Quick question on cap voltage

clarkefan

I want to audition a couple of bright caps in my DSL 50, the C12 cap on the red channel gain control.

Wondering if voltage is important? Some of the caps with the right values are 100v, one is 650v (this is Maplins).

Any advice greatly appreciated, want to pick some up on the way home later

ICBM

It makes no difference, there's no DC voltage on it anyway and the signal voltage is too small to matter. The stock cap is 50V.

clarkefan

Thank you as always @ICBM

Do you think there's any difference between the cheap ceramics and WIMAs?  Will probably pick up a few of each anyway.

MartinB

WIMA will likely be closer to their target value, since the tolerance is smaller.  They've got short, stiff legs though, so check the lead spacing if they're going on a PCB.  

clarkefan

Thanks, I'm thinking more about sound.

Yes the short legs are a bit tedious.

ecc83

There ARE measurable differences in the distortion generated by capacitors of different dielectrics but even the 'worse' ones are in the second decimal place and the best cannot be measured above noise. Even then the signal voltages have to be 10V rms or so, that's 28V pk-pk across the cap! The distortion produced by what in the guitar world we would call an 'uber clean' triode stage is at least an order worse.

If anyone has any 'night and day' claims for capacitors (of a certain Presidential colour!) can they please post very well done A/B recordings so that we may all  be amazed ?

Dave.

clarkefan

Very interesting, @ecc83 , thank you!  To be honest I gave up trying to understand this a long time ago, the amount of material people have churned out is mental, and people get so animated about being right and everybody else is wrong, you would really have to know your stuff to know what's correct and what's junk. So I gave up

ICBM

As Dave says there are real differences, but only when the caps are under high voltages, which only occur in the interstage coupling caps in an amp.

This cap is just a small-signal bypass cap with no significant voltage on it, so you can use whatever type is easiest to get and fit to the the PCB.

The same applies to guitar tone caps, about which there is more nonsense than you would believe...

MartinB

I've tended to use silver mica for bright caps in my own amps, if only because the places I've ordered parts from tend to stock them and they're a handy size to fit to the pot legs.  

ICBM

Just make sure you don't use a silver mica cap anywhere there is significant DC voltage applied to it - ie any coupling cap, and also the treble cap in a standard tone stack. Under high DC voltages the silver migrates through the mica and the cap will fail.

I repaired an amp with this fault not long ago from a maker with a reputation that would make you think he should know that...

RiftAmps

ICBM said:

Just make sure you don't use a silver mica cap anywhere there is significant DC voltage applied to it - ie any coupling cap, and also the treble cap in a standard tone stack. Under high DC voltages the silver migrates through the mica and the cap will fail.

I repaired an amp with this fault not long ago from a maker with a reputation that would make you think he should know that...

Not entirely convinced on this @icbm, there are original SM caps in my Tek scope with over 900v on them and they're still going after 50 years without even 1mA of leakage. Quality of the component is paramount of course.

A cracked coating on the axial ones is common if the leads are bent too close to the body leaving it open moisture ingress and eventually failure. All of the failed SM caps that I've replaced have had this.

ICBM

That's probably true, but I've certainly seen quite a few fail in older amps. It's possible that the quality of the assembly is to blame though.

It's also true that this modern amp had been in storage for a while, in Scotland… so perhaps moisture ingress would be more of an issue .

MartinB

Hmm, that's an issue that I hadn't been aware of.  I've used silver mica treble caps in a couple of amp builds, including one that's with a friend now.  I'd assumed that a 500v rated cap would be OK at the <200V they typically see.  So far they've been fine, but I might think about swapping them out at some point. 

jpfamps

The problem with silvered mica caps is that the silver can migrate through the mica and thus become leaky.

This is well known in the vintage radio community.

I suspect that older production caps with used silver foil rather than plating silver directly onto the mica are less prone to this effect.

Leaky sm caps seems to have got worse since the advent of RoHS.

I've replaced plenty of sm caps in modern amps that are leaking DC.

I suspect the increased use of sm caps in guitar amps is the "hi-fi logic" that they must sound better because they contain silver and are more expensive.

I would not use modern sm caps to block DC in an amp build.

If you believe that sm caps impart beneficial effects on the tone in your amp (personally I don't, but it may look good in the marketing blurb), then you would need to use a larger poly cap to block the DC from the cap.

RiftAmps

jpfamps said:

The problem with silvered mica caps is that the silver can migrate through the mica and thus become leaky.

