Iron Loss calculation methods

Iron losses can be calculated using either Bertotti (Classical or Maxwell) or Modified Steinmetz methods. The iron loss calculation method is selected in the General Losses (Magnetics - BPM, SYNC, SRM) section.

Iron loss data from steel suppliers is almost always measured with an Epstein frame subject to a sinusoidal flux density. Consequently, the provided iron loss coefficients are valid for laminated magnetic cores.

Bertotti iron-loss method (Classical)

The (Classical) Bertotti iron loss method has separate terms for hysteresis, eddy and excess loss:

where Lamination thickness is in [m], f is in [Hz], and B is in [T].

Kh , α and Kexc are determined using curve fitting techniques in Motor-CAD from one or two specific core loss curves.

It is possible to force α = 2 and adjust the other factors, but it is recommended to use the automatically determined coefficients for a better accuracy.

Bertotti iron-loss method (Maxwell)

The Maxwell Bertotti formulation is the same as the Classical formulation above, with an additional 2π2 factor in the eddy coefficient. This method should be used if wanting to validate Motor-CAD iron loss calculations against the iron loss calculations in Ansys Maxwell. These are the coefficients used when exporting a Motor-CAD model to Ansys Maxwell.

Kh and Kexc are determined using curve fitting techniques in Motor-CAD from one or two specific core loss curves.

Modified Steinmetz method

Another approach is to characterise the iron loss variation using the Steinmetz equation, with separate terms for hysteresis and eddy losses:

where f is in [Hz], and B is in [T].

Kh ,α, β and Keddy are to be determined using curve fitting techniques in Motor-CAD from three or more specific core losses curves.

Iron Loss Calculations

For single point on-load calculations, we use the equations above to estimate the iron losses for laminated cores, where we assume a sinusoidal flux density waveform (like an Epstein test frame). For a rotating motor, we have non-sinusoidal flux density waveforms, and so the above equations are modified to include the time derivative of the flux density components.

An additional term is also included for transient calculations to estimate losses from minor hysteresis loops, controlled by the parameter Kc. This coefficient is an empirical value and cannot be calculated. Based on previous studies it is set to 0.65 by default, but this value can be changed by the user. When calculating iron loss coefficients from steel loss data, this parameter is not fitted, since it is only relevant to rotating machines, capturing effects absent in Epstein test frames.

Note that when modelling solid cores (non-laminated), iron losses are calculated based on circulating eddy currents in the solid components, and the iron loss equations above are not used.

Losses are displayed in the Losses [Induction Machine Output Data] or Losses [BPM Output Data] output sheets.

See also Iron Loss Coefficients.

  1. G. Bertotti, "General properties of power losses in soft ferromagnetic materials," IEEE Trans. on MAG, vol. 24, no. 1, pp. 621–630, Jan. 1988
  2. D.M., Ionel, M. Popescu, M.; MI, McGilp, TJE Miller, S.J. Dellinger, R.J. Heidemann [2007] "Computation of Core Losses in Electrical Machines Using Improved Models for Laminated Steel", IEEE Transactions on Industry Appl. – Vol. 43, No. 6, Nov/Dec 2007 – pp. 1554-1564
  3. Allen N, Lai HC, Rodger D and Leonard PJ, IEEE Trans., Vol. MAG-34, No. 5, September 1998, pp. 2535-2538
  4. Hameyer K and Belmans R, Numerical modeling and design of electrical machines and devices, WIT Press, 1999
  5. Salon S.J. Finite element analysis of electrical machines, Norwell, MA: Kluwer Press, 1995