Using MagNet's Transient 3D solver, the current, torque, rotor pole flux and the air gap point flux density are calculated for a nonlinear transient problem. This example shows how the simulation results that were obtained closely resemble the measured data published in the TEAM Problem 24 definition.

TEAM problem 24 - Nonlinear Time-Transient Rotational Test Rig

This example shows a nonlinear transient problem, which was solved by MagNet's Transient 3D solver. This device is number 24 in the TEAM series of benchmark problems, and published experimental measurements are available for comparison. The problem was designed to be similar to a switched reluctance machine. Each stator pole has a 350-turn coil around it, and the rotor is fixed at 22 degrees relative to the stator. A step voltage of 23.1 V is applied to the series connected coils. A search coil is wound around a rotor pole to measure the total magnetic flux, and a Hall probe measures the y-component of the magnetic flux density at a specified position in the air gap between rotor and stator. Finally, a piezoelectric transducer measures the torque on the rotor.

Using MagNet's Transient 3D solver, the current, torque, rotor pole flux and the air gap point flux density are calculated. The simulation results agree well with the measured data published in the problem definition.


Results

Click image to see full size version
Material properties are critical to accurate calculations. The published benchmark specifies the material B-H curve, but there are not enough data points in the critical region at the "knee" of the B-H curve. By adding one more data point to the material curve, good agreement with experiment was obtained. The graph at right shows where the extra point was added in relation to the original published data. At the TEAM workshop in Tucson, Arizona in 1998, Dr. Takashi Yamada also observed that adjustments to the B-H curve were necessary to obtain accurate results, and published this in: R.S. Bains, H. Hashimoto, K. Tani, and T. Yamada, 鈥淪olution of team problem 24, nonlinear time transient rotational test rig鈥? Proceedings of the 6th Annual TEAM Workshop, Rio de Janeiro, November 7, 1997, pp. 36-38.". This has been confirmed in a paper by Kent R. Davey, "Working Nonlinear Transient Eddy Current Problems with Time Harmonic Solutions", Compumag 2003, Saratoga Springs, New York USA, July 13-17, 2003.

Click image to see full size version
This graph illustrates the comparison of the coil currents between the published measurement data and the results obtained from MagNet. The steady state current agrees exactly, since this depends only on the DC resistance of the coils, which is specified. The error in the transient current can be attributed to both experimental error and material modeling error (due to insufficient material data).

Click image to see full size version
This graph compares the torque values between the published experimental data and the MagNet results. Torque is one of the more difficult quantities to compute accurately, but these results show that MagNet's torque calculation is accurate.

Click image to see full size version
The comparison of the total flux through the rotor pole, measured with the open circuited sense coil, is shown in this graph.

Click image to see full size version
The flux density values obtained from MagNet are compared with the published measurement data in the graph at the right. The published experimental data were obtained using a Hall probe in the air gap. In MagNet, the field sampler tool allows quick and easy plotting of any field over either space or time, and of course it is even possible to probe inside components, something a Hall sensor cannot do.