Extinguishing the EV Battery Fire Hype
spectrum.ieee.org
spectrum.ieee.org
As such there's some decent statistics showing that it's much more likely for an ICE car to start burning than an EV[1][2], up to 4-5x.
Our EVs are on average newer, for the last few years the majority of sales have been EVs. As such our ICEs are on average older. Interestingly the number of cars catching fire[3] hasn't increased substantially since 2016. The number of EVs catching fire has doubled since then, but the number of EVs on the road has gone up 5x[4].
Here in Norway new EVs no longer come with the "emergency charger" that hooks up to a regular 220V socket, as charging using regular sockets has been identified as a potential fire hazard.
[1]: https://www.motor.no/aktuelt/elbiler-brenner-langt-sjeldnere...
[2]: https://www.elbil24.no/nyheter/myten-som-nekter-a-do/7821704...
[3]: https://www.brannstatistikk.no/brus-ui/search?searchId=9B135...
> Our EVs are on average newer, for the last few years the majority of sales have been EVs. As such our ICEs are on average older.
So wait, do the stats show that ICE cars are more likely to burn, or are they just older on average?
However back in 2016, only about 4% of our cars were EVs[2]. So the 2016 car fire statistics is essentially just ICEs, which would include young ICE cars as well.
In 2022, around 20% of our cars were EVs. Given that the age of our cars has remained roughly the same, if EVs were as likely to catch fire as an average ICE in 2016, you'd expect[3] around 140-145 EV fires in 2022. Yet there were only 29.
Of course, that's assuming I did the math right.
[1]: https://www.ssb.no/en/statbank/sq/10090898
[2]: https://www.ssb.no/en/statbank/sq/10090899
[3]: 661 fires per 2.4M ICE cars in 2016 vs 564k EVs in 2022
I tried finding the age distribution per power train, but the closest I could find was per brand, which isn't all that useful. Teslas is a decent proxy for EVs, being EV only and most popular for a long time, but not sure I can find a suitable, representative proxy brand for ICEs.
> But MSB’s Per-Ola Malmqvist has developed webinars that explain how to safely put out battery fires. In a 2022 webinar, he described the tools and techniques that were used to put out a raging EV battery fire in 10 minutes using only 750 liters of water. In another webinar about EV fire suppression best practices, Malmqvist interviewed a firefighter from Vestfold Fire Service in Norway, where the extinguishing method Malmqvist recommends was tried for the first time in battling an electric-vehicle blaze.
Not surprising. However, I would have liked to have heard a better explanation of the problem about EV fires self-reigniting, which can happen hours or days later. They touched on it, but passed it off as lack of firefighter training.
https://www.cnbc.com/2022/01/29/electric-vehicle-fires-are-r...
Simply dousing the fire with water as mentioned in the article is not sufficient.
Putting it out the first time so that it can be safely moved is important.
Mere words do not quite convey the issue.
How does one "isolate" a car? Does one take it somewhere? Where? Are fire engines able to undertake this task? It's there a concern en route to the somewhere?
There have been some pretty horrendous chain reactions of EV fires in China. I'm not sure storing a bunch of already defective ones in various states of burn together is safe at scale.
https://www.theverge.com/2023/4/20/23691743/ford-f-150-light...
The cars next to it were ignited, but nothing on the next aisle. Probably 10 feet is enough?
But it’s pretty easy to solve now. There’s already a kind of kiddie-pool like equipment that can easily be put up around the car to keep the bottom flooded with water for a few hours.
What does "per 100,000 sold" mean? Is that simply per 100,000 cars? This metric makes no sense unless you age-adjust vehicles; most EVs are <5 years old. Meanwhile, people drive 20, 30 even 50+ year old ICE vehicles all the time.
And dismissing firefighters lack of knowledge and training and equipment to put out these fires is strange. It will be overcome, but it's a legitimate issue.
Their stuff is ancient, so much so that they need to have "smoke drills" to put their very complicated kit on when they enter a smoke diving situation. This is all because the US helmet is the same shape it was in the 1800s, because tradition or something. They just refused to modernise.
The EU version? four clips attach to the helmet, tighten, done in under 5 seconds.
https://www.firehouse.com/safety-health/ppe/helmets/article/...
The timeline of social adjustment to a technology is just as important as the technology itself.
Meanwhile there are shops over here where people ship their EVs from around the country because they know their shit. They have the know-how and tools to fix EVs for a fraction of the cost the Official Solution (replace whole $thing with brand new part).
