On the effect of material density in dimensional evaluations by X-ray computed tomography of metal-polymer multi-material parts

An important aspect to consider in the evaluation of parts and assemblies by X-ray computed tomography (XCT) is the attenuation coefficient of the different materials involved, which are directly related to their density; depending on this coefficient, the X-ray penetration varies and, therefore, varies the contrast between different materials and with the background. This becomes more critical in those assemblies in which materials are characterized by a high difference in density, where the lighter material could be difficult to be characterised. In this paper, the effect of the presence of metals in the dimensional evaluation of polymeric geometries (having lower density than the metal parts) is studied, to evaluate the errors caused in dimensional measurements of different geometries and surface texture characterization. Based on a common geometry, four scenarios have been experimentally tested with variations of metal amount, in which macro geometries (precision spheres made by different polymers) and micro geometries (inclined ramps manufactured by fused deposition modelling (FDM)) have been characterised. Results show errors in the surface determination of the polymeric features directly related to the presence of metal: a high amount of steel makes significantly difficult to accurately determine the interface between background and material due to the noise and artifacts created, while aluminium has less influence on the irregularities of the features extracted. This effect is more evident for polymers with lower density due to the higher difference. Numerically, most affected parameters are those sensible to variations in surface determination, such as spheres’ form error and ramps’ maximum surface texture ( Sz ), while more solid features as spheres’ diameters, distances and ramps’ average surface texture ( Sa and Sq ) remain more stable. In conclusion and to sum up, it has been found that the quantity of metal present in assemblies made of polymeric and metallic materials is correlated with distortions in the dimensional evaluation of polymeric features by XCT.