In amps with the output taken off the emitters I used to use like ~100
ohm resistors from E to B and see how it runs with no load. The
problem is that with Darlington outputs disabled like that it may not
have enough current to even drive the feedback network.

But at quiescence there should be no offset and the collector voltages
should be quite low. Not sure how low but they should obviously track
the input signal. It can't be really detected at the output with that
configuration. You should be able to watch the audio on a scope. I
would think that if the voltage control uses the same feedback loop as
the main audio it might just clip up to max with low output, but you
can still ascertain if the regulators are working. If the voltages are
railed with no input, the failure was likely caused by a fault in the
regulator control.

This configuration cannot be a real class AB because for efficiency it
must turn the non driven end of the output stage fully off or very
close. In the linked article it claims a better damping factor than
class D, which I believe, but I don't think it is going to be as good
as good class AB.

All they have really done is take the final SESAPP stage out of the
class D domain. The "output transistors" actually operate as active
filters, as well as polarity switches. I figured someone would come up
with this sooner or later, and it's not much more (operationally) than
using a commutating power supply, except that there are a near
infinite number of commutators.

Since there is a limit as to just how fast this power supply can
respond to the input, I imagine that sustained high output at high
frequencies are hard on the outputs.

Personally I'll just stick with my Phase Linear 400 II and put a fan
on it, but business is business.

J