PSI - Issue 79

Alessandra Ceci et al. / Procedia Structural Integrity 79 (2026) 73–80

76

To validate the reliability of the Lost-PLA process, CAD-predicted weights were compared with the experimentally measured weights of the metallic specimens. As shown in Table 1, deviations were below 2%, confirming the high accuracy and repeatability of the process. From these values, apparent density, relative density and porosity were calculated for the two configurations. Table 1. Comparison between theoretical and actual weight of the structures and derived parameters (apparent density, relative density and porosity). *The real weight was calculated as the average value over three specimens. Structure CAD Weight ( g ) Real Weight ( g )* Deviation ( % ) Apparent density ( g/cm 3 ) Relative Density Porosity 1 49.9 49.3 1.2 0.48 0.18 82 2 90.6 89.0 1.8 0.87 0.32 68 The values reported in Table 2 were obtained from CAD data and experimental measurements of the metallic specimens. The apparent density was calculated as the ratio between the average real weight and the external cylindrical volume (Eq. 2). The relative density was determined by comparing the apparent density with the bulk density of the material (Eq. 3), while porosity was estimated as the percentage difference with respect to the fully dense alloy (Eq. 4). Finally, the percentage deviation between theoretical and real weight was calculated by comparing CAD-estimated and experimental values (Eq. 5). ∗ � ���� (2) ��� � ∗ � (3) � 1 � ∗ � (4) � ���� � ��� ��� ∙ 100 (5) where ���� is average real weight of the metallic specimens (g), ��� theoretical weight predicted by the CAD model (g), external volume of the cylindrical specimen (cm³) calculated as � �� /2 � � ℎ , ∗ is apparent density of the structure (g/cm³), � bulk density of the solid material (AA6082: 2.7 g/cm³), estimated porosity (%), percentage deviation between real and theoretical weight (%). Macroscopic observations (Fig. 3) further demonstrate the correspondence between the PLA models and the metallic structures. The layer lines from 3D printing are clearly reproduced in the alloy, highlighting the ability of the process to faithfully transfer geometric details.

b

c

d

a

Fig 3. Macrographs: a) Structure 1 in PLA, b) Structure 1 in metal, c) Structure 2 in PLA, d) Structure 2 in metal.

Compression tests were performed at a constant crosshead speed of 2 mm/min using a universal testing machine. Images taken at successive compression steps are reported in Fig. 4. Three specimens were tested for each configuration (Structure 1 and Structure 2) in order to assess the repeatability. The load–extension curves obtained are reported in Fig. 5. Both structures exhibit the typical behavior of cellular materials, characterized by three stages: