12AT7 reverb driver plate & cathode voltage problems

[David Root]

Here's my problem: 12AT7 reverb driver B+ is 410V off the screen supply node. Initially everything was fine, I got 409V on the plate and 7V on the cathode, more or less on spec.
Now I am getting 385V plate and about 40V cathode, with a standard 2k2 Rk value. In addition, it varies all over from about 15V to 50V on the cathode.
I have used a NOS Sylvania (branded FWIW) and a new EH, both with the same results.
The DCR from the screen node thru the driver transformer secondary is 1,070 ohms, about 10 ohms more than I measured the transformer only before I installed it.
I am also getting intermittent popping/crackling which seems to be coming from the reverb circuit. I have changed all 5 preamp tubes and the PI, to no effect.
I am also using a standard vintage Fender footswitch, but wired to a DC relay so the signal stays inside the chassis, all RCA jacks are isolated from the chassis, and I suspect there could be low level DC on the reverb tank screen(s) as they crackle when I move them. Relay DC supply is tapped off the DC heater supply.
All comments would be helpful.

[rhinson]

sounds like it possible your reverb trannie might have a fault. if you have a spare replace it and see what happens. also check your 2.2k resistor and make sure it's good---see if it's hot when in service. rh

[David Root]

Thanx for responding. The resistor reads 2k2 OK. I don't have a spare but I guess I could get one from Dave Allen. It was made by Heyboer so I doubt it's bad. The resistor is a 1W 700V rated milspec from Westlabs, which I've used dozens of. I can replace the resistor easily though.

I am increasingly wondering if the DC heater tap to the relay is getting in circuit somehow and screwing things up.

[rhinson]

i was going to say to dissconnect the whole relay thing but since you went to all the effort i decided not to with your cathode voltage and plate voltage acting in that very unusual fashion i'd suspect something strange like that. if it's not a great deal of trouble to put things back in stock fender style that's what i'd do. rh

btw, you shouldn't worry about the wet signal "exiting" the chassis with the old normal style switching---you're just connecting the tip to the sleeve to ground it out at the chassis. now if you were trying to do channel switching in this old style way, with say a dpdt footswitch and cable instead of a relay then then yeah your signal would be external the chassis and you could add noise. i wouldn't worry too much about grounding out the reverb wet---it's not even made it back to the recovery stage yet. rh

[David Root]

I think you might be right. I am mistaking voltage extension outside the chassis with actual current flow outside the chassis, right? If all I am switching is open/closed (grounded) circuit at the jack, I don't need a relay.

[David Root]

I discovered today I had neglected to ground the tremelo speed pot thru its 100K resistor. Duh! I put it all together, wire and resistor on the pot, but missed grounding it to the #2 star when I was doing the ground soldering. The ground end was not bared as I usually do, and it managed to hide itself behind the #2 star.

This caused the voltage problems on the reverb driver and the popping/crackling in the reverb circuit, not to mention of course the tremelo didn't function.

I took out the relays and reconnected the rev/vib footswitch per the old Fender layout, and now both work fine. The tremelo is awesome, especially with some reverb too!

[dehughes]

Woo hoo! Glad you fixed it, man. That's a good feeling, huh?

[David Root]

It sure is! And I want to say thanx to rhinson and drz400 too. You don't need relays for shorting reverb and tremelo circuits to ground and you can't be too careful about ground system design AND execution!