Here is my updated schematic with NFB & RAW control.
How I got here from the last set of mods: One of the input jacks went bad; the shorting leaf spring would not hold tension. So I built a new input jack assembly. While inside I replaced the old cap can. I had a 4X 20uF can on hand, so I used that and put 2X 20uF caps in parallel for the 1st filter stage. 40uF is the stock value for the AB764 VibroChamp.
Then 2 old problems kept recurring: a) intermittent squealing; b) nasty static like distortion on the attack of an overdriven note or chord. A voltage check showed oscillating DC voltage exceeding 300mV on the far side of the coupling cap to the power tube. This was a blue vintage cap shaped like a Sprague radial cap. To confirm, this cap tested bad on my cap tester. It was replaced with an orange drop, which I had on hand. Otherwise I may have been more creative with cap-type selection. The distortion was tamed with a heavy dose of NFB; also I re-inserted the local feedback suppression cap on the power tube. That smoothed-out the overdrive tone. (I tested NFB values with a 50K pot in series with the stock NFB circuit now tapped-off the 16 ohm OT secondary.)
Preamp gain was fine-tuned to 3.3K cathode resistors to match the 220K plate resistors on the gain stages. A/B tests were done with a Strat. That seemed to produce the best tone, but it's subtle differences.
The RAW control sounds very good on this amp; much better than on my Princeton which gets very ratty sounding with its new RAW control. I think I need to go back into the Princeton and play with NFB values in the hope of taming this rattiness. Will this ever end? :D