Help with Mesa Studio Caliber DC-2

[LarBal]

I repaired this amp a few months ago, that has apparently been a problem on/off for years.
When it came to me it suffered from a poorly performed cap job, where the main filters were actually mechanically broken. Needed new tubes, too.
Amp has been fine gigging etc for months. Yesterday it came back with a new issue, owner reportedly never has heard.

The signal will randomly fade in/out, over the course of 5sec or so. Tone is good, and even when the signal is "out", the amp maintains its normal amount of hiss/background noise.
So it's basically just not passing signal, on occasion. Problem occurs on both channels, essentially isoltaing V1, or V6(PI), or the EL84s as problem areas.

First thing I suspected was the heater string (all 6.3VAC, except V1, which is 6.3VDC). I reflowed all sockets on V1, V6, and EL84s, and the problem persists. I've got 6.3V everywhere I should too.

I've chopsticked, and pushed on the board, and sometimes can get it to act up, either way, when I do that.

Then I noticed that V1 is essentially not conducting, when in failure mode. I'm reading proper resistance to ground from both cathodes, but no voltage there, and no real voltage drop across the plate resistors. Otherwise, all high voltage is where it's supoosed to be. Swapping tubes doesn't help, and I've cleaned the socket, and reflowed it twice.
WHen I have been able to get it to act up by pressing on board, I got proper voltage on the cathodes, and it works fine until I press on it again.

I guess I've got a cold solder joint somewhere, but the fact that all my high voltage on the rail at the plate resistors is good, is baffling me.
I still can't help but think something is wrong with the heater rail, given V1 is the only one with 6.3VDC(to be clear, I'm reading 6.3VDC to ground, from pins 4,5,9, respectively). ALso, that "fade out" sound is consistent with a tube losing it's filament supply, there's no pop or anything strange sounding

I'm not sure where to start looking, and of course it's nearly impossible to try and make sense of the trace layout, plus, I'm not sure the schematic is even 100% accurate.

Any thoughts on what's next?
boogie-studio-caliber-dc2.pdf

[g1]

I still can't help but think something is wrong with the heater rail, given V1 is the only one with 6.3VDC(to be clear, I'm reading 6.3VDC to ground, from pins 4,5,9, respectively).

Agree with your suspicion, I think you might be missing something because you are measuring V1 heater voltage with reference to ground. Check the heater voltage across the heater, one probe to each end.

[drewl]

Had that heater problem on an old Magnetone this week.

Had to replace the socket.

[LarBal]

I'm back on this. I replaced the V1 socket, and no change.
I'm still reading continuity between pins 4+5, and from that to ground is 3.2VDC, and from pin9 to ground is the same, but across 4,5 to 9 and I get nothing.
Voltages good at the Rp, but still no voltage drop accros them or the Rk
FWIW, with no tube in V1, I'm getting 3.2VDC from pin 9 to ground, but nothing from 4 or 5 to ground, so that's probably the biggest clue in quite some time.
I'm also having a hard time understanding the schematic for the V1 DC filaments. It's on the right side of pg4

[g1]

Sounds like the -3V is missing or not getting to pins 4/5. According to pg.4 of schematic, right hand side, V1 gets +/-3VDC.
They don't show the complete circuit of how the DC is derived, looks like it is fed by the 6.3VAC.

[xtian]

Had similar problems with Triple Rectifier. 6vDC supply bad. I was dreading lifting the PCB, so I simply resoldered the four diodes in the FWBR for the 6vDC supply, and that fixed it! Bad solder joint, I guess.

[LarBal]

Thanks a bunch for the help, guys. Turns out I had lost the -3VDC supply. I ended up hardwiring the -3VDC spot on the rectifier, as there were bad joints by the socket as well as at the other side of the board by the DCV origin.