Hacker News Reader

Comment by 01100011 | Hacker News Reader

There seem to be two camps here:

- Bose is an evil megacorp who recklessly endangered its customers by selling dangerous and poorly designed products.

- Accidents happen and when a device explodes on your head you should accept it as a fact of life and move on. The chemical burns are nothing and OP is a pussy.

I think the reality may be different. My take on it is that Li batteries possess inherent dangers and it is incumbent upon manufacturers to take reasonable steps to mitigate the dangers while informing their consumers of the risk. Unfortunately, even when presented with the risks, most folks are unable to comprehend the probabilities and consequences and will nearly always choose convenience, features and value over the avoidance of danger. It seems we, as a society(not as individuals, as some will surely object) have chosen Li technology and deemed it safe enough to deploy ubiquitously.

I hope a review of this incident will shed light on the safety practices followed by Bose and will inform future engineers of unforeseen risks, giving us all safer, convenient devices.

Overall I find the discussions in this thread unbecoming of the types of folks I thought frequented HN. I hope we can all strive to elevate the dialogue and treat each other with a moderate amount of respect.

View on HN

> My take on it is that Li batteries possess inherent dangers

Yes, and most are entirely unaware of the dangers and best practices. I can see what may have happened is that even if the product was new, it may have sat in inventory long enough for the Li-ion cell to have self-discharged below 2.5V and remained below that critical threshold until purchased. Dendrites began forming immediately upon submitting to charge cycles. The dendrites shorted the cell causing thermal runaway. It could have burned the house down.

>Yes, and most are entirely unaware of the dangers and best practices. [It] may have happened even if the product was new [...] It could have burned the house down.

And yet, the calls to add labels that would allow consumers to tell which products contain lithium batteries at a glance have been derided here.

To add to this, my pair of wireless headphones can be also used with a 3.5mm cable, and I often do that, as it allows instant switching between devices. You can't tell they have wireless capability (or lithium batteries in them) just by looking.

According to comments, I'm the sole person on this planet who does read the instruction manual. And yet I am learning about thermal runaway in possibility in otherwise non-defective equipment from the HackerNews (thank you for explaining!).

Don't we want things to change for the better?

There's more than just the safety aspect, which, again, is to never allow your cells to drop below 2.5V, though dendritic formation is a function of how long the cell's voltage was below that threshold. At 3.6V, your cell is depleted, and I would expect at least smart devices will have low voltage detection and will shutdown at some low voltage threshold if low battery indicators are ignored. To be clear, it isn't the cell dropping below 2.5V that will cause the cell to turn into a bomb, it is afterwards attempting to recharge it that is dangerous. Stop using and never charge a Li-ion cell that's voltage dropped too low. Recycle it at a battery drop (most hardware stores have a battery drop).

The best practices are pretty simple. To get the most life out of a cell (and this is true of Li-ion, NiMH, NiCad, whatever), rest the cells for at least an hour after use and before charging, and after charging and before use. Adopting these rest cycles allows the cell to maintain its capacity. Abusing a cell by never resting it between use or charging will decimate its charge capacity and reduce the amount of amps it can provide compared to when it was new and unused.

Do not store Li-ion cells long term at full capacity, but instead at 3.7V. This is actually controversial, but allegedly (though I do not know why) storing at full capacity reduces the life of a cell.

And of course don't drop your cells, as it can and will cause internal damage, reducing the life of the cell.

>To be clear, it isn't the cell dropping below 2.5V that will cause the cell to turn into a bomb, it is afterwards attempting to recharge it that is dangerous.Stop using and never charge a Li-ion cell that's voltage dropped too low.

>Do not store Li-ion cells long term at full capacity, but instead at 3.7V.

These are great suggestions, and also they are absolutely impossible to follow given that we have absolutely no access to the sealed Li-Ion cells in devices that we buy.

How would the OP even know what voltage the cells inside his headphones are at? Even with replaceable cells, we can't expect people to have the tools to measure the voltage - unless the device itself displays it.

And even then, if we know that charging a cell after its voltage drops below 2.5V is dangerous, why do we allow chargers that do that (including devices with internal batteries) to be sold to the public?

With experience one can have an idea of what charge is left on a cell. One easy way to gain experience is with a decent Li-ion flashlight. It doesn't take a whole lot of tools to test resting voltage of a cell, a cheap multimeter will do. Modern chargers should be smart, have some function to automatically test for resistance, and refuse to charge a cell with resistance that is too high or if voltage under load is below 2.5V.

Frankly, I am a bit astounded these devices were developed. I'd recommend against wearing Li-ion cells on your head unless you really know the cells, know their age and history.

Again, as a tinkerer, I'm happily learning from your comments — thank you!

But the general public can't be expected to know all that without it being even in the booklet that comes with the product.

>Frankly, I am a bit astounded these devices were developed.

Well look at the comments here that say "risk of death is a part of life, deal with it". That kind of attitude is why.

Lihtium Ion cells are depleted at 3.0V, not 3.6V. The 2.5 is the lower threshold after which you should no longer use the device without replacing the battery. Most if not all BMSs on the market are well aware of these data points and will respond accordingly: hard shutdown at 3.0V, after that they will use a few uA until you charge them and around 2.6V they go into 'deep sleep' which means even indicators and other interaction stops until the device is recharged.

It is very hard to get a device to go below that 2.5V, unless you aim for it by leaving it for many months (and with slightly larger cells years), self discharge at that voltage is extremely low unless the cells have been damaged.

"Unfortunately, even when presented with the risks, most folks are unable to comprehend the probabilities and consequences and will nearly always choose convenience, features and value over the avoidance of danger".

Similar in regards to informed consent: if you're even told what the potential consequences are, a layperson may not, likely doesn't, have similar frame of reference to understand what say a potential "side" effect of a medication could be.

So how do we determine what's acceptable? What level of actually understanding, and perhaps confirmation of understanding, should be required? Unfortunately it's likely mostly those who were harmed that try to speak out, but as in the vast majority of cases they are the minority and their concerns drowned out - arguably only monetary losses via lawsuits being the impetus for organizations to change, however so long as it's extremely profitable with no jail time for executives who should be accounting and therefore be accountable.

>There seem to be two camps here...

Third camp reporting! Here's my take (as a Bose product user):

1) Bose QC-35 II headphones are a hazardous product that, in at least one instance, has spontaneously burst into flames and caused chemical burns while being operated according to the manual

2) This product does not come with appropriate warnings[0] that this is a possibility, while, according to comments here, hazards of the battery used are common knowledge among engineers

3) The product's manual instructs the user to take the headphones off if they experience a "warming sensation"[0], indicating that the engineers were aware of the risks, but neither the risks nor mitigation were not described in the instruction manual (compare this with the labels on something as common as epoxy resin)

4) The OP should report this incident to CPSC (as they did[1]!), since this is the body responsible for keeping track of such incidents and that will be able to act upon them if there's a pattern of them happening

5) The medical injuries sustained by the OP are unacceptable; and at the very least, Bose should pay for the medical treatment and resulting productivity loss, as they occurred at no fault of the OP, who trusted Bose's device to be safe to use as described by Bose. (Reminder for our friends outside the US: we don't have universal healthcare)

6) There needs to be some incentive for companies to ensure the safety of their products or actions. In the US, this seems to be (punitive) damages, whereas in Europe it's more through stronger regulations.[2]

7) If the OPs report is true, I also hope they sue Bose. It is important for all of us - people who use Bose products, their roommates etc - that they thoroughly check this incident and make sure it never happens again. They will not do it unless it costs them money and bad publicity.[3]

Please let me know what's so radical about it, as many people seem to get angry about it.

And fully with you regarding treating others with respect (particularly, it's sad to see the OP being called "hypochondriac" and their concerns dismissed as "paranoid"). It is, as you said, unbecoming.

____________________

Disclaimer: I love my Bose SoundLink Mini II speaker. I bought it, for myself, with my money, which was fully worth it. IMO, it sounds better than any other speaker of the same size that I tried.

That doesn't excuse this situation happening, nor how Bose handled it so far.

[0] Bose QC-35 II manual: https://assets.bose.com/content/dam/Bose_DAM/Web/consumer_el...

[1] As per https://news.ycombinator.com/item?id=29602614

[2] In the words of https://news.ycombinator.com/item?id=29603800

[3] In the words of https://news.ycombinator.com/item?id=29603321

Your phone, smart watch, smart glasses, extra battery pack, wireless earbuds, laptop, and essentially all other modern wireless consumer products can and do burst into flames on occasion. As a result, they are required to be shipped with hazardous material declarations like the UN3481 stickers you'll find on everything.

