Peavey Classic 30 signal bleed at zero volume

[FleshOnGear]

I am embarrassed to say I spoke too soon. *sigh*

After I finished the soldering job I had a problem with the overdrive channel cutting out. I could tap on the amp and the sound would return, but as soon as I hit a note on the guitar it would disappear. I thought the problem was unrelated, so I declared victory. Prematurely.

So I went in and touched up the relevant solder joints. This fixed the cutting out problem, but my crosstalk was back. It seems that V2A is inducing signal further down the signal chain, because I wasn’t getting the crosstalk when there was no signal on its grid. But why is it happening?

I broke out my scope, but I haven’t figured out much yet. The signal bleed seems to appear at the plate of V1B, but I don’t know why. There appears to be no signal at its grid or cathode. Granted I’m not an expert with the scope; I’m totally self-taught.

Anyway, the saga continues...

[Jazz P Bass]

Have we eliminated the possibility that K1B relay is at fault?

[FleshOnGear]

Have we eliminated the possibility that K1B relay is at fault?

There’s no signal at any terminal of K1B. The signal “appears” at C6, at about 22mVp-p, despite there being no signal at the input or cathode of V1B. It seems like the signal is being induced into the cap.

[nickb]

Lift the side of C6 that goes to the plate of V1B and see if the signal is still on C6.

[FleshOnGear]

Lift the side of C6 that goes to the plate of V1B and see if the signal is still on C6.

I did as you said, and there is still a little bit of bleed, but greatly reduced. Maybe 5mVp-p at the other side of C6. It’s enough to hear through the speaker, but much less than with the cap properly soldered in place.

So the signal is happening at the plate of V1B, not the capacitor. But I don’t know why. I’m starting to think that the PCB is conductive, which has me feeling a bit depressed.

[nickb]

22mV is small meaning the cuase at the cathode or grid is soo small to see easily. Crank the input up to get as big a signal as you can at the plate of V1B and then check the cathode and the grid for signal.

If nothing there check the other side of the plate resistor. You'll need to use a 630V 10nf DC blocking capacitor in series with the probe as most scopes can't handle that much voltage on the input.

[FleshOnGear]

22mV is small meaning the cuase at the cathode or grid is soo small to see easily. Crank the input up to get as big a signal as you can at the plate of V1B and then check the cathode and the grid for signal.

If nothing there check the other side of the plate resistor. You'll need to use a 630V 10nf DC blocking capacitor in series with the probe as most scopes can't handle that much voltage on the input.

I sent a 3V sine wave into the input of the amp, and I was able to see about 7mVp-p on the grid of V1B. The bizarre thing about this is that I still see no signal at the other end of the input resistor R14, where it meets the relay and R20. Also, there’s about 20mVp-p on the other side of R20, where it meets the zeners and the input capacitor to the op amp. It’s like the signal spontaneously appears across R14 and R20.

[nickb]

Probe everything around U1A on the component leads including power and grounds.

[FleshOnGear]

Probe everything around U1A on the component leads including power and grounds.

Do you mean everything physically near U1A, or just the components relevant to that gain stage? Is there anything in particular that I’m looking for? Incidentally, I tried popping U1 out of its socket, and it made no difference in the signal bleed, it just caused the reverb to hum.

Thanks again for stepping me through this!

[nickb]

The components relevant to that stage. You have more signal on the U1 side of R20 so it has to be coming from that area. If we eliminate all those then we are left with conductive PCB. At that point lift the U1 end of R20 and see what happens.

[FleshOnGear]

The components relevant to that stage. You have more signal on the U1 side of R20 so it has to be coming from that area. If we eliminate all those then we are left with conductive PCB. At that point lift the U1 end of R20 and see what happens.

I’ll do what you say, and I hope you’re right. I trust your expertise, but at this point I’m not left with much confidence that the amp will be ok, mainly due to my usual luck.

I just wonder, if something in the reverb driver circuit is the source, how is it making its way to V1B’s grid? There’s no signal at the node where R20 and R14 meet; that spot is grounded via either VR1 or VR3 (I’ve been doing most of my testing with those potentiometers at zero). So it seems that the signal can’t be going through R20 and then through R14. I’m also wondering if the feedback loop around V1B is the reason the signal at the grid is smaller than the signal at the input of U1A. Maybe I’m missing something.

Anyway, I’ll get a chance to check around U1 on Friday. Fingers crossed.

[Helmholtz]

Does the unwanted clean signal sound like reverb?

[FleshOnGear]

Does the unwanted clean signal sound like reverb?

No. But if I turn up the reverb, the unwanted signal comes through the reverb, a bit more pronounced than the dry unwanted signal.

[FleshOnGear]

The components relevant to that stage. You have more signal on the U1 side of R20 so it has to be coming from that area. If we eliminate all those then we are left with conductive PCB. At that point lift the U1 end of R20 and see what happens.

I couldn’t find a smoking gun while investigating U1A. I lifted the op amp end of R20 and nothing changed. What a shame.

[nickb]

I couldn’t find a smoking gun while investigating U1A. I lifted the op amp end of R20 and nothing changed. What a shame.

That tells us it's not coming from the U1A stage so that is progress. But you have an anomaly. The signal is still bigger ("nothing changed") on the U1A end of R20 than the V1 and that is with the resistor floated at the U1A end? You have to start lifting more (associated!) things to eliminate them. Then what is left, however improbable, is the culprit.