"After the first battery fire, GM officials complained that NHTSA did not drain the battery of energy as called for under the automaker's crash procedures. NHTSA normally drains fuel from gasoline-powered cars after crash tests, they say."

So did they drain the battery of energy this time?

There are some mechanical distruction safety tests that are done on individual cells to determine their response, if any (fire, short, etc.). There is the nail puncture test, the crush test, etc.

Different battery chemistries and cell designs behave very differently in these tests. So an issue with one particular type of cell does not mean the same behavior would be seen with other types of cells. I haven't happened to have seen these particular tests for LG cells (the type used in the Volt), so I don't know how they perform compared to the others. But there are people out there who do know because this type of data is given to potenial customers who ask for it.

It is really up to GM to come up with a response to this issue. No one else really has enough data to be able to say anything about it. It was LG cells, but in a GM battery. There are probably a number of possible things they could say depending on what the data says. But without this data it is difficult to say which of these things would apply.

One example is what they already said, namely that if a serious accident happens, totaling the car, that "someone" needs to drain the battery. They will need to specify the threshold of damage for battery draining and make the equipment and proceedures for draining available. And train a bunch of people on how to do this. It is a reasonable position, but one that does require some work and resources to implement.

But there are other things that might also be said to aleave the worries of the public depending on what the data actually supports. Such as how long it takes from damage to ignitian. Such as whether or not the fire is more smoke than danger. Etc.

It is also possible that this problem suggests some changes to the pack design to prevent the problem from happening and/or to sense when it does happen and let people know it is happening before it actually becomes a problem. The temperature and voltage sensors already in the pack might help with this. That would be good in that only a software upgrade would be necessary. But additional sensors may also be useful, such as IR or a chemical sensor that detects asperated compounds.

Certainly the NHTSA events seem to require a level of damage that would completely total the car and therefore people are not likely to be driving around in the car by the time, hours later, if there is a fire. And it seems that if such a fire were to happen that there would be warning of smoke some significant time before the heating would cause heat danger to anyone. A lot more time than a gasoline fire would allow.

But, again, it really has to be up to GM to look at all this data and pull together a complete package of what can be said about this. Because they are the ones who either have the data or can run the tests to get it. The NHTSA will also do their tests, of course. But GM can test more aspects of this issue on the cell, battery and car levels and can do the tests faster than the NHTSA is likely to. So GM should move quickly on this before the media distributes a small portion of the story too broadly, leading to the wrong impression.

I agree that the Volt is not an EV, it's a hybrid, but if the issue is in the battery pack it doesn't really matter much. Draining the charge from a battery pack can be done simply and safely with a large resistor, nothing expensive or complicated, though it will take some time depending on the size of the resistor.

The real reason the Volt is more dangerous than the Leaf is, as several have mentioned, that it has that tank of gasoline in it that can explode. A battery fire, while dramatic, like a safety flare, will generally be relatively contained and doesn't contain nearly as much energy as a tank of gasoline. The Volt's battery contains about the energy of a half gallon of gasoline.
If a highly flammable liquid like gasoline catches fire, it can (as we often see on the highways today) create an engulfing inferno.
Clearly, the thing to do after a crash that has damaged the battery pack on a PHEV such as the Volt is to remove all potentially dangerous chemicals as soon as possible. This includes draining the gasoline and removing the batteries.
Although testing and analysis with more information than I have at my disposal is necessary, IMHO discharging the batteries may not be the panacea that some discus and could cause more problems if done too quickly. If it does end up that discharging the batteries is necessary (such as being necessary to safely remove damaged batteries), that can easily be done, relatively slowly, without any extra equipment and safely by simply running the air conditioner for a long time - assuming it is still functional after an accident.
The BEV is clearly safer than the PHEV but I can't see how a Volt is any more dangerous than any of the billion pure-ICE vehicles on the road today.

It's going to take a large resistor dissipating a lot of heat to drain the energy in a battery pack in a reasonable amount of time. High current discharge can heat up a battery pack, not something you want to do to a damaged pack and not something that should be done by an untrained owner. If you think your pack is damaged don't start running accessories in an attempt to drain it. Frankly if you are in an accident severe enough to damage the pack you probably have other concerns.

