PSI - Issue 77

Boyu Li et al. / Procedia Structural Integrity 77 (2026) 316–322 Boyu Li et al. / Structural Integrity Procedia 00 (2026) 000 – 000

318

3

structural and thermal simulations. The G-code for a 10×10×10 mm 3 cube specimen and a dog-bone shape specimen, commonly used for mechanical testing, with varying infill rates and patterns, were reconstructed into CAD models. These reconstructed models were compared with the original CAD and printed specimens in terms of dimensional accuracy. The results demonstrated that the reconstructed models exhibited high accuracy and closely matched the original designs. This approach provides a reliable methodology for FEA of AM composites, enabling further investigations into the mechanical and water diffusivity performance under various loading or boundary conditions. Hachimi et al. (2024) retrieved the G-code using a JavaScript program to generate a Python script for constructing the geometry in ABAQUS. Scanning electron microscopy (SEM) was used to measure the dimensions of the oval rectangular cross-section of the extruded filament, which was then incorporated into the ABAQUS FE model. Two single filaments were printed and then tested under tensile loading to acquire the mechanical properties of the polymer. Dogbone-shaped tensile specimens were printed with varying orientations of 0°, 45°, and 90°. The number of wall layers in the specimens ranged from one to three. In FEA, the extruded material was assumed to be homogeneous with a perfect interfacial bond between filaments and layers. The numerical results showed a good agreement with experimental data, demonstrating the reliability of the model in capturing the mechanical behaviour of printed specimens. Furthermore, different infill patterns, including hexagonal, linear, gyroid 3D, and triangular structures, were sliced, reconstructed, and analysed. The FE model exhibited high accuracy in replicating complex geometric features of these printed structures. This preliminary study examines how the printing pattern of 3D-printed polymers affects the process of water diffusion, highlighting the necessity of geometry reconstruction when investigating the diffusion behaviour of 3D printed composites. 2. .Experimental study To examine the influence of printing patterns, two groups of 13 polylactic acid (PLA) cubic specimens, each measuring 10 × 10 × 10 mm 3 , were fabricated using an Ultimaker 2 3D printer with a line infill pattern, as shown in Fig. 1a. The detailed printing parameters are summarised in Table 1. After fabrication, each specimen was individually weighed using a KERN 572 precision balance with an accuracy of 0.001 g. The specimens were subsequently immersed in a tank filled with still water to a depth of 5 mm at a room temperature of 21℃, with their printing orientations positioned either parallel or perpendicular to the water surface (Fig. 1b).

Table 1. Printing parameters.

3D printer parameter

Value 1 mm

Layer height Line width

0.4 mm 0.4 mm

Nozzle diameter Infill pattern Infill density Printing speed

Line

100%

25 mm/s