Not heard back from owner yet, so no sound file yet.
Attacked with engraver and from that , can't imagine the G-forces involved,
passed A1. Whenever I do this with other amps there is always some
microphony showing up with one or more capacitors, but not with this amp.
But...
some things that are not right, if not actually wrong.
The main ps heatsink that carries the switching transistors, has two VR 7812
and 7815 with uninsulated tabs. The second pin of each is taken to ps ground
but also via the tabs and one Allen bolt fixing to the ps ground. Well it
would if not loosened to finger tight , requiring 1.5 turns to tighten. The
other fixing bolt is not bonded. Could some high frequency be induced in the
h/s just by proximity/miniscule leakage of those transistors and affect the
regulator/s ?

The 2 mains caps 200V, 1000uF are 105 deg C and measure excellent ESR and C
cold and hot, but one is slightly domed in comparison to the other. These
caps have ant-vibration mechanical cross bonding to each other with hightemp
hot melt but also fixed and so thermally connected to this heatsink. I can
see that bridge being cut away . Could a problem in one of those caps cause
the owner-observed fault.

The PA is grounded to chassis , but not the PREA, umbilical ribbon
connection only, input is plastic with no grounding ring.
With all the guts of the amp removed from the case then touching the Master
vol pot stem induces a lot of hum. There is no scraping away of paint on the
case for this pot , so could be making and breaking contact .