Hacker News new | past | comments | ask | show | jobs | submit login
As Batteries Keep Catching Fire, U.S. Safety Agency Prepares For Change (npr.org)
86 points by kawera on Nov 27, 2016 | hide | past | favorite | 75 comments



We need to convert over to lithium iron phosphate (LiFePO4) battery technology. Stores slightly less energy. More recharge cycles. Slightly more expensive. No thermal runaway problems. Does not blow up or catch fire even if a nail is driven through it.

Anything bigger than a phone should go this route. Boosted skateboards and the new generation of UL-approved hoverboards already have. Electric bicycles, which have a tendency to catch fire, need to.


That's a good heads up, thanks. Went in and read the intros in a couple of Wikipedia pages:

https://en.wikipedia.org/wiki/Lithium_iron_phosphate_battery - has the numbers specific to LiFePO4

https://en.wikipedia.org/wiki/Lithium-ion_battery - discusses the various variations of Li batteries. Mentions that "Handheld electronics mostly use LIBs based on lithium cobalt oxide (LiCoO2), which offers high energy density, but presents safety risks, especially when damaged." and goes on about the alternatives.


Many Black and Decker and DeWalt pro power tools use lithium iron phosphate batteries, and have since about 2008. Neither manufacturer advertises this. We're not seeing reports of power drill fires, which are devices that get hard use and many recharge cycles.

DeWalt had a recall about back in 2000 due to battery fires triggered by a charger that didn't turn off properly [1]. They apparently decided to switch to lithium iron phosphate, and no further problems were reported. If power tools were still catching fire, there would be Youtube videos. There are some videos of fires of early DeWalt models.[2]

[1] http://www.dewalt.com/support/safety-notices-and-recalls/201... [2] https://www.youtube.com/watch?v=8JJZfjzwuvM


The state of the art when it comes to LiFePO4 cells that you can find on the market right now has 2 to 3 times lower energy density by volume and weight (depending on form-factor).

Personally, I would prefer the increased energy capacity of traditional Li-ion packs, and when properly designed they are extremely safe too.


Can we please just bring back removable batteries?


That will not prevent batteries spontaneously combusting.

Replaceable batteries will only address the pain of safety recalls and increase the risk of fires due to consumers choosing to buy cheaper off-brand batteries.


Actually removable batteries should be less likely to spontaneously combust. Most battery failures seem to be caused by short circuits due to compression or flexing. Removable batteries are mechanically stronger because they have integral cases instead of just relying on the device casing for protection.

Personally I would never buy a cell phone with a built in battery. I always have spare batteries for my phones and frequently swap them out. It sure beats being tied to a charger, or plugging an external battery to an awkward and fragile USB connector.


You state absolutely valid facts, but removable batteries carry one huge additional risk: You can now go purchase a no-name cheap explosion-prone chinese one, which your phone manufacturer (barring recent incidents) most likely would have protected you from, with their QA.

Granted, you can do that anyway today if you go to an unauthorized shop for a replacement, or even get it yourself via Amazon/Ebay & co., but the general availability of shitty batteries for retail purchase would increase the size of the problem by an order of magnitude.


You can still purchase and connect "shitty batteries" (aka power banks) to your phone's USB plug with, potentially, same consequences. The only difference is that responsibility using external batteries shifts from a phone manufacturer to the user. I'm yet to see valid pro irremovable battery arguments from the perspective of a user, not manufacturer.


Actually I had completely forgotten about battery packs.

Point taken, absolutely. Hopefully the crappy ones are simply overstating their capacity instead of actually trying to reach it within volume/safety constraints (and cutting corners to do so), so they won't be as explosion prone - but indeed, valid point.


You are correct on both accounts, but I still think it's an overall safety benefit in the case of a recall. Portable dongle batteries have the same potential problems, so it's shifting the danger from one location to another.

My statement, though, is not entirely to address just safety, but also planned obsolescence, and consumer convenience. I've always carried a spare battery since even my HTC WinMo "smartphone" days. Same with my PowerBook.


I would prefer water-resistance to removable batteries...

(I don't think it's that simple, but given the choice, I would take the tight seal)


The Samsung Galaxy XCover I used to own managed both water resistance and a removable battery.


How would this resolve any issues?


It would have saved Samsung billions of dollars if they could have just replaced the batteries and not had to buy back 2.5M phones.


Don't be ridiculous. They would still have to recall every device, how else could they do it?

You can't send plain faulty batteries back.

And they still don't know why it happened, They tried 2 different batteries from different manufacturers and it was still exploding.


Er... isn't it obvious? They could mail you the replacement battery?


