Abstract
Layer-by-layer manufacturing imparts distinct material behavior to parts made of Additive Manufacturing (AM) compared to those made by conventional casting, forming, machining, and welding techniques. Fused Filament Fabrication (FFF) is a polymer AM process in which the intra-layer bonding is stronger than the inter-layer bonding resulting in directional mechanical properties. The angle between loading and build direction changes for different build orientations, and the parts display different resistance levels to fracture and its propagation. In the present work, a design of experiments (DOE) based investigation into the tensile strength and fracture energy dependence on part build orientation, build density, and layer thickness was carried out. Mechanical behavior of additively manufactured ABS tensile specimens as per the Taguchi design of experiments (DOE) as a function of four factors, namely, Layer Thickness (L), Azimuthal Angle (T), Meridional Angle (P), and Infill Percentage (D) was investigated. These four factors were selected after listing all the possible factors and comparative analysis of their importance. The fractional factorial design was chosen to optimize the experimental cost and, at the same time, obtain a nonlinear design. The additive manufacturing of the specimens (ASTM D638, 3.2 mm thickness, 57 mm gauge length, 165 mm overall length, gauge curvature R76) was made on FlashForge-Creator-3-Pro with a 1.75 mm filament. The specimens were then polished dry on the VERX Dual Disc and Belt Sander (VDS-640, 500 W) to improve the surface finish so that no abnormal crack initiation occurs due to surface disintegrations. The tensile tests were conducted on the Zwick-Roell UTM with an extensometer at a 2 mm/min test speed. Crack propagation paths were recorded using a high-speed camera (Phantom VEO 440L). Due to the brittleness of ABS, the fracture phenomenon was so fast that it was difficult to fully capture the fracture within the memory limitations of even the high-speed camera. To optimize the capture time (length), the correct triggering time was identified through some dummy experiments on sample specimens of the same material. The strain energy release rate was calculated using the ASTM standard method. It is observed that the meridional angle of build orientation, in-fill percentage, azimuthal angle of build orientation and layer thickness have, in that order, the most significant influence on tensile strength, which is evident through the responses of means and S-N ratios. The experiment is statistically valid by the normal probability plot. The tensile strength and the strain energy release rate were found to depend on the build orientation. In all the cases, the fracture was plane-stress, initiated at one and propagated to the other end in a zig-zig path. Based on the tensile strength, energy release rate, and build orientation, conclusions have been drawn on the nature of fracture propagation.