Power resistors are cheap, if bought in bulk. I don't think a desipation system should be complex or expensive. Nor would it need to heat the pack appriciably if the resistor is appropriately placed and thermally grounded. And it might be far more likely to survive the collision then the A/C. I have designed a lot of EVs over the years and I think it is doable at almost no cost, at least given car production purchasing power. Whether it is needed or not is another matter.

Drawing a kW or two off a pack this size is not high current. It is more likely to lead to a safer situation then to create more of a problem if the design of the disipation system is well thought out.

But we don't have all the info on this. How reliable is it taking many hours to get to an even? Did the coolant play a role? If so, that would be a completely different set of engineering solutions.

From what I've seen so far the coolant leaked out and created a short, which started a fire. I'm not sure if the coolant ignited or not. The real concern is not so much a damaged pack sitting around but a rolled vehicle with people stuck inside. Using a resistor to drain a pack in 24 hours does nothing in that situation and having such on board serves no purpose.

Actually, there are quite a number of design fixes if the coolant is shorting the broke battery it odd vehicle orientations. Going to a non-conductive coolant would allow GM to not have to change the current Volt design. But there are many others that could be used on future designs.

Most EV's are going to have the pack in the middle of the vehicle, either in a central tunnel such as the Volt or under the floor pan such as the LEAF and Tesla Model S. All would be subject to similar bending forces. I think the pack location is reasonably protected, it's the cooling that may need to be redesigned.

They specifically talked about the main pack as the source of fire. You make a good point that no accessory drain will fully drain the pack. I think the simple answer, which may or may not be possible, is non conductive antifreeze with a higher ignition point. Or just use air cooling as Nissan does.

The Volt uses a 50/50 mix of Dex Cool and deionized water, run through thin plates between each cell, made by Dana corporation.
More details:
http://media.gm.com/content/media/us/en/chevrolet/news.detail.html/conte...

The problem was a result of tipping the pack upside down, which means the coolant probably flowed down to what is normally the top of the cells, which is where the connections are. It might not need broken wiring to cause a problem.
Since you have the DexCool handy how about a flammability test as well? Maybe a few drops on the concrete and hit it with a torch? I've been meaning to try it with regular antifreeze.

"Do not try this at home".

I placed a 3" diameter ceramic crucible on a firebrick, placed the firebrick on a cement slab and poured in around 0.1" depth of Dex-Cool. I then lit a MAPP gas torch. MAPP gas torches are pretty hot. I evenly heated the liquid until the ceramic dish cracked. Then I continued to heat the liquid more. The liquid boiled but did not catch fire. When I removed the flame the liquid produced a very thick grey smoke, but again did not produce a flame. With additional heating, sometimes removing the torch to see the smoke, the liquid evaporated.

This coolant is not going to catch fire. But it produces such copious amounts of thick grey smoke that a casual observer who couldn't see what was going on at the site might think something had caught fire.

This may be an "all smoke and no fire" story.

When the car rolls over the coolant flows over various conductors, shorting them out. The resultant heating causes the coolant to put out copious amounts of thick grey smoke. There is some harmless but showy sparking.

Is that what the NHTSA saw? Is there less to this story than there appears?

It would help if we could see video or something.

So the Volt has an early warning system to tell people if there is about to be a fire: copious grey smoke. Anyone would be able to tell that it is time to leave the vehicle.

But if the coolant were not electrically conductive in the first place the short may not have happened and then the subsequent fire might not have happened.

@JRP3 - That might work. And if it seems expensive, keep in mind that the product you linked to is packaged in a small bottle for a niche electronics market. If GM bought it by the tanker truck it would be nearly free.

And there are non-conductive coolants good for freezing temperatures. Like the one GM has for their fuel cell program. Maybe even the one you linked to. I didn't see a temp spec there. But it did say it was compatible with aluminum and nickel, materials called out as being in the Dana system according to the link that you had in an earlier comment.

Battery powered cars are inherenetly safer than gasoline cars due to the large factor less energy onboard.

And this Volt failure mode seems to only apply to a roll over. And rolling a car with a battery pack in it is hard to do given the low center of gravity.

The reason, in my view, why we went to such an extend on this thread to get to the bottom of the problem and to find solutions is: 1. we want plug-ins to be as safe as possible since we will be driving them and 2. to get this problem understood and solved as soon as possible so that it doesn't trip up the plug-in effort as a whole.

