93 comments

[ 5.3 ms ] story [ 154 ms ] thread
I have to admit I under-estimated how problematic electric vehicle fires would be based on my knowledge with smaller lithium-ion devices like phones (that just pop and quickly fizzle out). But after reading about fire fighting them, you realize how big of a powder keg they are, in particular as they can spontaneously re-ignite themselves and water has limited effective (even if still the best we have).

Article on the topic:

https://www.nbcnews.com/business/autos/federal-regulators-wa...

Here's a firefighting training video (start at 3 minutes if it does not):

https://youtu.be/8n5Wf7TlGrU?t=181

Although there are some innovative solutions coming on tap, like this special modified shipping container that you put a still-on-fire EV into and then drive it to another safer location:

https://cfpa-e.eu/container-puts-out-inextinguishable-fires-...

Don't firefighters carry some kind of foam by default? After all there's a lot of grease fires(cars, kitchens) out there.
Foam is not as effective against electrical fires and the manufacturers recommend using water, not foam, on a lithium-ion electric car fire.
Not usually in the US unless it’s at an Airport. Although larger metros may have specialists. A couple friends of mine are firefighters in a very large city and they will help other counties or cities out with special situations. It all depends on what department has what specialists.
In Germany they often have special nozzles that add some foaming agent to the water to create a water-based foam. They might also have a couple of foam fire extinguishers for cases where you need something that isn't water.

I think the general practice for EVs around here (in larger cities at least) is to extinguish it with water, then put them in a container full of water for 24 hours.

What's done with the water afterwards? I assume it's a large volume of contaminated (in various ways) water?
Foam interrupts the "oxygen" side of the heat-fuel-oxygen fire triangle. Unfortunately, lithium-ion batteries make their own oxygen when heated, so it doesn't help to smother them.

The advice is to use copious amounts of water to cool the batteries, removing the "heat" leg of the triangle. This works but requires extended periods of cooling support due to the possibility of internal shorts re-supplying the heat, and starting the process again.

Maybe we can shoot liquid nitrogen- like in terminator 2 to freeze the battery
Pumping huge amounts of pure nitrogen into the atmosphere sounds like a suffocation risk.
Fortunately not, and we definitely don't want to increase the amount of PFAS getting into the environment and poisoning our water supplies.
I’ve been stricken with enthusiasm for hydrogen-powered cars since watching a video from the 90s depicting a backyard solar-powered hydrogen generator which hooks up to a garden hose [1]. It’s supposedly a lot safer than gasoline:

> “These tanks can be shot at with incendiary bullets, or cut in half with a chainsaw, and you could throw a match on them — they’d just smolder like a cigarette. You can’t say that about a gas tank. Here, only the hydrogen you need is released from the tank. When the tank’s heated, it produces hydrogen and the car burns it. So there’s never much gaseous hydrogen in this system at any given time.”

[1]: https://www.youtube.com/watch?v=BX2RxlXCZKA

My guesstimation is that we're just trading modes of failure and it'll roughly be a wash in the end.

Gas cars have leaky oil and combustion happening (my old high school car caught on fire under the hood). Users are regularly handling gasoline at stations.

Electric cars may be a bigger problem parked or at home. Overall less maintenance that might cause a fire but batteries so far seem more susceptible than gas tanks which are dangerous but don't generally ignite. Users are installing charging stations (although should be covered by electrical codes). Fire suppression in the garage and at charging locations might become popular (encouraged by insurance discounts).

Yeah, batteries are "dangerous". We're all still carrying them around in our pockets and purses.

I think it'll be a wash too but I think that the failures will move up the socioeconomic ladder so you'll hear more about it.
> Gas cars have leaky oil and combustion happening (my old high school car caught on fire under the hood). Users are regularly handling gasoline at stations.

It's worth considering that during the life of the car, far greater time will be spent charging as opposed to time spent fueling one, and even then, a fueling station can easily be outfitted with specialized safety devices that combat fuel fires.

> Yeah, batteries are "dangerous". We're all still carrying them around in our pockets and purses.

Stored energy is dangerous. The higher the density of that energy, the more danger it presents. Vehicle battery packs are designed with integrated cooling, and thus can be far denser than what's in your pocket.

Not to be completely negative, but to expect that there won't be a "safety learning curve" with larger and larger numbers of battery operated vehicles does not seem like sound planning. Met correctly, I'm sure we'll be able to manage it.

