[Boatanchors] Fun with old electrolytic caps

[Nick England]

Murphy’s Law #37(B)
The unique hard-to-find choke will fail in order to protect the
garden-variety cap.

On Sun, Mar 20, 2022 at 7:46 PM Charles via Boatanchors <
boatanchors at mailman.qth.net> wrote:

> An unexpected adventure this afternoon. I realized it had been a few
> years since I'd powered up my PP-7482/G (1 KVA 115v 400 Hz power supply)
> that I repaired 11 years ago, so I turned on the power strip and flipped
> its front panel breaker. A couple of seconds of 400 Hz transformer
> "singing" and the green NORMAL output light on, then the power strip's
> 15 amp breaker popped! Reset it, immediate trip. Thought it might be the
> cheap pushbutton breaker, so I plugged it directly into the 20 amp wall
> outlet. The front panel breaker didn't pop, but something started to let
> smelly smoke out so I quickly shut it off.
>
> It didn't take long to determine that the DC supply was a dead short.
> That consists of the non-isolated AC line entering through various LC
> filters (not Rifa caps), four hefty stud-mount rectifiers that share the
> H-bridge heat sink, and six paralleled 1100 uf, 200V Sprague 32D caps,
> which are short & fat compared to the more common 36D computer-grade
> caps. I quickly isolated the problem to the capacitor bank itself. After
> removing four caps and testing them for shorts, opens and leakage,
> finally the fifth was a hard short. The bottom /might/ be slightly
> rounded compared to the others, but the rubber plug vent had not
> ruptured. My nose tells me the smell was actually coming from one of the
> potted input chokes at the rear entrance. Everything else checked out
> since I hadn't kept trying it long enough to burn up the chokes or
> rectifiers. The caps were made in 1979 but all except the failed one are
> working perfectly. Looks like they either didn't need reforming, or
> survived the sudden application of 160 volts after years of lying dormant.
>
> Just for fun, I decided to see if I could blow the shorted cap clear
> with my 28 volt BFPS (on a variac, not wide-open!). I applied DC current
> up to 90 amps and it was still a very very low resistance short. I quit
> before it could rupture ;) It just seems like an odd failure mechanism,
> more like a transistor "punch-through" than an electrolytic that more
> typically draws excessive leakage current, gets hot, blows the vent, etc.
>
> Anyhow, the supply now runs fine with five instead of six caps (given
> electrolytic tolerances of the day, it's probably still in spec) so I
> put it back together and ran it for a while with a couple of light bulbs
> for a load. No further magic smoke escaping. I can get a replacement cap
> for $9 plus postage at Surplus Sales if I feel the need.
>
> -Charles
>
> ______________________________________________________________
> Boatanchors mailing list
> Home: http://mailman.qth.net/mailman/listinfo/boatanchors
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Boatanchors at mailman.qth.net
>
> List Administrator: Gary Harmon, K5JWK
> ** For Assistance: gharmon at idworld.net **
>
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to navy.radio at gmail.com
>
-- 
Nick England K4NYW
www.navy-radio.com