Reforming a capacitor

Thread Starter

EdGs

Joined Mar 10, 2011
17
Ran across this post on reforming a capacitor:

http://forum.allaboutcircuits.com/showthread.php?t=32195&highlight=forming+capacitor

I have some new old stock 6300uf 25v screw terminal caps coming, but they are pretty old. They are new, but lord knows how they have been stored over the last 20+ years.

When I do the calculations suggested in the post, I get approx 28mA leakage current. The Datasheet formula is a bit different. It's formula works out to less than 1.5mA. The datasheet also says leakage not to exceed 4 mA. ESR is .028 ohms @ 120Hz.

I don't have the caps in hand yet and wanted to know if the 10k resistor in series is OK here too. An engineer at CDE told me to use a 1k resistor. I have a good variable supply, so not a problem there.

Any suggestions on how slow to bring the voltage up on these caps and how long to let them run? I will end up calling the manufacturer back on this, but wanted to ask here also.

I'm not sure if these caps will be any good still, but they were cheap enough compared to new ones to at least give it a shot.

They are Mallory caps, part no. CG632U025R3C.

Thanks.
 
Last edited:

#12

Joined Nov 30, 2010
18,224
I think wookie did a great job of describing this, but it's always OK to be more conservative...it just takes longer to get the job done.

The data sheet description is a sort of guarantee of how the capacitor will perform if it's in good condition, not a description of how they formed the dielectric at the factory. They might have started with quite a bit more current than the finished leakage values. I bet they did.

May I suggest that you could dial up 25 volts and put 6.2k 1/4W in series with it so only 4 ma can flow, even if the cap is a dead short. Connect that up and watch for a while. If the cap has any spunk at all, the voltage will rise, the current will fall, and you can reduce the resistor value so it doesn't take a week to get the cap to a useful condition. Trial and error is OK as long as the error is on the safe side.
 

Thread Starter

EdGs

Joined Mar 10, 2011
17
Thanks for the info.

I will give it a shot when the parts arrive, should be in a few days. I just wanted to be careful, I had a big filter cap go boom once due to a shorted recifier diode. I was about a foot from it, scared the crap out of me. Part of it actually hit the ceiling. I'm just glad I wasn't over top of it when it went.

I think the desciption was great also, I just never have done this before. I will be surprised if the caps are still useable after all these years. I'll repost with the results.
 

#12

Joined Nov 30, 2010
18,224
You never know about old caps. I have a couple of 14,000uf @ 100V computer grade caps that I snagged out of a scrap box in 1975. They are still useful for bench tests, Ni-cad zapping, and other low quality experiments after 37 years!
 

Thread Starter

EdGs

Joined Mar 10, 2011
17
These are computer grade also. When I spoke to the engineer at CDE, he felt confident that they would be OK. But you never know.

We have a good LCR meter at my work, so I can take readings before and after the reform to see if there's any difference, just out of curiosity.
 

K7GUH

Joined Jan 28, 2011
190
A friend of mine lent me a Sprague Model TO-5 Tel-Ohmike capacitor analyzer. This wonderful antique tests everything one can test on a capacitor, and provides a special re-forming circuit. If you should ever run across such an instrument, collect it, if at all possible.
 
Top