Reverb bleed at 0 on RVB pot on a '65 blackpanel Fender Vibrolux

[Sansteeth]

Hello everyone,
I'm working on this beautiful blackpanel Fender Vibrolux (unfortunately all transformers have been changed and while no "mods"were made, lots of parts are not original inside).
It came in because the customer was complaining he could still hear the reverb when the reverb pot was turned all the way down.
However, I found out the original reverb pot was only going down to 700ohms and the 220k CC resistor to ground coming after that was measuring 272k (RVB recovery stage grid leak resistor).
I changed them both, now I get a solid connection from the wiper to ground when it's turned all the way down and yet I am still getting reverb bleed. It's subtle, but it's there. The footswitch gets rid of the bleed but I don't understand why the pot cannot put the reverb signal to ground all the same and get the same result.
If I lift one leg of the 500pF at the input of the reverb circuit to feed it a signal and look for it on the other side (without the dry signal), I need to crank up the resolution and my scope (analog) gets too hashy on the other side to find where the bleed happens on the circuit unfortunately.
I've had cases of conductive boards on Fender amps lately but mostly silverpanels, but now I'm paranoid the reverb pot works fine, but somehow the grid of the recovery stage picks up the wet signal somewhere on the board.
Have you folks had experience with Fender blackpanels where you just can't fully dial off the little bit of reverb tail left without the footswitch?
Thanks a lot for your input!

[sluckey]

This sounds like a non-issue to me. Teach the customer how to use the footswitch.

Connect a gator clip lead between chassis and the wiper of the reverb pot. Does this kill the reverb?

Also replace the 25µF cap that connects to pins 3 and 8 of V4.

[Sansteeth]

Thank you sluckey for taking the time to help me!

This sounds like a non-issue to me. Teach the customer how to use the footswitch.

  you are not wrong

Connect a gator clip lead between chassis and the wiper of the reverb pot. Does this kill the reverb?

Nope, which makes me think it's happening somewhere else....

Also replace the 25µF cap that connects to pins 3 and 8 of V4.

That also doesn't kill the reverb 

[pdf64]

Have the HT decoupling caps been replaced recently, especially the D node that supplies the preamp?

[Sansteeth]

Hello pdf64, thanks for chipping in!

Not by me but they have been changed by F+T caps not long ago it seems, 22uF instead of 16uF but I doubt that would be the problem.
You think the wet signal could have bled through this way?

[ac427v]

It's a stretch, but I wonder if your "solid connection from the wiper to ground when it's turned all the way down" was measured 0 ohms to the brass plate?      or to the actual chassis?
I'm thinking that corrosion on the brass or pot lock-washer is preventing the pot from sending all of the wet signal to ground.

[Sansteeth]

Hey!
not really a strectch, that's actually a pretty good idea, unfortunately it's 0 ohm to the actual chassis!
Thanks a lot though, considering how strange it behaves, the actual solution to the problem is going to sound like "a stretch" I think 

[stratomaster]

Since you said "all transformers have been changed" then it would be a good idea to investigate how the reverb transformer is wired.  The decoupling that has been suggested earlier in the thread may have been defeated by an inline resistor between HT and the reverb transformer.

[Sansteeth]

Thanks stratomaster for chippin in!
The power supply is as per Fender's schematics, the different nodes are respected, the Reverb transformer is tapped right after the choke

[pdf64]

Hello pdf64, thanks for chipping in!

Not by me but they have been changed by F+T caps not long ago it seems, 22uF instead of 16uF but I doubt that would be the problem.
You think the wet signal could have bled through this way?

Yes, if the D node doesn't have a very low impedance and to 0V common, then everything can mix together there.
Even though it’s recently replaced, I suggest to verify it anyway, by temporarily connecting another high voltage cap, any value 1uF or higher will prove the point.
Connect its positive lead to a relevant board eyelet, negative lead to the 0V common eyelet for the V4 cathode resistor.

[Sansteeth]

Yes, if the D node doesn't have a very low impedance and to 0V common, then everything can mix together there.
Even though it’s recently replaced, I suggest to verify it anyway, by temporarily connecting another high voltage cap, any value 1uF or higher will prove the point.
Connect its positive lead to a relevant board eyelet, negative lead to the 0V common eyelet for the V4 cathode resistor.

I just tried, still getting that reverb trail with reverb knob to 0... 
However, your advice might just help me with another amp I'm working on, so thanks a lot for opening my mind on the importance of HT decoupling (besides ripple filtering)

[sluckey]

V4A and V4B share a single resistor/cap on the cathode. This is another possible source of bleed thru. I would remove the jumper between pins 3 and 8. Then give each triode its own separate 1500Ω/25µF. Any better?

[Sansteeth]

Nope, exactly the same...
The customer mentionned it that the problem wasn't there before, he also feels that the reverb is more intense that it used to be.
I tend to trust him as he is an excellent blues player and that's the amp he brings to every gig.

[Sansteeth]

So I went ahead and lifted off the eyelet board the whole 470K-220K-3.3M-10pF and used jumper wires from the pot wiper to the 470K resistor and the mixer junction to the grid of V4b. I did this in order to rule out the hypothesis of a conductive eyelet board.
With the RVB pot to 0, and me shaking the reverb tank, I am getting no reverb on pot wiper side of the 470K and reverb on the mixer side of the 470K. It's VERY faint on the scope but I guess it's enough for V4b to amplify that...
Then I lifted the .003 coupling cap's leg on the RVB pot side  (I was getting signal reverb on the eyelet board around it) and jumpered it to the RVB pot.
Problem solved....
My hypothesis is that because of a slightly conductive eyelet board, the surroundings of V4a's coupling cap was bleeding reverb tail into the preamp side of the 3.3M/10pF mixer, bypassing the pot entirely.