Electrolytic capacitance degradation, linear?

Hi

Normally the lifetime of an electrolytic capacitor is defined as the minus

20% capacitance point

But it seems it continues to degrade linearly beyond the rated lifetime poi nt

formatting link

So, if a bigger capacitance is marginally higher cost, one could ?u se? the capacitance beyond the minus 20% point if the application w orks at that lower capacitance to get longer overall lifetime

Any one done this?

Regards

Klaus

Reply to
Klaus Kragelund
Loading thread data ...

And is the wear out mechanism linear to the end of life of the capacitor?

Some graphs with no markings on the axis states an exponential relationship, but seems this is maybe not related to facts

Reply to
Klaus Kragelund

Klaus Kragelund wrote in news: snipped-for-privacy@googlegroups.com:

For the reasons the EL degrades and even fails, I would not use one. I would instead bite the bullet and buy the newer technology ELs and install those. Most of those older EL failure modes are related to insulative failures and internal resistance becomes a problem even as you wish to tolerate the reduced capacitance. Very few (none) would be gaining capacitance.

It is like fitting your house with LED lighting. It has a high initial cost but pays for itself in longevity and reduced maintanence requisite, etc.

Reply to
DecadentLinuxUserNumeroUno

ESR is also a wearout issue. M

Reply to
mkolber1

torsdag den 14. februar 2019 kl. 16.08.14 UTC+1 skrev DecadentLinux...@deca dence.org:

unless the electronics and caps crap out ...

Reply to
Lasse Langwadt Christensen

We are using top brand and High quality caps, so I am not worried about pushing the envelope

I just want to know if it degrades linearly or if the reduction in capacitance takes off

The application is a converter that does not care if the ESR rises dramatically

Cheers

Klaus

Reply to
Klaus Kragelund

Klaus Kragelund wrote in news: snipped-for-privacy@googlegroups.com:

The way an EL cap degrades is by failure events. These are usually in the form of a pierced insulator layer (read electron tunneled). As the number of those failures increase, so too does performance. So it probably does get exponentially worse on some, and not as much on others.

I still say bad idea. It is like ESD events. They cause chip damage, but not always immediate failure. Then later on the device fails hard.

Reply to
DecadentLinuxUserNumeroUno

s 20% capacitance point

oint

?use? the capacitance beyond the minus 20% point if the applicati on works at that lower capacitance to get longer overall lifetime

That graph is for super/ultracapacitors. What you mention is common practice on PC PCBs. They install several caps: only one needs to work, having several reduces ripple current per cap & giv es redundancy.

NT

NT

Reply to
tabbypurr

** Stupid assertion.

** It takes off.

** An unsupported assertion that is non credible.

ESR rises first, cos electrolyte in the cap has almost vanished.

Microfarads are soon lost in a big way with no electrolyte.

FFS use a long life electro and run it cool, like normal folk do.

.... Phil

Reply to
Phil Allison

Why would expect reduction in capacitance? Broken foil will do that, but thinned oxide raises capacitance. Neither is an unlikely consequence of aging.

Reply to
whit3rd

Se page 4 of this publication:

formatting link

Does not look like it takes off to me ;-)

I would. But we are pushing lowest cost possible, multi million quantities per year

Cheers

Klaus

Reply to
klaus.kragelund

The capacitance lowers as the part is worn out

Reply to
klaus.kragelund

** Actually it flatly contradicts YOU.

Learn to interpret data carefully, instead of seeing what you want to see.

Plus: when you post a URL, point EXACTLY to where the relevant data is - not just imply that it's there somewhere - so go fetch rover.

Very stupid and very insulting.

** Your proposed dodge will not work.

Normal folk are not idiots - like you are.

.... Phil

Reply to
Phil Allison

whit3rd wrote in news:e342ad3e-d143-498c-89f0- snipped-for-privacy@googlegroups.com:

Never ever heard of ANY EL cap that gained capacitance with age...

Nope... not one.

Reply to
DecadentLinuxUserNumeroUno

DecadentLinux...

** In fact it happens all the time - it's called " depolarisation ".

The cap value in uF goes up and the max DC voltage falls, almost in reverse proportion.

Most likely to occur when the cap is in storage or the equipment is out of use for a while.

.... Phil

Reply to
Phil Allison

In my experience the ESR rise is *very* dramatic, hundreds or thousands times the new value. i.e. Ohms or tens of ohms! I've seen some where the capacitance value still read exactly the label value even though ESR made the device utterly unusable.

For hi-rel continous operation above 60degC I'd only use plastic film (HV) or Al-polymer (LV).

piglet

Reply to
piglet

Lowest possible cost means 5 year lifetime, perhaps you need to create a premium pump range for infrastructure with 20+ year life using only film caps?

piglet

Reply to
piglet

** 100% correct.

** Valve ( tube ) amplifiers use electros at ambient temps more than 60C and get very long lives out of them. Still good after 40 years of service is nothing unusual.

However, the caps involved are European made, large, bolted to the chassis types & very well sealed.

The OP says his " converter" app does not require the electro to have a particular value or a max ESR limit.

Perhaps he would be better off using a resistor ?

.... Phil

Reply to
Phil Allison

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.