So I have made a PM bearingless maglev motor. It is used to drive a pump. The issue I am running into is that it is not spinning fast enough when pumping a liquid.
I strongly believe that it is saturating because going from 150 to 200 windings did not improve the speed and also doing a speed versus current plot I get the exact saturation curve. Furthermor when I place the entire motor on a thick piece of low carbon steel that the stator is made of it goes faster by ~100Rpm (about a 5mm gap between the motor and the plate because of material in the way)

SO I did some stationary simulations in COMSOL of that design to see how to reduce any point that may be saturating. What I saw is that I needed to make the stator arms bigger so the low carbon steel was not near staturation. (see image)
Again no real improvements


So I figured I should do time dependent studies because it may saturate over time, which I am doing now. Its still not saturating. I am using a BH curve to describe the low carbon steel 1018. But My version does not have the ability to do hysteresis.
I am NOT good with magnetism but right now, I believe that the hysterysis is the issue in the sense that if I am rotating at 50Hz then the steel is staying in the non linear region of the BH curve most of the time.
I believe this has to do with the coercivity of the material? and a material with lower coercivity (like silicon steel) would not suffer from this as much? I likely need to switch to silicon steel anyways for heating reasons. there is also the maglev portion which is switching much faster than the rotation like in the kHz range but operating at a lower current(1 Ap-p vs 4 Ap-p for the rotation)

TLDR
I believe the speed of my motor is limited by staturating.
Simulations are showing its not saturating, but are not accounting for Hysteresis.
I believe a material with lower coercivity would help(from low carbon steel 1018 to silicone steel)

 

I believe that most motor designers try to avoid magnetic saturation.