'custom' RI Vibrasonic reverb driver oddity

[Alex R]

OK, this one had me stumped last night. I'll take another look today, but in the meantime...

it's a Fender reissue Vibrasonic. It came in with crackles and pops. On the 12AT7 reverb driver the cathode resistor was burned open, and the bypass cap had popped. I replaced them, and plugged in a new valve as it looked like a possible short in the old one. Brought the amp back up and everything seemed to settle down fine. But there was no reverb. Investigation showed a fat signal going in to the grids of the reverb driver, a tiny distorted one on the plates. Plates had well over 400v coming through the transformer primaries as they are designed to, cathodes had 22v (which is way more than the design 7.8v), grids zero. Looks like it's biasing itself into cutoff maybe? But why? Tried another 12AT7, same result. Checked the reverb transformer primary - about 10 ohms, looks ok, no leaks to ground. Checked the new cathode resistor - fine at 2K2.

My only idea is a transformer problem, but I don't quite see how/what/why. Any ideas?

[Enzo]

SHorted turns? I don't recall ever measuring one of those. Is 10 ohms a good reading?

A short to frame or to secondary wouldn't hurt the tube or cathode resistor, such a short would shunt the B+ off to elsewhere. The only way to burn up that tube and resistor is excess curent through them, and the only place that current comes from is through the transformer. Unless you see evidence of arcing somewhere. Or some loose hardware in the chassis causing shorts. But within the schematic, it comes through the tranny.

The current won't affect the bypass cap, but if the voltage across the resistor exceeds the cap rating, then it fails.

How about this: Tube shorts, puts 400v on the cathode, damage ensues. Now excess current melts insulation in the transformer. Transformer may not have melted, but all it takes is a couple shorted turns and it won't act like a transformer anymore, nor will it act like a proper load.

Also possible, but seems less likely to me, transformer fails, and stresses the tube, which THEN shorts and so on.

Transformers get away with relatively low resistance in windings because of the transformer action. But when turns short together, it drowns that action and nothing is left but the resistance - which you measured as pretty low.

Test the transformer. Cram a signal into the secondary and see what comes out the primary. If turns are shorted, then it won't pass much signal, it will load it down. And your description of the plate signal sounds loaded down to me.

Doesn't RG have a transformer tester on his web site? One shorted turn won't show up on your meter, but it will certainly load down a transformer.

Try subbing a different tranny in there. DOn't have another reverb tranny? SUb anything. It only needs to be on for a couple minutes, and all we want to see is does it make reverb drive. We don't care if the impedance is wrong, or if it doesn't make nice tone, we just want to see if it makes it work - as opposed to the one in there now.

I have a old Fender Bassman OT in the shop I use for just about anything. I'd connect it in place of the existing one in a heartbeat. I'd clip it into a MArshall if I wanted to see if the OT there needed replacing. If you had a CHamp OT, that would work. if you felt technical, run half the bassman primary insteaad of the whole thing, but really, I wouldn't care either way. It's just a go/no go test. Really, what would happen if the impedance of some oddball OT was "wrong?" All that would happen was the drive signal into the reverb pan wouldn't be as nice. I got to think if your signal comes back at the plate, then it was indeed the tranny. OT from a table radio - anything.

[Alex R]

Thanks Enzo. I keep for such purposes a 50W Marshall OT that I can try - didn't think of that. The reverb txfr in the amp measures 10 ohms primary, 1 ohm secondary; don't know what they're supposed to read. No I never found a bad reverb transformer, but there's always a first time and I never found a burned cathode resistor on a reverb driver either!

However...

I found something else that seems likely to explain it all. An old mod has replaced with a 470K the 3M3 (R32 - I'll attach the full manual) resistor across which the reverb signal is drawn and returned (the one that's usually a conjunctive filter, but not in this amp). Done in pursuit of gain I'll be bound. Can you think of any way the resulting feedback through the reverb circuit could have caused the damage at the cathodes of V3, and the current 22v overbiasing? (the 1 meg and the 500pf on its grid are fine btw).

[Alex R]

ok I put the 3M3 back and the cathode volts on the reverb driver V3 dropped from 22 to 9. So this must be caused by the fitting of that 470K instead of the 3M3 at R32 and the resultant feedback?

Then I clipped in the Marshall OT and back came the gain. So the reverb transformer is shot too. Mm - ok I can fix it now - but not sure I understand why/if that R32 mod - which must have made the reverb pretty unsatisfactory - could cause the overload. Any theories?

[Alex R]

High-freq oscillation, feeding back through the reverb, mm? Wouldn't it have to get through the springs though?

[52 Bill]

That's a weird one. I can't see why the mix resistor value change would cause the voltage disruption on the driver tube. If you experiment to find if it's high-freq feedback, try shorting the tank return signal to ground and then measure the voltages.

I didn't look at the schematic, but does this amp have the fet reverb switching? If it does, maybe a problem there?

[Alex R]

No, no FETs in this. The transformer is shot, but the cathode volts dropped before I clipped in the spare OT. Maybe when the new reverb txfr arrives I'll try the old one first and see if the volts stay away from the cathodes, just to check. Yeah, and ground the return signal. Anyhow the main thing is, it will fix, but I'm still puzzling.

Is that mix resistor mod a change anyone has heard recommended?

[Enzo]

No, and I can't think of how it could be involved offhand.