Can a split-load /Cathodyne PI be wired to also be a virtual earth mixer ?

[walkman]

I was wondering can a split-load /Cathodyne PI be wired to also be a virtual earth mixer ?

[loudthud]

It's theoretically possible but the gain will be very low because to the huge signal required at the cathodyne input. If you have the customary gain stage in front of the cathodyne, it would work much better at that tube's input. Just get the phase correct and give it a try.

[Chuck H]

You would lose a ton of gain. And the local NFB would be huge. Not sure how this would sound WRT clipped circuits. I've used a cathodyne to drive a pair of VE stages for their lower impedance output. The results were very good.

[walkman]

Would this be a reasonable way to reduce gain / give more headroom for the PI in a 5E3 type circuit then?

[Chuck H]

Yes, and no. As I understand it the 5e3 PI starts to break up right on top of the power tubes. A gain reduction in the PI could actually mean a loss in clean output watts since the power tubes may never see the maximum signal they can reproduce. That's not more headroom in my understanding. On the other hand the increase in local NFB and the reduced input impedance should make it more difficult to OD the PI stage. I haven't spent nearly enough bench time on this to qualify an answer. If you try it, start with relatively large shunt resistors compared to the feed resistor. The reduction in gain goes up and the input and output impedance of the PI goes does down as that resistor value goes is decreased.

[Dave H]

I don’t think you can make a virtual earth mixer out of the cathodyne stage as it is unity gain from grid to plate in the open loop state so there is no excess gain available to make a virtual earth when the feedback loop is closed.

As LT said it would be better to make the preceding driver stage the virtual earth mixer as it has an open loop gain of about 50 but making a decent virtual earth will use up most of that which will be too much gain loss for a 5E3 circuit I should think.

[Steve Conner]

I agree.

[walkman]

So for a 5e3 circuit it would be better then to add another triode stage as a VE mixer before the usual driver (12ax7) otherwise a VE added to the driver or PI would reduce the gain so much that the output tubes wouldn't be driven to their full potential. ( or ad VE to driver stage and add another triode stage after).

Could the gain loss of making the driver stage a VE be compensated for by putting a 12ax7 in place of the 12ay7?

Isn't there plenty of gain in a 5e3 already ? Doesn't the PI get overdriven easily I that circuit? How much gain is required to drive the output tubes to their full potential and at what point does it start to overdrive the cathodyne?

[Chuck H]

I don’t think you can make a virtual earth mixer out of the cathodyne stage as it is unity gain from grid to plate in the open loop state so there is no excess gain available to make a virtual earth when the feedback loop is closed.

That WOULD be a niggle. Good catch.

[Chuck H]

It might help us if we understood the goal a little better.