We could make different trade-offs. If we had electrified freeways (allowing cars to charge while moving), most people wouldn't need large batteries in their cars just for the occasional road trip and manufacturers could use safer, cheaper chemistries like lithium iron phosphate.
This is one reason I'm hoping that lithium iron phosphate batteries catch on for the low-to-medium end EVs that don't need huge range, as the batteries tend to be quite a bit safer. Not that it's impossible to catch LFP cells on fire, it's just harder.

I also wonder if it would help to have an emergency discharge port or something? In other words, a fire truck pulls up and they plug a cable from the car's discharge port that energizes some high-power heating elements to boil water or something. The idea being that the faster you can discharge the battery, the less stored energy you have to deal with catching things on fire again. (Though it might not actually help, since in the short term extracting power from the battery will just cause it to heat up more.)

(The "emergency discharge port" could just be a DC fast charger port with some method to force the contactors to engage.)

...another variation on this would be an emergency water cooling port. Like you have a fixture somewhere on the car that you can connect a hose to and it runs water through coolant channels.

A lot of EV batteries have something like this already, you'd just have the ability to run external water through it instead of just going with a closed loop. You'd also need an outlet, which I guess would spill ethylene glycol all over everything if it wasn't contained, but better that than a burning car.

Many race cars are plumbed with a port that can take pressurized water into the cooling system from the pits, and already have a temp/pressure valve to vent excess pressurized coolant [water in the case of a racecar, coolant mix for a road car] so the tech exists.
Well good news: LFP is likely to become the dominant chemistry in most transport applications that don’t have specific or very high capacity energy weight/volume density requirements, the lower end of EVs and passenger vehicles included. LFPs I expect will slowly eat up the higher end of the market as it specifically improves and scales.

Issues with the discharge port I can see include the issues that:

- broken circuit means energy will struggle to get anywhere even if you wanted. Solving for this increases complexity substantially and non-linearly.

- maximum discharge rate of the cells could mean even at max output it could take well over 10 minutes to depower if not much longer (and discharge also has a trade off of them heating up accordingly at higher discharge rates)

Ultimately, you have a lot of embodied “loose” energy and once the reaction has left your control and the walls of the cell are breached (and enough of them), it’s more akin to a collection of little firecrackers that “combust” when they touch each other rather than a fire that can be “extinguished”. Just in slow motions because they are metals.

I always liked the chemical acronym LiFePO4... it has "life" right in the name!
Yeah, and you also underestimated the cost of the subsidies, overestimated the effect on climate, etc.
Isn't oxygen a requirement for combustion ? I've seen usage of blankets to cover cars, this effectively makes it so very little oxygen can go in and continue the fire, why not use that instead ?
Once a battery is in thermal runaway it will keep heating up, oxygen or no oxygen. Putting the battery in an oxygen-low environment would stop the flames but not the heat. You'd still be sat there waiting for the batteries to be done reacting before you can reintroduce oxygen.

Which is why firefighting battery fires is such an irksome thing. You can douse the battery and get the flames out, but hours to even days later the battery can have collected enough heat to reignite spontaneously.

terrible look for GM. Who is going to be the guinea pig for the next iteration of GM EV product?

Has any notable numbers of Teslas caught fire while charging?

2019: https://www.thedrive.com/news/28420/parked-teslas-keep-catch...

That said, they haven’t had to do a recall so maybe it isn’t as bad (or maybe GM is just more aggressive about recalls)

Tesla addressed a fire risk with a software update for thermal management.

Recall is only needed for hardware fixes or if there is no 100% deployed software update network for the cars.

Yes, pretty sure anyways, things do happen. The rate of Tesla fires is quite a bit less than gas powered cars though. Tesla has failsafes to prevent fires, but nothing is perfect.

Doesn't seem that unusual, I wanted to replace the door between my house and garage, turns out there's a specific building code specification for that door, because apparently cars catch fire in garages.

Always seemed weird to me that building codes require fire-rated materials for attached garages, but there are not smoke alarms designed to be installed there.
People aren't supposed to sleep in garages, so it's not required.

What wouldn't a normal smoke alarm work in a garage?

I had the garage floor poured so it sloped slightly down towards the doors. This is so:

1. rain doesn't come in 2. gas leaks go out 3. it makes it a lot easier to push the car out if it catches fire while you're working on it

The slope is not enough to notice unless you pour water on it.

It didn't add any cost, it's just one of those little things I delight in.

