Repeatability and Reproducibility Assessment of a PolyJet Technology Using X-ray Computed Tomography

Featured Application From the beginning of the use of additive technology in small series, a major problem has been the repeatability of the results. This repeatability is most evident in different dimensions of the finished product. Additionally, depending on the time of printing during the year, there are large differences in the same products. Such a problem is particularly apparent in the processes of additive production in which the material is applied in the liquid form. Such repeatability is, of course, a big problem for the industry which produces small series (e.g., designers). The results from this paper can be applied to processes in additive production that belong to the group of material jetting, but also to other processes where the input material is in the form of liquid, or where acrylates, epoxy resins, etc., are used. PolyJet consistently produces parts within +/- 100 mu m, at a 95% confidence interval. Accuracy and repeatability in the x and y axis are significantly better than in the z axis. Form errors (i.e., cylindricity) are larger than positional or dimensional errors. Holes with diameters less than 0.75 mm cannot be reliably produced using PolyJet. From the very start of their use until today, processes in Additive Manufacturing (AM) have found a way to grow from prototype production to individual and small-series production. Improvements in machinery, materials and other challenges in AM development have improved product quality, its mechanical properties and dimensional accuracy. Research in the field of dimensional accuracy must be focused on achieving better tolerances. From the beginning of AM, there has been a big issue in assuring dimensional repeatability and reproducibility of a part being printed over the course of several days. In order to examine that, a test plate was designed and built repeatedly with PolyJet technology over the course of several weeks. Measurements of dimensional accuracy and shape deviations of several typical features were carried out using X-ray Computed Tomography. Measurement results were analysed and presented in order to indicate the repeatability and reproducibility of PolyJet AM technology. Results show that PolyJet technology consistently produces parts within +/- 100 mu m, at a 95% confidence interval, under reproducibility conditions of over a 1-month period. Accuracy for measurands (distance) in the x and y axis was significantly better than it was for the z axis which was from 56 to 197 mu m, i.e., in the x and y axis, it was from -8 to 76 mu m. Shape errors (i.e., cylindricity) were larger than positional or dimensional errors; this can be attributed to relatively large surface roughness and small feature sizes on the test plate that was used.