Application of magnetic Barkhausen noise for residual stress analysis – Consideration of the microstructure

Abstract

Abstract Magnetic Barkhausen noise (MBN) is a nondestructive method for residual stress analysis. A great advantage over most complementary residual stress analysis techniques is that measurements can be performed very fast. Barkhausen noise is affected by residual stress to the same extent as by the microstructure of the analyzed sample. Thus, a careful calibration considering the microstructure has to be carried out for quantitative residual stress determination. In the present work, the influence of slight changes in microstructure (i. e. grain size) was investigated for a mild steel. Calibration curves were determined by means of 4-point bending tests. Afterwards, the calibration curves were validated using tensile test samples with known applied stress. The tensile samples were manufactured from a different batch attaining a smaller average grain size than the calibration sample. The stress determined significantly differed from the nominal applied stress due to the differences in the grain size. After consideration of the grain size effect, satisfactory agreement between determined and nominal stress could be achieved. Finally, an MBN residual stress analysis was performed on a deep rolled sample. Complementarily, the residual stress state was determined by means of X-ray diffraction (XRD). Significant differences between the XRD and MBN results could be observed. The results indicate that the microstructure of the calibration sample must be identical with the microstructure of the sample to be analyzed to avoid erroneous residual stress results. Furthermore, the load direction during MBN calibration must be correctly chosen.