AN EVALUATION OF PHOTON TOMOGRAPHY AND OTHER NONDESTRUCTIVE TESTING METHODS FOR THE INSPECTION OF BRIDGE WELDMENTS (ULTRASONICS, RADIOGRAPHY, FRACTURE, FATIGUE)

The computerized tomography (CT or CAT-scan) imaging process is examined as an alternative technique for the nondestructive evaluation (NDE) of weldments. An extensive NDE test program including radiographic testing (RT), ultrasonic testing (UT), and CT was conducted on machine-flawed specimens (phantoms), intentionally flawed T and butt welds, and submerged arc welds from an in-service bridge. Destructive examinations (DE) were also performed on certain weldments to verify the results of the NDE. It was found that CT has a higher sensitivity index for detecting planar defects than the most sensitive RT, but lower than that of UT for defects less than 0.5 in. in length. In addition, CT provides data about the defect geometry required to evaluate defect severity that cannot be obtained by standard UT and RT. These results indicate that CT is a viable means of weld inspection when used in conjunction with UT and RT.;Proposed modifications to the current AWS Code are presented to increase the overall reliability of current weld inspection procedures and evaluation methods. These modifications can significantly improve the probability of defect detection by UT, dye penetrant testing (DPT), magnetic particle testing (MT), and RT. The proposed changes for defect evaluation can significantly reduce the variability of the results currently obtained from RT and UT.;Initial defect size calculations based on the principles of fracture mechanics were performed for four types of planar defects to determine the initial size implied by the provisions of the American Welding Society (AWS) Code for fatigue of butt weldments. These results and the research findings of others indicate that the NDE methods approved by the AWS Code have a very low sensitivity index for detecting the defect sizes implied by the Code.