Investigation into Influence of Tensile Properties When Varying Print Settings of 3D-Printed Polylactic Acid Parts: Numerical Model and Validation

Material Extrusion (MEX), particularly Fused Filament Fabrication (FFF), is the most widespread among the additive manufacturing (AM) technologies. To further its development, understanding the influence of the various printing parameters on the manufactured parts is required. The effects of varying the infill percentage, the number of layers of the top and bottom surfaces and the number of layers of the side surfaces on the tensile properties of the printed parts were studied by using a full factorial design. The tensile test results allowed a direct comparison of each of the three parameters’ influence on the tensile properties of the parts to be conducted. Yield strength appears to be the most affected by the number of layers of the top and bottom surfaces, which has twice the impact of the number of layers of the side surfaces, which is already twice as impactful as the infill percentage. Young’s modulus is the most influenced by the number of layers of the top and bottom surfaces, then by the infill percentage and finally by the number of layers of the side surfaces. Two mathematical models were considered in this work. The first one was a polynomial model, which allowed the yield strength to be calculated as a function of the three parameters mentioned previously. The coefficients of this model were obtained by performing tensile tests on nine groups of printed samples, each with different printing parameters. Each group consisted of three samples. A second simplified model was devised, replacing the numbers of layers on the side and top/bottom surfaces with their fractions of the cross-section surface area of the specimen. This model provided results with a better correlation with the experimental results. Further tests inside and outside the parameter ranges initially chosen for the model were performed. The experimental results aligned well with the predictions and made it possible to assess the accuracy of the model, indicating the latter to be sufficient and reliable. The accuracy of the model was assessed through the R2 value obtained, R2 = 92.47%. This was improved to R2 = 97.32% when discarding material infill as an input parameter.