Amp Build Report - Micro-Bassman tube amp in Marshall Lead 12 SS

[pdf64]

The (unshrouded) mains fuse on the board seems to be a (possibly fatal) accident waiting to happen, especially for a learner on an amp that's going to be the subject of much tweaking and tinkering.
Mains power is awful and it would be better to mitigate as much as is feasible the chance of contact with it.
Heatshrink, or other similarly robust insulation over all mains circuit terminals is highly recommended.

For the fizz, maybe the amp is oscillating at high gain settings; can you scope it? The cramped layout may not be helping things.
Grid stoppers of a suitable value (eg at least 8/gm), mounted at the tube socket terminal, for every grid, may be beneficial.

The original feedback circuit seems a bit naff, as it included the tone stack.

[Zozobra]

I've bumped the mains with my soldering iron once, and only once, before. Believe me when I say it really sucks. I now have a healthy fear of electricity.
+1 for putting the fuse in a shrouded holder and insulating it.

[mikepukmel]

Maybe found something, maybe not: in the other cathodyne phase inverter amp schematics I could find, the nfb is injected below the cathode resistor above a very small resistor to ground on the cathode of an input stage tube. E.g. on the Princeton Reverb, which your circuit is similar to, the divider is 1500 ohms to 47 ohms. So its injected 47/(1500+47) of the way up from ground to the cathode. But in the 5F6A-M, its tied right into the cathode of the input stage. Maybe its overwhelming the input stage with NFB. Have you tried something similar to what the Princeton does, moving the NFB injection point below the cathode resistor and above a small resistor from ground? Anyway, question for the gurus.