This is well known in the vintage radio community

Correct, commonly referred to as Silver Mica Disease.

However, it should be noted that the type of mica caps in radios that commonly suffer from this are found in the bottom of IF transformers and guess what? They have no insulation so are open to moisture, dirt, grime, and anything else that can work its way inside the IF can.

The other SM caps found in radios are of the sealed type and very rarely suffer, even those with substantial DC present.

Of course that's not to say that sealed SM caps can't fail, but I suspect that moisture ingress is the true catalyst. 

ecc83

First I have heard of SM caps having this problem so it must, as others have said, be modern stuff that suffers?

Never had one fail in an IF can (but then I seem to remember only EXTERNAL tuning caps?) SMs were use as RF decouplers on G2s, can't say ever a bother. The capacitors that DID  fail were Hunts (guess what WE called them?!!) 10,000pf brown fekkers in TV IF strips. Use to go 'weird' and cause patterning.  There were some bigger caps in waxed cardboard tubes, they were famous for going leaky/short, often the coupler to the 6V6 grid.

Then the bigger brown Wimas in Grundig tape recorders. Most unreliable things.

Dave.

jpfamps

ecc83 said:

First I have heard of SM caps having this problem so it must, as others have said, be modern stuff that suffers?

Never had one fail in an IF can (but then I seem to remember only EXTERNAL tuning caps?) SMs were use as RF decouplers on G2s, can't say ever a bother. The capacitors that DID  fail were Hunts (guess what WE called them?!!) 10,000pf brown fekkers in TV IF strips. Use to go 'weird' and cause patterning.  There were some bigger caps in waxed cardboard tubes, they were famous for going leaky/short, often the coupler to the 6V6 grid.

Then the bigger brown Wimas in Grundig tape recorders. Most unreliable things.

Dave.

Hunts caps were reputed to have done more damage to the British TV industry than the Japanese......

ICBM

jpfamps said:

ecc83 said:

First I have heard of SM caps having this problem so it must, as others have said, be modern stuff that suffers?

Never had one fail in an IF can (but then I seem to remember only EXTERNAL tuning caps?) SMs were use as RF decouplers on G2s, can't say ever a bother. The capacitors that DID  fail were Hunts (guess what WE called them?!!) 10,000pf brown fekkers in TV IF strips. Use to go 'weird' and cause patterning.  There were some bigger caps in waxed cardboard tubes, they were famous for going leaky/short, often the coupler to the 6V6 grid.

Then the bigger brown Wimas in Grundig tape recorders. Most unreliable things.

Hunts caps were reputed to have done more damage to the British TV industry than the Japanese...…

I never even bother testing them when I find them in old amps now - just replace the lot. It's a waste of time even checking, they're almost always leaky.

Those Wimas were total junk too…

I'm not sure the silver-migration problem is new - I first read about it a long time ago. I don't like them, I'll admit - given that there is a known issue, even if slight or dependent on conditions, I don't see any point in using them when they don't have any advantages in amp circuits as far as I know.

normula1

When I rebuilt my power amp chassis the Hunts capacitors inside had physically splits in them.

ICBM

normula1 said:

When I rebuilt my power amp chassis the Hunts capacitors inside had physically splits in them.

That's entirely normal!

jpfamps

ICBM said:

jpfamps said:

ecc83 said:

First I have heard of SM caps having this problem so it must, as others have said, be modern stuff that suffers?

Never had one fail in an IF can (but then I seem to remember only EXTERNAL tuning caps?) SMs were use as RF decouplers on G2s, can't say ever a bother. The capacitors that DID  fail were Hunts (guess what WE called them?!!) 10,000pf brown fekkers in TV IF strips. Use to go 'weird' and cause patterning.  There were some bigger caps in waxed cardboard tubes, they were famous for going leaky/short, often the coupler to the 6V6 grid.

Then the bigger brown Wimas in Grundig tape recorders. Most unreliable things.

Hunts caps were reputed to have done more damage to the British TV industry than the Japanese...…

I never even bother testing them when I find them in old amps now - just replace the lot. It's a waste of time even checking, they're almost always leaky.

Those Wimas were total junk too…

I'm not sure the silver-migration problem is new - I first read about it a long time ago. I don't like them, I'll admit - given that there is a known issue, even if slight or dependent on conditions, I don't see any point in using them when they don't have any advantages in amp circuits as far as I know.

I never bother testing the Hunts caps either; it is indeed a waste of time.

I agree to re silver mica caps. I can't see any point in using them in a guitar amp.