Microstructure Effect on the Lcr Elastic Wave for Welding Residual Stress Measurement

The ultrasonic residual stresses measurement is based on the acoustoelastic effect that refers to the change in velocity of the elastic waves when propagating in a stressed media. The experimental method using the longitudinal critically refracted (Lcr) waves requires an acoustoelastic calibration and an accuracy measurement of the time-of-flight on both stressed and unstressed media. The accuracy of this method is strongly related to that of the calibration parameters, namely the time-of-flight at free stress condition (t0) and the acoustoelastic coefficient (K). These parameters should be obtained on a free stress sample that has an identical microstructure to that of the stressed media. Our study concerns the ultrasonic evaluation of the welding residual stresses. This assembly process induces three distinct microstructures in the weld seam: the melted zone (MZ), heat affected zone (HAZ) and the parent metal (PM). Previously, the residual stresses evaluation in the steel welded plates, by the use of the Lcr wave method, was only possible in the MZ and in the PM zones. While in the HAZ, the residual stresses were incorrectly evaluated due to its small width impeding the extraction of the calibration sample. In this paper, we propose an original approach to solve this problem, which consists of reproducing the microstructure of this zone using a specific heat treatment. For the experimental part, P355 steel welded plates were used and the three zones were probed. The results compared with those obtained by the hole-drilling reference method show a proven potential of the ultrasonic method using the Lcr waves. The Lcr wave residual stresses measurements were made with sufficient accuracy, such as the variability of repeated measures was estimated on the order of ± 36 MPa.