Of course it's a bitch pushing the car back in to the garage, but getting a running start usually does the trick.
I know that a lot of people on HN have a lot of money tied up in products made by a couple companies that start with T and end with A and this being the internet many of them are chomping at the bit to put down the big three but before you make fools of yourselves running your mouths you might want to know that:

a) Only two vehicles have had issues.

b) It's unclear yet whether there's a design, manufacturing or procedural fault or if there was human error when performing a recall repair on these vehicles.

As a Tesla fan myself, I was surprised at how fast comments critical of Tesla comments get downvoted. I commented on a thread a couple of weeks ago about Model S deliveries and quickly got multiple downvotes for what I thought was a pretty innocuous comment.
The best part is that I'm not even being critical of Tesla.

(and frankly I find the other brand's fanboys far more insufferable, the Tesla guys are at least aware they're driving beta tech with teething issues)

I think there's been enough battery fires so far across all manufacturers to recognize a few fires as an actual problem. At least that's my perception of recent, as I'm going through a recall for our Kona EV battery.
I was looking at buying a used (obviously) 2019 Kona EV and now not so much due to the battery problems I've read about.

Hyundai is also famous for adding a warning light in their gas cars that lights up to indicate a catastrophic engine fire is occurring.

I'd buy the Kona, they're really great car. I'd buy it again, personally. The recall is a mild annoyance and the risk is fairly low. Also the recall here is painless. And newer models aren't susceptible to the flaw in the earlier design.
I didn't downvote you, and I own a Tesla, but it's not surprising that someone using a throwaway, convinced that they're going to get downvoted, throwing smug opinions around, then editing their post to complain about getting downvoted, is, in fact, getting downvoted.
Same reason I wouldn't even think about a Powerwall (or anything like it, not picking on Tesla specifically) inside my house. If it could be in an outdoor enclosure a safe distance away, maybe.
I wonder why don't installers put those batteries underground? If you have a front/back yard is it safer just to borrow a battery there?
Corrosion
We have metal pipes underground for 10+years. Can't powerwall handle that? Also you probably don't need full enclosure to be metal, can be some corrosion resistant material, and only metal heatsink.
Pipes leak all the time from corrosion or get penetrated by tree roots or get punctured by construction or get damaged in earthquakes.

Lithium batteries do not need external oxygen to burn and toxic smoke coming up from your backyard with no way to reach the source doesn't sound like a fun proposition.

How about corrosion resistant enclosure? And even it will be damaged (probability is probably low) you can have a sensor to detect that and replace the enclosure. Batteries burning underground should be less a concern compared to them burning on a wood wall of a house. In case of an incident circuit breaker will turn them off, and batteries will burn themself w/o damaging anything else.
If you're gonna have a fire a hole in the ground is probably just about the best place for it.
Actually, it’s a terrible place to have a fire because you can’t get to it to put it out: https://www.nationalgeographic.com/adventure/article/140716-...
The edge case is not the rule!

Unless you happen to have a coal seam in your yard or live on a peat bog it's overwhelmingly more likely that it'll run out of fuel before it hurts anything.

People have been digging fire pits (or building up the ground to the same effect) for millennia because it makes fires more controllable.

Underground is not dry.
Can you have a sealed enclosure with some air presure valve?
It suddenly costs a lot more than the garage.
Powerwalls are not (or should not) be mounted inside. They go on an exterior wall.

Edit: Apparently they can be installed in some interior spaces. I stand corrected.

Tesla marketing literally says 'indoor', on the Design pane:

https://www.tesla.com/powerwall

"The compact, all-in-one construction features versatile mounting options for indoor or outdoor spaces."

Both the installs that I’ve seen are on the inside: one in a garage and one in a utility/mechanical room.
Plug-ins are the new pinto.
The Pinto was no more prone to fires than other vehicles of the time. Its explosiveness is a myth.

https://jalopnik.com/why-the-ford-pinto-didnt-suck-5785674

People always bring up the Pinto case as if the lesson is not to build crap that explodes when in fact the lesson is that PR and public perception matters more than the reality as the numbers lay them out.
It was though? Ford put the fuel tank in the crumple zone between the bumper and axle to maximize interior space. It was found to leak gasoline after a rear impact in every crash test they conducted before recalling it to modify the design.

It was by no means the most dangerous vehicle on sale in its time—that was probably the Volkswagen Beetle—but it was absolutely more likely to catch fire in a collision than pretty much anything else on the road that wasn't a literal fuel delivery truck.

So no, it wasn't necessarily going to kill you in a collision, but if it did, it was a lot more likely to be with fire.

