Can definitely-bad capacitors test ok in some situations?

[OwenM]

Ok, so I opened up an old teac cassette deck that was completely dead.

The 3 power electrolytics, or at least one of them, had spilled a messy brown deposit over everything in the area. I removed them, cleaned up as much of the deposit as possible and replaced them with new caps.

Now it works, one of the cassettes has some other problem but that's another story...

BUT, all three of the removed electrolytics test ok on a cheapo ESR meter (including value), and pass the basic DMM test of rising resistance...

So I'm wondering, is this possible? It seemed to be CLEARLY the electrolytics, and it works now. Of course, I could have freshened up a dry solder joint in the area with the removals but then that doesn't explain the messy spillage.

Basically I'm just asking if the basic cap tests don't cover all potential failures!

[Enzo]

Basically I'm just asking if the basic cap tests don't cover all potential failures!

Of course they don't. Basic tests like that are like the dash board gauges on a car. A battery that can;'t supply current can still charge and turn on the lights, and yet not turn the starter. A lot of basic tests will tell you if a cap is BAD, but won't tell you it is good.

the basic DMM test of rising resistance...

This tells you the cap is trying to charge. A cap can be crappy and leaky but still wants to charge. If it reads flat out open or a dead short, well those are BAD.

Look at the old caps, is ther signs of venting? I mean is ther evidence the brown stuff came out of the cap? It is possible some other foreign material spilled on the board and caused problems. You might have solved the problem when you cleaned the board, rather than the caps themselves.

Your meters do not put real-world conditions on the caps. In a tape deck, we rarely see over maybe 20v, so leakage at voltage is not a problem often. But your meter only uses a volt or two, so it is still possible. This is a larger problem in tube amps. Caps that measure just fine on a meter can leak like a screen door at 400v.

Usually bad filter caps won't result in a dead unit. Usually it results in very hummy or low voltage conditions. But anything is possible.

[OwenM]

Of course they don't.

Thanks a tonne for your great reply! I have to admit on reading back my final/simplified question it did seem a bit obvious when I awoke the next day! I had reached the conclusion that my devices are only giving out a volt or two and this is nowhere near real-world testing but hadn't got much further into what that might actually mean for my tests.

It's an odd one. It's been opened before, but I couldn't see what had been changed. Theres no easy way for the liquid to get where it did on the board unless the unit was open. It is POSSIBLE it's the glue they used, but if so it was very viscous, and a bad job which covered nearby diodes and jumpers etc. I found another patch of the brown stuff elsewhere, so now I'm starting to wonder if it IS a crappy glue job. Could the glue have become conductive over they years? Seems a little unlikely! The brown stuff didn't dissolve in IPA (the cleaning fluid OR the beer) or detergent, and I could only remove by careful heat and scraping.


Oh well, just interested academically really, there's a mechanical problem where one tape just runs constantly and can't be stopped, and I have enough to think about learning about amps so I'm not too inclined to work on this any further! I'll probably just take the caps back and dismantle for parts!

Best
O

[g1]

I'm starting to wonder if it IS a crappy glue job. Could the glue have become conductive over they years? Seems a little unlikely! The brown stuff didn't dissolve in IPA (the cleaning fluid OR the beer) or detergent, and I could only remove by careful heat and scraping.

Sounds like it is the type of glue sometimes seen from that era. And yes it can become conductive. Yamaha SPX stuff used this kind of glue and it does cause problems with age (conductivity).
If that's one of those Teac W series dual decks from the 90's, they were not really worth fixing even then, so no great loss.

[Helmholtz]

Many DMMs can measure resistance up to 10 MOhm and more (see manual). So if resistance of goo measures above 10M, it shouldn't be a problem.

[OwenM]

Sounds like it is the type of glue sometimes seen from that era. And yes it can become conductive. Yamaha SPX stuff used this kind of glue and it does cause problems with age (conductivity).
If that's one of those Teac W series dual decks from the 90's, they were not really worth fixing even then, so no great loss.

I'm starting to think that's what it is, though I didn't get a reading through the tiny piece I had left. It didn't peel off in lumps though, I had to scrub it up and scrape away so hard to test...

Hehe, yeah it's one of those! Just having a look for learning purposes really, it was being chucked from a nearby school. It does work now as I say but it only fast forwards and rewinds, possibly something wrong with the tension or similar (though it's the same in both decks). I've had a clean of the area and pulleys but no luck, think it might be time to dismantle (any good bits) and dispose!