Physical Assessment of the Magnetic Path Length of Energy Loss Testers

Magnetic energy losses of SiFe sheets are determined by the standardized methods of single sheet tester (SST) or Epstein tester (ET). According to IEC standards, the total power consumption is measured in Watt-metric ways, and the portion that concerns the homogeneously magnetized region(s) is estimated on a nominal value l
_m
of the so-called conventional effective magnetic path length (PL). A clear definition of PL is lacking-a suggestion being provided here. Standards express l
_m
through the dependence between the magnetic field strength H and the magnetization current. However, they apply the same value l
_m
also for the determination of losses P. From the viewpoint of physics, two PL quantities l
_m,H
and l
_m,P
would be preferable. The problematic is enhanced by the fact that the flux distribution within the magnetic circuit of a tester may change dynamically during the period of time, with induction changes B(t). This means that the physical PL L is a function of time, according to a PL function L[B(t)]. Finally, dynamic changes in flux distributions prove to be distinctly influenced by hysteretic mechanisms which means that L[B(t)] exhibits loop character. The resulting complexity indicates that the concept of PL is not promising, if high accuracy of testing is demanded. However, for purposes of material comparisons, optimum PL values LOPT may be suitable that are determined for individual material families. Analyses of instantaneous loss power values p(t) indicate that energy dissipation is weak during high induction. Thus, determinations of LOPT should be focussed on moderate induction. The ET is characterized by strong modifications of 3-D flux distributions which means that individual values LOPT would be needed. On the other hand, the SST shows a clear design. For moderate induction, some energy dissipation is expected to occur within the yoke region. This indicates that LOPT exceeds the conventional value of 450 mm, in weak ways.