Abstract
X-ray computed tomography (XCT), long used in medical imaging and defect inspection, is now increasingly used for dimensional measurements of geometrical features in engineering components. With widespread use of XCT instruments, there is growing need for the development of standardized test procedures to verify manufacturer specifications and provide pathways to establish metrological traceability. As technical committees within the American Society of Mechanical Engineers (ASME) and the International Organization for Standardization (ISO) are developing documentary standards that include test procedures that are sensitive to all known error sources, we report on work exploring one set of error sources, instrument geometry errors, and their effect on dimensional measurements. In particular, we studied detector and rotation stage errors in cone-beam XCT instruments and determined their influence on sphere center-to-center distance errors and sphere form errors for spheres located in the tomographically reconstructed measurement volume. We developed a novel method, called the single-point ray tracing method, that allows for efficient determination of the sphere center-to-center distance error and sphere form error in the presence of each of the different geometry errors in an XCT instrument. In Part I of this work, we (1) describe the single-point ray tracing method, (2) discuss optimal placement of spheres so that sphere center-to-center distance errors and sphere form errors are sensitive to the different detector geometry errors, and (3) present data validating our method against the more conventional radiograph-based tomographic reconstruction method. In Part II of this work, we discuss optimal placement of spheres so that sphere center-to-center distance errors and sphere form errors are sensitive to error sources associated with the rotation stage. This work is in support of ongoing standards development activity within ASME and ISO for XCT performance evaluation.