventional internal combustion vehicles, the vehicle is heated by using heat being rejected from normal operation of the engine. Using battery power to heat the environment at low temperatures near zero degrees F will use a significant amount of available power. I suspect that heating the vehicle via battery power at temperatures near zero will result in a significant net loss of available performance. Indeed, some batteries will provide zero power at extremely low temperatures. Today, such problems are solved by using external power to keep the battery warm.


Larry Brasfield said at November 8, 2009 10:59 AM:

The discussion regarding use of battery power to heat a cold battery goes astray when it ignores these facts:
1. The battery open-circuit (zero current) voltage is not much affected by temperature.
2. The inefficiency at low temperature occurs as increased losses when current flows.
3. Those losses heat the battery.
In other words, there is no choice but to heat the battery, and there is no additional
loss of efficiency from letting the battery losses do it.

agesilaus said at November 8, 2009 11:11 AM:

Well looking at this paper, which is highly technical. Figure 5 implies about a 15% capacity loss when temp drops from 25 deg C to 0 deg C:


That is for lithium cells.

Camera batteries are severely affected by low temps and people need to carry spare batteries next to their skin (to keep them warm), under their outerwear, to use when shooting in sub-zero weather. A battery that might last all day at normal temps will last as little as 30 minutes when the temp dips to -20 or so. They will recover somewhat when warmed up.

XC said at November 8, 2009 11:12 AM:

I can heat my entire 36 foot trailer overnight in 25 degree temp with two mid-sized marine batteries.

So I am gonna guess that keeping a battery compartment warm in a smallish car is no big deal.


PS - The batteries are outside in the cold and do not get appreciably warm. Sounds like I should insulate them a bit and they'd work even better.

Insufficiently Sensitive said at November 8, 2009 11:18 AM:

In other words, there is no choice but to heat the battery, and there is no additional
loss of efficiency from letting the battery losses do it.

That's hiding the ball. There may be no loss if efficiency by letting the battery losses do it, as measured by battery efficiency - but there's certainly a loss of efficiency in national power consumption due to that 24/7 battery-heating use of capacity at all the power plants.

DCE said at November 8, 2009 11:41 AM:

I deal with lithium-ion batteries constantly for our products. Relative capacity does decrease with a drop in temperature though it doesn't become really apparent until the temp drops below 0 degC for 'standard' Li-Ion batteries. For the Lithium-Iron-Phosphate lithium-ion batteries used in cars like the volt, the discharge/curve is a little different, but the same problem exists. When it comes to charging the batteries there is a relatively narrow temperature window: 0 to 45 degC. Above or below that range charging becomes problematic. Charging the batteries outside that temp range can greatly reduce both the lifetime and capacity of the batteries, so temperature control of the battery pack is very important to ensure both good battery life and capacity.

Randall Parker said at November 8, 2009 11:46 AM:


Almost? GM knew going into it that the batteries would need to improve by the time they went into production. But GM thought the amount of improvement needed was within the range of what was doable. It appears they are right. Time will tell.


Thanks for useful information. 15% loss would take 6 miles off the range to go down to 34 miles. They say:

At 40C, the capacity increased 6% from ambient (7 Ah) to reach a maximum near 7.4 Ah, and at 0C, the capacity decreased 15% from ambient with a maximum capacity of 6 Ah (see Figure 5).

Looks like it went to 5.5 Ah at -10 C. So you'd have about 31.4 miles range at -10 C. That is acceptable.

John said at November 8, 2009 11:51 AM:

Electric cars may end up needing plug in heaters like diesel engines do in order to at least boost the batter life at start-up in cold weather. But on the other end, I'm not sure even with drivers who do less than 30 mile drives to work each day in urban/suburban area how they're going to handle a July traffic jam on the way home with traffic at a standstill the temperatures in the mid-90s. Crank down the window I guess, and breathe in all the other car and truck exhaust fumes while hoping the battery doesn't keel over in that stop-and-go traffic.

bbm said at November 8, 2009 11:56 AM:


GM has engineered the system to have a fair amount of excess capacity for this sort of reason, as well as to hedge against deterioration as the battery pack ages.

The pack is 16kwh... but the volt will only need about 8kwh of that for the 40 mile all electric range (AER). They want to keep the pack from going below 30% state of charge (SOC)to prevent excess wear on the pack. That leaves plenty of buffer to keep the AER at 40 miles in most conditions.

