is this a solution for a pcb ground loop??

[donovas]

i've been messing with a cheap ebay amp pcb. i'd rather not disclose the model for the fear of hurting the seller, and my construction is always less than perfect, so i don't want the blame shifted all to the product, but, this amp was suffering from an oscillation in one channel. (it's a stereo pcb) the channel farther away from the psu caps, as expected. the flaw shows up as a static noise in high frequency rich moments such as piano or... yeah, mostly just piano.

i try everything i known until on a whim i just run my finger over the pcb, and to my surprise the noise goes away when my finger's over certain spot. so on a further whim i lay aluminium foil over where my finger was, and now the noise is gone.

what could've been the issue and how did this foil solve it?

one culprit that i can think of is my resistors being inductive.

 

[Enzo]

Your finger or foil acts as a shield to block sensitive areas from picking up signal from elsewhere in the circuit.

 

[keantoken]

It's almost impossible to tell without seeing the PCB or schematic, but when a circuit is this sensitive to a nearby finger it's usually a construction or design flaw. Ground loops can cause oscillation, but if foil stops the oscillation then your problem is probably with FETs oscillating or the stability compensation needs fixed.

 

[DF96]

'Inductive resistors' are almost certainly not the problem. Capacitive feedback is the most likely cause. Your finger or foil alters the stray capacitance.

 

[AndrewT]

I wonder if the supply rail decoupling is in the wrong place, or maybe omitted?