Motor-CAD and experimental back-emf NOT MATCHING

[marcoslaviolas]

Hello everyone! I have 5 pp, 15 slots, conc. windings IPM motor, and I'm having issues at matching the results of Motor-CAD with the experimental ones, because this highly affects the On load tests  in correspondence of the rotor positions where the differences lay.

I attach the model and the comparison of the phase and line-to-line voltage on an Open circuit test, and what I noticed is:

1. the motor-CAD curve has "holes" at 90° and 270°, while in the lab test these are shown before 90° and 270°
2. this happens for every phase on every electrical cycle
3. the Motor-CAD back-emf is the same as the one calculated by "Simcenter MAGNET" that has been previously used to simulate the motor

I uselessly tried to:

• add static and dynamic eccentricity
• slightly incline the magnetization of the PMs
• add a skew angle
• modify the .dxf file imported, making the magnet pockets different from each other (not symmetrical anymore)
• double check all the materials and the geometry

but I don't know if I'm still missing some other parameters to modify, so I ask you whether I'd better let it go or try to do something else.