Signal bleed out of cathode follower and into phase inverter?

[DanielG]

So I've got two amps on my desk right now that I've put together. One has a basic Sunn Model T style preamp, and one has a Marshall JCM800 style preamp. Both are push pull, although one's 4x6L6 and the other is just an ECC99.

Both amps use the phase inverter from this old ax84 document: https://ax84.com/archive/ax84.com/static/corepoweramps/20W_PP/AX84_20W_PP_Poweramp_Schematic.pdf

The preamps are of course somewhat different, but they both do have a gain stage with cathode follower right before the tone stack. Both amps have separate power supply nodes for the phase inverter tube and the gain/cf tube.

Both amps fundamentally work well, but in both amps, I'm having a problem where signal that seems to be bleeding out of the cathode follower and into the phase inverter.

It seems to be some kind of impedance node issue - from playing with various volume controls, it basically seems the higher resistance to ground there is on the preamp side of C13, the more signal bleeds in. Even if there is no connection from the output of the tone stack to the volume control, or i ground that out, the sound still bleeds through.

The bleed signal is bedroom volume loud, and very thin.

So, the signal is either bleeding through the ground, or bleeding through the power supply somehow. The grounding schemes are pretty standard, and the power supplies are standard as well, with a cap and dropping resistor for each node.

Have any of you guys seen something like that before? What would I do to solve it? I thought maybe I had made a silly mistake on the first amp, but the second amp is exhibiting the identical problem, so there has to be something design related.

[sluckey]

Got a schematic that shows exactly what you have stitched together?