Re-tweaked Design

[Howlin' Mad Mac]

Once again here is my "American Girl" design from thread "Errors?" started yesterday. I'm looking for smooth & consistant operation as well as a good balanced gain from stage to stage. To any & all concerned, please review & leave your comments & suggestions for inclusion in its final design & as always, your input is greatly appreciated. Mac

American Girl v2.pdf

[loudthud]

Feedback for the output stage needs to go to the cathode of the 12AX7 before the Master Volume control.

[J M Fahey]

Ok J M, wasn't sure if I could add a presence control to a cathodyne PI or not... guess that's not allowed.

Well, you are still doing it.
The phase splitter shown has no gain (or a gain of 1, to say it in other words) and does not allow fedback.(1)
You must feed it into an earlier gain stage, more precisely the one driving the phase splitter.
Look at other classic schematics which do so.
You must move your volume control (which you probably call "master") 1 triode upstream, between tone controls and Triode#3.

(1) There is one way to do it, but you won't like it, because you'll end up with a very low input impedance, very hard to drive in the tube world.
Just check "inverting Op Amp"
If the final stage (including the transformer) has a gain of, say, 10X , the 56K feedback resistor will force you to use a 5K6 input resistor.
Now ..... which triode will have balls big enough to drive that?

EDIT: oooops !! , looks like we simulposted !

[Howlin' Mad Mac]

Ah ha! Ok guys, light bulb finally kicked in here. So because cathodyne has unity gain I can't tie in feedback loop like an LTP. When using cathodyne must run feedback to last stage in preamp. Makes sense really, just couldn't see that before. Thanks so much guys, you've been a great help!

[loudthud]

Having the Master Volume inside the feedback loop means the effect of the Presence control will be diminished when the Master Volume is turned down from full up.

[Howlin' Mad Mac]

Ok, so if I understand you correctly NFB loop needs to attach to last preamp stage before a cathodyne PI, however if located along with MV control Presence won't work so well so I take that to mean MV should be a gang pot design then between PI & power amp? Good info here... please keep it coming. I'll get this right sooner or later.

[loudthud]

Ok, so if I understand you correctly NFB loop needs to attach to last preamp stage before a cathodyne PI, however if located along with MV control Presence won't work so well so I take that to mean MV should be a gang pot design then between PI & power amp? Good info here... please keep it coming. I'll get this right sooner or later.

No, the ganged dual Master Volume between the phase inverter and the power tubes does the same thing. It reduces the voltage gain inside the feedback loop.

Feedback is a hard thing to understand. I never really got it until I started messing around with op-amps and descrete solid state power supply regulators. The biggest obstacle was the common misnomer that feedback is out of phase with the input. It depends on whether the stage is inverting or non-inverting. In this case the feedback is in-phase with the input but, being connected to the cathode, it reduces the grid to cathode voltage of the input tube. So if the input is +1 volt, something like +9/10 of a volt comes back from the output (thru a voltage divider) and the input tube only sees 1/10th of a volt change between grid and cathode.

[Howlin' Mad Mac]

It's like I'm better off doing away with the MV then & just attenuating the output which would also then allow running it full out (or close) where I'll get nice harmonic distortion at trailer park recording studio friendly levels. Sounds like a plan to me. :-)