92 comments

[ 2.9 ms ] story [ 159 ms ] thread
Note: this fire was actually last year when there was a spate of EV fires, but was not reported publicly until today.

The big takeaway from these fires is that proper safety for EVs might soon mean not leaving them in garages, especially those attached/part of residential structures. Expect insurance companies to adjust home insurance policies accordingly to account for EV fire risks.

> The big takeaway from these fires is that proper safety for EVs might soon mean not leaving them in garages

Just add a fire suppressant system to the Powerwall it's hooked on.

Not sure it's as easy as adding a fire suppressant system to a garage.

In many tesla crash reports firefighters leave the batteries to burn for hours, as there's no nice way to extinguish lithium fires.

Uh, someone correct me if I'm wrong on this... but aren't most Lithium Ion battery chemistries self oxidizing?

If that's the case, I don't think a fire suppression system would help.

You're correct. They are. That's why you can't put them out unless you drop the whole thing in a giant container of water.
It's possible to put the fire out by cooling the battery below ignition temperature, and depriving the fire of atmospheric oxygen certainly reduces the combustion rate, but the result for firefighters is that an enormous amount of water is required.

Typical petroleum fires (gasoline, diesel) can be smothered with foam to put them out. No amount of heat will ignite these fuels without oxygen.

Smother a self-oxidizing fire does not extinguish the fire. To remove heat, firefighters spray cold water on the fire, which is incompatible with smothering it (the water and steam push foam away), so the fire has atmospheric oxygen as well as oxidant from the fuel. Dropping the car into a tank of water accomplishes both - the water tank is a heat sink, cooling the car, and it also smothers the fire and removes atmospheric oxygen.

-I recall having read of at least two incidents where Norwegian firefighters simply dumped a truckload of sand over the burning vehicle.

This happened in places where simply leaving it to burn out wasn't an option.

Crude, but effective, apparently. Massive heatsink+depriving the fire of any oxygen not supplied by the decomposing battery.

Having to run a charging cable to your car parked away from the house/garage seems like a non-starter especially in areas of the country where it snows.
Lol. People plug in engine block heaters all the time in the winter.
When there's a snowstorm, ideally I like to get my car out of the way of snow plows.
Sucks you have to park on the street.
I don't park on the street. I have a (long) driveway and parking area with garage. If my car isn't in the garage when plowing happens, I have a whole lot of manual shoveling to do.
This is throwing a big wrench into the idea that EVs could help stabilize the grid in the era of relatively unreliable renewable energy sources.

At least for now until better battery tech will emerge. Hopefully insurance will keep track of this and update their policies regularly.

I wouldn't go that far, I'm going to guess that charging to 80% will be enough to reduce the risk.
Dread to think of a Tesla battery fire in a garage with poor access for the fireworks.
In the short-term this is partly addressable by not parking EVs in garages, and perhaps long-term by new safety measures on both EVs and on garages to contain potential fires.

But I have a related question: Are power walls also dangerous like this? They are also large lithium ion batteries.

> and on garages to contain potential fires.

I really hope this isn't the way this is solved.

Construction fire code changes all the time. The ones around here specifically for attached garages have added more requirements within the last decade or so, for instance.
Building code already mandated the garage being in a different/delayed 'fire envelope' than the habituated portion of my home when it was built, so at least in some jurisdictions we're already there.

Thicker, fire resistant drywall. Spring-loaded fire doors to the main house that default to closed. No air vent passthrough from the garage to the house, and so on.

I have an 8' separation between the two and it's all brick, that gives me a pretty safe feeling.
Change garages to suit a defective, dangerous piece of equipment? You've got to be kidding.
I think the question is, can you ever get to the point where storing a lithium-ion battery that large and energy dense is 100% safe? We are clearly going to see a switchover to EVs. If that's the case, building codes may need to adapt.

We have tons of fire and building codes, and they are updated all the time. Why would accounting for EVs be any different than other advances that required updating of codes?

I believe Tesla pitches the Powerwall as being suitable for indoor and outdoor installation, but I was speaking recently with a solar professional who expressed horror at huge banks of batteries being installed anywhere other than the exterior of houses.
I wouldn't be against treating a powerwall install like ammunition storage in an armored vehicle. Firewalled from the house with fire suppression installed, blowout panels to the outside, the whole shebang.
Indeed, but that undermines the whole idea which is that it's a super cheap self-contained unit you can just install anywhere and all you need is an electrician with a hammer drill and a masonry bit.
Agreed. When I installed a large battery backed PV installation I set it up with the batteries outside in a separate concrete underground bunker dug into the side of a (very convenient) hill.
Lithium Ion batteries are dangerous in general. I think they are more dangerous than gasoline: if I puncture my gas tank with an awl it doesn't explode in a 2000degC fire for 4 hours. It is scary, but necessary to adopt new tech IMHO, kinda like how the first Model T's could break your arm when you started them (by turning the crank).
I would disagree about lithium ion batteries being more dangerous - gasoline is far more energy dense for one. A slower release is actually a good thing for safety compared to the alternative, a very fast intense one. Besides, just a single mode of failure is not an apt comparison.

