Interesting to refer to motor mechanics in terms of digital distortion. The "pulsing" you refer to is known as cogging, which refers to the rotation of the rotor through the fixed poles (stator) on a synchronous motor. Depending on the rotational speed of the motor, the number of poles varies; a 300 RPM motor will generally have 24 poles, while at 600 RPM, there will be twelve poles. The overall torque of the motor is determined by the field strength of the magnets; stronger magnets have more cogging and hence more torque.
I find that with most 115VAC synchronous motors, run at their approximately nominal frequency (50/60Hz), it's usually better to smooth out the cogging by running the motor at approximately 65-75% of its rated voltage. For example, I run my 600 RPM/115VAC motor at 55Hz for 33RPM with a voltage around 75VAC on one phase, and around 69VAC on the other. This optimizes the torque to cogging ratio such that the motor doesn't introduce vibration through the drive chain, yet maintains enough torque to overcome all stylus drag during heavily modulated passages.
The SDS is a decent controller, but its lack of parametric adjustment (phase angle, harmonic distortion, and individual phase amplitude) makes it nothing more than a glorified speed controller (though nice for a VPI owner to not have to swap pulleys). It does little, however, to address the differences found even in sample to sample runs of the same exact motor. Without being able to optimize every parameter of the motor, you can never really null out cogging, nor get the best performance out of any given motor.
