Thread: Lifespan of Capacitors

Thinking about picking up a used Soulution 725. Curious if the concept of capacitor fatigue is a real issue?
Anyone actually audibly observe this phenomenon? Do companies Recap older pieces? Or have a way to test for need for this?

Originally Posted by Ricky64

Thinking about picking up a used Soulution 725. Curious if the concept of capacitor fatigue is a real issue?
Anyone actually audibly observe this phenomenon? Do companies Recap older pieces? Or have a way to test for need for this?

How to measure : How to Measure Capacitance with a Digital Multimeter | Fluke

caps last long time, typically 20 plus years

Many companies will bring a piece up to spec for a price. You'd have to check with the particular company to see if that is something they offer and how it works with them.

Yep , No more Blackgates , plenty choices abound ..!


Regards

The EE that rebuilt my amps and is going to do some work on my Pre has frequently talked about this subject.

He said that by testing most all caps start to deteriorate after 4-5 years though their life is longer. The caps in the amps were 17 years old and measured poorly.

The warranted life by most manufacturers is not very long.
We tried to purchase some longer life Caps for my amps but I would have had to purchase 2000 at around $250+ each. That math didn't work. Ended up with 8 year expected life around $150 ea.

So we have a number of different opinions on this. Anyone feel these are considerations for them when they make a purchase? Does anyone purchase only new because of legitimate concerns about this?

Originally Posted by brad225

The EE that rebuilt my amps and is going to do some work on my Pre has frequently talked about this subject.

He said that by testing most all caps start to deteriorate after 4-5 years though their life is longer.

That's interesting Brad. I wonder how long it takes before the degradation is audible vs. measurable?

Originally Posted by MichaelsMinute

I wonder how long it takes before the degradation is audible vs. measurable?

I guess that would totally depend on the environment around the capacitor. Continued increase in the internal core temperature due to electrical overstress will cause degradation within the capacitor. . If you can measure it, well depends on when you measure it. But of course a bad cap, well can hum, click or just flat out blow with a "bang". So it depends. I guess someone needs to try it out.

Originally Posted by Ricky64

So we have a number of different opinions on this. Anyone feel these are considerations for them when they make a purchase? Does anyone purchase only new because of legitimate concerns about this?

Not this per se, but I do purposely purchase gear I feel I'll have a better chance of being able to get repaired/supported in the future vs a more unique/obscure brand that may be gone in a few years.

Failure is different than optimal sound performance. I believe the original question was if there was a sound difference not the failure of the caps.

I have heard a difference between new and old caps in a re-capped amp. However, since the degradation happens over a lengthy time it is very hard to hear anything unless you have an amp of the same vintage with fresh caps to compare it to. I had 2 amps that sounded the same and was skeptical so I re-capped one of them and heard the difference. I went ahead re-capped the second amp.

Heat, and how close to maximum voltage parameters the capacitor is run. Of course there are caps that were not designed well in the past. Modern capacitors, under temp max, and under max voltage, will last a good while. (20-30 years).

Run hot, in temp, and voltage, will shorten lifespan. Exotic dialectrics make for a whole other issue.

Pretty much what Bones13 said.

It really matters what type of circuit the caps are in and how they are used. There are circuits that use really cheap, small caps that will last upwards of 30 years simply because they are being utilized well below their max ratings, which generally also means they're subjected to very little to no heat.

And of course, my thing is that I'm a strong believer of keeping gear powered up 24/7, which I have been doing for 20+ years now and I have yet to have a single piece of gear faulter due to capacitor failure. In fact, I have had next to zero gear failure of any kind since leaving the system running 24/7.

Originally Posted by atmasphere

Normally a cap of reasonable quality will go about 20 years even if there is heat (as long as there isn't too much heat).

If these parts were in a solid state amp I'd be concerned. Caps in solid state amps tend to last longer since there is less heat than in tube amps. So in a solid state amp 30 years is common. Caps failing in only 8 years were defective parts unless operated outside their recommended specs.

Ralph-Did you mean to say tube amp in the sentence I bolded?

Originally Posted by mep

Ralph-Did you mean to say tube amp in the sentence I bolded?

Not directed at me of course, but my 2 cents is that it shouldn't matter whether caps are in a SS circuit or a tube circuit. As long as the caps are within a safe operating margin and exposed to minimal heat, they should last just about as long in either.

Electrolytic capacitors are electrochemical devices, not purely electronic devices. Almost like a battery. Their lifespan is the limiting factor in most electronic devices today. A rule
of thumb based on the Arrhenius equation for temperature dependence of reaction rates is that electrolytic cap life halves for every 10 deg C rise in temperature. Applied voltage also impacts life but to a lesser degree.

So what actually happens, spec-wise?
The first thing that happens is Effective Series Resistance (ESR) increases. In power supply applications this increases the amount of ripple in the power supply’s output. Depending on the amplifier this may manifest itself as increased hum in the output. Or not. It depends on the amp topology.

Next the capacitance starts to decrease. This takes a lot longer. It will have a similar effect to increasing ESR.

Originally Posted by mep

Ralph-Did you mean to say tube amp in the sentence I bolded?

Perhaps he meant that if caps in a SS amp were bad after only a few years then there is a problem somewhere (e.g., circuit design, manufacturing quality, "bad" cap from the get-go, etc)?

Originally Posted by rbbert

Perhaps he meant that if caps in a SS amp were bad after only a few years then there is a problem somewhere (e.g., circuit design, manufacturing quality, "bad" cap from the get-go, etc)?

I'll let Ralp respond.

Originally Posted by puroagave

maybe someone can explain this phenomenon, but Naim owners seem to LOVE recapping their gear every 10-yrs, its almost a fetish. I ended up spending a small fortune recapping all the stuff I owned, albeit some of it was Olive and Chrome bumper pieces but, on their 2nd, and 3rd recap! By contrast I took a Marantz 8b with orig. 50-yr. old PS caps in for a check-up and the tech said they were fine, not to touch them.

From a DFSS perpsective, this is because different design embodiments driving different functional responses can result in different outcomes for the "Design for X" quality attribute, durability. Durability* for a capacitor is the total lifespan of that circuit component used under the product's nominal operating condiitons. It follows that different types of caps used in different applications (e.g. a DAC, preamp, tube amp or a SS amp, for example) will have different specifications for Durability. Personally, I've owned gear where the electrolyte in an electrolytic cap has dried up and cap is now junk, and it's continued use can put the entire circuit at risk of failing.

OTOH, the Teflon caps C-J uses, because of the Teflon material used as the dielectric in their constructon, take a very long time to burn-in and will likely function nominally for decades.

*– as contrasted to Reliability, which is the time interval of component functioning nominally between service intervals).

Originally Posted by W9TR

Electrolytic capacitors are electrochemical devices, not purely electronic devices. Almost like a battery. Their lifespan is the limiting factor in most electronic devices today. A rule
of thumb based on the Arrhenius equation for temperature dependence of reaction rates is that electrolytic cap life halves for every 10 deg C rise in temperature. Applied voltage also impacts life but to a lesser degree.

So what actually happens, spec-wise?
The first thing that happens is Effective Series Resistance (ESR) increases. In power supply applications this increases the amount of ripple in the power supply’s output. Depending on the amplifier this may manifest itself as increased hum in the output. Or not. It depends on the amp topology.

Next the capacitance starts to decrease. This takes a lot longer. It will have a similar effect to increasing ESR.

Excellent and informative post. 👍

Thanks, Tom