- You cannot create a fuel air bomb by throwing lithium ion batteries into your basement and igniting it when the concentration is correct. - You cannot suffer carbon monoxide poisoning from lithium ion batteries.

Air and oxygen are omnipresent.

Sparks and fires are not.

Yes, anything with the same chemistry is also dangerous. I would hope Powerwalls would have better cell isolation as they aren't as constrained by weight and size as in a vehicle.

That said, it looks like Tesla is going with LiFePo for powerwalls which is a much safer chemistry.

Supposedly Tesla uses NCA for its cars and NMC for Powerwalls. Most other electric car manufacturers use NMC. However the properties of a battery vary greatly depending on the structure of the cell, and even more so on the cooling of the battery, so you should not compare them on chemistry alone.

The reason why LiFePo4 (LFP) is safer is because it doesn't exhibit thermal runaway and is less energy dense.

https://www.powertechsystems.eu/home/tech-corner/safety-of-l...

"Autopilot was not engaged during this fire, and the user was not supervising the vehicle in order to extinguish it. We've determined this fire to be the result of user error." - Tesla
I wish they provided more information on what caused the fire since this happened last year. Remember there is at least 250 gasoline car fires a day in comparison

here is the source https://cleantechnica.com/2019/06/03/500-gas-car-fires-per-d...

I have literally never heard of a gasoline car fire, or at least not spontaneously in a garage (in response to an accident, sure). Source for 250 a day?
https://www.autosafety.org/nhtsa-consumer-advisory-ford-crui...

>Failure to have the switch disconnected could lead to a vehicle fire at any time, whether or not the key is in the ignition, and whether or not owners use the cruise control system. The safety agency said the fire danger is present regardless of the age of the vehicle, and could even occur while the vehicle is parked and unattended. Several dwelling fires have been attributed to the problem.

It can happen. That one was particularly bone headed design, but yeah. (I am in no way arguing that an old recall could have been responsible for the GP's claim of 250/day, just that spontaneous ICE vehicle fires have and do happen)

Not saying it could never possibly happen, just the particular claim of "this happens to hundreds of people daily" when I have not heard of this across, um, I've been a driver for more than a few hundred days, made me doubt that it was that common even if it's a worldwide stat.
> Remember there is at least 250 gasoline car fires a day in comparison

Citation please

Quick search gives a study on a subset of car fires:

https://www.usfa.fema.gov/downloads/pdf/statistics/v19i2.pdf

"Each year, from 2014 to 2016, an estimated 171,500 highway vehicle fires occurred in the United States, resulting in an annual average of 345 deaths; 1,300 injuries; and $1.1 billion in property loss"

That's 469 per day just on highway, given the date range probably 99% of them on fossil fuel cars.

Car fires on the highway pose different problems and dangers from car fires in one’s garage.
> The issue is causing mounting concern as a number of electric-vehicle makers have warned owners not to leave the cars charging unattended in certain circumstances, or sitting fully charged in garages.

Wow.

* Don't leave your car fully charged? Stop and think about the UX there. That's a completely new paradigm: don't completely fill you car with energy? Why not just stop the charging at 90% and call that 100%? What should you leave it at? (Drone batteries should be left at 50% if not going to be used soon.)

* Don't leave your car unattended while charging? So I'm supposed to sit there and watch it charge?

That's sort of an unacceptable usage model, IMHO.

The UI is already there. Tesla doesn't charge to 100% unless you configure it for a road trip
Then something presumably isn't working right, because cars are still burning.
I would like to know the extent of these problem and if there are other solutions. Seems this person did charge to 100%. Maybe this should be disabled for garages
I believe OP is saying that 80/90% or whatever should be displayed as 100% to the user, the same way fuel gauges in ICE vehicles typically leave a large amount of fuel in the tank while displaying as "empty".

This is mainly a U.S. practice, as many European countries like their gauges to be as accurate as possible.

> This is mainly a U.S. practice, as many European countries like their gauges to be as accurate as possible.

-Add the Japanese to that list. My Land Cruiser's low fuel lamp comes on when there's 22L/5.75 US gallons left.

