I need some help tracking down this noise (intermittent) in Marshall 50 Watt clone.

[Gtr0]

Good Day Gentlebeings.

I built this Marshall 1987 clone last year. IMO it sounds absolutely amazing but I can't seem to track down this noise. It is intermittent and sometimes rather explosive sounding. It is sort of a static sound, what I would relate to a bad resistor, but can find no culprit... yet. First, I attached a short 13 second long audio clip that starts with the noise and you will hear it abruptly stop after six seconds.... then I will list what I have done and things I have noticed...

SL50 Noise 3.mp3

First thing I notice is that the noise does not happen until the amp is at full operating temperature... so usually about 15 to 20 minutes after it is on and I have been playing. But it isn't always at 15 minutes... sometimes I can play for an hour before it starts and occasionally it doesn't happen at all, after two or more hours of use.

Sometimes it is quiet, sometimes louder (as in the above clip)

The tone controls effect the sound particularly the Bass control, the gain pot does not. This should more or less narrow it down to v2 - and when I pull valve v2, the sound stops. When I put it back in, it occurs.

I have changed the v2 valve two or three times - having used known good valves.I should probably do the same on v3 for good measure.

I have gone through with an oscilloscope. I have done this several times, waiting long enough for the noise to occur. But I have not had a lot of luck. I am not tracing a signal, just waiting to see some noise on the oscilloscope. The other day, after several attempts over several months, I got a noisy readout the pot side of one the 22nF caps in the tone stack... I replaced it and thought it worked, but the noise finally came back after a few hours of playing. This is what I recorded and added above.

Coincidentally, I previously had an issue with red plating this same amp, posted here...
https://music-electronics-forum.com/...ing-in-50-watt

Again after some months of troubleshooting that, I began to discover the amp was experiencing some type of slow moving bias runaway issue where the bias on the one power valve (EL34) in question was rising over time. But that is not to be part of this post, but rather to explain that I have a "problem child" on my hands. After changing the bias feed resistors at the PI and the grid stoppers to the EL34s, it seems to have halted for the time being but I did not perform any exhaustive tests to prove this with certainty. I only tested once that the bias was not rising on the one tube and left it there.

Extra info:
I had a cache of NOS Piher resistors that I used on this build. Caps are Sozo Next Gen. Power filtering using new build ARS 50uF @500V. Voltages in general look good... heaters at 3.3VAC which I know is a bit high... but unfortunately this transformer only came with a single 230V tap. Wall voltages here typically hover around 232V to 235V.

Here is the schematic... the only difference to my actual build is that I used a 250pF bright cap in lieu of the 500pF on the schem.


Any help or ideas is most greatly appreciated. Thanks!

[Jon Snell]

Is the background noise affected by volume setting, tone controls?

[xtian]

The tone controls effect the sound particularly the Bass control, the gain pot does not. This should more or less narrow it down to v2 - and when I pull valve v2, the sound stops. When I put it back in, it occurs.

I have changed the v2 valve two or three times - having used known good valves.

Let's take these statements as true. That leaves only a small number of suspect components: V2's 100K plate and cathode resistors, 820R cathode resistor and 0.68u bypass cap. Maybe start by retouching those solder connections, and if that doesn't change anything, replace those components.

[loudthud]

What do you have for a B+ bypass cap on V2 ? The schematic shows one section of a dual cap.

[Gtr0]

Thank you. Good idea... I retouched all of the tone stack, but haven't touched the plates or other cathode resistors. I was also wondering if I should pull v2 and shoot some Doxit in the socket (amp off and drained of course). I always use new sockets on builds but many have been in my little worksop for a few years.

I knew I was missing something in my original description, the only other difference in this circuit was that I left the 680nF cathode bypass cap out, as in older Lead/Superlead circuits.

First I will try the re-touching and Deoxit. Then if that doesn't work, I'll change all v2 components . Let us see where that leaves me.

[Gtr0]

What do you have for a B+ bypass cap on V2 ? The schematic shows one section of a dual cap.

The same, one section of a dual 50uF/500 volt, 100kΩ feeding v2a.

[g1]

What exact tubes have you tried in that V2 position? The cathode follower can be finicky about brands.

[Gtr0]

V2 had a TAD ECC83-WA. I have swapped it to a TAD 7025 high-grade, a JJ ECC83, and most recently an original Brimar. They all displayed the same noises. Again, pulling v2 kills the sound, but pulling v1 does not (and pulling v3 kills it of course as well). So I suppose it exists somewhere between the volume pot and the PI input.

However, this time since the amp chassis was out of the box and tubes were easily accessible, I started jiggling v2 when the noise came around... usually, but not always, the noise would stop. If it didn't totally stop, you could still hear the same type of noise when moving the valve around in the socket.

When I had the chassis out yesterday, I sprayed the sockets with dioxin (as well as tube pins). I also reflowed solder in all joints between the output of v1 and the PI input, including tone pots. I have not yet begun replacing components. Would it make sense to replace the valve socket as well? Or perhaps try that first? Having build a few dozen amps at this point, I have only seen one bad socket before, so I tend to think it would be rare - the last thing I would normally try. I would love to find it not being any of those old Piher resistors, but they may very well have a possibility of going bad since they were NOS, though very much in spec before I installed them.

Thanks

[Jazz P Bass]

Wow!
"I sprayed the sockets with dioxin"
I would sure hope that that is a typo.

[Gtr0]

Wow!
"I sprayed the sockets with dioxin"
I would sure hope that that is a typo.

haha. It's auto spellcheck that thought it was doing me a great justice. And I read my posts before posting them, and so my brain accepted that as well.

So after taken those steps above, it definitely helped, but the noise still returns. But now I just slam some guitar signal through and it usually "cures" it. If it doesn't, a little shake of v2 takes care of it. I suppose I will just swap the socket and find out...

[Gtr0]

Fixed!

It was the treble tone cap. It was an "NOS" RBS 470pF. Replaced it with an 470pF NOS Radio Spares... played for about 8 hours over three sessions so far - good as gold... for now :-)

Thanks all for your input! I appreciate it. Sometimes I feel a bit clueless when it comes to pinpointing little problems like that. I have been dealing with that for about 5 months!

[Helmholtz]

What a great and authentic sounding amp you built!

[Gtr0]

Thank you, Sir Helmholtz!