Amazon amp quirks - attenuator static

Hey folks, I have a Fosi Audio class D amp that's been on my PC desktop for a while. It's useful and wasn't outrageously expensive.

I have two gripes about this little feller, and I'm interested in seeing if I can tinker with it to smooth things over. Not a value proposition or whatever - I have it and I like to tinker so that's what I'm doing.

Problem 1: volume pot tracks unevenly between channels. Not a huge problem, I'll replace the pot.

Problem 2: Upon power-up there's always static on the volume pot. If I wiggle it around a little bit, it'll make staticky crunchy noises and channels will drop out intermittently. Then they get quieter as I move the knob until eventually the attenuator works quietly with no dropouts. Question: does this mean there's DC on the pot? What would cause this sort of noise?

It's a new sealed pot so I doubt dirt / grime in the contacts is the issue.

What do you think?
 
Its not that old! The pot will be ultra-cheap rubbish I suspect, replacing sounds like a no-brainer (if its standard footprint of course). Its also quite likely there's DC on the wiper too, this isn't the sort of product where there is attention to that sort of detail, but lots of attention to the BoM.
 
Tested with a multimeter. No DC anywhere on the pot while powered up.

Replaced the pot with a dual gang Bourns from Digikey. The only other non-SMD components were 4 off brand electrolytics so I swapped those out for Nichicons while I was in there. Cheap and easy. Replaced the cable from PCB to binding posts while I was in there too. All flux cleaned off (tons of flux mess from factory). I hate Rohs compliant non-Pb solder.

Everything is back together and working great. No noise, pot tracks great. Problem solved. This little thing is considerably more powerful than the F5 I just built, which is cool. It also puts out considerably less heat so I won't think twice about leaving it on.
 
Tested with a multimeter. No DC anywhere on the pot while powered up.
Note that it doesn't take much DC to make a pot noisy, a few nanoamps can be enough if those nanoamps are providing bias current to a bipolar opamp or transistor. The problem is even very short-lived drop-outs on the track removing the bias and causing large output swings. (FET inputs are much less sensitive to this).

This is a circuit design problem and I doubt most meters could see these tiny DC currents anyway (but the opamp sure can!). Its a problem that can gradually develop as a pot ages. Having a DC blocking cap on the wiper (and providing a different route for bias current) is essential with bipolar opamp circuitry to allow a pot to work a long life without crackles.


Or in otherwords look at the schematic to figure out if DC is an issue, not your meter.