Chevy will give worried Volt owners loaner cars.

But wait. Chevy Blazers and pickup trucks have had numerous incidents in which the vehicle burst into flames after a crash. Taking a loaner could be like jumping out of the frying pan. How about they just loan out some fire-proof suits instead?

I really doubt that there is undue hazard here. However, the timing of the two fires seems odd, suggesting a very slow reaction (maybe the electrolyte oxidizing in the region of a casing crack) that then exotherms at some point. Maybe cell balancers working overtime trying to balance cells with one damaged low cell? Coolant and electrolyte incompatibility?

I certainly hope they find the problem soon, and can fix it inexpensively.

When I first got out of college, with a mechanical engineering degree, my first job was as a rocket designer and aerospace tech developer. So everything I did had to do with developing high tech combustion devices. The first week on the job I stopped at stop sign at the rocket factory, looked both ways and slowly accelerated. My car spontaneously caught fire, sending black smoke everywhere. The rocket test site fire trucks sped out to put the fire out. It was very embarassing.

You aren't much of an engineer I guess. A vehicle with two systems is a hybrid, the Volt uses a combination of serial and parallel hybrid features and can switch how the vehicle is propelled. EV's don't have gas tanks, the Volt is never a pure EV, it's a hybrid that can run on battery power for part of the time.
Yes antifreeze can burn, the flash point of ethylene glycol is around 230 degrees. There have been instances of sprinkler systems with an antifreeze mix in them causing fires to get worse when they activate. However it appears that in this case the antifreeze caused short circuiting of the battery pack which led to a fire. I'm not sure it has yet been proven if the short caused the pack to catch fire or something else.
RC toys use LiPo batteries, a completely different chemistry than what is in the Volt, and other EV''s.
Talk about ignorance, your post was full of it, pun intended.

Three strikes and you're out. I never said flash point is the same as ignition point, nor did I say the glycol/water mix would catch fire at 230 degrees. I was just pointing out that if heated it has a flash point of 230 degrees or so and it can indeed burn at some point. Obviously a closed pressurized vehicle cooling system can operate at higher temperatures, so your points regarding such are meaningless.
http://www.sciencelab.com/msds.php?msdsId=9927167
The fire didn't start until antifreeze caused a short circuit, and didn't happen until many days after the initial test. RC LiPo's go thermal a lot faster than 7- 21 days after being punctured.

Yes the LGChem cells are lipo but they are a different chemistry than RC lipo and will have different characteristics. Just because a cell has lithium in it doesn't mean it's the same as another cell with lithium in it.

The Volt is never a pure EV because it's always hauling around a complex, expensive, gas burning engine. EV's are mechanically less complex than ICE's or hybrids, the Volt is more complex, hence not an EV. A standard Prius can drive about half a mile on battery alone, that doesn't make it an EV either. You can pretend you're driving an EV all you want but wishing doesn't make it real. Sorry. It's a hybrid, a plug in hybrid to be specific. Plug in hybrids are a good choice for many people, but that doesn't make them EV's. Fisker Karma is also a plug in hybrid, it's a full serial hybrid, but still a hybrid. Since at times the Volt ICE drives the wheels directly it's a parallel/serial hybrid combination, but still a hybrid.

You're insane. It doesn't matter how you drive the vehicle, it's the way it's designed and built that makes it a hybrid. I don't care if you never use the gas motor, it's still a hybrid, just as if you only used the gas motor and never plugged it in it wouldn't be an ICE. It's really simple. The Volt is a plug in hybrid, always was always will be, unless they pull out the ICE. I never back pedaled, talked in circles, or tried to mislead anyone, unlike yourself who is trying to convince people that a car with a gas tank is an EV. I've been 100% consistent in describing the Volt as a hybrid, plug in hybrid to be specific. You must work for GM marketing, who tries to pretend the Volt is an EV when it's not. Just because you bought into the marketing nonsense doesn't mean everyone does.
Just because an EV doesn't work for you doesn't mean it's not working for people right now, every day. That's right, some people use an EV for their only vehicle. You've turned this into some ridiculous moral debate about "pure" EV's and what's best, where the fuel comes from, if they ever drive an ICE occasionally, I'm simply pointing out the fact that the Volt is not an EV, it's a hybrid, and as such is mechanically more complex than an EV, and will have more maintenance issues than an actual EV.

