Output tube configs...

[lowell]

Has anyone tried running a 4 output tube amp with different plate voltages on 1 pair? I.E. the inner pair at 350V and outer pair at 450V? Would a second output transformer need to be used? I feel as though this would be a great way to get the high end response and attack of a high plate voltage output paired w/ the warm tone and "smoothness" of a high current output.

[mooreamps]

When I did my SE 36 watt, it was using two output transformers, but both running at the same B+. Could be done with different voltages on each.

-g

[loudthud]

Technically "Plate Voltage" is the voltage between the plate and cathode of a tube, not necessarily the B+ voltage. If all four cathodes are grounded, you would need a separate winding on the OT to have a different plate voltage on each pair of tubes (unless you put zener diodes in series with the plates). But if the plates are tied together, a different cathode voltage will give each pair a different apparent plate voltage. Many people have tried running one pair of tubes fixed bias and one pair cathode bias. Kevin O'Connor (TUT books) has a couple of designs like this with switchable bias on each tube. The next step would be simply a zener in series with the cathodes to reduce plate voltage even further. You could even make the zener variable. All these schemes require 4 coupling caps from the phase inverter because of the different bias voltages on each pair of tubes.

[jrfrond]

As an alternative, you COULD run one pair fixed-bias, and one pair cathode-bias. Don't know how it would sound, but it might be interesting. You would need to insert a DC blocking cap after the grid resistor to feed the second pair.

Hmmm.....

There are also Class G and Class H designs which use hi/lo DC supplies and modulated infinite supplies, respectively. These are geared more towards efficiency in solid-state designs rather than tone, but they could be adapted to tube designs, especially Class G.

[d95err]

Kevin O'Connor recommends running one pair as pentodes and the other as triodes. Says it works wonders to warm up a Twin or any other too-much-power amp.

[lowell]

Triode mode on pentodes sounds GREAT. I did that mod on my SF Twin Reverb. I also converted it to cathode bias and ditched the neg feedback, and removed the ultralinear taps on the OT.

This zener diode thing sounds cool... could you please explain what rating zeners can be used, as well as difference b/t cathode-zenering and plate-series-zenering?

[Russ]

Is it possible to connect the outer tube plates to the OT normally, but connect the plates of the inner tubes to the ultralinear taps?

[lowell]

Yes I am also curious what happens if the plates of output tubes are connected to ultralinear taps...

[Don Moose]

As it was exlpained to me a while back, all the tubes would be running at the same B+ (not an appreciable difference in DC resistance), but the tubes facing the UL taps would be running into 70% of the OT's primary impedance. Does that get you the 'high current' sound? Dunno.

Mesa Boogie Mk-IV outputs look like they run that way.

Hope this helps!

[lowell]

thanks Don. You heard a MkIV? Like it? How would you descibe it's sound?

[Don Moose]

I've had a couple of MkIVs at various times ... they weren't quite what I was looking for at the time - way too much power for where I was, even in triode mode. They just keep appealing to my feature creep tendencies - all those channels and modes.

The Mesa lead tones tend to get described as 'nasal', part of the reason you always see their graphic EQs set in a V.

The sweetest sounding rig I ever had was a SilverFace Princeton Reverb with an IceCube in place of the reverb, running flat out into a 12" Celestion, all driven with an Aria ProII guitar with a preamp in it, then an old TubeScreamer for a little more boost. There was only one good tone in that setup, but it was a darned nice lead tone. It couldn't keep up with a drummer without being miked and pushed through a monitor, though.

Hope this helps!