Date of Award
Doctor of Philosophy
The global production of metal, in particular, steel and aluminum keeps increasing. This material is used with various fabrication processes, such as, welding, forging, and rolling that can induce stresses in the material that can subsequently impact product performance and cause phenomena such as cracking and corrosion. When investigating plate materials it is necessary to map both texture and stress under a range of loading conditions. To address these needs a wide range of both destructive and nondestructive tools have been used. One family of methods are those based on ultrasonic measurements that relate ultrasonic velocity to properties, in particular stress. Two particular challenges are faced which are the relative insensitivity of compression and shear waves to stress and that there are also other factors which can also change velocity and these are temperature, texture and grain size.
This project focused on an analysis of ultrasonic velocity measurements and specifically ways to improve performance and capabilities for stress characterization. Two approaches were considered and are reported: the critically refracted ultrasonic longitudinal (LCR) wave and higher order Lamb waves.
The LCR wave method was modelled and optimized based on the fact that the sensitivity between waves and stress can reach maximum when they propagate in the same direction. However, in reality this wave typically propagates at an angle to stress, which will decrease its sensitivity. This thesis reports a numerical model used to investigate the transducer’s parameters that can influence the directivity of the LCR wave and hence enable performance optimization when used for industrial applications. An orthogonal test method is used to study the transducer parameters which influence the LCR wave beams and this method provides a design tool that can be used to study and optimize multiple parameter experiments and identify which parameter or parameters are of most significance. The example considered simulation of the acoustic field in a 2-D water-steel model is obtained using a Spatial Fourier Analysis method. The significance of the effects of incident angle, the aperture and the center frequency of the transducer were studied. Results show that the aperture of the transducer, the center frequency and the incident angle are the most important factors in controlling the directivity of the resulting LCR wave fields.
The second method considered Lamb wave propagation in the direction perpendicular and parallel to an applied stress. Sensitivity, in terms of changes in velocity, for both symmetrical and anti-symmetrical modes was determined. An available model due to Gandhi, was extended to higher order Lamb modes which were discovered to be more sensitive to stress than either bulk waves or fundamental Lamb modes. The study considered the case of an aluminum plate both analytically and experimentally. Dispersion characteristics were investigated. The experimental system used a pair of compression wave transducers on variable angle wedges, with set separation, and variable frequency tone burst excitation, on an aluminum plate 1.6 mm thick with uniaxial applied loads. The loads used were up to 600 µε, which were measured using strain gauges. The measurement was taken in various locations on the plates to investigate the effects of small changes in plate thickness, the grain size and texture. Model and experimental data are in good agreement. It was discovered that the change in Lamb wave velocity, due to the acoustoelastic effect, for the S1 mode exhibits about 10 times more sensitive, in terms of velocity change, than the traditional bulk wave measurements, and those performed using the fundamental Lamb modes. The data presented demonstrate that there is potential for the use of higher order Lamb modes for on-line characterization of stress in plate materials and that these methods offer potential for higher sensitivity than that reported previously.
Pei, Ning, "Analysis of critically refracted longitudinal and Lamb waves for stress characterization" (2017). Graduate Theses and Dissertations. 17286.