How did you figure out that the battery is the problem? Samsung has yet to figure out a root cause, so it may be mechanical design, or ...


No, I mean, instead of recalling 2.5m phones, you replace 2.5m batteries. Which you can do. Because they're replaceable.


Samsung tried shipping different batteries in the replacements but those still acted up. It's pretty unlikely replaceable battery which have done much.


Sure, but that doesn't change how much easier, cheaper, and less invasive to customers the replacement can be when you can take the batteries out.


No, it would ultimately have caused more fires with no way to blame them on anyone in particular. If we're going to treat batteries as safety-critical components, then it's counterproductive to encourage end users to replace them with the cheapest ones they can find.

Mobile phones with replaceable batteries aren't coming back, and good riddance.


Wouldn't a better solution be to just make sure dangerous batteries don't get sold? Just regulate them for safety like tons of other dangerous goods already are


After the first recall, when a few of the replacement phones caught fire again, Samsung couldn't even reproduce the issue in a lab. Regulation isn't going to prevent bad things from ever happening. You need a fallback plan.


So your solution is to say "no, only large companies are allowed to pick batteries". Well, the Note 7 seems to show how good an idea that is. Plenty of devices have replaceable batteries without them exploding, and if anything, a replaceable battery would let you say "hey, this battery is getting old, let's replace it" rather then continuing to use a device with an old, possibly failing battery.


Well, the Note 7 seems to show how good an idea that is.

Yes, as a matter of fact, that's exactly what the Note 7 debacle demonstrated. There was a quality problem that could cause fires. There was a single party who could be expected to take responsibility. Finally, there was a coordinated worldwide recall.

Problem solved, system working as intended. Sucks to be Samsung, of course, but the end result will be safer batteries for everyone.

Plenty of devices have replaceable batteries without them exploding, and if anything, a replaceable battery would let you say "hey, this battery is getting old, let's replace it" rather then continuing to use a device with an old, possibly failing battery.

Sorry, that's not going to happen. Get used to disappointment.


Or, the manufacturer of the batteries could recall them.

Also, plenty of phones have replaceable batteries, including new models, so I'll just keep buying them. Means I won't be buying Samsung, but that's just an extra bonus.


Yeah, it's sad. I'm fond of my Galaxy Note 4, and plan to keep it running as long as possible -- yes, the battery is replaceable, so I should be able to get a few years out of it -- but once it dies I won't want to buy another Samsung, and nobody else is making a phablet with a stylus.

Maybe some other manufacturer will see the opportunity Samsung has given them to break into this niche.


Well it would have likely solved the Note 7 recall, and any futures battery recalls. It would also allow a user to remove an unsafe battery so that it's not in use while waiting for a replacement.


I thought the problem with the note 7 was with the power controller, and not the batteries -- didn't phones continue catching fire after the first recall?


There is still no root cause for the Note 7 issue, apparently.


Didn't the phones with replacement batteries also catch fire?


Less than .01% of the phones caught fire. The recall is crazy.


If the Note 7 hadn't been recalled then eventually many more would have probably caught fire. We'll never know what the cumulative failure rate would have been if the devices had been left in the field for years. Eventually people would have been killed.


The fraction of phones that caught on fire was similar to the fraction of hoverboards that caught on fire. Normal products with similar batteries have a much lower failure rate.


.10% within a few weeks of release.

Without a recall, that number keeps on climbing and nobody has any idea where it stops. 1% of phones exploded? 10%?

You obviously can't unexplode the ones that exploded already, but a recall is about the ones that haven't exploded yet.


When you decide whether a risk is unacceptable, there are two parts to consider: the likelihood of the event, and the potential worst-case scenario should the event come to pass. .01% is a very low likelihood, but the worst case scenario that one of these phones maims the user is an extremely bad outcome. In this framework the recall is the most sensible thing to do.


Do you have a source for that number?


No, because of reliability. Either you have a connector or a spring neither is good for reliability


When was the last time you've heard of a phone that failed due to a broken battery connector?

Phones are already designed as throwaway trash to be discarded in 18 months (and if you somehow keep it for longer, the software industry will happily make it almost unusable within the next 2 years). Reduced reliability of a battery connector isn't a meaningful topic, and it never was - not even in the old dumbphone days.


My Galaxy Nexus would often shut down due to a loose battery connector. It was a widespread problem and the workaround was to fold paper or tinfoil at the bottom of the battery to push it against the connector.


I've personally had two phones fail because of the battery connector. Anec-data for sure, but those spring clips are not robust at all.


One of the reasons I am getting an LG V20.


