Moving the PI to earlier in the circuit

[Markusv]

Hello all.

Last night after 3 whiskeys and a bunch of Uncle Doug videos I came up with this question. Please indulge me- I have not built anything serious for about 6 years and am getting back into the game. So I am rusty as hell:

What would happen if you moved the PI to directly after the 1st gain stage and then added another gain stage or two to each of the split lines toward the output tubes?
Here you may find novel tone shaping options but also a bunch of gain permutations.

Has this ever been done?
If so- do you have examples?
If not- why not advisable?

Pros \ Cons etc

Can't wait to hear opinions and thanks so much in advance

Markus V

[Phil_S]

The PI inverts half the signal (phase inverter) so the power tubes are in a constant rotation. This is the essence of push-pull. It also provides the appropriate amount of drive to the power tubes. There's no point to introducing the PI too early in the circuit. If it worked at all, it would be a challenge to control the drive to the power tubes. Pass the whiskey, please.

[Markusv]

Thanks Phil

Yes- I do get the rotation of the Power tubes and the point of the PI. That I am very familiar with.
But I was thinking if a single preamp stages preceded the PI. And if that preamp stage had modest gain- ( or if the PI was Cathodyne for less gain) Surely there would be space for hitting the Power tubes a little harder post inversion? Not to mention different tone shaping options for each output tube post PI?

Disclaimer- not being a smart ass. Just want to hear a variety of opinions.

Markus

[pdf64]

In hifi, I think it was fairly common to have a gain stage after the phase splitter, eg see the 2 designs in the KT66 circuit supplement https://tubedata.altanatubes.com.br/she ... 66_GEC.pdf
More common with bigger valves that need a higher signal level at their control grids; cathodyne has the best balance but lower max output before clipping , so common cathode stages (with local NFB to maintain balance and linearity) after a cathodyne provide gain and increased max available signal swing.
With the extra gain comes the requirement for very high spec OTs, if there’s going to be an NFB loop around the power amp.
But then independent tone controls acting on each half of the nominally balanced signals may make NFB impossible.
Independent tone controls would mess up the balance though, because they’ll cause differing phase shifts. So the output valves and OT would no longer be handling mirror image signals, there will be phase / magnitude disparities. That will reduce the power output, but I’m not sure where the wasted power will get dumped, either the OT, output valves, or both.

[Suckseeder]

Marshall did that with their 200w amp, Cathodyne PI followed by a 12AU7 driver for the KT88/6550

[Markusv]

PDF64
Amazing answer. Thanks- lotsa things I did not consider

Suckseeder-
Thank you I will look for the schematic

You folks rock!

Markus

[SoulFetish]

Silvertone does it in their 1474. They use a Cathodyne phase inverter feeding 6CG7 drivers for a quad of 6L6s. I think it's a great output stage.

[Markusv]

@SoulFetish
Whoa!!
That is different.
And on another note- did u see where the standby switch is located?

Thanks so much!

[SoulFetish]

Yeah, I like how Silvertone implements the standby as well. They have this in a lot of models.