Induction machines – Motor Factory – Test environment

Run the basic test twice consecutively can lead to wrong results.

When you run the test Characterization / Model /Basic twice consecutively by only changing the line-line voltage, the results you get at the end of the second run are wrong.

Indeed, in that case, you will see that the line-line voltage is inconsistent with the phase voltage meaning that our internal process considers that the voltage has not been modified especially for computing the no-load performance leading to wrong results for the equivalent scheme parameters (ref.: FXM-16302).
  • If you run the test only once,
  • or if you modify the voltage and the frequency at the same time,
  • or if you modify other user input parameters but not the voltage,

In these three conditions the results will be right.

So that, the workaround consists in performing the second computation with not only a new value of the voltage but also a new value of the frequency very close to the one initially considered.

For instance, 50.001 Hz instead of 50.0 Hz.

In that case our internal process will consider that the line-line voltage has also been modified.

The computation of power balance for IMSQ in “accurate mode” is not well balanced.

In the test “working point – sine wave – motor – U, f, N”, while computing the power balance with the accurate mode (i.e., with the transient application) the results are not well balanced. Indeed, the difference between the electrical power and the power on the shaft is not exactly equal to the total amount of losses.

Depending on the considered slip the difference can be about a few percent (ref.: FXM-16121).
Note: The number of computation points per electrical period and the number of considered electrical periods (user’s inputs) have an impact on the accuracy of the results.
Note: The displayed value (in FluxMotor) of the mechanical torque is based on the Finite Element computations and considers the iron losses and the mechanical losses.
Note: The rotor Joule losses (squirrel cage) result from the Finite Element computations.

Computation of working point using the accurate mode is not possible for slips higher or equal to 1.

The internal process is not compatible with such values of slip (ref.: FXM-16163).

The computation of the efficiency map (U, I) with mechanical losses can fail.

This issue raises a null-pointer exception (ref.: FXM-16157).

The flux density is not displayed in accurate mode computation.

While computing a working point (U, f, N) for an induction machine with a skewed squirrel cage and outer rotor the flux density inside the airgap is not displayed. (ref.: FXM-16154).

Error when exporting and solving a project in Flux Skew – Transient application.

This issue occurs when the user input “Represented coil conductors” is set to All phases (ref.: FXM-15877).

IMSQ - Scalar Maps or Efficiency map (U,f) tests fails with hairpin winding technology

Sometimes, the tests Scalar Maps and Efficiency map (U,f) are not correctly solved with a hairpin winding configuration, like for the Motor M1 of the reference catalog (ref.: FXM-15843).

Power balance of No-load working point

Sometimes, computation of the no-load working point (slip=0.1%) leads to a NaN (Not a Number) result. The computed amount of iron losses is not consistent with the power balance (ref.: FXM-12600).

Torque slip curve

Test results are not continuously consistent over a torque slip curve. This occurs with the test Performance mapping T(Slip) - induction machines with a skewed squirrel cage. When the user targets a working point as an added value to be computed

with the whole Torque-slip curve, sometimes this additional working point doesn't belong to the curve.

(ref.: FXM-12599).