Peavey VTM 120 Bias Voltage

[tdlunsfo]

I have a VTM 120 on the bench in which the customer requested a full recap and a bias pot. Done. He brings me NOS Phillips ECG tubes and the plate current is high ~ 55ma plate current at 482v plate volts (measured with my euro tubes bias probe). Even if I give it all of the -56v bias the amp can provide, it is still above 70% dissipation. Well, I just so happen to have a second one he dropped off with old caps and when I put the same tubes in this amp, I'm at 28mA plate current and 428v plate volts. These new F&T caps seem to have raised my plate voltage to the point where I can't bias these NOS tubes low enough. Thoughts or suggestions on what to do? The bias is halfwave rectified, but I can't tell from the schematic if I have a center tap available to consider a full wave setup.

https://schematicheaven.net/newamps/peavy_vtm_120.pdf

[g1]

That is a big difference in plate voltage considering the hot bias. If you the idle current down, the plate voltage will be even higher.
I'm wondering if something other than the caps is responsible for the difference in B+ between the 2 amps you compared.

[tdlunsfo]

I did some looking around at older posts and saw where others referenced a plate voltage in the 480v range. I guess I can recap this other one and see what it does.

[g1]

Is the -56V making it to the 47K grid resistors of the power tubes?

[tdlunsfo]

Pin 5's are -55v.

[tdlunsfo]

Recapped amp #2 and the same result. 480v and 53mA plate current. 85% dissipation. Starting to think this amp needs to be cathode biased.

[Pedro Vecino]

If the bias circuit does not allow the voltage to be increased, one possibility to redirect this adjustment could be to install a zener diode (48 or 60V - 5W, for example) in series with each screen grid resistor. You could install it on a single tube previously to check the result.

[Delta362]

It seems the most negative you can currently achieve is the -55V and you're pretty much getting that at the pin 5 grids. How about increasing the screen grid resistors from 100ohm to 470ohm or 680ohm?

[tdlunsfo]

I'll try the screen resistors. These do seem low to me.

[Helmholtz]

Yes, I think screen voltage is the key.
But with a screen current of maybe 5mA even adding 1k to the screen resistor would drop screen voltage by only 5V.
I like the proposal of Pedro Vecino.

[tdlunsfo]

So we had already drilled the amp for a bias pot is the only reason why I'm not considering the zener. My used russian Tung sols seem to bias up pretty normal. ~40mA.

[Helmholtz]

So we had already drilled the amp for a bias pot is the only reason why I'm not considering the zener.

What has the bias pot (which can only decrease bias voltage) to do with screen zeners?

[tdlunsfo]

1k screen resistors made a difference, but not a ton. 51mA and 486v plate

[tdlunsfo]

What has the bias pot (which can only decrease bias voltage) to do with screen zeners?

Sorry I didn't read Pedro's post clearly. He suggested zener on the screens.. for some reason I glanced at it and read 60v zener on the bias supply or pin 5. Am I reading the suggestion correctly in that the zener would drop 60v out of the screen voltage supply?

[tdlunsfo]

I have some 16v 5w zeners I can try for proof of concept.