5 ms·
Let’s Talk About Capacitor Failure
- Golfkid2Gadfly 8y agoAfter a recent problem with failures of electrolytics, after a overhaul that included changing all electrolytics, the problem was tracked to stock storage. The manufactures specify a 2 year shelf life, but advised this could be extended by regularly charging the capacitors.
- hoorayimhelping 8y agoDiscussion from yesterday: https://news.ycombinator.com/item?id=19600852 https://news.ycombinator.com/item?id=19600852
- snarfy 8y agoThey are the only thing not solid in your solid state devices. I've successfully repaired 5 out of 6 dead monitors and TVs by replacing all the big capacitors. No diagnostics skills required. Just replace the big caps.
- Cthulhu_ 8y agoI'm confident this is an intentional act; using substandard or slightly the wrong capacitors to ensure they'll die just outside of the warranty period. I had the same thing with a monitor of mine, it started dying after a good three years. My dad soldered in some new capacitors, good ones that can take more voltage or something (as opposed to not enough voltage like the older ones) and it's still good, almost ten years later? I don't even remember. However, this is not true for all monitors; this was a relative budget monitor, Fujitsu I believe. Not long before it started acting up, I invested in a (twice as expensive) Dell monitor, and it's not given me any problems in the nearly 8 years I've had it (just looked up the receipt, bought 23 September 2011)
- qntty 8y agoBuying the cheapest part that will last longer than the warranty period is the standard practice at most companies. It's not malicious, it's just what happens when life of a product isn't something that consumers want or are able to judge.
- JimBrimble35 8y agoWait, are you blaming consumers for the short life of the products they buy? Sort of a chicken and egg problem I guess. What came first, the short lived product, or the consumer's desire to buy something newer and shinier. People definitely want to buy new shiny things, but I fully disagree that product cycles and lifespans are the result of that desire. Companies have positioned themselves to scale uncontrollably at that cost of product lifespans. They make more money if people buy a new thing every 18 months, so they make the thing fail every 18 months instead of making something that has a longer term life. This practice can only be characterized as wasteful and greedy. Spending less on components that are guaranteed to fail _is_ a malicious act. Apparently we're so far down the rabbit hole now that there are actually people who think that planned obsolescence is a feature. You're blowing my mind here in the worst possible way.
- qntty 8y agoEverything is guaranteed to fail eventually. At no point in the process of designing something does an engineer or manager decide whether to make something that's going to fail or something that's going to last. You make something that you know will fail after n years. n depends on how much you are willing to spend on the most failure-prone parts. What's a reasonable value of n? Well, it depends on a lot of things, doesn't it? You say that spending less on parts you know will fail is malicious, but everything fails eventually. If you know that your resistors are likely to last 100 years, and your capacitors (even the best ones) are like to last 20 years, is it malicious to not spring for the resistors that will last 200 years? You have to make practical decisions about the level of quality that you're aiming for at some point. I know that planned obsolescence is a thing, and it should be illegal (in certain forms). How common is it? I don't know. Do you? Maybe I should have said "it's not necessarily malicious", because it's not. You can make a cheap product just because you think there's a market for a cheaper version of a good product. Does that mean you want the thing you're making to fail? Not necessarily. Maybe you just want to serve a market that wants a cheaper version of something. I sometimes find myself wanting to buy the cheapest version of something because I only need it for a brief period of time. Is this wasteful? Sometimes. Should it be illegal for a company to sell me a cheap version of a thing? Maybe for environmental reasons, but not for consumer protection reasons.
- fogetti 8y agoWhile I agree with what you said I would warn everyone who decides to pursue this endeavour and has no experience that capacitors can store a good lethal dose of voltage for quite some time. Also I enjoy this YouTuber's explanation about capacitors quite a lot. Uncle Doug is a terrific presenter: https://youtu.be/l-u8J-yh9ZA https://youtu.be/l-u8J-yh9ZA
- ungzd 8y ago(Note that there’s lots of information in comments section)
- blakespot 8y agoThanks -- yes, I hope people read the comments on my post (OP) and leave their own experiences in the comments. The purpose of my call for advice was to turn that post into a go-to for people searching the web with capacitor concerns. (The entire blog is a retro computing blog, up since 2004.) Some GREAT comments came with yesterday's exposure. Great to see.
- contingencies 8y agoPretty interesting stuff. Out of interest, is there an overview resource specifically focused on longevity expectations that is relatively comprehensive for common modern components?
- JimBrimble35 8y agoSomething to consider regarding longevity expectations for things like capacitors would be the input tolerances they're designed for. In many cases the capacitors that fail don't fail because they have reached the end of their physical life. Instead the manufacturer chose a component that could only handle the lowest nominal load not accounting for things spikes in electrical input due to normal power system anomalies. A number of years ago I pulled an otherwise perfectly good Samsung 46" TV out of the dumpster. While researching how to fix it I found that if they had simply installed higher rated capacitors to begin with, the TV would not have stopped working.
- dajonker 8y agoAbout 15 years ago I had a motherboard with several "bulged" capacitors. I think at that time I wanted to upgrade anyway and just bought something new, but I read lots of people had similar issues around that time. Rumor was that some manufacturers used stolen plans to build these capacitors, and those stolen plans actually contained some errors. The related wikipedia page: https://en.wikipedia.org/wiki/Capacitor_plague https://en.wikipedia.org/wiki/Capacitor_plague
- jonawesomegreen 8y agoI think the most interesting thing from the discussion on this yesterday [1] was the implication of industrial espionage in the problem from an article on wikipedia [2]. > Industrial espionage was implicated in the capacitor plague, in connection with the theft of an electrolyte formula. A materials scientist working for Rubycon in Japan left the company, taking the secret water-based electrolyte formula for Rubycon's ZA and ZL series capacitors, and began working for a Chinese company. The scientist then developed a copy of this electrolyte. Then, some staff members who defected from the Chinese company copied an incomplete version of the formula and began to market it to many of the aluminium electrolytic manufacturers in Taiwan, undercutting the prices of the Japanese manufacturers. This incomplete electrolyte lacked important proprietary ingredients which were essential to the long-term stability of the capacitors and was unstable when packaged in a finished aluminum capacitor. This faulty electrolyte allowed the unimpeded formation of hydroxide and produced hydrogen gas. [2] [1] https://news.ycombinator.com/item?id=19606691 https://news.ycombinator.com/item?id=19606691 [2] https://en.wikipedia.org/wiki/Capacitor_plague#Implications_of_industrial_espionage https://en.wikipedia.org/wiki/Capacitor_plague#Implications_...
- honestoHeminway 8y agoIts also very neat, that placing a capacitor near a heat source, one can move the bell curve of failure, to a peak just after the end of warranty.
- JimBrimble35 8y agoYou're a bad person and you should feel bad. You won't, but you _really_ should.
- qntty 8y agoWhen disagreeing, please reply to the argument instead of calling names. "That is idiotic; 1 + 1 is 2, not 3" can be shortened to "1 + 1 is 2, not 3." https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
- JimBrimble35 8y agoThat is idiotic
- sctb 8y agoPlease don't keep going; we'll ban the account and we'd rather not.