15

According to Gitnux:

“Every year, 1.5% of lithium-ion batteries are linked to overheating, explosion, or fire incidents.”

On the face of it, this is worrying—especially since overheating may go unnoticed and lead, via thermal degradation in a positive-feedback loop, to conflagration. (Li-ion batteries generally do not explode; but they do burn with great intensity.)

Does anyone know of an authoritative source for this? My guess is that the 1.5% has been overstated by a huge factor: perhaps a thousandfold.

9
  • 18
    Just had a look at a random other page on Gitnux. gitnux.org/candy-crush-statistics. The paragraphs read like they were written by ChatGPT. Also at the bottom of this page the FAQ also mentions frequency of battery fires, and contradicts the line you quote. Mar 2 at 14:43
  • 31
    Yup: gitnux.org/editorial-process Our internal algorithm organizes these statistics into themes ... The articles are created using our internal database, with the assistance of artificial intelligence and our editors. IOW a human might have looked at it before it gets published. Mar 2 at 14:49
  • 26
    If that's true, given the number of lithium batteries in my house I'm surprised I've survived this long.
    – Schwern
    Mar 3 at 19:14
  • 18
    The good news (grin) is that other bots will read that page, "learn" that, and report it as true in turn. "AI" is getting closer and closer to mankind, capable of repeating false information without any discrimination.
    – jcaron
    Mar 4 at 12:06
  • 3
    "overheating" seems a bit undefined - guessing they've included any time a battery gets excessively warm while charging.
    – Criggie
    Mar 4 at 20:45

1 Answer 1

71

On the face of it, this is worrying ...

On the face of it, this statistic is ridiculous, and it does not jibe with other statistics on the same page. The web page also claims

In 2020, the US Consumer Product Safety Commission reported an average of one battery fire every two days in the U.S.

In 2019, there were 360 million personal devices in the U.S., almost all of which used lithium-ion batteries. (This number has only grown since then). That corresponds to 0.00005%, not 1.5%. About 270 people in the U.S. are struck by lightning every year, which is more than one every other day. You have a greater chance of being struck by lightning than having your cellphone's battery catch on fire.

Note also that this 1.5% statistic bunches together "overheating, explosion, or fire incidents." Leave your cellphone flashlight on and put it in your pocket (or butt dial the flashlight while the cellphone is in your pocket) and it will get slightly warm. Strictly speaking, that's an overheating event. I myself have butt dialed my cellphone's flashlight, and yes, my cellphone got warmish. Conflating overheating events, the vast majority of which are mild, with explosions and fires makes for a highly dubious statistic.

I'll next look at electric vehicles. IEEE Spectrum reports that

In the United States, according to a 2023 study citing recent data from the National Transportation Safety Board and the Bureau of Transportation Statistics, gasoline-powered, internal-combustion engine (ICE) cars were involved in about 1,530 fires per every 100,000 sold. On the other hand, pure electric vehicles (meaning those powered only by batteries) were involved in just 25 fires per 100,000 sold.

The IEEE Spectrum article goes on to look at statistics from Sweden where 40% of the new vehicles are now EVs. The IEEE Spectrum article reports

that in 2022, there were only 24 EV car fires in Sweden, representing 0.004 percent of battery-powered cars there. For cars running on gasoline or diesel fuel, the fire rate was 0.08 percent, or 20 times the frequency.

10
  • 11
    This already answers the question nicely, but maybe point out the information about the source uncovered in the comments, i.e. that the source is a click-mill based on LLM?
    – Nobody
    Mar 3 at 13:24
  • 15
    @Nobody First off, it doesn't really whether it was a human or an AI that pulled a statistic out of thin air. It was pulled out of thin air, period. Secondly, the question as written by the OP doesn't address that issue. That was a comment that I did not feel at all compelled to address. Mar 3 at 16:21
  • 15
    @Nelson Who knows? The author of the blog in question did not clarify the meaning of "overheating". There are several other "statistics" on that page that are highly dubious, for example that "48% of cell phone fires are caused by lithium-ion batteries." I suspect that the correct statistic is well into the 90th percentile. Cellphones do catch on fire (extremely rarely, but it does happen). In the very rare events where they do catch on fire, it's almost always the battery that is to blame. The 48% is a made-up statistic -- as is the 1.5% statistic. Mar 3 at 17:24
  • 9
    A lot of the page is actually concerned with fires at waste plants. The website is very unclear, so I'm not going to make any excuses for them, but actually I wonder whether they're referring to batteries discarded in landfill or other waste treatment facilities. Firstly, the batteries shouldn't be there in the first place (so that massively reduces the pool of candidate batteries to combust) and secondly, I could imagine a battery being smashed around in a processing facility could be expected to pose the level of risk they state. Still, badly worded, even if that is the intended context
    – roganjosh
    Mar 3 at 19:08
  • 6
    @roganjosh A lot of discarded batteries don't even make it to the waste plant: A vape battery the size of an adult's thumb caused a recycling collection truck to catch fire in Melville Mar 3 at 21:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .