Quasi-static simulation and fatigue life estimation of fused filament fabrication of polylactic acid specimens using finite element method

Abstract

The additive manufacturing (AM) technique has recently attracted more attention in rapid prototyping methods due to its unprecedented performance in engineering applications. Fused filament fabrication (FFF) is the most common technique to manufacture sophisticated engineering components by depositing the polymeric material layer by layer. Since the produced materials exhibit different mechanical behaviors, it is required to perform adequate mechanical tests and engineering analysis to attain a product with an acceptable performance under the designed loading conditions. This research deals with a series of tests to achieve the mechanical characteristics of the FFF-made samples printed in different raster orientations. Due to the high level of anisotropy achieved compared to using conventional manufacturing techniques, the effects of printing directions should be studied. In this study, the mechanical properties of FFF Polylactic acid (PLA) specimens, e.g., tensile and fatigue tests were investigated experimentally. Then, the finite element (FE) model for estimating the fatigue behavior of notched samples was developed. To do so, two specimen sets, including unidirectional and double-directional (cross) samples were prepared for experiments. Tensile test results show that in specimens with parallel loading direction and raster orientation, can sustain higher load; however, ultimate strength values in the specimens with perpendicular raster orientation to the loading direction were reduced by 18 %. Furthermore, the strain-life diagrams were obtained through the pure bending fatigue tests, for three specimen sets (unnotched, single-notched, and double-notched) fabricated with a ± 45° raster orientation. The fatigue life of the notched samples was predicted using the FE model and the Coffin-Monson equation. Comparing the experimental and predicted results shows that a good agreement exists between the experimental data and numerical results.