TBH if you're in your 60's and a car mechanic, there's no point in learning EVs unless you're personally interested. There's gonna be enough work for the next 20 years easily just with ICEs
But recognizing that this is normal doesn't mean it isn't going to require work and investment to make it all pencil out. People - like me - who want to see these technologies succeed need to be just as attuned to this and not downplay it or take it for granted that people will come along for the ride if they don't yet see the benefit to themselves.
https://www.reddit.com/r/dataisbeautiful/comments/123niq7/ag...
e.g., https://www.jstor.org/stable/44631649
I'm not an expert on this though, and these are just my general impressions, which is maybe the point of the article. On the other hand, it doesn't seem to address the nature of EV fires in a scientific, public-health kind of way.
I think the thing most people are concerned about is a question like "how likely is my car to cause my house to burn down just sitting in the garage, while I'm sleeping?" Maybe the answer to that is the same with ICEs and EVs but that's the question that needs to be addressed.
Your general impressions don't reflect reality. Out of all the cars I've owned my ICE ones have been way more likely to burn down my house while sleeping than my EV. Your own article even starts off with "most occur in motor vehicles not involved in collisions."
The ICE car combusted while the engine wasn't running? Like, it was parked, they'd turned off the car as usual, and hours later it randomly combusted? Do you have any information on what the likely cause may have been? (e.g., hot weather? unexpected sparks?)
Here's an article on a recall Hyundai/Kia are going through right now: https://fortune.com/2023/09/27/hyundai-kia-recall-nearly-3-4...
They're recommending you park the vehicles outside and away from structures.
> the anti-lock brake control module can leak fluid and cause an electrical short, which can touch off a fire while the vehicles are parked or being driven.
> [...] Dealers will replace the anti-lock brake fuse at no cost to owners.
How can replacing a "fuse" prevent fluid leakage?!
In the mean time, swap the fuze, and while the car is going to become immobilized or throw ABS warnings when the leak happens its not going to burn someone's house down.
I'm not that illiterate about cars or basic physics. I was trying to figure out what actually went wrong in their situation. Because I'd never heard of any specific cases of these actually happening. Previously these sounded incredibly less likely to me than what I'm now discovering here.
But yeah, vehicle fires happen to parked cars all the time. As mentioned I've had multiple cars have "don't park in the garage" kind of recalls, one car has even had multiple (Hyundai). In the article posted above it opens with:
> WHILE SOME FIRES in motor vehicles occur as a result of collisions, most occur in motor vehicles not involved in collisions. These non-crash-induced fires are relatively frequent (one for every 1,000 registered vehicles) but in general are less hazardous to occupants and bystanders than crash induced fires.
It later states:
> They occur on an average of one out of every 1,000 registered vehicles and can result in significant property damage
Possibly one of the above plus gasoline vapors which are a lot more dangerous and explosive than gasoline itself and can be easily ignited by a spark not necessarily due to a short circuit.
That means the odds of it catching fire on your garage while you sleep are lower.
EVs by their turn are highly stressed while charging.
This is not backed by any evidence while the opposite is shown in the referenced paper. Most ICE fires happen when parked according to actual data.
Since ICE vehicles operate at much hotter temperatures, they inherently have a combustion potential due to leakages (brake fluid, gas, oil, etc). It's pretty common to have a bit of smoking from a fluid change, purely from accidental drips (even a little bit). If it's a continuous drip and hot enough areas such as exhaust, you can get a fire.
EVs just don't have anything that hot during normal operation. The operating temperature of both the electric motors and the batteries is much much lower than the temperature of engine operating temps (usually 200+F) or exhaust gas temperatures (something ridiculous like 1200F on catalytic converters, and 500F on exhaust piping).
I otherwise agree, but it's complicated because while what you say in the first sentence is true, it was true for ICE at one point too.
So for example, this is why gas containers have pressure releases, etc.
I mean, for an ICE vehicle you go to a gas station and fill it up. It's hard these days to think about how really dangerous that is because of how safe we'd be able to make it. It's much more dangerous than charging an EV car in theory.
Even storing the gas is dangerous!
That said, what you are pointing out is that the other difference is more around the mechanism. Gasoline can't ignite as a liquid. At high temperature it just vaporizes quickly into a gas and then ignites :)
As a gas, it has a lower explosive limit of about 1.5% and an upper of about 7.5%. Within that concentration level it is flammable. Outside of it, no.
Because you can get it to combust within this range, and we've become good at avoiding this happening in "normal" circumstances through safety mechanisms developed over the years.
This was not always true, and you saw more fires in lots of situations as a result.
EV fires on the other hand are usually self-sustaining chemical reactions[1] that got out of hand. Once triggered, they result in fire unless some safety mechanism stops them.
