Strong oscillation in Reverberocket Echo circuit

[fredcapo]

An Ampeg Reverberocket R-12-RB came in with a problem. This is the 6SL7 version. Nice amp.

After the amp is on 20 seconds or so a strong oscillation develops when Dimension (Reverb) pot is anywhere off of fully CCW. Even if the reverb is switched off
(input to tank grounded) it will oscillate. If I disconnect the tank no oscillation. (I did not try disconnecting either in or out...will try tomorrow).

The reverb is working as can be heard in the few seconds before the oscillation starts. I only checked so far for DC leakage in the circuit.

Amp otherwise is stable and clean sounding with tank disconnected.

I am not sure what to check next... tank?

Thanks

[rf7]

Does putting your hand physically on the reverb tank stop the squeal? If so, then you need to put it in a bag.

[fredcapo]

You are suggesting that something has degraded in the tank...? I would guess it has been working OK as existing without bag for 40 years. I will try to find a tank with similar impedance to swap in.

[drewl]

Pull the tank out and check if any of the springs are loose and resting on the tank.
Always pull the tank and look for anything not right.

[Randall]

Have you done a visual inspection to the tank to see if all the springs, transducers, and wires look OK?

[Enzo]

The inner reverb spring asswembly hangs withing the thing from four short corner springs. One of those somes loose and it loses isolation and will feed back. Or the inner assembly could have hopped up onto the locator post that some have, then same result.

A broken long spring can do this.

If the rubber grommet "feet" break down over time, it can wind up sitting on the floor more or less, and that can make it feed back. Rubber that was good 40 years ago may not be now.

The suggestion to grasp the tank with your hand is a good one. It can tell you that indeed that is the part feeding back.

[fredcapo]

Thanks. I did not feel the tank itself reverberating when it oscillates...in any event I had another almost identical tank (178 ohm in and out)
that I swapped in and the feedback problem is fixed.

In comparing the tanks I see that both in and out rca jacks have the shield side/grounds tied to the tank chassis, EXCEPT the tank that was causing the feedback the measurement was 11 ohms between rca ground and the chassis on the out side. Not sure where the resistance had developed.

In general I seem to have noticed that some reverb tanks isolate the shield side from the tank chassis on either the in or out, and some dont.

If anyone can shed some info on the reasoning in the reverb circuit design I would appreciate. Thanks.

[52 Bill]

Not sure where the resistance had developed.

Probably corrosion between the jack and the metal casing. You could solder the two parts together to reestablish the ground connection.

In general I seem to have noticed that some reverb tanks isolate the shield side from the tank chassis on either the in or out, and some dont.

For older tube driven circuits, one jack is usually isolated to prevent a ground loop from using shielded cables running to and from the tank, like a typical Fender amp. In amps like Mesa Boogies, they use a non-shielded wire to carry the drive signal to the input of the tank, so the input jack must be grounded in order to complete the circuit.

In modern solid state designs, the input coil is often placed in the feedback loop of the driver opamp circuit so one side can't be grounded.

[Amp Kat]

Know this sounds petty but did you swap the reverb tube ? That tube does take the most voltage and is prone to going south before the others. Also agree with the jacks being corroded and some steel wool on them make knock some rust off. Try moving the cable around while it's on and see if you get any variations or change in dynamics.