Evaluation of printing parameters in additive manufactured samples using fractal geometry of computed tomography images

Abstract

Additive Manufacturing (AM) has already attained a reliable level of maturity, specifically Fused Filament Fabrication (FFF), emerging as the most widespread process. Concurrently, the industrial demand for these parts has increased, requiring the analysis of their internal geometry to determine the level of similarity achieved concerning the expected structures. This work aims to provide tools to characterize FFF parts by relating printing properties to geometrical variables. For this purpose, three samples were printed in Polylactic Acid (PLA) with three different layer heights and analyzed by X-ray Computed Tomography (CT). After processing the images, fractal analysis was carried out using the box-counting method on the voids that appear between the filaments in order to obtain the fractal dimension. The porosity of the voids was also calculated. The analysis identifies the parameters characterizing the voids as number, size, shape, and location. In contrast to traditional porosity studies, the novelty of this work is that fractal analysis provides information about shape and distribution of voids in a single value (fractal dimension). It was corroborated that the fractal dimension depends not only on porosity but also on the shape and location of the voids. Additionally, it was found that not all void parameters influence equally the geometrical variables; variables related to porosity (number and size of voids) are more relevant than shape and location. Finally, it was demonstrated that by knowing the parameters of layer height and extrusion flow, the ideal porosity and fractal dimension can be determined, and any deviation from these parameters indicates the geometric printing error incurred.