Design and performance of a compact high-energy computed tomography system for the study of metal solidification
Author(s)Jureidini, Imad Maurice
MetadataShow full item record
The aim of this work is to study the feasibility of a computed tomography (CT) sensor useful to the metal casting industry to improve product quality and productivity. X-ray CT allows the non-destructive measurement of the two-dimensional density distribution in a given plane through an object. The solid and liquid phases of aluminum differ in density by approximately 10%, allowing CT to discriminate the solidification front. A compact high-energy x-ray CT system was designed and built to observe the solidification front in an aluminum sample 15-cm in diameter. X-rays photons were produced by a compact 6 MeV linear accelerator (linac) with an average energy of 1.3 MeV. The linac was pulsed at a rate of 180 Hz, with a 4 Rs pulse duration. The photons were emitted from a 2-mm spot in a 320 fan-beam configuration. A 128-channel detector system, placed on an arc 84.5 cm away from the source, performed x-ray attenuation measurements. Channels consisted of 1.8 mm wide cadmium tungstate scintillation crystals coupled to photodiodes. Tungsten anti-scatter plates were used to reduce the influence of scattered photons on the signal. The imaged object was placed on a motion system in order to perform the 3600 rotation necessary for tomographic reconstruction. The system achieved a resolution of 1.6 mm with a slice thickness of 5 mm and a sensitivity of 1.5 %. Data acquisition time was 2 minutes, but was limited by the maximum speed of the rotary stage. The solidification front in a pure aluminum sample was successfully imaged. A study of the feasibility of identifying the solidification front in a metal cast using measurements obtained by translation alone is presented. A model-based reconstruction algorithm was implemented assuming a rectangular liquid zone surrounded by a rectangular solid zone. A sample object's geometry was correctly reconstructed from experimental data, demonstrating promise for this technique. An analysis of its theoretical performance is presented in terms of the experimental parameters.
Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Nuclear Engineering, 1998.Includes bibliographical references (p. 137-138).
DepartmentMassachusetts Institute of Technology. Department of Nuclear Science and Engineering
Massachusetts Institute of Technology