You can see this is really not dissimilar from ICE - we spend almost all energy/safety mechanisms on preventing the ability to cause a fire in the first place.
However, that said, these chemical reactions are more "omnipresent" than ICE for sure - once they are both "off", EV vehicles are more likely to explode than ICE ones.
A lot of this is the fact that they are not really "off" most of the time.
Regardless, however, we will do the same thing we did for handling of gas in general - we will figure out how to make that safer.
If we can make hand-filling your gas tank with an explosive fuel safe i've got faith we'll be able to make EVs sitting around doing nothing safe. We actually already can, just not at the energy density we want yet :)
[1] I understand that as a nitpick, so is burning gas, but let's leave this alone at this level :)
What’s interesting to me is that most people talk about NMC batteries in EVs, but from what I can tell LFP batteries should be safer and are becoming more popular. Would be good to see if there are stats to back this up.
I'd assume that average ICE car catching on fire is older that average EV just by EV not existing when ICE was made
> Furthermore, fires in electric cars are declining. The MSB says the number of fires in electric cars has been around 20 a year over the last three years
The number of fires should increase if age was a linear factor. Could be non linear of course.
article directly says that there's an exploding growth of new vehicles, thus "average car" hasn't aged yet
I'm not arguing against EVs being more safe, I'm arguing against awful statistics of comparing "new EVs against aging ICE" while only saying "EV vs ICE"
The difference is so large that it’d be incredible if age would close the gap.
New vehicles have some risk as well: I think defects in battery cell manufacturing can be a significant risk in the early years of EVs.
I've personally experienced laptop and cellphone batteries catching on fire and can see how something with 100x+ the capacity could cause some serious fucking trouble.
Anyone who has one of those 1kWh+ emergency power backup things sitting inside their house right now really needs to think twice about contingencies. Extend this thinking to an EV with 2 orders of magnitude more storage.
Can a gasoline car set itself on fire inside your garage entirely unattended? Sure. But I think it is much less likely to occur and the failure modes are more acceptable to me - I.e. I can inspect and anticipate if my gasoline car might be unsafe with more clues than around an EV car. You can smell gasoline. You can't smell a manufacturing defect in a battery pack.
You think. The point of gathering data and studying it is to attempt to replace our animal-brain intuitions with real evidence.
I honestly haven't looked into the evidence on this enough either way, since I haven't yet seriously researched purchasing an electric vehicle.
But I know that I have a big tank of gasoline sitting directly underneath a portion of my childrens' rooms, and a big pipe of methane running into my basement running to a couple little tiny flames that are always lit and sometimes turned into a big blue flame to heat the household's air, or a slightly smaller fire to heat its water.
I trust these clearly dangerous things to be safe enough, but I haven't ever researched the evidence on the safety of these either. The only difference is that I've always lived with these dangers without ever really thinking about them; they're just part of the water I swim in.
One day EV batteries will be no different. It's rational for novel dangers to receive more scrutiny than old ones, but new things become old before long.
Stored energy is stored energy.
You do have a bit of a point about large lithium power bricks. I trust those things less than I trust an EV, especially if they are the weird ChinaCorp off brand type from Amazon.
There's plenty of stored energy in a slab of solid steel, but that doesn't make it remotely dangerous. What makes lithium-batteries dangerous is the thermal runaway, enabled by the liquid capacitor.
Source? Are you confusing energy expended to make steel with energy being stored in it?
Here’s an example:
> Tension on the transmission wiring harness could lead to wire insulation pulling back from the electrical connector. As a result, water from external sources could penetrate the connector. The presence of water may create a short circuit over time. As a result, the short circuit could lead to thermal overload if the vehicle's ignition is off for longer periods of time. Subsequently, the risk of fire cannot be ruled out.
https://static.nhtsa.gov/odi/rcl/2022/RCLRPT-22V533-6872.PDF
Details can be found in the fire departments evaluation[1] of the fire, specifically page 28, where they note they did not find evidence of any thermal runaway in any EV.
They also note that cars with liquid fuel often have plastic fuel tanks, which can get damaged due to the heat and start leaking, spreading flammable liquid over a larger area...
[1]: https://www.rogbr.no/Rapporter%20og%20utredninger/Evaluering...
An example, how many decades of natural gas use were there, before we purposefully made it smelly? We've had more than a century of improvements to ICE cars, to learn how to make them safer. And even how to make safety systems and methods more cost effective. The issues with Pintos exploding upon being rear-ended, resulted of course in fixes, etc.
So I do wonder what will, eventually, come of battery tech. Say.. 40 years from now, or some such. I imagine it will be much, much safer. We certainly try to prevent such disasters, but it seems we're also commonly having an issue, people almost or sadly do die, and then we fix that problem.
