PSI - Issue 78

Vittoria Borghese et al. / Procedia Structural Integrity 78 (2026) 1229–1236

1233

Model initialization

Materials Geometry

Geometry definition

Displacements Modeshapes

Inputs

Outputs

Material assignment

Analysis typology

Modal analysis

Fig. 2. Flowchart of the structural framework analysing the inputs, the model creation on DIANAFEA environment and the outputs.

to the selected typology of analysis, yielding the values for each lamella used in the finite-element model. Finally, the analysis typology was represented by an integer from 1 to 3, corresponding to the moduli assignment strategy used for model initialization enumerated above. The input data drive is a fully automated Python routine that constructs and analyses the CLT panel within DI ANA FEA . First, a new 3D structural model is initialized and orthotropic material properties are assigned to each lamella. The routine then iterates over all lamellae, using their corner-node coordinates ( x 0 , y 0 , z 0 ), local dimensions ( l k , i , w k , i , t k ) and layer parity to generate solid elements, impose linear-elastic behaviour, and apply the appropriate sti ff ness matrices. A structured mesh is established by seeding each element along the CLT directions ( X , Y , and Z ), and lamella axes are set to align with the material orientation in each layer. Orthotropic mechanical properties are defined for each lamella based on the obtained longitudinal modulus of elasticity (Table 1), using fixed ratios given in Table 2, which are based on (Senalik & Farber., 2021). The mesh was generated and a modal analysis ( EIGEN )was configured with the desired number of eigenmodes (here, 15 to ensure capturing the desired eigenmodes). Finally, the solver computes nodal displacements, natural frequencies and mode shapes.

Table 2. Orthotropic property ratios assumed for each wood specie, based on Senalik & Farber. (2021). E l / E t E l / E r E l / G lr E l / G lt E l / G rt ν lr

ν lt

ν rt

Ash

12.5 16.9 20.0

8.0 7.8

9.2 8.1

13.0

107.5 100.0 142.9

0.37 0.42 0.29

0.44 0.46 0.45

0.68 0.53 0.39

Spruce

8.3

Douglas-Fir

14.7

15.6

12.8

Note: Subscript conversion l = longitudinal = x , t = tangential = y , r = radial = z

3. Results and Discussion

3.1. Experimental

Based on the Frequency Response Function (FRF), obtained from the modal tests described above, the natural frequencies and associated mode shapes were determined. The natural frequencies present themselves as dominant peaks within the FRF, with each frequency being associated to a mode shape. The order of frequencies and their mode shapes in the FRF is influenced by the CLT layup and material properties. Therefore, the determination of mode shapes is crucial in order to determined the global sti ff ness values. In order to determine the mode shapes, the imaginary part of the FRF and the associated location of the sensor were used. Figure 3 (a) shows the determined natural frequencies f (2 , 0) , f (0 , 2) , and f (1 , 1) and the associated mode shapes for panel 3. It becomes clear that f (2 , 0) is associated with flexure in the longitudinal panel direction, f (0 , 2) is related to flexure in transverse panel direction, and f (1 , 1) is related to twisting. The obtained frequencies can then be used for the determination of E X , E Y and G XY using Equation 1 of which the results are given in Figure 4.