40 Miles? I do 36 round trip daily. This vehicle would be useless if anything less than perfect (only 10% reserve) would strand me
for however long it would take to get a tow and then charge 50% to get home.

As was noted above, the Volt's TOTAL range is effectively unlimited because once the battery drops to a 30% SOC, the gasoline engine "range extender" kicks on to run a generator to provide the electricity to run the car. You can fill up with gas like any other car for long trips. The idea is that most driving in the US is less than 40 miles, allowing most trips (70-80%)to be accomplished on electric power.

bbm said at November 8, 2009 11:59 AM:

Plus, it seem like it would be pretty easy to run the range extender for a few minutes or so to heat up the pack and/or interior in truly cold conditions.

Bob Elkind said at November 8, 2009 12:03 PM:

Lithium Ion batteries have limited useful lifespan, roughly 2 years.

Anyone who owns a laptop computer knows than Li-Ion batteries have a limited useful lifespan, roughly 2 years, after which battery capacity declines significantly. Reference:

"Lithium-ion batteries lose capacity through cell oxidation, a process that occurs naturally during use and aging. The typical life span of lithium-ion is 2-3 years under normal use. Cool storage at 40% charge minimizes aging. An aged lithium-ion cannot be restored with cycling."

If Volt's storage cells are Lithium-Ion based, longterm cost of ownership should be a concern. Anyone out there have better info on battery type, replacement cost, etc. ?

Indiana Johnson said at November 8, 2009 12:08 PM:

Personally, I think the engineering design of the Volt is pretty ingenious - makes a lot more sense than the Prius. The pack is big enough to cover the daily mileage of about 80% of the cars in the States, and in terms of "energy cost per mile," centrally generated electricity is hard to beat - would be even better if we allowed more nuclear. If the battery life is reduced due to temperature, the engine comes on - which could be used to heat the battery right, as well as recharge the battery.

One thing I have wondered, do they have any plans to run the engine while the car is parked - for purposes of recharging? Seems like there would be some advantage to that.

I just wish the States were getting the Opel design instead of the Chevy.

mndasher said at November 8, 2009 12:43 PM:

Battery efficiency vs temperature is quite severe actually at -4 F only about 55 percent as 77 F, and 70 percent loss at -40 F.

Randall Parker said at November 8, 2009 12:47 PM:


What kind of battery are you using for your numbers? Got a link for this?

Liz953 said at November 8, 2009 1:14 PM:

I have a couple of somewhat off-topic questions. If these batteries have about a 2-year life, what is the cost of not only replacing them (as Indiana Johnson posed) but also the cost to dispose of the dead battery. Also, if battery-driven cars are supposed to reduce greenhouse gases, can anyone define the tradeoff between reduced air pollution and increased use of, say coal-generated electricity, the costs of replacement and the costs of (and possible hazardous) disposal?

Randall Parker said at November 8, 2009 1:36 PM:


GM is using a lithium iron phosphate battery that is supposed to last much longer than laptop or cell phone batteries. A Volt's batteries might last 10 years.

Bob Young said at November 8, 2009 2:34 PM:

What heats the passenger compartment in winter when temperatures are below freezing? Do these electric vehicles have hot coolant circulating through a radiator like a gasoline engine? Do the electric motors themselves generate enough heat for passenger warmth? Electric motors that run hot usually aren't efficient.

Or is the equivalent of an electric space heater used? That would flatten the batteries pretty quick.

Chuck said at November 8, 2009 2:45 PM:

Point taken about the booster engine. I was thinking about the Tesla. Mea Culpa.

OTOH -- What is the milage on the 1.4 liter engine during a long run with the battery down?

Rob Peterson said at November 8, 2009 6:21 PM:

This is Rob Peterson from GM Communications. Maybe I can offer some additional insight into the Volt's battery performance.

Just about all batteries struggle in cold temperatures, whether it's range or performance - it's a matter of physics. A cold battery will struggle to deliver its full power performance and cabin conditioning will draw energy normally reserved for vehicle range. In extreme temperatures the Volt's battery can be warmed by grid electricity (if plugged in), or by starting the engine-generator first to power the vehicle and warm its 16 kWh liquid cooled battery to ensure performance. The flexfuel engine-generator, which creates electricity on-board, will extend the range of the vehicle to ensure you always arrive at your final destination or until you can find the nearest gas station.

