McIntosh MC2100

[xtian]

Amp is reassembled (top cage and bottom cage), and has been surviving repeated power cycles.

I'm reconsidering the idea of oscillation (since I was never able to see anything odd at the output, nor scope the driver cards while operating). Also, consider that, every time during "fail" mode, the amp was still passing clear, undistorted audio.

Consider the thermistors R201, R202 in series with wall voltage. I noticed that these heat up to about 150°F soon after startup. Are these devices known to age poorly? Could these be failing short, drawing excessive current before it ever gets to the power transformer? That would explain a lot.

[Helmholtz]

I'm reconsidering the idea of oscillation (since I was never able to see anything odd at the output, nor scope the driver cards while operating). Also, consider that, every time during "fail" mode, the amp was still passing clear, undistorted audio.

As said, oscillation should show at the output.

Consider the thermistors R201, R202 in series with wall voltage. I noticed that these heat up to about 150°F soon after startup. Are these devices known to age poorly? Could these be failing short, drawing excessive current before it ever gets to the power transformer? That would explain a lot.

As the thermistor are series wired, they can't cause excessive current.
They are supposed to get hot and low resistance.

[xtian]

As the thermistor are series wired, they can't cause excessive current.
They are supposed to get hot and low resistance.

Oh. Are they a "soft start" component?

[Tom Phillips]

Oh. Are they a "soft start" component?

Yes. They are negative temperature coefficient (NTC) resistors. As they heat up, the resistance goes down until the amp is basically unaffected by their presence except for the heat they give off.
They must be sized correctly for the application for proper performance.