PSI - Issue 42

C. Boursier Niutta et al. / Procedia Structural Integrity 42 (2022) 1449–1457 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

1453

5

Finally, two rigid walls, one fixed and one moving in accordance with a prescribed motion law, are retained for modeling the quasi-static crushing test. In particular, in order to limit the inertia effect, the velocity of the moving rigid wall is smoothly increased up to the constant value. 3.2 Modelling AM defects For the investigations of the sensitivity of the structure to the presence of defects, the lattice structure with diameter d =1.5 mm and cell size l =8.35 mm is considered. The number of cell replications is 3. The two typologies of defects are: i. diameter variation; ii. lack-of-fusion. Indeed, as shown in Tables 1 and 2, the actual values of the beam diameter were consistently different from the nominal ones, as a result of the manufacturing process. Here, we conservatively assume a 50% reduction of the diameter with respect to the nominal value of 1.5 mm. A specific part, with reduced cross-section, is defined in LS Dyna. The lack-of-fusion defect is among the most common defects in AM components [8]. As the lattice structures are produced by FDM and given the small diameter of the beams, it is here assumed that the unfused portion results in a discontinuity in the beam. As the beam is discontinuous, loads cannot be transferred from one node of the cell to the other. The beam discontinuity can be modeled by duplicating the nodes within the defective beam. As an alternative, we modelled the material discontinuity by assigning the *MAT_NULL card to the defective beam. This material card prevents the load and moments from transferring from one node to the other, thus simulating the lack-of fusion defect. The unit-cell models with diameter variation and lack-of-fusion are shown in Fig. 4.

(a) (b) Fig. 4. Unit-cells with defects: a) diameter variation; b) lack-of-fusion modelled through the *MAT_NULL card.

The influence of each defect on the SEA of the lattice structure is investigated by randomly distributing the defective beams within the unit-cell. Four different percentages of defectivity are considered: 5%, 12%, 25% and 50%. This means that, of the 36 beams constituting the unit-cell, 2 beams, 4 beams, 9 beams or 18 beams are defective. Thereafter, the unit-cell is repeated to construct the specimen and perform the simulation. As the defective beams are randomly distributed in the unit-cell, three simulations are performed for each percentage of defectivity. This allows to average the effect of the location of the defects within the structure. 4. Results In this section, the results of the numerical investigations are presented. Firstly, the validation of the FE model is with the experimental results is shown. Thereafter, the sensitivity of the lattice structure to the diameter variation and the presence of lack-of-fusion is analyzed.