(comment deleted)
That is low for an off-road vehicle where civilisation could be quite far away. My Suzuki and Chevy off-road vehicles are calibrated the same. My Buick thinks low is 24 miles of range.
While that's undeniably true, if venturing off-road, I'd expect the driver to plan better than simply heading back when the light comes on.

(I'm probably - make that 'almost certainly' erring on the side of caution - having fitted a 90L/24USg auxiliary tank and also often carrying a jerry can or two when heading off into the boonies - heck, even my field kitchen runs on diesel!)

That's why you have five topped off jerrycans in the truck bed.
> I believe OP is saying that 80/90% or whatever should be displayed as 100% to the user

My BMW i3 does that. When my car is fully charged, it's almost exactly 80% of the actual full battery capacity. When it says it's empty it's about 20%.

I verified this using some REST API calls someone had figured out reported capacity (actual and available) in kWh and correlated with the 94Ah specification.

Not a Tesla user but how well it integrates with Siri? Sounds like type of thing you’d want good voice interface.

Changing AC settings while driving would be great too.

Tesla doesn't integrate with any personal assistants natively. There are 3rd party solutions (Alexa skills, etc.) available that can report on and control things ("What is the state of charge?", "Turn on the heater", etc.)

Generally, I find these solutions not very user friendly at all and not well integrated. And I'm personally a bit wary about the security of it all. Giving a 3rd party access to my calendar is one thing; letting them control my car is something different.

Well it’s changing your temperature or opening window at most, not driving off a cliff
Tesla suggests only 80/%90% anyways as complete fill/drain cycles degrades the battery faster.
I'm sort of perplexed that this is even a problem? Surely it's possible to dynamically disengage the charging mechanism once full?
The problem is that batteries themselves don't have a concept of "full". Measuring the state of charge in Li-ion cells is not easy as the voltage curve can be quite flat, so you have to use all kinds of heuristics to determine what a safe charge level is.
I see. Thanks. I take it thermal differential between battery and outdoor air won't work for some reason?
Batteries heat up depending on the rate of charge and while it is easy to measure and control to avoid damage from overheating, I don't believe it can be used to measure the state of charge. Charging at home in a garage would likely be using a much slower (and thus cooler) charging method than a commercial charger, so I don't think the fires are caused by the rate of charge, but perhaps by over-charging potentially worn or damaged / defective cells.
Are you sure that's correct? The BMS should have a high voltage disconnect and not allow further charging.
I'm sure it does - the problem is that the disconnect trigger voltage could be a normal state of charge for one cell but a high / dangerous state of charge for another.
i think most chargers (phone etc) have a "don't leave unattended while charging" warning as well just to cover their bases.
Again, people go for the Tesla story. What about this list?

https://allev.info/2021/07/list-of-known-chevy-bolt-fires/

Edit: It get it, here come the hate replies and down votes.

Yes I know the article "mentions" others, but the title is again about Tesla and not EVs.

I get it. The Tesla articles always get the talk though. I honestly hear "Oh, Teslas light on fire. Electric vehicles are bad." Ask about the Bolt "Those are electric? I thought they used gas."

Let's see how many comments this article gets about this fire compared to the Bolt. The original Bolt talk here from 9 months ago had 106 comments. Heck even the latest post where GM had to recall the Bolt, again, had like 2 comments. Let's see how fast this one grows due to branding.

It's so rare, but they (Tesla) get the real headlines.

You do realize there were articles about those as well, right? Articles specifically from WaPo.
Generally parked automotive fires are started by a 12 volt electrical fault that progresses to fuel lines and interior plastics and upholstery. This is not optimal but a lot easier to contain than parked lithium battery powered car fires which require 20 tons of water and several hours to extinguish, followed by removal to a safe place in case of reignition.

Accident impact damaged battery gang fires are a lot more difficult to control due to unstable stored energy

If you read the article, it also mentions the bolt.
Tesla is the market leader. More people have heard of Tesla. Tesla makes a lot of promises about safety.

For all these reasons, something involving a Tesla is news in a ways that something involving a rare car like the Bolt is not (but note that the article also mentions Bolt and Kona fires as well).

If Tesla can't take the heat, then maybe they should cut back on the self-congratulatory marketing.

Rudolf Diesel is referenced about all that is bad about his engine, similary Tesla will naturally receive most of the exposure. In addition WSJ is a finance paper, and tesla is by far the highest value EV manufacturer
I searched for your username in the Bolt fire comments and didn't see it. So your big problem is that that article got fewer comments than this one, while you comment on this article and not that one?
I don't mean to be a Tesla apologist, but what's the rate of cars with ICE engines catching on fire in garages?