Can we just go back to the 80's when you could chain smoke your way through duty free cigarettes whilst reading a highly flammable and freely provided newspaper, pissed, oh, and flying at 70000 feet, twice the speed of sound on Concorde? Check in was so much quicker then and nobody had exploding batteries.


Failure rates for rechargeable Li-ion batteries are on the order of one in 10 million cells.

http://cen.acs.org/articles/91/i6/Assessing-Safety-Lithium-I...


"Failure" including everything from fire and explosion to peaceful, boring death.

At a rough guess, the number of cells that fail excitingly are probably a fraction of those that fail overall.


One in a million odds breaks down when you're talking about a population size of billions. A couple hundred cellphones every year catching on fire is still pretty noteworthy.


If that's noteworthy, what does one call a class of product that causes >1.25 million deaths per year? (https://en.m.wikipedia.org/wiki/List_of_countries_by_traffic...)

Definitely worth fixing if it's not too hard/expensive, but probably not worth spending much time worrying about it.


How many cells in a laptop? I would assume 3 at the low end, but it can go much higher and many people have multiple laptops, cellphones etc. So, 20 per person on an aircraft is probably a solid round number.

How many people in an aircraft say ~200. How many aircraft in the air say ~5,000. Now that 10 million is down to 2-3 per however often those 1 in 10 million chance is supposed to stand for.


Replace aircraft with car and continue this conversation.


Cars typically have battery packages that can easily survive a cell failure without anyone noticing.


If your phone battery explodes and starts a fire you have better odds surviving a dive out of your car as opposed to an airplane.


You must have a pretty big phone to take down an airplane. If you are driving the car you are much more likely to loose control and cause deaths then a phone bursting in an airplane causing anything more then nasty burns.


Fires in a luggage compartment may be an issue even if they start small.


The unfortunate reality is that battery and solar panel manufacturing processes are dirty chemical affairs, even with exploding batteries aside. Just look at the process for solar panel manufacturing......

------------

1. Solar panels are made in China, Malaysia, the Phllippines, and Taiwan. (I think around 40% comes from China)

2. Solar panels are made from Quartz.

3. Quartz comes from mines, is abundant, can be found everywhere, and causes silicosis for miners.

4. Quartz is turned into silicon. —metallic grade silicon. We use this to harden steel. (requires a lot of heat) output = CO2 and Sulfur Dioxide. — not too bad. ;)

5. To purify the silicon into polysilicon you end up mixing hydrochloric acid with metalic-grade silicon = trichlorosilanes.

6. The trichlorosilanes mix with hydrogen that you add to the process.

7. Then bam, you get your polysilicon, and as well some nasty stuff called silicon tetrachloride.

8. For every 1 pound of polysilicon you make, you make 4 pounds of the nasty stuff-silicon tetrachloride.

10. Okay, so some green fanatics might tell you that you can just recycle this silicon tetrachloride stuff into new polysilicon because it requires less energy, but that’s unlikely because it costs a lot to do so.

11. If you’re a prudent manufacturer, you say “Fuck it.” and you just dump the silicon tetrachloride in an oil well.

12. It acidifies the soil and demolishes any water nearby.

13. Maybe one day we’ll manufacture the solar panels with ethanol instead of chlorine-compounds. — this would forego the nasty stuff — silicon tetrachloride.

14. That’s just the process to get polysilicon, there’s still a ton more to do and a ton of risks afterwards. But anyhow, let’s progress.

15. There’s got to be a better way, right?

16. Well there is, or so the hippies will proclaim.

17. There’s thin-film solar cells.

18. Most thin film cells are made from cadmium telluride and copper indium gallium selenide — often called CIGS

19. Usually you’ll layer these and splice in some cadmium sulfide.

20. Cadmium is carcinogenic.

21. You don’t want to be dumping cadmium all over junk yards, but that’s what happens to post-consumer waste.

----------- I'd even go as far as to say the nuclear disposal waste risks are an order of magnitude safer than a Solar future marked by silicon tetrachloride toxicity.

If there are any ChemE's in here, I'd love to hear their take on it. I'm polymer & Systems engr, so I don't go deep enough into the mfg process.


> Quartz comes from mines ... and causes silicosis for miners.

Quartz is not mined in a way that causes silicosis. It's so abundant that you can literally just pick it up off the ground, you don't need to go underground.

Silicosis comes from mining other things, and crushing rock in order to do so. Also from manufacturing that grinds or crushes rock.

> some nasty stuff called silicon tetrachloride.

Which if you simply mix with water makes hydrochloric acid and silicic acid, both of which are harmless in the environment, and are only a very short term problem when in high concentration. You could mix them with some water and drink it and it would not hurt anyone.


