De-biasing a Peavey Classic VT (6L6 output stage)

[iep]

Thanks ChuckH. You've spotted one of the odd features of the amp (or at least unusual it seems to me). If you check the schematic I posted at the top of the thread you'll see that there are no individual screen resistors but that any screen current is limited by the output resistor for the 490V screen supply (shared single 400R 10W, IR1).

Since the amps are known to be quite reliable I guess this is not too much of a problem but I don't know what implications it might have if I want to add the triode switching option.

Does anyone have any suggestions on this?

Thanks,

iep

[Chuck H]

I would add them for triode operation like this. 1k 5W should be ok. Watch for over dissipation on the screens by measuring current across the added resistor. Bump the value (2.2k) if dissipation is too high.

[Tom Phillips]

... there are no individual screen resistors but that any screen current is limited by the output resistor for the 490V screen supply (shared single 400R 10W, IR1)...

Not quite. High instantaneous demands for screen current are not limited by R1 because the C8-C9 stack is connected directly to the screens. For that reason we consider the schematic that you posted a “no screen resistor” design.

If you decide to try the triode mode I suggest that you first just make a temporary mod without expending all the effort to mount a switch. I suggest this because:

1. The amp will sound and respond differently in triode mode and you may not like the change.
2. As Enzo pointed out, the loudness won’t be reduced much by changing to triode mode.

Tom

[iep]

Chuck and Tom. Thanks again for the advice, this has been a really helpful thread. I agree that a 'mock-up' before drilling the chassis for switches etc would be a good idea. Also, a large part of my ambition here is just to learn more about valve amps. Obviously, I know enough to know how dangerous they can be and will take precautions. On that score, the PV helpfully has bleed resistors on all high voltage caps so returns to a safe state in about 2 mins after power down.

I have also sourced a copy of the amp schematic for the Deuce/Mace amps which use a near identical power amp except with 4/6 output valves respectively. See attached.

Curiously, on the 4 valve Deuce, there are no screen resistors but on the 6 valve Mace they include screen resistors on 4 out of the 6 output valves. I can't figure out why this would be the case and if it would have any implications for the triode mod that we have been discussing here.

Deuce Mace VT schematic.pdf

FYI, I have ordered a basics book on valve amp design so that I cna become a bit less dependent on advice from you guys (ie, take up less of your time)

Cheers,

iep

[iep]

Screen dissipation for the 6L6 seems to have a max of around 3.9W. I guess anything south of 3W would be safe?

Out of curiosity, what limits screen dissipation in the existing design? Would that still be R1?

Cheers,

iep

[iep]

Out of curiosity, what limits screen dissipation in the existing design? Would that still be R1?

I think I've figured this out. The screen dissipation listed on many sites for the 6L6 is actually 5W (rather than the 3.9W I had believed).

Reviewing the first schematic I posted you can see that the voltage drop to the screen voltage (490V from 500V) is 10v over R1 (400R) res. So, 25mA. The voltage drop over the two 22k res to the 300V supply requires just over 4.3mA so this would suggest 20.7mA goes to the valve screens (10mA per valve). At 490V this would be 4.9W which seems close to the max spec for the valve and so seems plausible.

Cheers,

iep

[Tom Phillips]

You are calculating the steady state values. However, the screen dissipation issue with guitar amplifiers is that the screen dissipation greatly increases when the output tubes are over driven. It's not a linear increase and will cause the screen dissipation to greatly exceed the spec. It's not a linear increase. Couple that with less robust modern tube designs and you have trouble.

Screen resistors have been added to guitar amps over the decades because of the following trends:
1) Players driving the output sections into distortion
2) Increases in operating voltages used in the amps and old amps being run on higher modern line voltages.
3) Modern tubes less able to survive the abuse

[iep]

Sounds like you are saying the original design is a sailing a bit close to the wind (with 4.9W of screen dissipation steady state)? I guess this may be safe enough in the PV since the output tubes were never intended to be overdriven.

If so, then it does seem prudent to use slightly larger screen res values in triode mode since this will limit max output power and so increase likelihood of overdriving the output stage.

'Watch for over dissipation on the screens by measuring current across the added resistor. Bump the value (2.2k) if dissipation is too high.'

Since it is only really practical to measure this at quiescent, are you suggesting simply adding margin (maybe aim for 2.5W)?

Cheers,

iep

[Chuck H]

I guess this may be safe enough in the PV since the output tubes were never intended to be overdriven.

That's the key right there.

Since it is only really practical to measure this at quiescent, are you suggesting simply adding margin (maybe aim for 2.5W)?

If the quiescent measurement is within spec I would expect a 1K to keep things safe under signal, BUT, I'll wager that amp originally ran a fairly cool bias with a big strapping drive voltage for max power. Just to be safe you could simply add full time 1k screen resistors and run the triode mod through another 1k for each screen.

And you CAN actually measure current across a screen resistor when conducting. You just need a dummy load and a signal generator. It shouldn't be necessary though.

[iep]

Well, I've collected together all the bits i need (inc a lovely period peavey switch for the back panel) so all I need now is the time to dismantle the amp and make the changes.

One last thing that occurred to me though. If I switch the power stage into triode mode, this presumably increases the output impedance (hence the lower efficiency). Does this cause potential problems with impedance matching with the output transformer/speaker load?

Cheers,

iep

[Chuck H]

One last thing that occurred to me though. If I switch the power stage into triode mode, this presumably increases the output impedance (hence the lower efficiency). Does this cause potential problems with impedance matching with the output transformer/speaker load?

It does mismatch the impedance. I've never had, or heard of it causing a problem.

[iep]

Thanks.

[nsubulysses]

Anyone able to repost the schematic?

I found Enzo Schematics online but there is a fatal error

vhttp://www.ampix.org/thumbnails.php?album=4&page=3

[nsubulysses]

No matter where I find this schema online it is just 4 blank pages. Maybe it is me somehow.

[Chuck H]

If the VT schematic will suffice, the one at the top of the page in post #1 still seems to be working.