I talked to the tech at work and he has never heard of a problem like this. My biggest issue is that the servo amp is made specifically for the driver. I cannot fit the beast in my car. Even if I get it in for the tech to look at it, I don't know what kind of source I could play for him to see the problem. I think I will stick a multimeter on the right input terminal of the sub (receiver is plugged in left) and see if I can get anything abnormal.
I agree this is a really strange fault.
However it is a one way analog connection, and the sub can not send an audio signal back to the receiver, but it could send DC or voltage spikes back back, if there is a power supply problem or a failure of an isolating cap in the sub
You won't get anything meaningful unless you have a high impedance amplified meter. A scope would be best.
What I would do the find the problem, is put a Y at an audio generator or sub end.
I would then connect the sub to one limb and the scope to the other. I would watch the scope as I tried reproducing the fault by turning up the generator output, and see if there was a DC offset or voltage spikes that could be precipitating the fault.
If all else failed I would I would replace the generator with the receiver and see what happens when I caused the drop outs.
One simple thing you might do is put a big 50 or 100 mfd cap in series with the hot line between the receiver and sub.
If you could not then cause the drop outs, that would be very good evidence that the sub is presenting DC to the input terminal at load.
If the problem is transient spikes though, the cap would not stop the problem likely. That is why a scope would be best. Also the damping of a meter would likely miss fast transient spikes.
I'm really curious to find the cause of this. It is one for the books. You have to find out what is going on, as I think your receiver could well be at risk connected to that sub.