I only ask because (n=1) I recall a major house fire across the street from me that started in the attached garage when the car caught fire.

The ultimate problem is the same: it takes a lot of stored energy to propel a car 300 miles, regardless of that energy source. Controlling that energy will be challenge regardless.

(comment deleted)
According to the IIHS[1], the S burns more often than than the average vehice, and as does the X.

[1] https://www.iihs.org/media/c93b98d8-6a7d-44a1-810e-4468ec539...

(comment deleted)
From tables 1 and 2, about noncrash fire losses / insurance claims from cars built in 2016-2018 if I'm skimming this right:

     IVY  RFC  Make and series
     36k  192  Tesla Model X 4dr electric 4WD
     43k  140  Tesla Model S 4dr electric 4WD
     50k  288  Dodge Charger HEMI
    508k  101  Ford Fusion
    151k   49  Honda Fit
     21k    0  Chevrolet Sonic
RFC = Relative claim frequency which has an asterisk in the document saying "100 = all-passenger-vehicle result". If 100 is all passenger vehicles... result? I can't parse this sentence, but I take it to mean that 100 is the average of all cars.

IVY = Insured Vehicle Years, basically sample size

Given that a car with 21k exposure years can have 0 claims, and there are also combustion vehicles with higher values, I'm not sure how statistically significant it is that these Tesla values are >100.

It sounds like "relative claim frequency" is a weirdly expressed ratio, ie Charger owners make fire claims at a rate 2.88x that of the average across all passenger vehicles.
I have had one ICE car catch fire (in the driveway, not garage, thankfully) but it was caused by the electrical system (looks like the alternator spontaneously shorted somehow). No fuel caught fire despite the fire burning for upwards of 15 minutes before the fire crew arrives.
This is terrible, obviously. Where possible, all EV makers should protect against this.

Gas cars may not catch fire when they sit in a garage, but they're not perfectly safe. My parents' friend left the car on one night, by accident. The carbon monoxide killed half the family.

There's low-probability tragedies around us, each a little different. There's no silver bullet.

leaving a gasoline car running is a foolproof and expected way to gas your room. you would not expect disaster when leaving an EV charging
This is a fine hit piece coming from Jeff "can't compete with SpaceX" Bezos' WaPo.

It starts with the headline:

> It’s a risk many automakers are taking seriously.

Implying Tesla doesn't take this seriously. And the argument: other car makers recalled cars for fire risk.

What if the reason is that Tesla didn't find a defect that warrants a recall?

If we assume that regulatory bodies are not somehow favoring Tesla, more recalls by other car makers show that Tesla cars are safer in that regard.

It would also imply that Tesla took this issue seriously before they released the cars, not after.

And also this gem:

> Tesla in the past has argued that its cars are a tenth as likely to catch fire as gas-powered vehicles, drawing on data from the National Fire Protection Association and U.S. Federal Highway Administration. > Still, safety experts note that the fires can burn more intensely and last much longer.

So, the government data show that Tesla cars are safer than gas cars wrt. fires. WaPo doesn't dispute that.

Still, they manage to spin that superior safety of Tesla with some shade.

> The Washington Post has documented at least five fires

You don't say. Five out 1.5 million cars.

> Tesla had come under scrutiny over concerns it allegedly manipulated battery software in older vehicles to lessen the risk of fire.

So when Tesla releases a software update to lessen the probability of fire (something GM also did recently after many Bolt fires), WaPo says it's "manipulating battery software".

This whole article is just trying to make people think that Tesla is uniquely bad at fires when the reality is that not only electric cars are safer than gas cars AND Tesla cars are safer than other electric cars.

So if you're supervising your vehicle being charged and it catches fire, what do you do? Are you stuck waiting for the fire brigade anyway?
Yep but you can call them early rather than finding your whole house in fire and then calling them.
not much you can do really since the standard way to put out an EV fire is to let it burn completely
Tesla detected charging problem and send a message, but a battery pack crammed with temperature sensors didnt notice being on fire?
> Gasoline driven cars don’t catch fire in the garage when they’re sitting there. And that’s the difference,”

Lol. I was 14. I was working on something when I was suddenly startled by a really loud noise out of nowhere.

Our neighbors Caddy Escalade spontaneously caught fire. Out of nowhere.

> Gasoline driven cars don’t catch fire in the garage when they’re sitting there. And that’s the difference,”

Lol. I was 14. I was working on something when I was suddenly startled by a really loud noise out of nowhere.

Our neighbors Caddy Escalade spontaneously caught fire. Out of nowhere.

That said, a car that combusts out of nowhere is terrifying as hell