Capacitive coupling?

[Helmholtz]

Does R69 actually measure 100k when lifted from circuit?

[GainFreak]

Yes, it measures 97k which is within 5% of that CF resistor.

[Chuck H]

Post 15 is much better reporting. Thank you. Now, here's the weird thing...

R69 has NOTHING to do with channel isolation or muting. It's there to mitigate blocking distortion in the high gain channel and doesn't share any part of the clean channel circuit. So this is a really strange realization. Now I better understand why you suspected capacitive coupling. This is odd behavior and may actually be signal induced because of board design either by signal path proximity or grounding scheme. If it's the ground scheme then maybe the problem can be solved by eliminating some suspect grounds and running flying leads to the nearest node for the board ground. Sort of trying to eliminate the daisy chain (if that's the case). But I need to be more sober to realize this (I'm already two beers in ) Sorry. I didn't know this would come up. I'll look it over AM and see what I can find.

[Chuck H]

Rethinking this now. I think we should isolate exactly where the clean bleed signal is coming from. If you'll indulge me please lift the cathode at V1 pin 8 and report on whether or not the clean signal bleed is still there. V1B is exclusive to the rhy channel so that channel will no longer function. But the lead channel should.

[GainFreak]

If you'll indulge me please lift the cathode at V1 pin 8 and report on whether or not the clean signal bleed is still there.

Done. The "improved" leak is still there.

I checked the PCB ground layout as well. It's a ground bus/daisy chain or other type of common ground for all stages.

[Chuck H]

This may seem like an insulting question, but have you tried other preamp tubes?

[GainFreak]

This may seem like an insulting question, but have you tried other preamp tubes?

Yes, several times.

Here's a new development though. As already noted when Lead Volume is up (Lead Gain at zero) there's still some leak however when Gain terminal going to C9 is shorted to ground the leak is gone.
I replaced C9 with a known good one but nothing changed. Then I lifted one end of the cap going to Gain's terminal and ran a wire directly to the pot. The leak level dropped considerably and is now negligible. Tried that back and forth and it looks like that PCB trace running to the Gain pot was problematic. I shorted it to ground just in case. With that I'll consider the problem solved.

[Chuck H]

Well since disabling V2B, which essentially makes the amp ONLY the lead channel, still allowed for some bleed it seemed to me that cleaner signal from earlier in the preamp was getting into signal chain later in the preamp. And that only left me suspecting gounds again.

That lifting C9 stopped the bleed is an indication that the bleed is getting into the preamp at V2A. That would have meant the two halves of V1 were coupling somehow. That would make me suspect grounds or decoupling again, normally. But with your jumper off the board discovery I think I have to conceed that this does seem to be capacitive/inductive coupling on the PCB.

Is this a one off custom build? Boutique builder effort? Prototype? If this is a fundamental design problem, which looks probable at this point, I don't expect this is a commercial design since such a flaw would be unacceptable for market.

[GainFreak]

Is this a one off custom build? Boutique builder effort? Prototype? If this is a fundamental design problem, which looks probable at this point, I don't expect this is a commercial design since such a flaw would be unacceptable for market.

This amp is a clone of a let's say boutique amp and was built based on a PCB designed and sold by somebody/company on the internet. The owner of this amp (that somebody else has built for him) contacted the guy some time ago but was told that he wasn't aware of such problems with his PCBs. I know who the guy/company selling the PCB is but won't say because I'm still not sure if the person who actually built the amp didn't make any mistakes and I don't want to spend any more time investigating. It took me more time than I anticipated anyway.

[Chuck H]

This amp is a clone of a let's say boutique amp and was built based on a PCB designed and sold by somebody/company on the internet. The owner of this amp (that somebody else has built for him) contacted the guy some time ago but was told that he wasn't aware of such problems with his PCBs. I know who the guy/company selling the PCB is but won't say because I'm still not sure if the person who actually built the amp didn't make any mistakes and I don't want to spend any more time investigating. It took me more time than I anticipated anyway.

Yep. If the designer says there's no such problems reported and they're still selling them I'd lean toward some kind of build error. But good luck finding it, right? I know you're through with this one but I may look at the docs again with this in mind.

[GainFreak]

Thanks to all for your suggestions.

I know you're through with this one but I may look at the docs again with this in mind.

Thanks but I had enough of this amp already The sound is usable now and that's the end of this story.