As metal additive manufacturing (AM) continues to gain relevance for use in industrial settings,
some of the inherent problems with the technology must still be uncovered. Traditionally, the roughness of parts produced through the electron beam melting (EBM) process has been associated with detrimental effects on the mechanical properties of the components. It has been established that the roughness arises from partially melted powder particles that stick to the boundary or contours of the parts as the electron beam scans and melts each layer. Observations indicate that the crevices formed at the intersection of these partially melted particles serve as initiation sites for cracks. The higher the roughness, the higher the number of potential crack initiation sites. Therefore, great efforts in both industry and academia are currently exploring the effect of the roughness on AM parts, as well as processes to mitigate this effect.
The aim of this research was to investigate the effects of surface condition on the tensile properties of Ti6Al4V EBM produced specimen. The specimens studied had three surface
conditions: as deposited, machined, and treated with REM’s proprietary ISF® process. The specimens fabricated were in the X, Y or Z directions as per the ASTM F2971-13 standard. All the specimens were fabricated in a single run in an Arcam A2X machine at Addaero. The specimens were divided into two further categories (edge or core) corresponding to the build location within the build envelope; edge specimens were built at the periphery of the 150x150mm start plate used for fabrication, whereas the core specimens were closer to the geometric center of the plate (Figure 2 details these two regions). After fabrication, the corresponding specimens were either machined from blanks, treated with the ISF® process, or used as-deposited for testing. Tensile testing was performed at a NADCAP certified vendor following the ASTM E8 standard. The specific orientation during fabrication of the specimens tested is given in the table below.
The values obtained from the tensile test for yield and ultimate tensile strengths are summarized in Figure 3 below. In all cases, the strength values obtained meet or exceed the minimum requirements dictated by ASTM F2924 for TI6Al4V produced by powder bed fusion AM. Interestingly, the ISF process shows the greatest improvement in mechanical properties, having a statistically significant increase for both XY and Z specimens. As can be read from the chart, the values of UTS and YS for the ISF treated components were 1036.5MPa, and 984.8MPa for the XY specimens, and 1000.9MPa and 941.1MPa for the Z specimens. These results can likely be contributed to a slightly larger diameter gage length for the ISF specimens vs. the machined specimens. The ISF specimens were printed with additional material added to the gage length and the ISF process was used to bring the diameter of the gage length down to the correct dimension.
With regards to ductility measurements, the test results are shown in Figure 4. The chart shows that overall, the as-fabricated specimens performed the worst for both XY and Z specimens, followed by minimum improvements by ISF treated specimens. In this case, the best ductility, as measured by the percent elongation and percent reduction in area was shown for machined specimens. The representative values of % elongation and % area reduction for the machined specimens were 17% and 27% for XY specimens and 17.83% and 47.4% for Z specimens, respectively.
For the experiment performed, the tensile strength and yield strength results were driven by the diameter of the gage length. As discussed previously, because the dimensions of the ISF specimen and As-fabricated specimens were dictated by the accuracy of the additive process the final gage diameters of these specimens were larger than the machined specimens. Achieving dimensional accuracy in As-fabricated or chemically milled tensile specimens equivalent to a machined specimen is not currently possible in EBM and will, therefore, drive variability in the results. The machined specimens did perform better in elongation due to the improved surface finish achieved from machining.
Addaero is planning to do additional experiments in relation to this work in the coming months. Be sure to check for updates on our blog and LinkedIn.