@richardbrand
Re feedback, IMO it is fundamentally the same. We are comparing an output to an input and correcting if needed. Sure in a TT we are looking for a constant output, while as, you say, in an amp we are looking for a constant multiple of the varying input. But in both cases we are correcting, if necessary, a difference in output compared to input. I think that effectively we are saying the same thing.![]()
I have very limited experience with idler drives. so temper what follows with that in mind.
This is how I see it. If the idler is using a synchronous motor, there is what could loosely be called feedback in that the rotor is following the stator's rotating field synchronously. The rotor lags behind this field in phase but travels at the same average speed. If a higher torque demand event occurs, this phase angle increases a little in response but on average the rotor still rotates at the desired speed. Reduce torque and the angle decreases but, on average it still rotates at the desired speed. The rotating field is thus compliant. So microscopically this angle is modulating about the mean torque demand which also modulates the current demand. The result is tiny momentary speed changes. The platters inertia helps smooth this but does not totally eliminate it.
The same principle applies with a BD or DD. The drive is preconditioned to the present torque demand and the energy stored in the platter is fixed, at that moment in time. Take energy out and the platter must slow down because all drives are in some way compliant, feedback, if used, is not instant and the platter's moment of inertia is not infinite. The core difference being a BD has an extra compliant link in the drive chain. You could argue either way whether an idler wheel is compliant or not.
Years ago I observed this vicariously by scoping the motors current demand in a Goldmund Studio. It was not steady when simply spinning a platter and was less steady when playing a record.
You will not see this speed change by looking at a strobe. And the GMTs speed errors are measurable.
Cheers.

