Best way to revive old capacitors

[ostie01]

Hi, I have a bunch on old caps and would like to use them for a project. They are band new but maybe 10 years old. What is the best way to bring them to live slowly. Thanks for any advice.

[poobah]

Connect them to a DC source. Bring their voltage from zero to rated maximum gradually over the course of a day or two.

If you have a lab supply you can apply a constant current equal to the maximum spec'd leakage current... and then let the voltage come on its own. A faster method but more complex... either will achieve the same end result, but method 1 is far simpler with multiple caps.

[Christer]

Originally posted by poobah
Connect them to a DC source. Bring their voltage from zero to rated maximum gradually over the course of a day or two.

And then bring the voltage down again, just as slowly, according to most of the credible sources.

[poobah]

Hi Christer,

Do you recall the reason???

[anatech]

As opposed to shorting them, sure. Leaving them charged would be a hazard.

I just connect a bleeder across the cap.

-Chris

[forr]

I would measure capacitance and ESR first. I have tested a great number of electrolytics, some quite old and having not been in service for more than ten years. My conclusion is that most electrolytics have a very extended life without showing any misbehaviour.

[poobah]

I usually just hand them to my kids to discharge them.

[Christer]

I don't think I have ever read any reason for the formatting process at all, just several different sources saying this is the way to do it, although there are also several other sources suggesting variouse more or less reasonable methods.

Anyway, I tried some googling to see if I could find any explanation, but none of the documents I found seemed to advocate what I recommended, so maybe that is based om some misunderstanding.

The recommendations for reforming time seem to vary a lot though. One source said 5 mins plus 1 extra minute per month of storage. Other sources said 12-24 hours, or even more. The important thing is anyway to limit the current so the cap won't heat and produce gases that may leak out or even cause it to explode. Better take it slowly to be on the safe side.

It might be worth noting that the manufacturers use a voltage of 135 - 200 % of the rated voltage for the initial formatiing, and don't seem to bring it up slowly. However, they form the film itself before rolling it to single capacitors, so they probably don't have the same heating problems. I have found no evidence, thugh, that reforming should go higher than the rated voltage.

[poobah]

The oxide film is actually broken down over time and temperature by the electrolyte. Capacitance value and leakage resistance drops.

I'll bet the manufacturers exploit overvoltage on a cap with fresh phlogiston to buy extra shelf life... perhaps also to weed out weak units.

[Christer]

Actually, it seem not yet clearly understood why storage breaks down the oxide layer, although there are suspicions that the electrlyte itself may be one contributing factor in som cases.

"It is a characteristic of wet electrolytic capacitors that the anodic film becomes weakened or less effective during sustained idling periods. It is not clearly understood just why this should be so but it is thought that possibly the electrolyte hydrates the outer boundaries of the oxide film to some degree and in that way temporarily reduces the effective thickness of the dielectric. An indication that this is true is the fact that anodic films, formed initially with an appreciable outer coating of aluminum hydrate, show very poor idle shelf or recovery leakage characteristics."

from
http://www.faradnet.com/deeley/chapt_06.htm