Also, the Volt's battery is a purely automotive design - from the chemistry (li-ion mangnese spinel) based), cell design (prismatic as opposed to cylinder), pack management which restricts overcharging (which impacts calendar life)and deep discharges (which impacts battery power) to automotive quality manufacturing at both the cell and pack levels (both of which will eventually be performed in Michigan). We're increasingly confident - based on test results from both our battery lab and in the nearly 100 Chevy Volt pre-production vehicles - that the battery will meet our internal targets of 10 year, 150,000 miles of life.

Hopefully this information is helpful.

Rob Peterson
GM Communications

Dave said at November 8, 2009 7:07 PM:

I thought I read somewhere earlier that the Volt would burn some fuel if needed to "warm up" the batteries and electric motor to optimum level. Rob P above suggests keeping the Volt plugged in will also heat 'er up to desired temp. But then, can that be a user-defined off-on option like a block heater, or will it burn up the juice for heat all night during charging, even though it's not pulling out of the driveway till 7:45AM?

Rob Peterson said at November 8, 2009 8:14 PM:


We're still refining the details, but owners will be able to define grid charging time (ie 11 pm - 7 am) among many other vehicle attributes. We'll have more information to share once we've returned from our cold-weather testing in Kapuskasing, Ontario early next year.


Paul said at November 8, 2009 8:26 PM:

Electric cars would do better to store energy in a thermal mass rather than using a battery for resistive heating (or, likely, even using the battery to drive a heat pump). The heat energy/mass of a very hot insulated mass could be quite high. This "heat battery" could be recharged resistively when the car is plugged in.

Mike McDaniel said at November 8, 2009 9:21 PM:

Keep in mind that the battery used to start a car in cold weather cannot be compared to the enormous battery pack that will power an electric vehicle. A common 12 volt battery need produce only enough power to turn a starter motor with specifically designed gearing and is charged continuously as the vehicle is operated. With modern ignition systems and computer controls, that is not the chore it once was, but in cold climes, it is still common for vehicles to be garaged or engine block heaters to be used to lessen the strain on the battery, which does lose substantial power as the temperature drops.

A Chevy Volt battery pack will need to propel an entire vehicle and power all of its electrical components, including a heating device and fan for cold weather, both of which are high drain components. Add anything else that will drain electricity from the battery such as headlights, taillights, turn indicators, stereo, etc. and one might well be able to make it to work, if it's only 10-15 miles, but after a day in the cold, all bets would likely be off in terms of getting home. In very cold areas of the nation (to say nothing of Canada and Alaska), one might not even be able to make it out of the driveway. Remember too that added weight does decrease mileage and range. This is barely noticed in conventional vehicles, but would be a significant and immediately apparent factor in an electric vehicle. In the cold, the effect would be multiplied and glaringly obvious with very negative effects on range and function. At the very least, much of the year the abilities of the vehicle would likely be so weak that the vehicle would be, essentially, a driveway ornament.

Notice that Mr. Peterson from GM did not specifically address these issues, but merely outlines some fun to know facts and presented a very optimistic picture. One would, for example, hope that a battery pack lasted at least ten years, because at last glance, GM was suggesting replacements would cost upwards of 1/4 to 1/3 the price of a new vehicle. One might very well, after several years, end up with a low mileage vehicle worth less than the battery replacement cost.
And yes, one can plug an electric vehicle in to keep it warm, but who is going to invest in the massive infrastructure necessary to plug in thousands of EV's in workplaces and other public spaces? Absent that, we're still stuck with the essential problems of physics, as Mr. Peterson so cogently informed us. Batteries really don't work in cold weather, particularly batteries that have to propel motor vehicles.

Let's remember too that in cold climates, vehicle reliability can be a life or death issue. Even if we assume that an EV could be sufficiently practical to work in cold climates--and that seems unlikely with current technology--the range and power reserve limitations would relegate an EV to very light, short range errand or commuter duty. Getting stuck in a snowbank during a snowstorm could be a death sentence in the wrong circumstances as a battery pack quickly weakened and died.

All in all, there is no known way around these issues.

Randall Parker said at November 8, 2009 9:31 PM:

Mike McDaniel,

Getting stuck in a snow bank? This car also has a gasoline engine. What we are debating here is how far the car will go purely on battery in extreme weather before it has to activate its gasoline engine.

