Gamma-ray tomography instrumentation for investigating carbon steel pipe corrosion under insulation

• Main Author: Susiapan, Yvette Shaan-Li
• Format: Thesis
• Published: 2016
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/60726/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:93974
• Institution: Universiti Teknologi Malaysia

This thesis details a non-destructive type of tomographic technique to monitor corrosion under insulation. This method requires a radioactive source, where a detector-source pair is mounted on a circular gantry located around the periphery of a steel pipe. A radioactive source with an appropriate activity was chosen based on the thickness of the pipe under test. In order to detect gamma rays, photon scintillation detectors are required. A Thallium-activated Sodium Iodide (NaI(Tl)) detector crystal was used. Stability tests were conducted on the detectors before they were ready to be used. A portable hardware to house the detector and source was then designed and built. The system is able to operate without the need for an outlet socket to enable it to be used in remote locations where power supply may be unavailable. For testing purposes, artificial corrosion defects were created by grinding the surface of the pipeline under the insulation by a minimum of 2 mm in depth and by placing foreign objects (wood splinters and pebbles) under the pipeline insulation. A modified parallel beam sensor geometry was implemented to measure the intensity of the incident beam after passing through the pipeline under test. The Backprojection method of tomogram reconstruction without the need for sensitivity matrices was used. The software written generates three tomograms, which are the ideal pipe tomogram, the tomogram of the pipe under test and the error tomogram which compares the first two tomograms and then highlights in red the suspected corrosion under insulation. The system was able to achieve an average accuracy of 96.283% when a comparison was done between actual dimension of the objects under test and the dimension based on the intensity profiles. The error tomogram constructed was able to detect and highlight artificial corrosion of a minimum of 2 mm in depth.