Vibration-proof bias pots?

[glebert]

I am working on a homebrew combo tube amp that a friend of mine built. One of the problems is that with the internal speakers it breaks up very badly, external speakers it doesn't happen. Tapping around the chassis shows that it seems like the bias pots are being bounced. This is a top-load amp, and the bias pots are mounted on the vertical side of the chassis facing the speakers. The chassis seems like it is aluminum and has a bit of flex and the chassis/cabinet mounting is not super solid (chassis supported mostly from top). Don't have pics right now but can get them.

Bias pots are Alpha 50k linear panel mount pots. Cleaning them did not help. I was thinking about changing them to multi-turn but finding panel mount multi turn pots of that value (or close enough) doesn't give a lot of options, they are expensive, and I don't know if that will actually help. Looking for suggestions either for pots, or if I should try to either reinforce the chassis/cabinet (add crossmember or something), or somehow decouple the pots from the vibration of the chassis. Other option is to give up on pots and just hardwire resistors in.

Thanks in advance,
Greg

[J M Fahey]

I think your problem lies in the general construction flubbyness, rather than just the pots ... if at all.

Tapping near them leads to noise but actually whole chassis is flexing , the real bad contact, because thatīs what it is, can be anywhere.

Besides tapping, wrap long nose pliers tips with insulating tape and pull/push/twist everything, and I mean everything.

Besides components and wires, every single nut, bolt, rivet, terminal strip, etc.

Refresh all solder joints, I very much suspect cold solder or that case where component wire is rusty/dirty and solder makes a pool around it (so offering *some* contact) but not "wetting" it , so no real soldering.
Also check tube sockets, jacks, etc. Even power switch and fuse holder.

Besides that, replacing bias pots with a fixed resistor chain is fine, millions of non adjustable bias amps have been working forever.
Just set one particular bias voltage you like and replace pot.

Good luck

[mozz]

They do make a pot (screwdriver adjust)with a locking nut, or at least they did 40 years ago. I have some here pulled from military equipment. Have you tried a locking compound such as nail polish on the shaft?

[glebert]

I think your problem lies in the general construction flubbyness, rather than just the pots ... if at all.

Tapping near them leads to noise but actually whole chassis is flexing , the real bad contact, because thatīs what it is, can be anywhere.

Should have been more clear in my original post that tapping the chassis with no signal sends out a bit of a "cracking" sound, like the sound when it breaks up with signal, and as I tap closer to the bias pots it gets worse. Do agree on general flubbyness of the chassis though. Construction quality definitely dodgy in spots, but he did good work in others. In general pretty good considering he is a lawyer. Sounds great too (it is a 22 watt plexi clone).

[g1]

I'm guessing it's set up so you can bias without chassis removal?
You would need to sacrifice that, but how about 'shock-mounting' the bias pot(s) on a glob of silicone/caulking? (or some other type of rubber shock mount for the bias pots)

[Chuck H]

Are we even sure the bias pots are the problem and not their wiring or connections. I'm having trouble reconciling that the pots internal structure is failing due to vibration. More likely a cold solder joint or intermittent short or fault. Maybe the lug to track connection is failing due to overheating. I've seen that happen a bunch with Alpha pots. Even an internally shorted lead wire. Consider that ALL combo amp pots are technically mounted to the chassis and the whole amp is vibrating when you play through it.

[glebert]

I'm guessing it's set up so you can bias without chassis removal?
You would need to sacrifice that, but how about 'shock-mounting' the bias pot(s) on a glob of silicone/caulking? (or some other type of rubber shock mount for the bias pots)

I was thinking of that an an option as well.

Chuck H, no, I am not actually sure that the pots are the problem, just that the area where they mount is the most sensitive on the tap test. I could just swap them out with similar pots and see if that changes anything.

[Randall]

I'm also having trouble accepting that the bias pots are failing, unless it's actually broken. Seems to me all you would need to do is monitor the bias voltage with a scope or DVM and see what happens when you tap the chassis in that area. My guess is not much. And even if the pots were somehow fluctuating, I don't see how that would make a bad breakup sound in the speakers.

[Enzo]

Yes^^^^

To me the thing that sticks out is that if it were the pots, how odd they both broke in the exact same way at the same time.

[g1]

And since we're considering whether they could even cause the problem, what is the circuit configuration, standard rheostat in series with resistor from bias line to ground?
Or some other arrangement where the bias voltage passes through the wiper?

[glebert]

Yes^^^^

To me the thing that sticks out is that if it were the pots, how odd they both broke in the exact same way at the same time.

This is an odd amp, it is built to have two pairs of tubes that can be configured separately (like a pair of 6V6 and a pair of 6L6) so each pair has a bias pot, not one per tube. I am only using one pair because there is one socket that seems to have a short on the other pair. So there are two bias pots, but I am only using one currently.

Based on responses I am thinking that I will 1.) replace the existing pots with something similar, perhaps not Alpha, and reflow any questionable looking joints while I am in there 2) make a crossmember to address chassis flex (I am building a different cabinet anyway so will have the tablesaw and kreg system out anyway) and if those don't take care of it figure out a #3.

Thanks,
Greg

Oh, here is the schematic I got with the amp LOL.

[J M Fahey]

Ouch!!!!

[Chuck H]

Oh, here is the schematic I got with the amp

Wow!!! This is proof there should be a way to light things on fire in cyber-land. We seriously need to do some trash burning and this goes on the pile.

[glebert]

Wow!!! This is proof there should be a way to light things on fire in cyber-land. We seriously need to do some trash burning and this goes on the pile.

To be fair, this is something he built for himself and never planned to sell, so the schem and notes were just for his own use, and he probably had some other schems he was working with since the plexi clones are popular. After he had problems with it sat on a shelf for 5 years and finally I bought it for very little money.

[J M Fahey]

Based on this mess which should be (presumably) more or less accurate, at least 90% of it, spend 1 hour drawing your own.
When in doubt, you have the real thing in front of you.

*Then* we can roll up ur sleeves and get to work.