Thomas Vox V1032 troubleshooting help?

[MWaldorf]

Shorting R103A didn't make a difference.

I really appreciate your help here. When you built #1, which version of the circuit was it? A V1031? V1021? If there's something that you know worked and I can reproduce it, then at least we'd know on some level if the issue is something to do with my amp's specific parts (or me) or if the issue is something about the interaction of, say, the e-tuner or the reverb and the rest of the system.

[R.G.]

I built it as a Cambridge V1031. No joy there.

Curses about R103a. It would have been nice that there was one place to point.

I may have to go populate another board and do some testing here.

[MWaldorf]

I just disconnected the e-tuner circuit and I'm still getting the same issues.

You know, instead of populating another board, I could just send you mine, chassis and all...

[R.G.]

I just disconnected the e-tuner circuit and I'm still getting the same issues.

You know, instead of populating another board, I could just send you mine, chassis and all...

I'd do that - except that my bench time is so very limited that it might be months - really! - before I could get to it. My life has a lot of interruptions, and that keeps me from focussing on one thing and getting it fixed.

It may come to that, but you're likely to be happier if I can figure it out in the short intervals I can think about it while facing the screen.

[MWaldorf]

Oh well, it was worth asking. Whenever you have more thoughts on things to test, let me know!

[R.G.]

OK, here's today's desiderata.

Logic tells me that you're still having a ground problem. I know, nothing we found showed that. But the symptoms scream it. On the Cambridge, the grounding was - well, unusual. The controls panel and controls themselves were grounded only by a couple of wires to the input jacks. The control panel itself wasn't really grounded. It occurred to me that you were testing with an open chassis. Does the control panel show continuity to the main ground point, the negative side of the first filter cap?

[MWaldorf]

I'd noticed the odd grounding, and as I've got the chassis open, I put an alligator clip on the control panel and the chassis to make sure there's continuity. There's no resistance between the negative side of the filter cap and either the chassis or control panel.

[R.G.]

OK, this is officially making me nuts now. Thinking, thinking, thinking ...

[R.G.]

Mel, can you re-run the test where it goes into oscillation at higher volume control settings, but with one of your 'scope probes on the +31V supply and the +27V supply?

These things can be grabbed with a scope probe on the north end of R53A, immediately under C34 (+31V) and the + pin of C34 or C34b. I speculate that perhaps something might be happening in the power supply to starve the regulator of voltage, and the oscillation may be the regulator shutting down in protection, then coming back in. Not sure why that might be happening, but it might produce the symptoms we see.

[MWaldorf]

OK, I scoped it two ways, first with probe A on C34A+ and probe B on R53A(N):
https://youtu.be/4i69sipeEDk

Next, with a single probe, having the tip on C34A+ and the ground on R53A(N):
https://youtu.be/yhYq-KTc1FQ

[R.G.]

Say, Mel, what was the amplitude of the drop in +27V on those dips when the oscillation was running?

Background:
We know from earlier stuff that the oscillation gets back into the preamp. There are only about five wires it can do that on: three inputs, one ground and one power supply. We've pounded on the inputs and ground, that leads me into being suspicious of the power supply. In a single-ended amplifier like this, there isn't much if any power supply rejection, so power supply noise feeds right into the stage's output transistor collector.

The idea of using a voltage regulator to replace the resistor-capacitor combination that Thomas used was to force the supply to the preamp stage to be constant no matter what the rest of the power supply did.

[R.G.]

And thinking on it some more, does the break into motorboating oscillation happen with no speaker attached?
(translates as "maybe this is an issue with power supply loading getting transferred back to the preamp)

[MWaldorf]

At C34A+ the voltage changes from 27V to about 24V when the oscillation starts. At R53A(N) it goes from 31V to fluctuating in the range of maybe 27-30V. The scope shows no motorboating with the speaker disconnected. Maybe that's the fix - no speaker!

[R.G.]

OK. That shows that somehow that full-blast power supply sag is riding through.

OK... thinking...

[MWaldorf]

I gather sag is the reduction in voltage caused by transient current spikes. Should I try measuring the current at various points in the amp?