I checked all the grounds for loose connections. Nothing changed. I thought the tranny bolts were tight enough but I really reefed on them. Hum gone. It's like my vibro champ now. Thanks.
Here's what I've done now:
Return everything to stock values, components, and circuitry (no grid stopper nor screen resistors on the power tube) except for
a) all eletrolytics are new but the 2mf/25vdc is now 25mf/25vdc (another site suggests this will reduce the "fartiness" of the bass response)
and the 25mf/25vdc bypass cap (in parallel with the 2W 470 ohm cathode resisitor) is now 25mf/50vdc
b) the 470 ohm cathode resistor is still the 2W pot set to 470 ohms (for measurements as suggested)
c) no 330pf cap on the power tube
Note: this restoration includes
a) removing an extra leg of filtering using a 270 ohm 10W resistor in parallel with a 20mf/450vdc cap immediately after the rectifier
b) reusing the old coupling capacitors (because I want to compare them to the new ones when the bias is set right)
OT primary resistance: 288ohms
After 15min of warming the amp up, (voltages measured to ground unless noted)
red wire of OT -- B+: 378v
pin 3 -- plate: 364v
pin 4 -- screen: 374v
pin 8 -- cathode: 22v
voltage across cathode resistor: 22v
Please let me know if you need more info and note that I will likely put back the extra leg of filtering, use grid-stopper (1k as you suggested) and screen resistors (??) unless you strongly think otherwise. I'm assuming its more important to get the bias set properly first but I'd really like to get the screen voltage down below plate.
I appreciate you taking the time out to help.
The other thing I'm planning to (re)do is to reduce the voltages going to pins 1 and 6 of the 12ax7a. I read that the champ has a "sweet spot" distortion-wise (ie. more blackface like) when the voltages are around 180v. Currently they are 225 and 230, respectively, but were at 183 when I used a pair of 180k resistors instead of the 100k ones. Comments anyone?
Sounds like everything is fine with the output stage of your amp. Don't worry that the screen voltage is higher than the plate. From your voltage readings it appears that the screen is drawing about 2 to 3 mA and that is what you should expect from a good 6V6GT. I just checked an NOS RCA and found the screen current to be 3 mA with plate, screen and grid voltages the same as what you have. The voltages on the Fender schematic are approximate, and are not going to be exactly what you have.
I wouldn't add any additional screen resistance, just the original 1 kOhm 1 Watt. You could add a g1 stopper of maybe 1 to 10 kOhm, but it really isn't needed with a SE output stage. It might reduce blocking distortion a bit.
As far as the plate voltage on the 12AX7's go ahead and play with those values all you want. Maybe put pots in and adjust them to get the tone you like, it won't hurt anything. If you install 180k plate resistors, the 2nd order distortion would be about half of what it is with 100k. The gain would be about 20% higher per stage with 180k plate resistors.
Regarding the original caps, I'd be concerned about the coupling cap feeding the grid of the 6V6. If it's leaky, it will increase the grid bias and plate current. It doesn't look like it's much of a problem now because your plate current is about right with the correct 470 Ohm cathode resistor. An easy way to see if the caps are leaky is just lift the end not connected to the plate and measure the DC voltage with a high impedance meter. If you have more than a volt or two I'd replace it. If you know the input impedance of the meter you can calculate the leakage resistance by knowing the voltage drop across the cap.
About the additional RC filtering, I don't see any problem with that at all it will just lower the B+ by about 10 to 15 V.
Have fun hope you get the sound you want. I've always found that the speaker makes the most difference with combo guitar amps. I'm a big fan of Weber speakers.