A few thoughts covering several recent posts:
Stability with tetrode amps should not be a problem if you use ceramic
tubes with the proper bases. More than anything, I think it's the lead
length in glass tubes that brings problems.
As opposed to rf -ve feedback, you might look at envelope -ve feedback.
Here's my take on it: if you build your basic amp, it will probably 40+
dB IMD for starters (and be one of the cleanest signals around anyhow).
At these IMD levels, the distortion is primarily amplitude
non-linearity. Once you head into gain compression, you start to get
AM/PM conversion, so there's phase errors as well. While it's mainly
amplitude distortion, you can apply envelope feedback to get improvement.
If you haven't already, have a read of:
for an overview. Amplitude detectors with lots of linearity are easy,
two diodes and an opamp. Somewhere I think I still have a pile of stuff
from Gosling and Petrovich from the 80s.
Amps mailing list