Electrolytic Capacitors – an Electronics Industry Scandal?

For repairing my own devices, I usually scavenge capacitors from an old device.

If I can’t do that, or I want to do a slower, more robust repair then I will buy on ebay. It’s not always easy to know which supplier offers a good balance of cost and quality (including spotting fakes) but it seems to work for me.

I would not be upset if common electrolytic capacitors were excluded from this, especially because appliance manufacturers might not have access to the same economies of scale as an electronic component supplier and therefore might not offer good value anyway.

My 2 cents.

1 Like

Hello dear Ugo,
I think the most difficult issue is to find the bad electrolytic capacitor, for this I think that the "Nuova Elettronica "ESR meter project should be useful.
They are generic components, their values are clearly written as number of microfarad, you don’t need a schematic to find if the capacitor is the problem, if you can measure the ESR value.
In TV and other big electronic items they are often in the first circuit to create several voltages, sometimes high voltages, and one important thing is to take care to avoid injures or to damage tools.
Often the bigger problem is to open the device without create damages. After this step Electrolytic Capacitors are the worst component, with a medium lifetime of 10 years, so someone as first step of a repair changes the bigger ones.
I think that the fact that they are economics, essentials and fast-aging are all nice characteristics for a vendor, not for the users.
:cry:
Ciao,
Sergio.

PS
Have you ever seen this site?

1 Like

To weigh in on this topic; electrolytic capacitors are not new and are found in virtually every device that handles mains voltage (possibly very simple devices like blenders and hair driers excluded). They have always had a limited lifespan due to their construction technique (electrolyte liquid in/on a medium, contained in a shell). It’s commonly known for instance, that caps should be replaced in valve amplifiers “every ten years”. In reality this would typically mean refurbishing a 30 year old amp requires a recap; if you come across a 50-60 year old amp, it’s likely been recapped once in the 80s or so.

The above is intended to make 2 points - 1: in typical conditions electrolytics don’t die after 2-3000 hours, they will last 10-20 years happily. 2: it’s part of ordinary maintenance to check capacitors and replace as necessary. We simply don’t tend to see/own electronic devices with a lifespan in decades anymore.

In my opinion the issue is not the capacitors - they’ve been the same for 70+ years, and almost all circuits require them (so changing to another product would impact billions of products). They have standardised sizes, ratings, etc - so should be easy to replace.

The fact that so many devices fail due to bad caps largely points to the fact that devices need to be easier to open and diagnose, and consumers need to be willing to do so. They should be the equivalent of “My car’s headlight isn’t working, I’ll check the bulb and then the fuse”. I don’t think there’s a problem with the design of the components or their use.

The comments on ongoing fallout from the capacitor plague (must be almost 20 years ago and still causing issues!) are very interesting though. Of course it’s in the interests of sustainability to make and use longer-lasting components, but I think we’re stuck with electrolytics as a category.

N.B. tantalum capacitors were more popular in 60s and 70s circuits but have fallen out of favour for reasons mentioned including issues when they fail, consistency of spec over time, and also cost.

2 Likes

This provides a concise explanation of the one capacitor failure issue and something I see quite often a repair cafes - the industrial espionage paragraphs says it all!

As for manufacturers knowingly producing short lifetime components well I don’t believe reputable manufacturers would. I’ve used and have electrolytic capacitors that are 40+ years old and are still working fine. Failure is more likely related to poor circuit design considerations and keeping mass production costs down (ie profits up).

For example, if you use a 50V capacitor on a 48V power supply rail then it is being used near it’s limit. Voltage spikes and surge currents may stress it and shorten its life. In critical systems all components are de-rated so eg a 50V capacitor may be rated for use only at 25V to ensure a robust design. Any robust circuit design must consider these factors.
The problem with de-rating is the higher voltage rated component may be more expensive, spread this across a mass produced items and costs can rise.

This article explains de-rating and uses capacitors as an example:

In aerospace systems (spacecraft for example) all components are de-rated, some quite significantly to ensure good lifetime probabilities since repair once in use is not possible!
NASA / ESA carry out extensive tests and research on component lifetimes.

3 Likes

Thank you, @Benski. I agree with everything you’ve said - but if capacitors are the equivalent of a car headlight needing replacement, shouldn’t they be made just as easy to identify as faulty, and to replace? Of course, this would cost more, and we now live in a world in which companies have no incentive to add to manufacturing cost - an issue much wider than capacitors!

