PSI - Issue 56

Laszlo Racz et al. / Procedia Structural Integrity 56 (2024) 3–10

6

4

Author name / Structural Integrity Procedia 00 (2019) 000–000

Belgium), where the length of the contact areas, both inter layer and intra layer necking was measured as presented in Figure 2.b. In the CAD model, considering the flattening effect, the shape of the filaments (ellipsoidal section) is adjusted with the overlapping necking areas, to adjust the perfect geometrical model to a more realistic model. To extract the cross-section area of the tensile specimens, the created geometry was meshed with Tetra 2 nd order elements. Finite element analysis (Epilysis solver) that simulates a tensile test is conducted on the different numerical specimens corresponding to the 6 analyzed infill patterns. The same boundary conditions were applied to the FE model as for the real specimens during the tensile test. The selected material model considered in the simulation of individual filaments was isotropic elasticity model, the goal of the research being the evaluation of E-modulus change with the infill pattern. The simulations were run for a tensile force producing only elastic deformations. 3. Results and discussions To validate the method, real life tensile tests are necessary, where the experimentally determined tensile strain will be compared to the result of numerical simulation. If they are comparable, the area of the cross-section is correctly established and can be used for further computations. Uniaxial tensile tests on standard ISO 527-2-2012 specimens (type 1A, featuring a gauge length of 75 mm with a 4 mm thickness) with different infill patterns were carried out on a universal testing machine type INSTRON 3366, 10 kN capacity at a loading speed of 1 mm/min, a uniaxial extensometer being used to measure the tensile strain. The results are summarized in Table 1 where the experimentally measured strain is compared to the results of the FE simulations, the results were selected from the elastic domain of the stress-strain curve.

Table 1. Strain values obtained by numerical simulations vs experimentally determined.

Strain EXP (mm/mm)

Strain FEA (mm/mm)

Relative Dev. FEA-EXP (%)

Specimen’s infill pattern

Grid 0-90

0.00259 0.00267 0.00279 0.00333 0.00274 0.00223

0.00277 0.00269 0.00285 0.00351 0.00294 0.00205

6.95 0.75 2.15 5.40

Grid +45-45

Fast Honeycomb Full Honeycomb Triangular 60°

7.2

Wiggle

8

The results showed a range of minimum 0.75% and maximum 8 % difference, fact that validate the proposed methodology for simulation of 3D printed specimens. In Fig. 3 the cross-section of the tensile specimens is presented for grid 0°-90° and ±45°, triangular 60°, fast honeycomb, full honeycomb and wiggle infill patterns.

a)

b)