Date of Graduation
The primary objective of nondestructive testing and evaluation in metallic structural members is to identify and quantify damages and predict member behavior. In this aspect detection of fatigue cracks is of great importance. Steel bridge members are subjected to heavy loads and large number of cyclic stresses. Weak zones, such as near a weld, may eventually develop fatigue cracks. Currently, a major obstacle for ultrasonic field inspection for detecting fatigue cracks is surface preparation (removal of paint and rust). This research describes the development of an innovative ultrasonic testing methodology to detect fatigue cracks in uncoated, painted, and rusted steel members. Optimization of measurement condition was achieved using suitable ultrasonic couplants and constant clamping force. A comparative study was conducted using critically refracted longitudinal waves, symmetrical and unsymmetrical Lamb waves, and Rayleigh waves. Attenuation studies of these waves indicate that critically refracted longitudinal wave is the least affected due to the presence of paint. However, the signal strength for this wave was very low. Rayleigh wave was the most effective in detecting fatigue cracks in uncoated, painted, and rusted members. Cracks were detected using a combination of time and frequency domain analysis. Further studies using Rayleigh waves on 6 m long beams showed their ability to travel long distances and detect cracks in both painted and uncoated members. The study also explored the ability of Rayleigh waves to detect fatigue cracks with various orientations. The study showed that detection of surface cracks, inclined on the surface plane away from the normal to the incident wave, was difficult, especially for the painted specimens. In such cases only spectral analysis indicated the presence of cracks. An expert system was also developed for assisting the user in choosing suitable ultrasonic testing procedures for different engineering applications. VP-Expert knowledge based tool was used to develop the artificial intelligence (AI) system. The AI system includes several complex geometries and NDE characterizations. ASTM standards and other useful information are also incorporated in the AI system.
Franklin, Reynold, "Innovative ultrasonic methodology for fatigue crack detection in steel members." (1998). Graduate Theses, Dissertations, and Problem Reports. 8869.