Different lithium ion batteries have completely different properties depending on their chemistries (what is in there other than lithium), depending on the electrolyte and depending on the design. They have different power and energy densities and different characteristics in a failure, such as the nail test. Some lithium batteries will immediately go snap, crackle, pop, some will autoignite hours later and some will just slowly and gently self disipate. A lot of which path is taken is the internal impediance of the cell and a lot of it is the thermal stability of that particular chemistry. These issues and the cell configuration also have a big role on whether or not the battery needs to be liquid cooled or if it can be air cooled.

@ex-EV1 driver.
"There are 3 kinds of cars that we are discussing:
- Electric
- Internal Combustion Engine (ICE)
- A combination of both, traditionally called a "hybrid""

Not to defend Anonymous (because I believe there is a little truth in everything that is being said by both sides) but you can't put all cars into 3 categories. EVs, yes. ICEs, yes. But hybrids are difficult. The issue is that hybrids come in all ranges from start/stop (I don't consider S/S to be hybrid but OEMs like Honda traditionally do) which are the closest to an ICE to EREVs which are about the closest to EVs you can get. Where you draw the line is somewhat opinionated depending on your usage, background, and perspective. Any suggestion that one system is better or classified a certain way will only lead to the petty arguments observed here.

BTW, even ICEs aren't black and white. What about Hydrogen ICEs where all the hydrogen is generated from electricity. Are H2 ICEs electric? Just another thought starter (or argument starter)

It's not being a silly purist, it's using accurate terms. I'm sorry but there need be nothing complicated about regen, my DIY EV has regen, I can adjust it's strength, but it's still just using the motor as a generator, like every other EV with regen. Mine is setup like the Tesla Roadster, regen on the A pedal, and I like it that way.
I said nothing about the Volt not being able to run down the road at any speed on battery only, I'm quite aware that it is capable of that, just as I am aware that capability does not make it an EV. Any more than the various kits that increase the battery size of a Prius, add a charger, and allow it to run on battery alone for an extended range, turns it into an EV. It turns it into a plug in hybrid. You seem to be overly sensitive to some perceived dismissal of the Volt when in fact all we are trying to do is accurately describe what it really is. I never said it was a bad vehicle, or anything negative about it, other than the reality of it being more complex than either an EV, or an ICE.
I've seen nothing from Nissan that says they will be offering the LEAF with a trailered genset. I've seen individuals experimenting with such, which is nothing new.

As a new Volt owner and enthusiast (1,100 miles and 4 weeks), I have really enjoyed this discussion. Although I have an advanced degree in electronics and understand the general design philosophy of the Volt, what amazes me most is how well executed the car is from a driving experience perspective. The true test of how well Chevy did in this respect is how easily my technically-challenged wife adapted to the car and doesn't need to know (or care) whether it is an EV, an ICE, or a hybrid. She just knows that she pushes the blue lighted button to turn it on and off. The biggest transitions for her from her beloved 2002 Honda Accord are the EPB (electric parking brake) and that the rear wheels are set really far back requiring some care to avoid running over curbs.

As far as I'm concerned, my Volt is the neatest car I've ever owned, including my 2003 50th Anniversary Corvette. I think I've used 4 gallons of gas in my 1,100 miles of Volt ownership and most of that was testing to make sure the ICE really would start and do its job when needed. We've made a couple of 90 mile trips and it was fun to save some EV capability for the destination by using Mountain Mode in the middle. The Volt has many dimensions and many personalities. In Sport mode, it really has a nice kick in the back; maybe not quite as enthusiastic as my 5.7 liter Vette but pretty close.

A final comment about phrases like: "battery depleted" and "charge depleted." It is my understanding that the EV range estimates are not to battery depletion but to probably 20-30 percent capacity remaining. One can confirm this when driving off from a full stop after being in EREV mode (battery meter on zero). The ICE frequently is not running and it feels and sounds just like being in EV mode until a lot of acceleration is requested.

Thanks to all for your comments and insights. I have really learned a lot.

Well the H2 ICE is really fueled by whatever fuel produced the hydrogen, which could also be coal powered electricity.

@regman,
Are you just trying to be argumentative here? You're going so far off topic that I had to go back to the original article to find out the subject.

