Influence of contour parameters on the fatigue properties of AlSi10Mg produced by L-PBF

Details

Supervisors

Simar, Aude

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil mécanicien, à finalité spécialisée

Abstract

The presence of porosities and low surface quality of samples manufactured by L-PBF is one of the main challenges faced by the additive manufacturing industry. These surface and sub-surface defects have a detrimental effect on the mechanical performance, especially under cyclical stress. Traditional post-treatments aim to refine the microstructure, improve surface quality, or alter the stress state of the surface in order to improve the fatigue performance. The main drawback of these processes, however, is that they are time-consuming and may not lead to significant improvements in fatigue lifetime. An alternative to these post-treatments is the optimization of the contour process parameters. By increasing the contour track energy density, it is possible to reduce the surface roughness, but at the expense of the creation of keyhole porosities. In this investigation, the influence of roughness-optimized contour parameters on the fatigue behavior of net-shaped, vertically-built AlSi10Mg samples is studied. Results show that a high track energy density will lead to an improved surface roughness, but the appearance of keyholes will depend on the melting mode. Operating at a TED of 912 J/m (laser power: 272.6 W, scanning speed: 300 mm/s) leads to a keyhole melting mode, and results in a surface roughness of 4.2 μm and keyholes in the contour. In contrast, operating at a TED of 456 J/m (laser power: 272.6 W, scanning speed: 600 mm/s) leads to the onset of a keyhole melting mode, and results in a surface roughness of 5.1 μm and no keyholes in the contour. The lower roughness was attributed to an increased wettability of the lateral surface, and the keyholes to deeper and more unstable melt pools that trap gas bubbles. An optimization process has shown that, for the same track energy density, the geometry of cylindrical fatigue sample will influence the roughness level and appearance of keyhole porosities. Larger cross sections are associated with a greater total heat input and less heat dissipation, which leads to variations in roughness levels and porosity distribution along the height of the sample. A stress-relief heat treatment of 250°C for 2 hours will lead to a decrease in the ultimate tensile strength of approximately 450 MPa in as-built samples to 410 MPa in heat treated samples for all contour conditions. The heat treatment caused limited globularization of the Si network. Furthermore, it is likely it affected the contributions of Si precipitates within the Al matrix to the strain hardening capacity. These factors could explain the drop in ultimate tensile stress before and after the heat treatment. No differences in the tensile properties in either AB or SRHT samples could be attributed to the contour parameters. Overall, the heat treatment did not alter the microstructure significantly. Stress-relieved samples manufactured using a contour TED of 912 J/m (100-300) and 456 J/m (100-600) had a better fatigue performance than the reference samples, manufactured using a TED of 193.8 J/m in the studied stress range. This improvement can be attributed to the lower surface roughness in both sample groups due to the contour processing parameters. The fact that the best-performing samples were the 100-300 (best surface roughness and keyhole porosities) indicates that these porosities do not negatively impact the fatigue performance. Further improvements in fatigue performance were achieved by polishing the samples. However, polishing exposed the keyholes in the contour of the 100-300 samples, causing a lower performance than polished 100-600 samples. Sandblasting produced the best fatigue performance in 100-300 and 100-600 samples due to the introduction of residual stresses into the surface. However, it presented a low repeatability and high scattering of the results.