Reverb oscillation fixed

[Chuck H]

I wasn't sure if this should be posted in "theory" or "troubleshooting". Well... I chose one.

I have an amp on the bench that suffered a reverb oscillation when the reverb volume was set over 3/4. An actual circuit oscillation presenting as a high pitched whine. Not an acoustic feedback. At first I tried all manor of top end bleeders at different points in the circuit. I narrowed it down to the net gain of the output being the trigger but nothing I tried would stop the oscillation short of a cap to ground so large in value as to be unacceptable. I checked grounds for both integrity and best location. I was just about pulling my hair out as I was moving the pan around to see if by some odd chance it would help. And as I was doing this I layed my hand on one of the speaker baskets and the oscillation stopped! So... Grounding the speakers at the basket or the magnets would stop the oscillation. This amp has two speakers and the oscillation occurred with the pan on either side of the cabinet unless I grounded that speaker. With one speaker grounded, if I scooched the pan near the other speaker the amp would oscillate. I could have just grounded one speaker and left the pan on that side, but I grounded both just cuz.

Interesting observations...

This amp has two different type speakers. One G12H30 and one Vintage 30. The G12H30 has a resistance between the magnet assy and basket of over 100k. The V30 has a resistance between the magnet assy and basket of 7 ohms. Yet grounding either the basket or the magnet assy on either speaker would stop the oscillation. I grounded both baskets.

Using a capacitor of only .022uf in the grounding lead would stop the oscillation. I just used a lead wire from the basket to the negative lead of the spider with no cap, but I did experiment.

Wrapping the dual RCA reverb pan cord around the lead from the amp to the speaker ALMOST stopped the oscillation (did I mention I experimented ).

My reason for posting is that I would like to better understand what I've observed. The problem is solved, but why? And why did it happen in the first place?

[The Dude]

This sounds like (POSSIBLY) one I had in the shop a while back. The magnetic field of the speaker was causing the transducers in the tank to vibrate. Maybe too late since you seem to have already solved the issue, but if you put your finger on the transducers to stop them vibrating, does the oscillation go away? If so, my solution was to apply some glue between the lamination and the coil so that it wasn't loose. It worked well. No more oscillation.

[drewl]

I've seen it alot on older amps with no tank bag with the speaker right next to the tank.

Good fix, good info.

[Chuck H]

I'm still "upgrading" the amp. It's a custom I built for a customer So I'll definitely trouble to unground the speaker baskets, unbag the reverb, touch the transducers and report. I don't have high hopes though. I replaced the two spring long pan in this amp with a three spring short. The reverb seems just as full and the acoustic stability is better. The oscillation occurred with both reverb pans, so, what are the odds that the coils physical resonance was inappropriate for both pans? But I'm checking anyway. Even if I never get a definitive cause, I'm coming away with all the info I can

[trem]

I do not THINK it was Reverb, but I remember Juan Fahey discussing "older amps" that grounded the speaker frame to the amp chassis for some reason.
Wish I could remember the details.......

[Chuck H]

Thanks trem. I've found some similar cases reported when searching on line. I just wondered why this should be the case for some amps but not others. I suspect it's a phase issue with the EMF of the speaker winding talking to the reverb pan output transducer. That might explain why all those old BF Fenders didn't need to ground the speaker baskets. Their particular circuit, which was the same in all the amps, didn't have the bad phase relationship. I could probably also solve the problem by reversing the phase of the speakers!?! Which I'll try and also report on when I unground the baskets.

FWIW I grounded the baskets with a self tapping screw into one of the unused spider assy mounting holes. I sanded the finish off the frame right there and used one of those crimp ring connectors. The other end of the lead wire is simply wrapped over the negative spider terminal and the speaker lead crimp fitting was jammed on. Worked great and only one small lead per speaker with nothing extra to unhook when removing the chassis for service. Just thought I'd share that as an easy way to do it.

[J M Fahey]

It was seen on Japanese amps of the 70/80's , think Roland or Yamaha, some even showed it in the schematic.

The original idea came (I guess) from PA speakers, those overhead ones fed by a long "500 ohms" or 50/70/100V line, up to a block away from the amplifier.

It was found that sometimes static electricity developed and small sparks jumped from frame/magnet to voice coil, and that was stopped by grounding the frames to amplifier.

But in your case I guess the frame is acting as a shield .

[trem]

Ask and ye shall receive.....that was it.
Thank You JMF

[52 Bill]

Roland JC series amps have a ground wire as do Musicman amps. And I think that some of the Yamahas had them as well.

I've also seen a grounding tab that connects the speaker frame to the negative terminal at the mounting rivet.

[Chuck H]

UPDATE

I un-grounded the speaker frames for more testing.

Touching the transducers with the intent of diminishing vibration doesn't help.

Reversing the speaker polarity doesn't help (I was surprised at this).

Rotating the pan 180* doesn't help, but...

Flipping the pan upside down stops the oscillation!?! WTF?

This I don't get. I thought sure that reversing the speaker phase would do it, but it didn't. However, flipping the pan upside down did. Now I have to wonder if reversing the polarity of the pan output transducer would fix it. I wouldn't think so since speaker polarity has no affect.

I re-grounded the frames. The problem is solved but I still don't know why.?.

[Jazz P Bass]

When the pan was flipped, could that have flipped one of the transducer fields?

Just thinking out loud.

[The Dude]

I un-grounded the speaker frames for more testing.

Touching the transducers with the intent of diminishing vibration doesn't help.

Apologies for wasting your time and guiding you down the wrong path.

[Chuck H]

When the pan was flipped, could that have flipped one of the transducer fields?

Just thinking out loud.

I was thinking the same thing. So, if proximity to the speakers is causal, as it seems to be, shouldn't flipping speakers phase have solved it? I thought so, but maybe I'm missing something.

[Chuck H]

Apologies for wasting your time and guiding you down the wrong path.

Not at all! I'm experimenting with this as a learning experience. The problem is solved. I'm still trying to find out why.

[Enzo]

reversing speaker polarity might have an effect if the problem was acoustic, but the electrical fields don't change, The wires have no idea what direction the speaker is moving. In fact I'd probably suspect the speaker wires before the voice coil. Just like grid wires too near the OT in an amp chassis.

Ever have one of those amps with a hum, and the hum is there at zero, and minimizes about 2-3 on teh control, then gets louder as you continue up? That is often caused by two sources of hum out of phase, so you hear the one at zero, then as the other is potted up it raises to the level it cancels the other, that is the 2-3 spot, then as you continue higher, the second hum source takes over.

So just a wild hypothesis, perhaps flipping over the pan exposed the transducer to some field that was in a phase relation to counter the oscillation... or something like that.

Ever try to measure voltage on the plate ends of an OT primary, and one side gives you a proper reading, but the other side starts a HF oscillation with the meter probe on it? Maybe sorta like that.


or it might be something else entirely.

I better go home to bed.