It wasn't. Provide a citation, please. I provided one:

> "But despite the memo's cold calculations, was Ford characterized fairly as the Kevorkian of automakers? Perhaps not. In 1991, A Rutgers Law Journal report [PDF] showed the total number of Pinto fires, out of 2 million cars and 10 years of production, stalled at 27. It was no more than any other vehicle, averaged out, and certainly not the thousand or more suggested by Mother Jones."

So again, it was no more likely to catch fire than other cars at the time.

Mother Jones is rightly a revered publication now, but their reporting on this was seriously flawed to the point of being journalistic malpractice.

1991 was 13 years after the recall during which Ford went out of its way to reinforce the fuel tank to mitigate further criticism. There was a legitimate and well documented problem with the original design of the Pinto, and no article about how it's fun to race them now is going to change that. Here's an NHTSA report: https://www.autosafety.org/wp-content/uploads/import/ODIPint...

"In nine staged collision tests of 1971-1976 Pinto 2-door sedans and 3-door runabouts impacted by 1971 Chevrolet Impalas at closing speeds of 30 and 35 miles per hour, two tests resulted in fires. In all of the remaining seven tests, fuel tank damage occurred with fuel leakage rates ranging from 6 to 700 ounces per minute, with an average rate in excess of 240 ounces per minute."

Does having two out of nine low speed collisions result in fire seem normal? Does having nine out of nine low speed collisions result in significant enough damage to the fuel tank to cause gasoline leakage seem normal? I would certainly hope not.

The original Pinto design was flawed and dangerous, unnecessarily killed people, and Ford was rightfully and successfully sued for releasing it.

P.S. That Rutgers PDF in your citation is about the myth of the Ford Pinto legal case, which is that whole thing about how Ford allegedly calculated that it would cost less to settle on wrongful death lawsuits than to fix their cars, not about the safety of the Pinto. The link in the article is broken but you can find it here: http://www.pointoflaw.com/articles/The_Myth_of_the_Ford_Pint...

The contrived collision tests have become irrelevant because we have actual data of people driving the car.

It's like you're saying that a lab test is more relevant than years of field testing with millions of miles traveled.

> The original Pinto design was flawed and dangerous, unnecessarily killed people, and Ford was rightfully and successfully sued for releasing it.

Again: it did not kill more people than the average car at the time. That was my original contention, that it was not more likely to explode than any other car.

Could it have been safer if they had designed it more responsibly? Yes. Could those deaths have been avoided? Probably. And should Ford have been sued for that failure? Yes.

But that doesn't change the raw fact that it did not explode more than the average car at the time.

> That Rutgers PDF in your citation is about the myth of the Ford Pinto legal case

The Rutgers studies provides data, which are you are still ignoring, that showed the Pinto was 1.9% of all vehicles on the road and accounted for 1.9% of all fatalities.

If the lawsuit was suggesting that Ford was negligent for not getting that number even lower, then it's still a valid lawsuit. But I wasn't referring to whether Ford should've been sued, only that the "exploding Pinto" myth is a myth in real-world observational data.

The "contrived collision tests" are actual, standardized, government monitored collision tests, with verifiable results. Your "actual data of people driving the car" is not from the period when these claims were made. Again, the vast majority of early model Pintos produced that had not already combusted had at that point already been recalled.

The Rutgers report is about the lawsuit, not the safety of the automobile, as mentioned. It's also just an opinion piece. Though it does cite some data!

Can you check the PDF? Upon its publication the Pinto accounted for 1.9% of vehicles on the road, 1.9% of vehicles involved in fatalities, and 4.1% of vehicles involved in fatalities involving rear impact. So maybe the Pinto was actually overall a pretty safe car! Except for exploding upon rear impact. Which is, you know, the defect in question.

500 people died by being burned to death in a ford Pinto after an accident that otherwise would have survived.

They oftentimes couldn't escape the vehicle due to crumple zones pinching the doors shut.

Sure, the "exploding Pinto" is a myth, but only on a technicality. Those people experienced your real-world observational data first hand.

https://www.motherjones.com/politics/1977/09/pinto-madness/

The link, and quote, provided goes to a car website, which, unlike Mother Jones, didn't earn a Pulitzer for its reporting on the Pinto.

That site then linked to the PDF on a website that, at time of writing, returns 404. Some googling produced an article in the now defunct "student run law journal" that appears to be a hit piece or history rewrite. Perhaps it is an application for work at the law office of a major car manufacturer.

