Correlation Between AC Core Loss and Surface Magnetic Barkhausen Noise in Electric Motor Steel

Abstract

Core loss is a significant source of energy loss in electric motor steel laminates. Therefore, there is interest in monitoring the quality and consistency of laminates at various stages of manufacturing. The purpose of this study was to investigate the feasibility of using surface magnetic Barkhausen noise for the evaluation of AC core loss, and further, to examine potential origins of magnetic loss in nonoriented electrical steel. Core loss values were measured by a single sheet tester and Barkhausen noise measurements were performed using pole flux control on eight laminates with various grain size, texture and composition. Magnetocrystalline energy was calculated from X-ray diffraction data to quantify texture. Results demonstrated higher surface Barkhausen emissions for samples with lower core loss. Barkhausen noise analyses were used to examine the interplay among core loss, grain size, magnetocrystalline energy and B-H characteristics. The inverse correlation between core loss and Barkhausen noise emissions was qualitatively explained in terms of the orthogonal vector contribution of microscopic eddy currents to losses associated with bulk currents arising in the sample during magnetization.

Local Magnetic Properties in Non-oriented Electrical Steel and Their Dependence on Magnetic Easy Axis and Misorientation Parameters

“…In addition, work by Samimi et al has shown that for NOES, the MBN energy correlated inversely with loss. [15] A similar result was shown by Birsan et al for GOES. [21] Consequently, it is believed that MBN energy can also be applied to NOES to study the effects of misorientation and orientation, and be correlated to loss properties.…”

“…[15] Similarly, the TF variations for the test locations also showed variations on the order of variations observed between samples of different manufacturers. [6] This demonstrated that local variations in microstructural parameters and magnetic properties were present in the NOES strips.…”

“…[23,25] More details about the testing method and apparatus can be found elsewhere. [15] In simplified terms, the MBN measurement involved the application of an alternating magnetic field in the plane of the sample, and measuring the induced voltage caused by the vertical magnetization components projected out of the sample's surface. The rms envelope of the Barkhausen signal is measured, and the MBN energy, defined as the time integral of the voltage squared signal, is evaluated.…”

“…Each location also showed a texture component running from the cube fiber to the gamma fiber at a u 1 of approximately 25 deg, referred to as the a* component. Based on the ODFs, it is expected that the A2 test location would exhibit the highest MBN energy (indicating lower losses according to [15,21] ), since it possessed more easy axes present in the plane of the sample, as indicated by the higher cube fiber component.…”

“…These have included the magnetic texture factor (MTF), [6] the A parameter, [7][8][9][10][11][12] and texture calculations based on the formula for magnetocrystalline anisotropy. [13][14][15][16] However, these texture parameters are typically directional, in that they are calculated with respect to a magnetization direction of interest, typically the RD. This can allow for the better prediction of measured magnetic properties, since the magnetic prop-erties of NOES are typically measured in one direction using Epstein frame or single strip testing.…”