Enginuitor

Tore off some of the heat shrink and tested between the leads and the can. None of the leads have continuity to the can. I stand with my position that this is a single cap.

It's possible it had something to to with the component density and the fact that this is a single-layer hand-routed board (before the days when computers figured it out for you!). From the way the traces are laid out, perhaps they just thought it would be convenient. I am pretty convinced that this is a single cap, though, based on the following: Found this quote on the Power Electronics trade journal website: Personally I think it'd be pretty darn unlikely that two seperate capacitors in a can would somehow fuse together in such a way that there was perfect continuity between the two. In addition, I have seen double capacitors, and they usually have the leads on the same side, with a common ground. I'm going with the single-capacitor-with-extra-leads theory. What do you all think?

I have confirmed perfect continuity between the two sides, on both leads, with both the multimeter and the "lightbulb" test. Can I reasonably expect that this is a BAD thing, and possibly the cause of my problems?

And here's the whole thing...

Here we are! This is the capacitor...

I'm afraid not. However, if I feel up to it I might draw one out. It would probably take most of the day, but could help a lot with the debugging process. I'll try...

All four leads go different places. However, because this is an old-fashioned single-sided board (hand-routed traces, etc.), it is possible that they just used this particular capacitor instead of the two-lead variety because it doubled as a convenient set of high-power jumpers. I will take the board out again and look more closely at where all the traces go.

I recently stopped by HSC (a local surplus place) and found a big old lab power supply that came from the Lockheed Martin labs in Sacramento. Standard 4-knob controls (coarse/fine on both voltage and current), analog meters. 0-40 VDC @ 0-10 amps! Only problem was, it didn't work. So I managed to talk them down to $25 (from the initial $50 price tag), and took it home to the bench. Lucky for me, this thing is 80% discretes, plus some common transistors and two very standard IC's. 100% replacable! There were three components inside which showed heat stress: a 12V zener diode, a 1.5k resistor, and this strange electrolytic capacitor with 4 leads. I went back to HSC and bought higher-wattage replacements for the resistor and the zener, but the capacitor stumped me. They didn't have a single capacitor like it. It's a metal-can type, with glossy plastic ends and transparent shrinkwrap over it. There are two leads coming out of each end; each end has a + and - lead. It's marked "CDE" "UFT 570-75", rated at 570MFD, 75VDC, and 105.oC. There is obvious heat stress around the points where the leads enter the body. and the ends are bulged slightly. Here's where I'm stumped... I brought the cap to one of the guys who ran the store, and he said it was probably two caps in one can, as was a popular strategy a while back. However, when I got back home, both my multimeters and an old-fashioned lightbulb continuity tester indicated that the leads on each end were continuous with each other (meter read 0.00), making it appear to be a single radial cap with the leads going in one end and out the other, perhaps for better power/heat handling. Or perhaps it truly was one of those old-fashioned duplex capacitors, and somehow it has failed in a way that by some twist of fate has caused both ends to become connected? I would appreciate any help identifying this cap, and designing a suitable setup to take its place. As a last resort, I might just slice this baby open over the acid tub, and see what it looks like inside. Thanks in advance.