Headlights, tail lights, stereo, etc: How much current does all that pull? A car heater is going to pull more current in a PHEV because it can't count on heat from the gasoline engine. But the other electric power using devices in a car won't be any worse in a PHEV than in a normal car.

Engineer-Poet said at November 9, 2009 4:43 AM:

Indeed, vehicle reliability can be a life-or-death issue.  This is why the vastly superior hard-starting capability of a vehicle like the Volt (it can crank the engine much longer than any ICE-only vehicle) plus the emergency-limp mode implicit in the unused battery capacity make it inherently safer than today's cars.

The discharge characteristics of the A123Systems battery are here.  You'll notice that the performance at -20 C isn't all that bad.  Further, the battery would not stay that cold under load; the major effect of low temperature is to reduce ion mobility and increase internal resistance, which causes the battery to heat up.  If the resistance of the 2.3 AH cell rises from 10 mΩ to 200 mΩ due to cold, it would dissipate 1 watt at 1 C discharge rate (roughly highway cruising power for a Volt; acceleration would be more).  That would bring the cell temperature up at a very reasonable rate.

If the battery is so cold that its performance is truly marginal, the management system can take 2 kW for a few seconds to start the sustainer and use it both for extra power and for battery and cabin heat.

Sean said at November 9, 2009 8:44 AM:

How about a gasoline-fed high efficiency catalytic heater, with a simple gravity feed from the existing gas tank, that turns on in cold weather to heat the battery bank? There would be a mileage penalty for sure, but it would be a much more wise use of energy compared to using precious battery power. Sometimes low-tech can be used to empower high-tech.

rob said at November 9, 2009 1:00 PM:

What kind of warranty is GM going to offer on the battery? That would give a sense of how long they expect them to last.

John Moore said at November 9, 2009 8:21 PM:

How about the 5KW required to provide air conditioning for much of the year here in Arizona?

Engineer-Poet said at November 10, 2009 8:31 AM:

So, Sean, you would burn scarce petroleum in an open flame to provide heat, instead of in the sustainer engine to provide motive power plus heat?  You'd pre-heat the battery with gasoline instead of grid power?  Funny set of priorities you have there.

The cooling power of the Prius system is listed as 3.4 kW, so it is doubtful that the electric power consumption is even half that.  The electric A/C is more efficient than the engine-driven system, so even if you have to start the sustainer you'll burn less fuel.

The situation in a fuel shortage is radically different.  If you had a 20-mile one-way commute in a Volt, you could shut off your A/C and drive the entire route on electricity if you had to.  You could go days or weeks this way.  A conventional car would eventually leave you stranded no matter how much you feather-footed it.

BBM said at November 10, 2009 8:48 AM:

Reportedly, the battery warranty on the Volt will be for 10 years. The pack has been engineered to still supply 40 miles of AER at that time (because it is "overbuilt" with excess capacity). GM is being very careful here to avoid any type of perceived failure on its first EREV.

Here's an (independent) fan site with lots of good info and articles:

Tim Andersen said at August 6, 2010 2:07 PM:

In the Twin Cities temps sometimes stay below 32f for months. Diveing below zero nightly. Batteries are destroyed quickly by the extreme cold.

Throw in rolling resistance added by thick greese and snow on roads adds more challenge to an electric car.

I wonder how much juice will remain in the Volts batteries after sitting outside at work for 10 hours while at 20 degrees?

Oh ya the "GAS" engine will probably start and pick up the slack but what does that say for an electric car in Minnesota?

The Nissan Leaf is very doubtfull for use in the cold.

Kelly said at September 21, 2011 8:36 AM:

I have a Volt. Admittedly, I have owned it for only four weeks with about 700 miles so far on the vehicle. When the car was received it said .3 gallons of gas had been consumed. We drive up to 35 miles a day yet we have not used any gas so far. Part of the trip is on the freeway at speeds of 65 mph. The car drives like a normal car except it feels sportier in response and is too quiet. It says there is eight gallons on board and that the range for combined is always somthing over three hundred miles but I have never heard the gasoline engine come on yet.

I have no idea what the "huge infrastructure" comments are about the charging. I just plug it into a 110 volt outlet and it takes four to six hours to charge. I am thinking of hooking up a $475 220 volt charger to my dryer circuit but actually so far there seems to be no reason to do this. Either way it is not an expensive process for either a 110 volt or 220 volt outlet.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright