Hacker News Reader

Comment by asafira | Hacker News Reader

Can anyone on hackernews comment on the specific issue in the Note 7? And why was it (seemingly) easy to screw up the replacement phones as well?

Are we taking for granted the lengths that other hardware manufacturers go to to make sure such issues don't occur with their batteries? (At least, those innovating in regard to battery capacity, chemistry, and shape)

View on HN

Indeed, it just seems inconceivable that a company with Samsung's experience and institutional knowledge about Li-ion technology could so thoroughly and irreparably botch something like this.

Li-ion in 2016 is, outside of the occasional 3-5 years-away breakthrough story, boring, a mature technology. One would think the best practices for safely building power supplies around it would have been settled on years ago.

Or: Am I wrong and there are in fact several competing standards in the industry right now? Is there actually some present innovation in Li-ion that Samsung may have utilized, in retrospect prematurely, in the Note 7?

While li-ion is widely used in practice, I don't think it's fair to say it is a mature technology (as a sibling commenter points out, there are still lots of research papers coming out in the field) or that it is well understood.

For example, Boeing had problems with battery fires on the 787 in 2013 and 2014. "The causes of the battery failures are still unknown."[1]

[1] https://en.wikipedia.org/wiki/Boeing_787_Dreamliner_battery_...

Li-ion is a widely deployed technology that isn't very good. Like coal.

This battery chemistry is inherently fragile. Drive a nail through a Li-ion battery and it will explode. Tesla had to put a titanium plate under their battery after some fires from punctures by road debris. If a Li-ion battery overheats, it will go into thermal runaway and increase its temperature until it catches fire. Overcharging alone is sufficient to do this. It takes about six safety devices to make a Li-ion battery reasonably safe. Leave some of them out, and you get the hoverboard debacle.

There are safer battery chemistries, such as LiFePO4, but you give up about 14% energy density.

> Drive a nail through a Li-ion battery and it will explode.

It's misleading to say that it "will" explode. As the following video shows, they don't always.

https://www.youtube.com/watch?v=pqPFuzRIANs

That's LiPo though, not Li-Ion

There are far more boring things that have created immeasurable problems, around 2001/2 capacitors began leaking and rendering almost every electrical device that was fitted with useless after around a year of use. By the time the problem had been discovered, the capacitors were in widespread use all over the world.

https://en.wikipedia.org/wiki/Capacitor_plague

That was not a slip - that was clearly intended failure.

Apparently it was because they used an incomplete copy of an electrolyte formula developed in Japan, but didn't understand the chemistry of the formula properly.

https://en.wikipedia.org/wiki/Capacitor_plague#Industrial_es...

Isn't the problem here that they don't know what the issue is? Or that there are a multitude of issues? I've yet to see a statement from Samsung that wasn't vague or unconvincing.

(And in at least one case, it seems they're more concerned about covering up their incompetence than explaining their incompetence: http://www.theverge.com/2016/10/9/13215728/samsung-galaxy-no...)

If they do actually know what the issue is, they certainly haven't actually told anyone. They've only given generic and sort of hand-wavy "we fixed it don't worry" kind of explanations.

Well today's statements seem fairly straightforward - don't use it, don't try and exchange it, and to retailers - stop selling it.

If I had to bet it'd be that they don't know what the issue is, given the PR damage of combusting replacements.

You're right, their statement now it's pretty clear. Initially, things were worse. They give out conflicting information to different people, no one was really sure if it was a recall or how to handle it. That was weeks ago.

My father-in-law is a PhD chemist that reads various research papers for fun. I don't pretend to understand the field, but he has talked about Li-ion batteries some.

His main concern about Li-ion is that it is a still heavy researched field. There are lots of papers and studies coming out about the material and how it can be used in application. The volume of papers around the topic is a sign to him that it's not as well understood of a material as the public believes.

Couldn't this be because Li-ion is currently a hot field with lots of commercial applications?

I'm not a chemist (my dad was, but got out of the field right after I was born), but that's what I see in computer science research. When OOP was hot we saw a huge volume of papers in subtyping systems, faster method dispatch, design patterns, etc. When XML was hot we saw a lot of papers on tree-diffing, compression of tree structures, etc. After BigTable, Dynamo, Riak came out we saw a lot of papers on distributed systems and CRDTs.

It makes sense that as a technology catches on in industry, the amount of research grants related to that technology increases, which incentivizes scientists to study that topic in more depth.

You need to keep in mind that Li-ion does not describe just one type of battery it describes the specific method by which a battery operates (intercalation of Li Ions). There is continuous experimentation to improve the performance.

Additionally the materials which make up the battery can vary. The cathode of a Li-Ion battery is typically some type of lithium intermetallic compound (typically Lithium-Cobolt based) and the anode can vary (from simple graphite to intermetallic compounds to more exotic things like Sn nano-particles, polymers etc).

In 2010 my undergraduate thesis in materials engineering was on developing a prototype anode made from graphene dispersed Tin-Cobolt nanoparticles. This was back when graphene was first discovered and was being pretty widely hyped as a 'wonder material'. It was hoped that by dispersing the metal particles in graphene matrix we could better accommodate the mechanical stresses that come from cycling the battery (when you intercalate and deintercalate Li ions there is a volume expansion and subsequent contraction - this is what causes battery to degrade after charge/discharge cycles). Results were mostly inconclusive largely due to unreliability of synthesis technique we used for manufacturing graphene - Electrostatic Spray reductive Precipitation (ESRP).

But yes Undergraduates are messing around with Li-Ion technology its not surprising corporations are too.

Any recommended resources (textbooks, course lecture notes/videos) for someone with moderate science background (say intro college courses in chemistry, electromagnetism, circuits) to learn about batteries?

https://www.amazon.com/Principles-Electronic-Materials-Devic... This is the textbook I used at Uni. It's a little pricey I remember it being pretty good.

I talked to an acquaintance at Apple about the new iPhone recently, and the Galaxy Note 7 topic came up. I wasn't aware that the problem hasn't been publicly revealed. So, here's what I remember of that conversation; though maybe it is just speculation and hearsay.

According to said acquaintance, the Note 7 was designed with profile requirements that limited the thickness of the battery. The battery designers had to make the battery thinner than what was previously considered safe. Lithium batteries (technical portion that I likely didn't understand correctly) are built in layers and each layer has to be separated from each other. To thin the battery profile, they minimized these inert layers between the lithium layers. However, they didn't make the battery thin enough. So, when it is installed, the rest of the phone compresses the battery and compromises those thin layers around its edges. (I am guessing here.) When the layers touch that is completing a circuit (or some chemical reaction) which is the reason the phone is producing so much heat.

I'd be interested to hear from someone that could verify these speculative details.

The conversation I had was a couple weeks ago, but I did my best to explain the details to the best of my ability (from memory and on a topic that I have a minimal education).

Reply on news.ycombinator.com