PSI - Issue 63

Petr Lehner et al. / Procedia Structural Integrity 63 (2024) 43–50

44

and applications around the world. On the other hand, a separate chapter is 3D printing using plastics, metals, and wooden materials (Nicolau et al., 2022). This possibility of applicability in construction is not publicly known because there are many unresolved research questions (Arrêteau et al., 2023). For example, inaccuracies in 3D printing that need to be investigated and new techniques and procedures are being developed for this (Kozior et al., 2024). While we find some practical applications, there is a lack of in-depth logical integration of standard construction practices and needs with the benefits of 3D printing. The first step in such a deeper understanding is to find the goal. In this case, the goal is to create a numerical model of a scaled-down physical model of a truss arch bridge that combines wooden elements and 3D printed joints. This goal is based on the idea that the shape of the arch bridge can be generated by nonrepeating angles of the member connections and, moreover, that there is great potential for design variability. The reduced physical model was chosen so that the results could be experimentally verified with the numerical model in the future. When a goal is chosen, the path to that goal must be clearly defined. The path to the mentioned goal is set out by not simply understanding the behaviour of the base material used in the 3D printer, but through the correct definition of the numerical model of the joint, the numerical model of the whole structure, and the linking of the results into the most realistic output. This paper focusses on one of these steps, namely, the modelling of a truss connection example. The aim is not to present or evaluate a specific case, but to highlight the actual process of creating a detailed numerical model (based on finite element method (FEM) software) (Bathe, 2008), to demonstrate the differences when using two different sources of material information and to highlight some interesting points. It is worth noting here that the presented study focusses purely on numerical modelling and does not include information on future real-world printing of samples and their testing. 2. Materials characteristics The material and its properties are very important for models. In the case presented, three variants were analysed. The first material model adopted information on the behaviour of the material printed by the 3D printer from the manufacturer and supplier (Hu, 2021). In the second variant, tensile tests were performed on test specimens in the lab, and material properties were derivated on the basis of these tests. In the first two cases, it was a classically supplied (see also (Dedek et al., 2024)) polycarbonate (PC) blend, and in the third case, it is a polycarbonate blend carbon fiber (PCCF), which is also supplied by the printer manufacturer (Hu, 2021). In the first material model, the following material properties are set for the 3D printed element: the modulus of elasticity in tension is 1.9 GPa, the tensile strength is 63 MPa, the Poisson constant is 0.4 and the density is 1220 kg/m 3 (Hu, 2021). This model is marked as PC01. In the second model, the first phase consisted of the analysis of test results for 5 pieces of samples for the tensile strength test dog-bone specimen. The samples were 170 mm long and the cross-section at the neck was 10 x 4 mm. The resulting tension and deformation diagram was converted to a stress and strain diagram. From these data, a tensile strength of 50 MPa and a modulus of elasticity at tension of 1.5 GPa were determined. This model is marked as PC02. It should be noted here that the manufacturer himself states that the values are guaranteed due to the nature of 3D printing based on the specific size of the printed layers and precisely defined conditions. From that point of view, it is interesting that in this case the differences are around 20%. As a complement to these two models of the same material, but with parameters from different sources, a model where the material was different was added. PCCF have this material property: the modulus of elasticity in tension is 3.5 GPa, the tensile strength is 64 MPa, the Poisson constant is 0.4 and the density is 1220 kg/m 3 (Hu, 2021). This model is marked as PC03. For all models, the C24 wood class was used as the material for connected beams.

Table 1. Material characteristics Model No.

Fly ash [%]

Modulus of elasticity [GPa]

Poisson constant [-]

Density [kg/m 3 ]

PC01 PC02 PC03

1.9 1.5 3.5

0.4 0.4 0.4

3.7 3.7 3.2