Tweet
It's been a good experience reopening my amp with the intent to make things better. I've found a couple of errors that I completely missed the first time around.
I've been looking at the reverb section (taken mostly from the blackface deluxe reverb AB763) with the intent of cleaning up the wiring, especially with respect to possible ground loops through the cable shielding present on signal carrying wires. I had noticed a couple of places where I had grounded the shield at both ends.
And now that I'd just about eliminated overall hum in the amp, I'd noticed that the reverb was still pretty noisy.
With my grounding scheme (trying to follow Aiken, more of a tree than a star) I had a ground for the reverb transformer going back to screen filter cap ("B" in the schematic) and another ground for everything else in the reverb circuit, grounded at the decoupling cap for this circuit "D". The idea being to keep each section's power supply and ground together, tied at that section's filter cap. I had not realized that the reverb circuit is actually two sections separated by the reverb transformer.
I recently read the advice to treat the ground for the reverb driver section separately from the rest of the reverb circuit, the reverb transformer being the dividing line. So I now have a ground wire for just the driver section (the 12AT7 cathode) going back to the "B" filter cap and everything else, including the output of the reverb transformer, collected and going back to the "D" filter cap.
I hadn't been terribly rigorous in my shield wiring which is complicated by the reverb pedal and reverb tank connections. All pedal and tank jacks are isolated from the amp chassis and the tank itself. The pedal (combined reverb/vibrato) is now grounded to the reverb board ("D") at which point are also tied the shields for the vibrato pedal wire, the reverb pedal wire, and both reverb tank wires. The isolated reverb tank box is grounded to the amp chassis through a wire and lug screwed nearby.
With all this wiring cleaned up, I measured ground resistances to verify I hadn't forgotten anything before turning the amp on and discovered that there was now something very wrong, the entire second half of the reverb circuit no longer had a zero resistance ground.
After quite a search, I discovered that when I had reorganized the power supply filter stages (to add a couple more stages) I had forgotten a ground wire along the 'ladder' so the last two filter caps (preamps and reverb) were not grounded to the rest of the ladder. But everything seemed to work. It turns out that both preamps and reverb (all jacks isolated) got their grounds through a combination of signal wire shields. But their decoupling caps went nowhere and so weren't doing any filtering.
Turn the amp on, reverb still pretty noisy. Only the "reverb" knob controls the level of noise and the pedal cuts it out completely. I finally decide to move the reverb tank far away connected through some rca audio cables. No more noise. It turns out that I had hung the tank close to the amp chassis more or less where it sits in the cabinet, but turned around so that the open side of the tank faced all the electromagnetic parts of the amp. Rehanging the tank with the closed side facing the amp and the noise is gone.
All this to say that debugging is fun and things can seem to work even if they are terribly wrong.