Effects of magnification and sampling resolution in X-ray computed tomography for the measurement of additively manufactured metal surfaces

Recent studies have shown that X-ray computed tomography (XCT) can be used to measure the surface topography of additively manufactured parts. However, further research is necessary to fully understand XCT measurement performance. Here, we show how magnification of the X-ray projections and resolution of the volumetric reconstruction grid influence the determination of surface topography in the XCT data processing pipeline. We also compare XCT results to coherence scanning interferometry (CSI) measurements and find that by increasing the magnification of the X-ray projections, smaller topographic detail can be resolved, approaching the lateral resolution of CSI. Results show that there is an optimum setting for magnification, below and above which XCT measurement performance can degrade. The resolution of the volumetric reconstruction grid has a less pronounced effect, but in general, adopting higher or lower resolutions than the default leads to degraded repeatability in surface determination. The problem of determining sensitivity of XCT surface measurement as a function of setup parameters is complex, and it is not yet possible to provide optimal setup configurations that work regardless of object geometry. However, the methods presented here, as well as the results obtained, represent a useful contribution to good practice for XCT measurement of surfaces.