Ansys Maxwell: Transient generator simulation torque fluctuation

[piyushverma]

TL;DR Summary

while simulation for a PM axial flux generator at no load 1000RPM, shows moving torque fluctuating between -0.7 to 0.7 Nm, which should have been 0, as there is no load attached and it is coreless design so no cogging torque as well. Need someone to help me out.

I am working on an axial flux generator design, I consisits of 2 rotors and one stator, there are 6 pole pairs and 9 coils. Here is my simulation setup

The three phase windings are connected in star configuration through Maxwell circuit.
I created two bands enclosing each rotor and then united them to assign a single band for both rotor which is set to rotational motion at 1000RPM constant. I am getting satisfactory results for output voltages, what I don't understand is this torque curve

currently the stator material is air ( I switched to air after setting it initially to electrical steel laminated, which resulted in 2Nm peaks in torque).
I don't understand the reason for this torque as there is no core so no cogging torque and also there is no load connected.
I am stuck, Kindly help :(

 

[Delta2]

Any generator that relies on motional EMF has a very low current when rotating even in the presence of no external load.

On second thought it puzzles me that the Torque switches sign, according to my explanation it should 've been only negative (or only positive, depending how you define positive or negative here).

Something else, the coils are assumed to be ideal or have ohmic resistance? I think I might be able to explain this if the coils are not ideal.

 

[Delta2]

The poles of the generator act as a capacitor that serves as some sort of external load. Depending on how you have setup your simulation software, it might be able to "see" this capacitor and include it in an implicit way in its calculations.