I'm not going to debunk the "long tailpipe" myth here. You're wrong, there is no pure gasoline-ICE vehicle that is cleaner than a comparable sized BEV fueled even from pure coal. Even the Prius only comes close depending on your assumptions about gasoline production. Its been discussed ad nauseam on other threads.

Why is it hard to do EV garbage trucks?

@alt-e,
Interesting. This is the first time I've heard that. My impression is that they do require a lot of power since they start and stop a lot but they don't actually use very much energy in a day since they don't drive very far. I've got a friend who works for the city. I'll ask him how much fuel (they use CNG) they use each day.

@ ex-EV1 driver - The garbage trucks may not go far or fast but they are heavy (almost like armor plated) and they have to power the hybraulics to put in the garbage and to crush it. They also run a long time each day. So the battery pack would be big and add even more weight. And be very expensive.

I realized my mistake about the starter after I posted, never pretended to be an expert, but didn't bother to correct it because my initial point that cranking over an engine to discharge a pack is not what they would do, and it's frankly a stupid idea.
As for regen, again, it's not complex, it is simply the reverse of throttle input for acceleration. GM may have had to do a bunch of monkey business to make it work smoothly with their overly complex hybrid system, something an actual EV does not need to do.

I liked this recall blurb regarding Pontiacs and Buicks that I stumbled upon:

>> Both models apparently have a problem with oil leaking onto the exhaust manifold under hard braking. If the oil makes it underneath the manifold's heat shield, it's possible that it could ignite a small flame which could spread to the spark plug covers. and GM is suggesting that all of the 207,542 vehicles affected shouldn't park inside garages, carports or other parking structures until the required fix is completed. The recall notices were sent out today, so we'd suspect that dealers should have the necessary parts in stock by the end of the month. <<

From here: http://www.autoblog.com/2008/03/13/recall-watch-207-542-pontiacs-and-bui...

I especially like this "The engine fire normally happens five- to ten-minutes after the vehicle has been shut down... " Seems to be saying that engine fires are normal, right?

When I occasionally need an ICE I use another vehicle, instead of carrying it around with me all the time. Just as I sometimes need a truck, I don't drive one every day.
Regarding carrying around extra battery, the Volt has to do that since by it's very nature the smaller pack will probably be depleted every day, so the Volt carries around much more battery than you are allowed to use, to preserve pack life. So not only are you carrying an ICE you try not to use you carry around extra battery you aren't allowed to use.
I agree if you regularly take longer trips you are better off with a hybrid for now, be it Volt or Prius, but if you are like myself, and millions of others, it's quite practical to use an EV99% of the time and occasionally use an ICE when you need it. I don't own a pickup truck though I occasionally need one. I rent or borrow one when necessary.
Hybrids are a stepping stone to the future, EV's are the future, which work quite well for many today. Too much importance is placed on rarely needed range. Unlimited range is a luxury left over from the days of cheap abundant oil, those days are behind us, as we'll be seeing quite clearly in upcoming years. When gas goes back to $5 and beyond few people will care about taking long trips and will just want to get to work and the store every day without spending a fortune.

@Anonymous,
Both have produced production vehicles with lead acid, Nimh, and Li-ion batteries
What Li-ion production automobile has Ford or Toyota built?

. . . and yes your Volt is a hybrid. So is a Moped and one could even argue that any manual transmission ICE vehicle without a clutch safety switch is as well since one can drive them a short way using the starter motor and battery.
. . . and there is nothing evil about that.

You still miss the point. It matters not where your electricity comes from or how you use your vehicle, it's the design and construction that makes it a hybrid. Get over it already, you drive a hybrid. There is nothing wrong with that if that's what you need, but obviously you think there is. If your Volt remains parked in it's garage forever and never uses gas or electricity it's still a hybrid. When it comes time to change the oil in your Volt you might finally realize that it's a hybrid, not an EV.
Of course batteries last longer when not fully discharged, that's why the Volt has to restrict usable capacity so drastically, because the pack is so small it would be fully discharged on a regular basis if they did not. EV's with a larger pack can allow a larger portion of the pack to be used on occasion because most of the time it is shallow cycled. The new Toshiba Scib cells that Honda is using in their EV can be more deeply cycled than most, as can A123's.