This work aims to provide a methodology for producing Inconel 718 specimens ISO 527–2-5B with a tailored mechanical response, by Selective Laser Melting (SLM). Since the macroscopic mechanical behavior of 3D printed metals is dictated by their microstructure, the effect of the printing parameters such as the laser power, the scan speed, the hatching distance and the layer thickness was considered. To evaluate the maximum dimensional accuracy and optimum Volumetric Energy Density (VED) value, the specimens were printed with a constant layer height, hatching distance and spot size and with a varying combination of laser power with scan speed. The 3D printed specimens were subjected to uniaxial tension. The experiments have shown an increased reproducibility of the specimens’ microstructure that led to a more consistent mechanical behavior, when lower scan speed and power were used. Specimens printed with higher speed and power seem to have a more heterogeneous microstructure leading to a more varying mechanical response. While the differences in the mechanical response between specimens of different batches are attributed to different laser power and scan speed, different response between specimens of the same batch is attributed to the inherent inhomogeneity of their microstructure. A simple elasticity model was used to describe their elastic behavior, while a 1D gradient plasticity model was utilized to describe their post-yield tensile response, in which the effect of microstructural heterogeneity on macroscopic behavior was taken into account through a respective internal length. The combined elasto-plastic model was used to capture mechanical response until failure, in good agreement with the experiments. The predicted mechanical characteristics were related to the specimens’ VED to come up with simple relations providing optimum 3D printing parameters for producing specimens with a certain mechanical response. For a complete characterization, the specimens’ roughness was studied and its VED dependence was performed.
Read full abstract