Abstract
With steadily increasing use in dimensional metrology applications, especially for delicate parts and those with complex internal features, X-ray computed tomography (XCT) has transitioned from a medical imaging tool to an inspection tool in industrial metrology. This has resulted in the demand for standardized test procedures and performance evaluation standards to enable reliable comparison of different instruments and support claims of metrological traceability. To meet these emerging needs, the American Society of Mechanical Engineers (ASME) recently released the B89.4.23 standard for performance evaluation of XCT systems. There are also ongoing efforts within the International Organization for Standardization (ISO) to develop performance evaluation documentary standards that would allow users to compare measurement performance across instruments and verify manufacturer's performance specifications. Designing these documentary standards involves identifying test procedures that are sensitive to known error sources. This paper, which is the third in a series, focuses on geometric errors associated with the detector and rotation stage of XCT instruments. Part I recommended positions of spheres in the measurement volume such that the sphere center-to-center distance error and sphere form errors are sensitive to the detector geometry errors. Part II reported similar studies on the errors associated with the rotation stage. The studies in Parts I and II only considered one position of the rotation stage and detector; i.e., the studies were conducted for a fixed measurement volume. Here, we extend these studies to include varying positions of the detector and rotation stage to study the effect of magnification. We report on the optimal placement of the stage and detector that can bring about the highest sensitivity to each error.