advertisement


4 terminal capacitors

Multi-section electrolytics were widely used on valve kit but very rarely if ever on SS. I don`t remember ever seeing a low voltage multi-section cap.
 
It appears I need one of these Vishay MAL205148103E3 which appears to be a pretty exact match even down to the 5 'terminals' correctly spaced, although only 2 are used electrically. They seem to be available abroad from distrelec which I guess is where I'll head unless anyone can point me to a source in the UK?

My problem is both channels, when I first turn the amp on for maybe 15secs or so it's noisy / crackly. After that it's fine. I can turn off / on and it's fine, but leave overnight and it goes crackly again, the longer I leave it, the worse it is. These are the caps linked to where the power from the transformer attaches to the board.
Thanks.
 
I agree, pretty unlikely. I would expect the noises to be equal on both channels if the power supply is involved.
Try the tooth pick / lolly stick and freezer spray methods. With it 'quiet' carefully go round all the components tapping and pushing lightly with the lolly stick. You might find a poor joint or fractured resistor or track. Freezer spray each component in turn (carefully again) and you might set the noise off again. Use cheap speakers and be ready to switch off quickly or better still use a 'bulb tester' - just in case...
 
I'm not suggesting the power supply per se is the fault, but that the caps smoothing it out on the PCB have a problem.
I'm curious to understand why a fault with resistors etc. would go away once 'warmed up'. I know that failing caps can regenerate once power is applied and will start to go faulty once left without power for a while so that bit makes sense to me which is why I was following that route.
It also makes sense to me that 2 identical caps would start to fail at a similar age and exhibit similar symptoms, I find it harder to believe that a random fault on a resistor / track would cause similar symptoms on both tracks at the same time. Sure, other components go wrong randomly but don't as far as I'm aware tend to age in the same way caps do.
I may be incorrectly applying logic?
 
Whilst it's not actually impossible for it to be the smoothing caps it would the first time I've seen such a thing in 35+ years if it was.... It's just not the type of symptoms you would get with faulty smoothing caps. You would get mains hum if they are really knackered and probably even then only if they were very badly worn and very high ESR. They would have to maybe only 20% functional before you would notice.... The PSRR (power supply rejection ratio) of the amps themselves can cover up a multitude of sins! Only remotely similar thing I've known is when metal fatigue has caused a cap terminal to physically break but the end in the board is still nicely soldered in.... usually pretty dramatic when this happens and accompanied by sparks from the dodgy terminal at the break as it intermittently makes contact... rare, and even rarer in your case with all those extra fastening points to the board from the caps!

The fault you describe could be so many different things that it's not possible to even suggest anything I'm afraid.... but Snowman Als advice is good.

Is it in both channels and identical in each channel? If so that's your biggest clue as so few things are common to both channels.

It just occurred to me that Roksan fitted IEC socket mains filters to a lot of their stuff and they were really dodgy. Usually went with a loud hissing and pall of acrid smoke and then once the internal caps had blown open circuit all worked again as if nothing had happened. Could be starting to break down possibly if one is fitted in your case.
 
Is it in both channels and identical in each channel? If so that's your biggest clue as so few things are common to both channels.
Thanks for that explanation, appreciated.
It is both channels and a similar sort of crackle in both, but not as best I can tell identical in each. That's why I was thinking (agreed, possibly incorrectly) that the same component was going on both channels or at least at a similar point on each channel.

I have noticed that although I've replaced a load of leaking caps, there are a couple more on each channel right near the audio input I'd not recognised as caps so hadn't replaced yet (little red towers at 3.3uF rather than black cylinders). Before I replaced the first lot of caps the amp was making lots of noise and the caps I replaced are just after those, so I'd be quite prepared to believe it's those. There are also a couple of small green boxes I don't recognise.
The mains socket looks very basic, I can almost see the backs of the mains pins connecting to the wires going to the transformer (not literally, but it's that basic).

I do intend to go with Snowmans approach, just waiting for clarification on bulb.

Again, thanks.
 
What sort of bulb would you recommend please? I take it you're suggesting connecting bulbs between the output terminals and watch for it flickering.

Not quite. It is an old incandescent (filament) bulb in series with the mains live feed. You plug the amplifier into the socket and if there is a fault (output devices fail or severe oscillation etc.) and lots of current is drawn it simply lights the bulb brighter but normally saves the amplifier from further harm. 60 and 100 Watt bulbs depending on the amplifier.