Once all vehicles are electric, would such fires spread in a parking lot?
The future looks green for insurers and the premiums.
I don't know the likelihood of EV's in the future using these Lithium batteries. You would think people would be working hard trying to find something more dense and more safe. It seems there'd be a lot of money in that, but what do I know?

If it is all lithium, well, make sure you have good insurance.

Aren't these issues inherent to being a battery?

All the reactants are stored in the same place, unlike a fuel, where you have to mix the reactants before there's energy to release

A battery is just a tepid chemical reaction that we can easily reverse, which is why we accept the poor energy density. If we could find a dense fuel that didn’t throw off a lot of carbon dioxide or methane, synthesizing that would probably be better.

If only storing hydrogen weren’t such a problem …

Yes. That's happened with fleet vehicles already. Five electric busses parked at a university in China had this happen:

https://www.youtube.com/watch?v=T71cVhxG_v4

Yikes. Thanks for the clip.

Hmm. About 250 million vehicles in the is. If you assume maybe 100 million parked closely in lots during peak load times. Wonder what the average daily odds of a battery fire are.

This’ll probably be one of those things society will figure out, but I certainly never considered this aspect of electric cars.

FYI: Most of the small electric planes mentioned in articles from 2017-2019 caught fire in the air, killing the test pilots.

https://electrek.co/2018/06/04/siemens-electric-plane-protot...

https://www.planeandpilotmag.com/article/electric-plane-catc...

https://www.rnz.co.nz/news/national/432551/in-flight-battery...

The 787 originally didn't use a complete battery box, and there was a major fire on board that grounded the 787 fleet for a year. The fire destroyed a rack of electronics, and was serious enough to bring down an airliner.

https://en.wikipedia.org/wiki/Boeing_787_Dreamliner_battery_...

In those cases, the batteries were all Lithium-ion.

Disclaimer: Did some work on a FutureCar/FutureTruck team back in the day.

GM: throwing shade on EVs since the EV1.

In my city you have to park in giant parking garages manned by valets that park the car for you. The garages hold hundreds of cars. There are more and more electric cars - including mine. Should be interesting if a fire blows up the entire floor of a parking garage one day, and that's caused by a malfunctioning EV.

And I was already getting stressed over the NY Post article about rats chewing soy-based wiring chemicals in Teslas. Of course, rats are everywhere here.

These electric car fires pose an interesting problem for communities that don't have a lot of water available. Putting out a burning electric car can easily take about 30000 gallons [1] of water. Putting out a traditional car only takes about 300 [2].

Car manufacturers need to find a way to get these batteries to easily extinguish before they can overtake traditional petrol cars without dry areas getting massively impacted.

I can see countries and states facing droughts putting a ban or extra tax on electric cars if manufacturers do not find a way to extinguish these fires. Rescuing people from electric vehicles is already complicated enough by the instructions for disconnecting the high voltage lines being different per model and per brand. I don't think it's reasonable for every fire department to be expected to have several shipping containers on standby to just dump and drown burning cars into.

This stuff makes me wary of the future of electric cars. Maybe the hydrogen people have a point, despite their clear connections to big oil.

[1]: https://www.popsci.com/story/technology/electric-vehicle-bat... [2]: https://www.nbcnews.com/business/autos/federal-regulators-wa...

For anyone who may had the same thought as I did.

There was a company that was trying to sell water tank services via a boeing 747.

"Reese says his SuperTanker can be dispatched from its Colorado Springs base to anywhere in North America in 4h 30min – or to almost any location globally in 20h."

"tanks able to drop 72,700 litres (19,200USgal) of water or 66,300 litres of retardant"

-https://www.flightglobal.com/flight-international/meet-the-l...

That company folded. Looks like there is another active one. https://www.coulsonaviationusa.com/

If you race your car, it's required to have an easily accessible battery disconnect switch. If I spend any time working on a car, I usually wind up installing one.
No wtf that's when the NHTSA and FTC needs to step in and tell them to force recall the cars and either offer them money back with interest or a new car.
Turns out this kind of thing can happen with ICE cars, too. In my case, due to some kind of faulty wiring harness. The best part is that the remedy wasn't actually available for 2 years or more after the issue was identified and announced. In the meantime, the only advice was the same as this: don't park inside.
they had fires just from a parked car that is off? that doesn't seem right.
Yes, in fact there were two different recalls, announced within days of each other, that both led to that result:

1. An electrical short can cause the parts within the PCV valve to melt, increasing the risk of a fire, even when the vehicle is not in use.

2. Wiring that overheats could cause the electrical connectors to melt, and increase the risk of a fire, even when the vehicle is unattended.