Marshall TSL 122 positive bias voltage

[Danglin' Fury]

Hey Greg. Yeah, I've done a lot... Lol. It's funny, on the cap side (C36) of R67, I'm getting full bias voltage (-50 to -35 ish as I adjust the pot) But on the other side of R67, I'm getting no more than -20 Volts DC. I've changed that resistor (220k) and it stays the same...

[Danglin' Fury]

And to Leo Gnardo, are you talking about the 22 pf, 500volt? Lol! I replaced it with a 1000 volt cap. It was smoked when the amp first came in.

[Greg Robinson]

So is there a voltage drop across R67?

EDIT: If so, and you've already isolated pin 5 from the PCB with a core drill, then the tubes are drawing grid current, and are part of the problem.
If instead there's no voltage drop across R67, and rather there's a larger voltage drop across PR1 than PR2, then leakage is getting in somewhere else, like around CON1, R77, etc.

[Danglin' Fury]

Hey Greg...The measurements that I told you are with all the tubes removed. I also noticed that r77 and r68 are not what the schematic says they are. R68 is a 22k instead of a 33k... R77 is most definitely not a 10k as it says in the schematic. I'm gonna change it to the correct value and see what happens. Also, yes. R67 definitely has a voltage drop. Approximately 20 volts. On the cap side of R67, I read full bias voltage..on the opposite side of R67, I'm reading -18 to -20 volts and I can dial it down to -8 volts, but I cant get it higher than -20. No tubes installed.

[Greg Robinson]

Ok, so you've now confirmed without a doubt that this amp is suffering from the board leakage problem.
You now have to decide if you'll attempt the board modification as described in earlier posts, or replace the board.
You can attempt most of the proceedure required for the drtube kit before ordering it to see if it might be viable before purchase (ie, lift those components from the board and suspend them). I've already expressed my opinion that such modifications are not a long term solution. YMMV.

[Greg Robinson]

Oh, and C46? 1kV is not adequate in this position. It really should be 3kV rated or greater. Most examples I've had across my bench were stable without it present though, so it can also just be removed from circuit.

[Helmholtz]

Oh, and C46? 1kV is not adequate in this position.

Why?
I can't imagine a situation during normal operation where the difference between plate and screen voltage would be larger than 500V.
Exception would be oscillation.

Most examples I've had across my bench were stable without it present though, so it can also just be removed from circuit.

Did you do stability testing using a square wave signal with a resistive as well as a speaker load?

[g1]

As far as changing part values to match a schematic, is it the exact same version as the unit you are looking at? They are often tweaking values through the various revisions. If you are changing to a newer revision I guess that would be ok, but I would want to make sure it does not correspond to other differences that may not be present.

[Greg Robinson]

Why?
I can't imagine a situation during normal operation where the difference between plate and screen voltage would be larger than 500V.
Exception would be oscillation.

I haven't probed directly at the plates - didn't have a sufficiently rated differential probe when I last had one on the bench, this has been determined through experience. I presume there's enough leakage inductance that allows it to exceed 1kV.

Did you do stability testing using a square wave signal with a resistive as well as a speaker load?

Yes. I've observed a degree of ringing that was not altered by the presence or absence of this cap.

[Danglin' Fury]

Sooooo has anyone seen Syntax Error: Unexpected Token '<' " when trying to order Dr. Tube's stable bias board? The site just will not let me order it. I've tried 3 times now to enter my information into my account and when I get to checkout, it throws up the Syntax Error message every time.