PSI - Issue 42

Saveria Spiller et al. / Procedia Structural Integrity 42 (2022) 1239–1248 Saveria Spiller/ Structural Integrity Procedia 00 (2019) 000 – 000

1244

6

* Ultimate tensile strength of the green parts (prior to debinding and sintering phases)

Fig. 4. two 1.5x1.5x1cm 3 specimens with a magnification of the top layer and the side. Solid a) obtained with the parameters named trial 7 in Table 1, solid b) obtained with trial 5. 4. Tensile tests Using the knowledge obtained from the preliminary trials, some tensile specimens were designed and printed, according to the ASTM standard E8. The parameters configuration chosen was the one named trial 5 in Table 1. The specimens were printed with the flat configuration depicted in Fig. 3c. Since the parts were more extended than the boxes and the tensile specimen that were tested before sintering, a sort of scale effect was observed, and some specific strategies were used to improve the quality of the prints. It was not possible to set the layer thickness below 0.1mm. When 0.05mm was set, the printer gave an unstable extrusion considering that the nozzle path was longer. Some adhesion problems were also observed for the same reason. To increase the adhesion to the platform, the temperature of the platform was kept at 100°C, while for the boxes it was 60°C. In addition, the first layer parameters were altered with respect to the rest of the specimen: the print speed, as well as the cooling fan speed, were reduced, and the layer height was increased, to facilitate the extrusion of the polymer. A wide brim was also printed to avoid warping and detachment of the specimens from the platform. Furthermore, to improve the appearance of the walls, both the inner and outer walls were printed with a reduced speed to avoid deformations at the endpoint of each wall. For the compactness of the infill, it was also beneficial to increase the overlap of the rasters up to 30% with respect to the default value of 0%. According to BASF, the filament producer, the specimens were supposed to shrink during the sintering process by 19% along the X-Y plane and 21% along the Z-axes ( BASF Ultrafuse 316L ). Those values were therefore used to over-dimension the specimens. However, a higher shrinkage was measured from the fabricated specimens, as depicted in the bar plot in Fig. 5. The highest reduction in percentage was observed along the thickness of the parts, the Z-direction. This was expected considering the flat orientation of the print, and it is related to the non homogeneous distribution of the powder in the deposited raster of softened material (Kurose et al., 2020).