I suggest it because of your description of the original fault - which may just be part of the current problem still.
You need the speakers connected to listen for the noise of course.


308wjue.jpg

9sgf2c.jpg


Mine is from my scrap box, but a 'bathroom' type ceiling holder mounted on a piece of wood together with the socket would be better...
 
Just for interest this is a sound file of the noise from the right channel https://www.dropbox.com/s/t4c8sjch5e6jwp9/Voice_180726.3gp?dl=0
The click at the start is me turning the amp on at the mains, there's a bit of transformer hum followed by some cracking. The hissing around 33-45secs is background (fridge). It mostly goes quiet after a few secs although there is a little more noise around 1'07".
The freezer spray should arrive for collection at RS tomorrow.
Jeddy
 
Purely speculation from recent experience, but are there any LEDs in the circuit? I had issues with random crackles and pops from a phono stage recently and after changing all the caps, opamps, transistors it turned out to be a single red led on one channel that was the issue.
 
Purely speculation from recent experience, but are there any LEDs in the circuit? I had issues with random crackles and pops from a phono stage recently and after changing all the caps, opamps, transistors it turned out to be a single red led on one channel that was the issue.
Thanks for the suggestion but the only LED is the one on a tiny PCB by the power switch (i.e. the on / off LED) which is earthed to the main PCB but gets its power from a terminal block which connects the main PCB to the transformer.
 
That was a very useful trick. When I spray one of the green Wima caps on the right channel it gets very noisy - less so with the same cap on the left channel but still a bit of noise so it appears it is a couple of caps which need replacing and not one of the big Roksan ones! Who knew! ;) Spraying the rest of each channel doesn't do much so I suspect those are the main if not only problem.
However there are also another couple of similar green wima caps which I may as well replace so long as I'm going to have to take the PCB off again (pain disconnecting from the heat sink).
Any thoughts as to whether I should replace the 3 orange drop caps? There are also another couple of blue components I'm struggling to identify.

I wonder if someone could assist though please as googling the following isn't finding a replacement, so I suspect I'm going to need an equivalent.

  1. The problem caps are labelled 330 110-U WIMA B3 5%
  2. 330 110-U WIMA B4 5% on the other channel
  3. The ones I may as well also replace are labelled (as best I can work out) 6800 63-U WIMA B6 5%
  4. The writing on the blue component is hard to read, but possibly 100 V MKC-HQ 344-PH 8207 220n k I'm guessing that's not a cap? Or is it, if so I'll replace that too - but what with.
Please can someone assist me in finding replacements for the caps 1-3 above and 4 if appropriate?

Apologies for my apparent refusal to believe it wasn't the big caps, the symptoms suggested to me it was a cap problem and I'd not recognised the above as caps thinking the only remaining caps were the big ones!

Thanks.
Jeddy
 
I will admit I am more than a little surprised... WIMA caps generally have a good reputation.

These are guesstimates without seeing a close up picture of each component to be sure:-
330 110U is likely to be 330pF at 110 volts (110U is strange - sure it is not 100-U?) B3/4 is irrelevant.
So 6800 63-U is likely to be 6800pF (same as 6.8nF) at 63 volts.
The blue component will surely be a cap too. Maybe 220nF at 100 volts? If it not a WIMA I would leave it if the spray has no effect.
Leave the orange drops.
 
Thanks snoman_al.
You are right, that was a typo, 330 100U. That agrees with what I was thinking but I didn't know the B3/4 etc. was irrelevant - I guess it's a version number or similar.
I am guessing these are polypropolyne / polyester.
From quite a bit of searching I've ended up at
http://uk.farnell.com/wima/fkp2d003301d00jssd/cap-film-pp-330pf-100v-rad/dp/1005986
http://uk.farnell.com/wima/fkp2c016801d00jssd/cap-film-pp-6-8nf-63v-rad/dp/1005989
I also noticed another pair of red caps 3.3uF 50 N WIMA B0 which I believe takes me to
http://uk.farnell.com/wima/mks2b043301h00kssd/cap-film-pet-3-3uf-50v-rad/dp/107423
The other component doesn't say Wima - I'll post a picture in a minute.
I'll leave the orange drops alone as suggested.
 


advertisement


Back
Top