Gibson GA19 RVT tremolo issue

[hylaphone]

Boy, I'm on a roll here..
Just resurrected one of these and it is sounding nice

However, there is scratchy DC present on the volume pot only with the trem engaged.
I swapped associated coupling caps, and relocated the trem/PI filter cap away from the preamp, no change.

Suspecting coupling/lead dress issue but prodding hasn't revealed anything.

How could stray DC sneak all the way back to the volume control?
Here is the clearest schematic I could find, albeit with a few mods:

[g1]

Gibson's are notorious for not always matching the schematics. Have you verified that this actually matches your circuit?

[The Dude]

I'm inclined to agree with g1. I don't see how that's possible unless something is connected differently. Do you actually measure DC on the control?

[g1]

If the drawing is correct, the only possible path I can come up with is some real weird fault in the footswitch, and the DC getting through via the reverb return circuit.
Does it make any difference if the reverb footswitch is on or off?

[Enzo]

Agree, is there a measured DC on the pot? Or are we just inferring it from the scratchiness?

[hylaphone]

I've measured a dancing .5V on the volume pot, V1 grid 2, V3 grid, and around .7V at V2 grid 2.
It's not present at V2 grid 1, the reverb recovery.
It ceases with the trem switched off.
It is unaffected by the reverb switch.

C23 is new, shouldn't be leaking.
Only departure from the schematic I can see is at the trem speed control - in this amp the wiper goes to the footswitch and the 47k is at the top of the pot to ground

[Tom Phillips]

...C23 is new, shouldn't be leaking...

Shouldn't be but measure it for voltage leakage anyway. New is not a guarantee of good. Just lift the "0V" side and measure the DC voltage on that side of the cap with respect to ground with the amp turned on. Are you familiar with that test method?
Did you try a different V2? You could just swap V2 with V1 if you don't have a spare.

[hylaphone]

No leakage there, and yes I've swapped V1 and V2.
Just scoped the preamp B+ node, and it dances along with the tremolo.

It is oscillating the whole plate supply.

Now I recall having a similar issue with a 5G9 build last year. I ended up leaving it as-is, maybe this time I should try a separate filter node for the tremolo tube --

Or can anyone suggest a simpler solution? The tag board on this amp is very delicate, I am trying to tread lightly..

[hylaphone]

Anyone else experience this before ?

[The Dude]

With the tremolo off and no signal applied, how much ac do you measure on the B+ line. I'm wondering if you don't have a filtering problem/bad caps.

[hylaphone]

If the drawing is correct, the only possible path I can come up with is some real weird fault in the footswitch, and the DC getting through via the reverb return circuit.
Does it make any difference if the reverb footswitch is on or off?

Well, I was mistaken (and/or crazy) -- sure enough, last night the wiggling DC only appeared with the Reverb engaged.
I did triple check C23.

I'll try bypassing the footswitch --