[Amps] How to know when filter caps begin to fail
Jim Barber
audioguy at q.com
Wed Mar 28 17:37:58 PDT 2012
Many moons ago my boss at the time was probing about in a live chassis
when an aluminum cap blew. His glasses kept the stuff out of his eyes,
but when he lifted his head I saw he had a chunk of the aluminum can
sticking out of his forehead. When I told him, he calmly reached up and
pulled it loose, then put me to work cleaning up all the paper and gunk
that had showered the shop.
Since then, I've been a bit shy about working directly above a string of
electrolytics...
73,
Jim N7CXI
On 3/28/2012 5:19 PM, Gene May wrote:
>
>
> 1. I, too, have shared Bob K6UJ's experience with an electrolytic cap exploding, and the mess was awful enough to make me conservative about the issue of replacement. The goo between the foil seems to be somewhat corrosive, was slow to clean off, and damaged some of the metal it got on.
>
> 2. I'm cheap, and have used many surplus electrolytic caps. Along with Glen K9STH's suggestions, I suggest that anyone also using surplus caps connect them in series and place some DC across for a few hours, then see how much they leak. I toss anything that gives off any more than a few watts -- I've seen leakages of 20 - 30 mA at voltages of around 300V (6 - 9 watts) on surplus caps -- and know others who are even more conservative than I about throwing them out. They can be so much cheaper than new that I think it is cheap insurance being conservative. Along with Glen, I have found that frequent usage tends to make electrolytic caps last longer, not go bad faster.
>
> 3. Finally, particularly if you use surplus caps, I suggest that AFTER THE CAPS ARE THOROUGHLY DISCHARGED, every so often it is worthwhile to simply feel them after they have been under power for a few hours, and replace the whole batch if any one of them is particularly warm.
>
> Gene May
> WB8WKU
>
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
More information about the Amps
mailing list