Rejuvenating condensers

[Diabolical Artificer]

How do they perform electrically Sue?

Bees wax may be a useful addition and would make the wax less brittle, make the caps smell nice too. : )

Andy.

[mole42uk]

Bees wax may be a useful addition and would make the wax less brittle, make the caps smell nice too. : )

Perhaps that's a technique for re-sealing RIFA caps, at least they'd smell better!

[suebutcher]

Originally Posted by Diabolical Artificer

How do they perform electrically Sue?

What I've done is test them on a Lafayette TE46 analyser. The leakage has gone down from about 5mA @ 400V to unmeasurable, that is, less than 5uA, and the capacitance is close enough to the marked value. But I don't know how to measure effective series resistance. Only one of the three caps is giving a strong peak on the magic eye when I test for capacitance, the peak is weak on the other two. Would that be caused by series resistance?

[GMB]

Weak peaks on the bridge is indeed a bad sign.

I think 5uA leakage is quite poor for this kind of capacitor if that is at room temperature. Only electrolytics are allowed to leak a little bit.

I suggest you measure the leakage while processing them - while hot. This is much easier as the temperature massively increases the leakage, and allows you to follow their progress. Looking for bubbles is not a good guide for the end point.

If you can get it down to a few uA at 100C then it will be very good at ordinary temperatures. See my earlier thread on this (link above).
With the molten wax method you need to be more careful not to over-cook them, so keep an eye on the temperature.

[suebutcher]

Originally Posted by GMB

I think 5uA leakage is quite poor for this kind of capacitor if that is at room temperature. Only electrolytics are allowed to leak a little bit.

The leakage may actually be less. I'm just saying that I can't measure below that level on the meter I have. I'll try again at a higher than room temperature. 5uA at 400v is an insulation resistance of 80 Megohms, which would be "serviceable" for most applications according to "Receiver Troubleshooting And Repair" by Ghirardi (my main reference for this sort of stuff). Maybe not good enough for inter-stage coupling, though.