Consumers in general are unaware of the hazards and unwilling to learn, mostly because it's very rare for anything particularly bad to happen. You can find thousands of examples of dangerous situations here: https://www.reddit.com/r/spicypillows/

edit: as a tangential story, I once ended up at a battle robot live stream. Turns out lithium battery fires are a normalized part of that, with arenas filling with toxic smoke and volunteers with no PPE collecting actively combusting robots while breathing billowing clouds of thick smoke. It happened to have some affiliated moderators; after pointing out this was stupid and very, very unhealthy, they basically told me to shut up and donate PPE if I didn't like it. I thought it was a fascinating example of the disconnect between OSHA-regulated industry and end consumers.

>As a result, they are required to be shipped with hazardous material declarations like the UN3481 stickers you'll find on everything.

These modern hieroglyphics a)mean little to nothing, and b)stop existing once you remove a product from its packaging. That's not enough.

>Consumers in general are unaware of the hazards and unwilling to learn

Unwilling? Well, perhaps they would be more willing if the instructions that came with the product contained the information that you'd want them to learn.

According to this thread, this includes:

* Awareness of Li-Ion cell chemistry, normal operating voltage ranges, critical low voltage after which the battery should not be charged, and the possibility of formation of dendrites and runaway thermal reaction;

* Awareness that the said runaway reaction, in practical terms, means that charging a device with a "dead" battery can result in spontaneous combustion;

* Awareness that, given lack of access to internal batteries, and any information about its state, we solely rely on regulating circuitry to prevent that from happening, and that this circuitry might be faulty, resulting in the possibility of spontaneous combustion under normal use conditions;

* Awareness how to handle a Li-Ion battery that caught on fire, and how to dispose of a device with a burning battery safely, especially when it's operated in highly flammable environments (e.g. using headphones in a bedroom/in a dry field outdoors/at a gas station);

* Awareness that the battery catching on fire may result not only thermal, but also chemical burns if it is not disposed of in a safe manner while it's on fire - in addition to toxic fumes;

* Awareness of how to deal with thermal and chemical burns should they happen (e.g. for chemical burns, rinse with cold water for half an hour), as well as consequences of inhaling the fumes;

According to commenters here, all of the above is "common knowledge", "high school chemistry", "middle school knowledge", "easy best practices", etc. - which, at the same time, are virtually unknown by the general public (due to unwillingness to learn, no less).

Also, the manufacturers are under no moral or legal obligation to inform their users of these hazards and procedures, and this information isn't (and shouldn't) be included in product manuals.

Holy self-contradictory Jesus this thread has been quite a read.

>2) This product does not come with appropriate warnings[0] that this is a possibility, while, according to comments here, hazards of the battery used are common knowledge among engineers

should it? especially when the risk is so low and warnings like "contents may be hot" on coffee cups or " This product contains chemicals known to the State of California to cause cancer and birth defects or other reproductive harm" are widely derided?

>should it?

Yes, it should, because people expect wireless headphones to be as safe as wired ones, which have exactly 0% (zero, zlich, nada) chance of spontaneously catching on fire.

Wired headphones are still widely used, made, and sold. This would allow the consumers to make an informed choice, and will give the manufactures of safe devices (wired or wireless) the deserved edge in the market.

Also, consider the comments just from this thread about the risks:

>In the end you're carrying a high energy density power source on your head that would love to just catch fire. Don't they teach kids anything in chemistry these days?

>I mean I’m glad it didn’t happen to me, but a billion people are carrying around billions of Lithium-ion batteries. Those batteries sometimes catch fire because Lithium is highly flammable. It’s gonna happen.

>Nothing is perfectly safe, that is how lithium batteries fail. You should know so you can properly deal with problems.

>Well, at lease EVs are spontaneously catching fire.

>Everyone is aware of the risks, there were a bunch of stories about airplane cell phone fires a few years ago.

>lithium battery fires are really nasty, and that's why he got a chemical burn. LiPF6 is a contact irritant, PF5 it decomposes to is a gas, and also a respiratory irritant, and HF PH5 decomposes in the air is a dangerous poisonous acid. This is what everybody visiting a lithium battery factory is told on safety orientation. In case there is fire in the factory, run, preferably until you are few blocks away.

So, paradoxically, the risks are both negligible and inevitable, something that every consumer should be prepared for because "everyone knows" batteries explode, but no manufacturer should warn about, mitigate, or account for in the design because who'd buy their product then?

I hope you agree that the above is self-contradictory.

Now, ultimately, warnings are not a panacea, but they do work. At the very least, having their product labeled as "unsafe to use in bed" would make the manufacturers invest in research that would result in a better product that doesn't merit such a label.

