False bi-directional anisotropy in two-dimensional rotating field experimental set-ups

Abstract
Different types of geometrical configurations have been used in experimental set-ups, to measure the magnetic behaviour of electrical steel sheets submitted to magnetic rotating fields. Commonly, the magnetic samples have the forms of discs, squares or crosses. In our laboratory a cross-shaped sample is used in rotating field experiments. Due to the geometrical configuration it is difficult to obtain a rotating field with constant amplitude, if the material approaches saturation. In this paper a 2D-finite element processor developed by us is used to determine the magnetic field in the central square region of the cross sample. The results have shown that the magnetic field does not rotate as in the linear case, being stronger along the 45° directions. It is also shown that there is a lag angle between the field direction in the linear case and the actual orientation. This angle, also present in the magnetic flux density, is similar to what would be observed if the sample material presented a bi-axial anisotropy.