SOUND ON SOUND

[ef37a]

Failure of capacitors in switch mode supplies cannot be ALL that common?

Take the countless millions in PC supplies and chargers, had a few over the last 15 years or so, never had a fail.
We had 20 or so really old desktops testing network modules running 3.1 and in ten years I cannot recall a PSU croaking.

Yes, FSTVs could be an exception, I am now on my third in about ten years (well, the Sony 32" still works but has 'issues') Probably crap caps too close to hot bits.

Dave.

[wireman]

No, I used 8600 to get a similar waveform.

But my (new) capacitor tester gave up claiiming the leakage was too high so I think all bets are off with any simple measurements. We can say that a theoretical failure model can result in the sort of waveform I saw, I'm not sure I will be able to pin down some equivalent circuit to the failed capacitor but might give it a go at some point but I will be pretty busy for a few weeks now. I suppose that applying a DC offset to check for leakage and also putting a square wave across it will reveal something.

[N i g e l]

Failure of capacitors in switch mode supplies cannot be ALL that common?
Dave.

The only capacitor scandal ive heard of was the Dell PC one
https://uk.pcmag.com/news/101461/report-dell-concealed-optiplex-capacitor-issues

[ef37a]

Failure of capacitors in switch mode supplies cannot be ALL that common?
Dave.

The only capacitor scandal ive heard of was the Dell PC one
https://uk.pcmag.com/news/101461/report-dell-concealed-optiplex-capacitor-issues

Yes, heard of that but did not know the details, so some 12 years ago? That I bet fueled 'craze' for re capping?

In the domestic electronics trade there was a definite problem capacitor waaaay ago, The dreaded "Blue Tant" so not even the Aluminium capacitors we are discussing here. These could give a variety of faults depending upon failure mode ( oc, leaky, sc) but most often they went dead short. So common was the fault that we did not bother to test for them, just swapped then en bloc. I am also pretty sure the replacement advice was just an Ally type of the same value and voltage.

BTW the value of the old leaky cap' could be found by putting it in series with a good capacitor of similar, known value then calculate using the formula for parallel resistors.

Dave.

[Folderol]

Years ago I read one of 'those' stories about why tants explode so violently.

It's because they are low noise. Well all that noise has to go somewhere so what they do is compress it while the unit is powered, and then when it's switched off the noise gets gently released so you don't notice.

Well the problem is that they didn't expect people to leave kit on continuously. So obviously the limit is reached and all the noise is released at once with a big bang.

[James Perrett]

The only capacitor scandal ive heard of was the Dell PC one
https://uk.pcmag.com/news/101461/report-dell-concealed-optiplex-capacitor-issues

Yes, heard of that but did not know the details, so some 12 years ago? That I bet fueled 'craze' for re capping?

It was a thing amongst US technicians a long time before that. It was very common advice on US newsgroups back in the 90's.

If you ever venture into the vintage radio forum you'll also find it discussed although, as they're often talking about 60-80 year old gear, it is probably more justified. Even then, there are a large number of people who don't advise wholesale recapping - just replacing the types that are known to go bad.

However, I have a couple of late 90's Alesis ADATs with power supplies that sometimes don't start which are probably down to dodgy capacitors - but I probably won't get round to finding out for sure for a while.