> So, paradoxically, the risks are both negligible and inevitable, something that every consumer should be prepared for because "everyone knows" batteries explode, but no manufacturer should warn about, mitigate, or account for in the design because who'd buy their product then?

I think this can all be true. ER doctors say avocados are one of the most dangerous fruits because of knife accidents.

Yet avocados don’t come with any warnings or mitigations. And for the general public, you probably don’t know any one who was severely injured slicing an avocado.

Surely avocado farmers should be working on cultivars that have a small or no pit so people don’t get injured preparing these fruits? (I mean that only half sarcastically, I would enjoy a seedless avocado if it tasted the same.)

Another example might be bagels, which are manufactured products, hence more putative liability.

> “Americans ate an estimated 3 billion bagels at home in 2011, an average of about 11 per person ... in the course of slicing up all those bagels, almost 2,000 people cut their fingers so badly that they ended up in an emergency room.

> By the finger-cut-to-E.R. metric, that makes bagel-cutting the fifth most dangerous activity in the American kitchen.

https://www.newyorkerbagels.com/blogs/bagels/bagel-safety-do...

Why are we discussing this here?

A bagel sitting on a shelf isn't going to cut you up.

Bose QC-35 II headphones, apparently, may catch on fire and cause chemical burns when touched.

Analogies are great for illustrating a point, but if you say "this thing is just like that other one", you need to argue why that's the case first.

Product, in ordinary usage, might be implicated in injury to user.

I don't have an argument here, except that bagels might be a better example than avocados, because bagels are a manufactured product.

In either case, a knife is required for the injury to occur, so I agree that the comparison to spontaneous headphone combustion is not great.

> Bose QC-35 II headphones, apparently, may catch on fire and cause chemical burns when touched.

Not if never charged (IIUC). Of course they probably come with chargers and are never used without them, but does anyone buy uncut bagels and not cut them?

The charger needs not be present for injury.

If you really want to stick to the bagel analogy, it's as if the bagel stabbed you back with a knife after you cut it and set the knife aside.

I hope you would find such backstabbing bagels problematic.

Bagel is a bad example here if fingers were cut by a knife. Most people thought by parents how to use a knife safely, but probably not all. Thought I doubt that anyone would read a manual for a knife and then follow it.

I wouldn't mind a sign over the avocado bin at the supermarket warning consumers about knife injuries and an infographic showing the correct way to cut them. The avocado lobby has enough money to help fund it.

I kind of like the giant seeds. It reminds me that this bizarre fruit is a leftover from megafauna eating and pooping out the giant seeds over long distances (you need a big seed to survive dinosaur digestion)

Maybe, is there any evidence Prop 65 warnings have reduced people buying carcinogenic foods in California supermarkets?

I don't think you need evidence. People who are concerned about cancer would be looking for the cancer warning and avoid those products, just like people concerned about the rainforest will seek out Rainforest Alliance-certified products. People who don't care, don't care, but for people who do, they make their decisions based on the information provided to them.

And I just think it would be cool if we dropped little hints to people about the best way to do common things, like cutting an avocado. Imagine if the world were filled with little tips and tricks everywhere we went. Tiny changes, but added up, might elevate life as a whole.

I feel like the people we're arguing with are going around at night tearing down street signs:

>We have GPS, nobody reads directions anyway. You should know your way around these parts. Getting lost is a part of life, deal with it. Using a sextant is easy, just follow these 25 steps in clear weather, and you know precisely where you are, no need for "you are here" displays. Kids these days expect everything to be spoon-fed to them. Nobody gets lost anyway, why should tax money be wasted on this? etc.

The Prop 65 warnings apply to basically everything, so it's difficult to avoid them. That doesn't mean that the warning is bad, just that the threshold is too low.

Giving radiation warnings at the level of a banana would be useless. Does that mean we shouldn't use Geiger counters?

You seem to ask the question in bad faith, as clearly this is not the right place to find out these statistics. Google is your friend. There is evidence that warnings, in general, do work.

There is also evidence that wired headphones, which don't have batteries in them, do not self-combust and therefore will not require such a label. A label that differentiates products is certainly useful.

There is also plenty of evidence that most people are unaware of the potential dangers of lithium batteries.

There is no parallel with Prop 65 warnings here.

I believe it was giant ground sloths as the seed dispersal fauna.

>I think this can all be true. ER doctors say avocados are one of the most dangerous fruits because of knife accidents.