My objective in initially raising this issue was to increase consumer awareness of how low-cost electronics is likely to be short-lived, and to encourage people to pay more for quality, where as you rightly point out, it is likely that manufacturers will derate their components more conservatively. It is a sad fact that even as an expert it is impossible to assess the degree of derating when buying new electronics, so we have to rely on well-known manufacturers’ desire to protect their brand reputation. This is a very imprecise way to establish quality!

2 Likes

Glad this has led to such continued comment and discussion. I’m continually learning. Also reminding ICYMI that our partners working on regulation in Brussels are interested in the matter.

And here’s an important reminder that @Chris_Moller is presenting on this topic at Fixfest UK on Saturday, 11am UK time.

Please join us for his demo and commentary, and be prepared to come with your questions and comments.

https://app.konf.co/event/2G5gAFnTn/session/255

My company makes electronics and we pay a great deal of attention to electrolytic lifetimes. The actual lifetime is extremely temperature-dependent, those figures like 3000 hours are at the maximum rated ambient temperature in combination with the maximum rated current. If you look at the full data sheet you will see curves showing much longer lifetimes at lower temperatures and currents. Most electrolytic caps in cool places with low currents last decades. Some of the caps in switch-mode supplies are working hard with high currents and in hot places because of the other heat-producing components. A good designer looks very closely at these conditions and ensures the parts are selected to give a sensible life expectancy (in a hot environment, and taking account of likely usage). But of course there are pressures to reduce cost and to reduce size. Also the power supply design can be overlooked and the caps can be hotter than expected due to a lack of thorough testing. I don’t think this is intentional life reduction, but it can be penny-pinching and lack of attention to quality. There is a trend towards using ceramic caps, which become more viable as technology allows the use of higher frequency switch-mode supplies, but they too have life limitations if they carry high current. They also use rare elements. Aluminium caps are fairly benign for the environment. I think it is a matter of design quality rather than the choice of aluminium.

2 Likes

Hi Colin. It’s great to hear that some people take longevity of electronics seriously!

A long time ago, I worked in the USA for a manufacturer of electronic lab equipment, designing test equipment. I remember taking a completed design to the CEO, and being told, “This costs too much. I want you to go away and start pulling components out until it stops working. Then put the last one back, and that’s what we’ll manufacture”! I refused, but it didn’t help my promotion prospects!

BTW, does anyone know what chemical is used for the liquid/gel electrolyte? (None of the documents seem to say.) I suspect it’s not very good for you.

Not sure how typical, but this names Ammonium adipate and Ethylene Glycol - and has it as what I would non-technically call “irritant” rather than actually poisonous https://www.illinoiscapacitor.com/pdf/Papers/Material%20Safety%20Data%20Sheet%20AL%20lytics.pdf

Plenty of information including electrolyte composition here - follow link for full text:

Chris’ session is starting now! Please join us

https://app.konf.co/event/2G5gAFnTn/session/255

1 Like

@Chris_Moller Thank you again for a really interesting session!

And to everyone else: It’s time for a quiz! During the session, Chris shared this image. How many capacitors do you find here? Our followers on Instagram guessed anything between 5 and 14, I wonder how many you are spotting!

2 Likes
  1. Definitely 10.

I also got 10, under the assumption that none of the blue components are inrush NTC thermistors

@Chris_Moller inspired us to order an electronic component tester after his session. It arrived yesterday and for $12 [p+H incl] Danny is impressed/ It should take a lot of guesswork out of the problem solving and fixing of the secondhand electronics we are given to bring back to life. Well, providing the little device holds up!

4 Likes

Oh yes! they are great, in some cases they even work in circuit but make sure capacitors are discharged and batteries/power supplies disconnected.
I decided to get the boxed version with rechargeable battery, more rugged for those repair cafe events: https://www.ebay.co.uk/itm/New-LCR-TC1-Transistor-Tester-ESR-Capacitance-Meter-Electronic-Component-NPN-PNP/402288581009?hash=item5daa449991:g:0KYAAOSwvCRe3eat
Charging lead and 3 clip leads also included. It even confirms IR remote handsets are at least functioning.

1 Like

Thanks for sharing Roger and for the link. Agree, more rugged for repair events. Interesting, as Danny says there is not much need for these devices at the events we attend in Victoria. He repairs more electrical items in the community and at home he tinkers with electronics. Go well. Karen

1 Like

Hello,
Is the @Chris_Moller available online?
Thanks,
Jesus