DC cathodyne splitter as a gain stage

[Gaz]

Chuck, thanks the for the simulations. Super helpful to compare the different configurations, and thanks for showing it clipped (might be more subtle results otherwise). Right now, the loaded with no grid stop is looking like the ticket. If I'm seeing it correctly, more swing on the output having the unused output loaded, and more compression and weird waveform with no grid stop.

Per my reply to Hemholtz, I'm curious if the output off the cathode would look any different, or how just a DC coupled cathode follower would compare...

[teemuk]

I would feed the two outputs from phase splitter to a differential amp (an opamp is great for this app) and simulate the power tube grid effect with just simple shunt clipping diode clamps. You can generate a negative ("bias") reference to increase clipping threshold. ...What do ya know, with LTP phase splitter in place of the cathodyne we have built ourselves the input part of either Vox Valvereactor or Hughes&Kettner hybrid power amp.

[Helmholtz]

is it really any different than just using a DC coupled cathode follower?

I think a typical DCCF is biased hotter.

[Chuck H]

Well here is a spice shot showing both cathodyne outputs. Grid in green, plate in blue and cathode in red. Both outputs are loaded and no grid stop used. It looks like the cathode output doesn't demonstrate the same reaction as the plate output and is basically a mirror of the what's happening at the grid. If you want the asymmetry you'll have to use the plate. I don't expect any loading issues since you'll be running it into a high impedance circuit.

[Helmholtz]

I would feed the two outputs from phase splitter to a differential amp...

Or just feed the outputs to both inputs of a LTPI .

[Helmholtz]

Well here is a spice shot showing both cathodyne outputs. Grid in green, plate in blue and cathode in red. Both outputs are loaded and no grid stop used. It looks like the cathode output doesn't demonstrate the same reaction as the plate output and is basically a mirror of the what's happening at the grid. If you want the asymmetry you'll have to use the plate. I don't expect any loading issues since you'll be running it into a high impedance circuit.

With symmetrical resistive loading I have problems to understand how the same plate current = cathode current could produce totally different voltage shapes at cathode and plate.
The only explanation I can think of would be considerable grid current adding to the cathode current only.

Edit: I just found that Merlin gives exactly this explanation in his book. Using a large grid stopper for the cathodyne should limit grid current and thus improve output symmetry.

[teemuk]

Do also note that it's one thing what comes out of those two phase splitter's outputs and another thing what comes out when one extracts the difference of those signals, which is what the preceding push-pull stages would do.

I'm not sure what is the specific goal here... do you want to amplify just one signal output from the phase splitter, which is very asymmetrically distorted, or difference of its two outputs, which has more asymmetric qualities? (Excluding deeper overdrive when phase splitter outputs actually become asymmetric in other ways than just being "mirror images" / reverse phase). If the answer is the first option (which is very different than what a complete power amp circuit produces as output), then which one is the "right" output assuming outputs differ from each other as described?

[Helmholtz]

Do also note that it's one thing what comes out of those two phase splitter's outputs and another thing what comes out when one extracts the difference of those signals, which is what the preceding push-pull stages would do.

Did you mean the subsequent push-pull stage?

[teemuk]

Yes, sorry. That's what I meant.

[Helmholtz]

Here is an earlier thread/discussion about the DC coupled cathodyne:
https://music-electronics-forum.com/...yne-perfection

Just being curious: Where does the attraction of the DC coupled cathodyne come from? Which desirable amps use one?

[mhuss]

If you want to get crazy, you could put a resistor and Zener (or string thereof ) to ground on each plate on the fake PI to simulate forward grid conduction of the output tubes.

[Gaz]

Here is an earlier thread/discussion about the DC coupled cathodyne:
https://music-electronics-forum.com/...yne-perfection

Just being curious: Where does the attraction of the DC coupled cathodyne come from? Which desirable amps use one?

The old Orange amplifiers. Not sure how desirable they are, but they have a unique sound I want to capture of. I have compared AC coupled and DC coupled cathodynes in the past and preferred the DC when overdriven - it was smoother and punchier. Again this is with a PPIMV turned down, so the power tube interaction is not relevant to me like it would be in a low watt non-master volume amp.

Or just feed the outputs to both inputs of a LTPI .

I had considered it, but then I can't use the standard style presence and resonance controls I wanted to.

[Gaz]

Do also note that it's one thing what comes out of those two phase splitter's outputs and another thing what comes out when one extracts the difference of those signals, which is what the preceding push-pull stages would do.

I'm not sure what is the specific goal here... do you want to amplify just one signal output from the phase splitter, which is very asymmetrically distorted, or difference of its two outputs, which has more asymmetric qualities? (Excluding deeper overdrive when phase splitter outputs actually become asymmetric in other ways than just being "mirror images" / reverse phase). If the answer is the first option (which is very different than what a complete power amp circuit produces as output), then which one is the "right" output assuming outputs differ from each other as described?

Yes, good point. The goal is approximate the sound of the cathodyne with PPIMV turned down, so that's why I'm not too concerned with the power tube interaction. The idea is to be able to use a regular pre-inverter master volume after the faux-cathodyne into a cleaner power amp with presence and resonance controls. The amp will also have a second channel, and I don't want that one to have the overdrive effects of the cathodyne. Plus, it keeps the channel switching scheme much simpler.

You're right that whichever output I choose will not be the whole picture, but hopefully closer. Afterall, this aspect of the amp is only part of the recipe to begin with. It still seems closer than using an ordinary gain stage in the preamp, but I'll have to do some A/B testing to see what I can hear...

[Gaz]

Well here is a spice shot showing both cathodyne outputs. Grid in green, plate in blue and cathode in red. Both outputs are loaded and no grid stop used. It looks like the cathode output doesn't demonstrate the same reaction as the plate output and is basically a mirror of the what's happening at the grid. If you want the asymmetry you'll have to use the plate. I don't expect any loading issues since you'll be running it into a high impedance circuit.

Turns out the anode is the more exotic output after all! Fortunately, it will be easy to rig this on a switch to hear the difference.

[Chuck H]

With symmetrical resistive loading I have problems to understand how the same plate current = cathode current could produce totally different voltage shapes at cathode and plate.
The only explanation I can think of would be considerable grid current adding to the cathode current only.

Edit: I just found that Merlin gives exactly this explanation in his book. Using a large grid stopper for the cathodyne should limit grid current and thus improve output symmetry.

Yes. You can see the effect of the grid stop in the first two simulations in post #14. That's why I employed it as an option. Here's a shot of the cathodyne plate and cathode while using the large grid stop resistor. Green is the signal behind the grid stop, blue is the cathodyne plate and red is the cathodyne cathode. Note that I'm still intentionally clipping the phase splitter.