This analogy doesn't apply. Avocados don't lunge at unsuspecting people with knives while they're not looking.

I would say a step further: instruction manuals are inundated with worthless warnings, and there is no sense to the user which are real threats. If you believe instruction manuals, no device should ever be used for its intended purpose, all filters should be changed once every 14 hours, and anything with a fan should be placed on the floor in the exact centre of a 20'x20' room.

There is maybe a quarter page of useful text in the average 20 page manual.

Warnings in instruction manuals about novel threats are completely worthless.

>I would say a step further: instruction manuals are inundated with worthless warnings, and there is no sense to the user which are real threats...

This is not the case. Neither in general, nor, in particular, in the case of Bose QC-35 II headphones.

The worst danger the manual explicitly wans about is the potential of "warming sensation"[1], in which case the user is to take the headphones off. Nothing about an imminent fire hazard, or potential chemical burns.

Compare this to the labels on, say, two-part epoxy glue about the potential hazards, and what to do.

Finally, there are only three possible scenarios:

* Including this information in the manual and packaging will scare away people from buying this product because they are uninformed

* Including this information in the manual and packaging is pointless, because people don't care

* Including this information in the manual and packaging will allow people to make informed choices

Arguably, it should be include in either case:

- If people become scared, the manufacturers will be pushed to innovate on safety, which is a great outcome;

- If "nobody" reads these warnings, there is no reason to add one more about an actual danger that does happen, if rarely;

- Correct product labeling is what makes a free market work.

I urge you to peruse the Bose QC-35 II manual[1] before responding, and come back with the scariest warning you find there. It is a counter-example to all of your claims.

[1] https://assets.bose.com/content/dam/Bose_DAM/Web/consumer_el...

They're a direct consequence of the sue happy American consumer. It's all about legally covering your ass, not about actual risk.

There is no such thing as "sue-happy American consumer".

There is, however, an American consumer whose only way to get their medical bills paid in case they are hurt by negligence or harmful actions of a corporate entity, is to sue that corporation.

Europe has regulations. We have lawsuits.

Lack of regulation is justified by the possibility of taking the offending party to court. Stop shaming people who are doing just that.

> " This product contains chemicals known to the State of California to cause cancer and birth defects or other reproductive harm" are widely derided?

The problem with those warnings is not their existence, but the fact they are unspecific. If the ingredients on food containers just said "This contains an allergen", people would ignore them too. The way to make the California warnings useful would be to list which specific chemical is the problem, so that it is actionable.

Similarly, for the headphones case, listing the specific problem allows the user to take into account their specific situation (climate, usage pattern etc.) to decide if it's applicable to them. Obviously if the warning just said "this is hazardous", it's worse than useless.

> according to comments here, hazards of the battery used are common knowledge among engineers

How many people here are "engineers" (Licenced P.E.s with EE degreee)? Most people here string npm modules together and aren't qualified to assess LiON battery risks -- myself included.

Most licensed EEs are completely unqualified to weigh in on product safety and liability too! Licensed EEs sign off on building power, basically.

EEs that work on products fall under the "industrial exemption" - their employers are responsible for having appropriate processes in place for product safety.

>How many people here are "engineers" (Licenced P.E.s with EE degreee)? Most people here string npm modules together and aren't qualified to assess LiON battery risks -- myself included.

The point of the remark you are quoting was that it's unreasonable to think that the engineers that designed that product were unaware of dangers associated with this kind of battery.

I.e. if people here know about these dangers, so does Bose.

The "Bose is evil" camp are probably the same ones who would damn anyone straight to hell for ever giving money to them, simply because a percentage of their products are "overpriced" and "don't sound as good" as similarly-priced alternatives. Except when it comes to ANC, there really isn't a better solution for the price, I would argue maybe a single Sony model could even attempt to compete in the exact same space and price range.

>The "Bose is evil" camp are probably the same ones who would damn anyone straight to hell for ever giving money to them

I think the money I paid for my Bose SoundLink Mini II is very much well-spent, and I don't know of any other speaker of the same size which can beat it in quality of sound and battery life.

At the same time, some commenters here believe that wanting to hold Bose accountable means I have a bone to pick with them.

I feel like the problem here is that people think that the brands they use represent their personality. I love the product from Bose that I own, and that's why I want to see them held accountable for the self-combusting one.

That's to ensure I can enjoy using their products going forward, and recommend them without being worried about whether their newest gadget may set my friend's house on fire.

Reply on news.ycombinator.com