Quantitative Assessment of Voids' Impact on Mechanical Properties of Standard Dogbone Model Versus End-User Component in Non-Linear Geometry.

Additive manufacturing (AM) offers advantages such as design flexibility and reduced production times, but defects like voids impact mechanical performance and limit its broader adoption. This study quantitatively examines the relationship between void characteristics (volume fraction, distribution, and size) and mechanical properties in both linear and non-linear geometries, represented by a dogbone model and an end-use component, respectively. Samples were produced using Fused Filament Fabrication (FFF) with varying overlap levels to control void content. As the overlap increased from 0% to 99%, voids transitioned from large linear gaps to smaller point-shaped voids. In non-linear geometry, void reduction from 12% to 2% led to a threefold improvement in mechanical response, while in dogbone samples, voids decreased from 12% to nearly 0%, improving the elastic modulus by only 1.5 times. This disparity is due to differences in void distribution, as voids in non-linear geometries affect both margins and internal layers, significantly influencing structural integrity. The findings highlight the importance of the void location in determining mechanical performance and emphasize the limitations of using linear dogbone models to assess void-property relationships in complex 3D-printed structures.