Do you have any material describing and/or examples demonstrating usage of the (new at r12) ARMINT command to model rotation in even and odd periodic sector electromagnetic models of rotating machines?

The ARMINT procedure is described in the attached presentation and implemented in the 4 the attached input files. The system used to test the feature consists of 4 radially poled permanent magnets spaced 90 degrees apart. The PMs are suspended in "air." A cylindrical sliding interface is created a small radial distance away from the radially outermost surfaces of the PMs. Beyond that lies a modeled representation of more "air." The domain inside the sliding interface can be regarded as the "rotor," that outside the interface the "stator."

There are 2 investigated cases:

1) PM polarity alternates (so that a 90 degree periodic sector of the system exhibits even pole periodicity).

2) PM polarity does not alternate (so that a 90 degree periodic sector of the system exhibits odd pole periodicity).

The 4 input files model both cases using (a) full symmetry and (b) 90 degree periodic sector models. The "rotor" position is parameterized. This allows you to test if the fields are properly calculated at the periodic boundaries and overhung portions of the sliding interface in the periodic sector models.

This simple system was selected because it allows instant visual verification of the "stator" fields calculated by the periodic sector models (the field in the "stator" should have the same shape regardless of PM position).

Although these models calculate the steady state field created by the 4 permanent magnets, this modeling procedure may also be used in transient simulations; the objective of these examples is to demonstrate/isolate/verify the ARMINT modeling procedure.

You should find that the full and quarter symmetry model results compare well for both even and odd periodicity.

Please note that the CYCLIC command was not used. Presently, development suggests using macros (such as those supplied in the attached zip file) to establish periodicity with the edge-flux formulation. As of this writing, there appear to be problems using CYCLIC in combination with the ARMINT procedure.

Hopefully this will serve to instruct you on using this strategy to model electric machines with even and odd pole periodic sector models.

Show Form
No comments yet. Be the first to add a comment!