The market dumps today.


Would you have a source (book or webpage) to recommend if I wanted to get a better grasp on the current solar panel manufacturing process and its environmental hazards?


I'm going off memory here. eek. :(


Literally everything here is scare tactics and if you have a chemical sciences degree you should know better.


The Galaxy Note 7 recall is reported to cost "at least $5.3 billion"[1]. Even at a chaebol as large as Samsung's, that's big enough to get people to notice.

That cost, more than the recent election, will bring about change for the better.

[1] http://www.latimes.com/business/technology/la-fi-tn-samsung-...


Change for something that occurred in less than .01% of the devices? It is incredible the kind of irrationality that is brought about by fear. It's the same reason that, here in Australia, there is talk of culling sharks after a single person is attacked.


Would you fly to Seoul from Australia if there was a .01% chance of your plane catching fire? That's assuming just one phone on your flight.

.01% is crazy high.


You just went from "the risk for any given Note 7 to catch fire is 0.01%" to "the risk for any given Note 7 to catch fire in the time it takes to fly from Seoul to Australia is 0.01%". Those are very different numbers.


> Would you fly to Seoul from Australia if there was a .01% chance of your plane catching fire?

Given the sheer magnitude of all the other things that may kill you before you even arrive at the airport, let alone the very act of hurdling through the sky at over 650mph in a thin aluminum tube filled with 36,000 gallons of highly flammable/explosive fuel... a phone catching fire is really not much of a factor when deciding to book a flight.


(devices that did explode + devices that would have exploded in the future) = risk

Over three years what percent of the devices would have had the issue?


So the takeaway is, batteries will continue exploding because Trump won the election. The whole "Trump is responsible for everything bad" thing is getting ridiculous.


Not as ridiculous as your tortured interpretation of the story. Given the GOP's stated hostility to consumer regulation and the fact of significantly lower agency budgets during a previous period when the GOP held both White House and both Houses of Congress, observing that operating budgets may shrink at this agency is entirely rational. See, I was able to construct an argument without even mentioning the President-elect, based solely on past behavior of the soon-to-be majority party.


The article barely mentioned Trump at all, and basically just said that things "might change". That is self-evident with a new administration coming in.

Regardless of that, the article describing the organization's role and approach was interesting.


The article barely mentioned Trump at all

Oh come on, the spin in the story was much worse than that. The direct quote from the chairman of the CPSC was: "My hope is now with the election and potential leadership change here, that that work is not scuttled".

Using loaded words like "scuttled" makes it quite clear that he's throwing shade at Trump.


Does quoting a person relevant to the article who has an opinion spin the story? I'd think that's the journalist doing their job: reporting what those related to the topic said.

There can be bias in the choice of story, or the choice of quotes, but sometimes the story is that there are opinions that disagree, which seems relevant in this case.

There have been a lot of concerns with changes that Trump might make. He himself has been very clear that he's going to make changes. In particular, Trump has been vocal about reducing regulations, and as I understand it, the CPSC is a regulatory agency. Seems pretty straight up to me.


You're reading a lot into Kaye's comment. I'd imagine he would have made the same remark about pretty much any incoming conservative administration. And maybe even some liberal administrations, since weakening consumer protection laws can certainly cut across party lines when the money lines up.


The prez isn't in charge of budgets. They have been cut over the years anyway. So I don't see any great changes there. He could put a crony in charge, someone would apply the science selectively, but there are legal bulwarks against this. They couldn't, for example, place restrictions on only foreign goods. That wouldn't fly constitutionally. And if they just stopped testing anything, then companies that product dangerous products would be hit with lawsuits. Frankly, most would rather have the governments looking at the safety of products that they would jurors. A mandatory recall is always better than a mass class action.


> He could put a crony in charge, someone would apply the science selectively, but there are legal bulwarks against this

There are some, but past Presidents have had great influence over regulatory outcomes.

> most [companies] would rather have the governments looking at the safety of products that they would jurors.

That would seem rational, but that's not how many companies think. For example, Wall Street, despite recent experience with catastrophic consequences, still actively resists regulation.

That's also a good example of the influence of Presidents. Bush deregulated Wall Street; for example, the SEC is supposed to be the consumer advocate but Bush put Wall Street insiders in charge of it (IIRC). Obama restored at least some regulation.


This (the CPSC) is the agency that should focus on lead painted baby toys and BPA-tainted toothbrushes, but instead mostly takes away well-loved toys like lawn darts and little magnet balls. Things have gotten to the point that I reflexively feel the urge to quickly buy anything they ban. The batteries might be worth banning, but I sure don't trust the CPSC to determine this.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: