SE multi tube design

[Lynxtrap]

Thanks Martin, I'll have to read that a couple of times more
In what way is the continuous current limit critical?

12 or 15W output doesn't matter, but approaching 10W with the big bottles does begin to feel a little like a waste of potential. If I had to choose I'd skip the option to use a 6V6, they seem to make the equation problematic in all sorts of ways.

I do have a PT in the donor chassis that would be nice to use for monetary reasons. If I remember correctly it used to put about 380V on the plates of two 6L6GC in cathode bias (push-pull).
Variable plate voltage might not be a reasonable/feasible possibility. I guess to do it right I would need a PT with two optional B+ secondaries.

Screen voltage, didn't even think of that... I take it that one might end up with the screen voltage being higher than the plate voltage?
That doesn't seem to be taken into account on the AX84 I was looking at, the documents just show a bias option with a 1K pot in series with a 100R cathode resistor.

[martin manning]

The continuous current limit is there to avoid overheating and core saturation. For the screen voltage, it would come off the plate node, so it will be at or below plate voltage.

Some numbers to illustrate the concept of adjustable screen voltage for different cathode resistors:

Assume a 6L6GC, 5k load, 360V on the plate to get ~345 Va-k, and quiescent plate current at 69.2 mA by adjusting Vg2 (to center the operating point on a 5k load line).

150 ohm Rk: screen voltage is 232, and Vg-k is 11.3.
300 ohm Rk: screen voltage is 343, and Vg-k is 22.5.

With half the idle grid bias voltage, the 150 ohm Rk will break up sooner than the 300 ohm Rk.

[Stevem]

While we are on this topic of SE amps I think it's good info to get out there that every class AB amp will output a certain amount of class A power.
A 100 watt RMS amp could output as much as 20 watts of class A power when designed with that in mind!

A 400 watt or more per side SS amp can output a bunch of class A power!