Edit:
Googled, curious. Wow.
https://www.gasodorizer.com/gas-odorization-history/
New London School Explosion ... The explosion left behind a collapsed building, with as many as 295 deaths.
As a result of the disaster, the United States and Canada began regulating the use of odorants in gas.
This is the sad way it is, often. Everyone will say "Oh, it's fine!", sometimes with logic, sometimes with a desire to not accept the potential risk, thus feeling safer, and then a massive disaster occurs.
My guess would be some sort of global warming induced flooding event, with salt water, as a key for many cars at once going BOOM!
As such, it's important to take all possible reasonable steps to mitigate the risk. Those vehicles are still great overall, if only because, between air quality problems and exacerbated extreme weather events, fossil fuels cause deaths, injuries and property damage just through their normal usage. But the danger is there, and needs to be considered.
There is statistically one fire a week on passenger aircrafts in the US now. ⁽¹⁾⁽² ⁾ New York City is trying to deal with a large number of fires started by ebikes og escooters. ⁽³⁾⁽⁴⁾
To dismiss this problem by finding a fire that happened in Vietnam that was incorrectly attributed seems quite biased.
When it comes to the stats for cars, they are probably correct but somewhat misguided. Most electric cars are pretty new, a large percentage are nearly brand news.
To compare them to the entire fleet of fossile fueled car fires is a bit unfair. I think if you compare all new and brand new combustion engine fires the stats would look different.
May well be much higher for combustion engines, I dont know, but the comparisons used in the article are in my opinion biased.
¹ https://explore.dot.gov/t/FAA/views/LithiumBatteries/Inciden...
² https://www.forbes.com/sites/suzannerowankelleher/2023/03/03... ³ https://www.nytimes.com/2023/06/21/nyregion/e-bike-lithium-b... ⁴ https://www.theguardian.com/us-news/2023/jun/22/ebike-batter...
According to some, our neighbourhood should be a EV-fire hellscape and we should have blackouts every week. The roads should have to be replaced every month from increased wear. Nobody can go on road trips anymore. Yet everything is fine. Better than fine. As the statistics are pretty clear on: there’s fewer car fires when you have more EVs. Also air pollution is down
When commenting on this topic online it feels like we’re living in the future, trying to argue with people from the past. If you’re wondering what a future with lots of EVs look like, you don’t have to speculate. All you have to do is come to Oslo or Bergen and take a look around. Yet people talk with absolutely certainty about how everything will go to hell when there’s too many EVs.
(Yes, Norway has some qualities that makes the transition easier. Like a strong grid to start with, low speed limits and a strong economy. But then the technology was much more primitive and expensive when Norway got started. Also, Norway has a winter that is brutal on the range of EVs, and an insanely high share of Norwegians will drive hours to their cabins high in the mountains every other weekend in winter.. not the most ideal market for EVs)
My neighborhood isn't exactly new, it was mostly built in the 70s to very early 80s. There are several households on my street which I know have EVs. Given how many argue about the load on the grid you'd assume the power company would have had to massively upgrade the infrastructure here to support it. But in the end our actual peak power usage largely hasn't changed; we're just pulling a bit more power in the very late evenings and super early morning hours. Time when the grid around us has plenty of capacity.
Those batteries are handled, charged, and manufactured/repaired with much less care....
Car manufacturers actually care about reliability, because they're usually bigger corporations with deep pockets.
Pretty much anyone with a contact in China can start "manufacturing" white-label e-scooters from the cheapest bidder. Then just shut down the corporation and disappear if any issues arise.
Meanwhile car battery is safety related part covered by dozens of regulations. Some YouTube channel was showing them under direct fire for minutes and they didn’t ignite. Maybe Rimac batteries, I don’t remember.
A gasoline fire is a well understood event. Gas ignites and burns until there is no more fuel, or no more air. We know to extinguish this type of fire, we need to remove the air. Once the reaction has stopped, it would need another fresh ignition source to reignite.
A lithium fire is, on the other hand, is almost completely misunderstood by most people. It is a chemical chain reaction. It it an uncontrollable state in which the lithium becomes so hot that it perpetuates it's own combustion. You can remove the oxygen, but that's not the chemical that's keeping this reaction going. So the reaction will not stop. The lithium will keep heating itself, which will cause that heated lithium to heat more lithium, which heats more lithium, until there is no more lithium remaining.
It is disingenuous to compare apples to oranges like what is being done by comparing ICE car fires to EV car fires.
Lithium combusts in the presence of water, so fighting a lithium(not-ion) battery fire requires special care.