PSI - Issue 78

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

1230

of large dimensions (up to 30 × 3 m) with sti ff ness and strength in the major and minor panel directions (Swedish Wood, 2019). Figure 1 (a) shows a drawing of a 5-layer CLT panel with alternating layer orientation. CLT is commonly formed from structurally graded timber. The assigned strength classes of the lamellae provide the input properties for the estimation of the overall properties of the CLT. Currently available CLT is usually made from coniferous timber (e.g., Spruce), but recent developments within industry and research consider other materials for the individual layers, such as engineered wood panels, such as laminated veneer lumber (LVL) and oriented strand lumber (OSL) (Wang et al., 2017; Niederwestberg et al., 2018), deciduous timber (Gong et al., 2015), or reclaimed timber (Stenstad et al., 2021; Llana et al., 2022; Vonk et al., 2024), each with their own levels of material properties and variability. CLT made from layers of di ff erent materials are often referred to as hybrid CLT. Due to the large variability in used materials, the estimation of the mechanical properties of CLT is challenging. Non-destructive methods such as eigenfrequency analysis can o ff er a solution for estimating sti ff ness properties, such as bending sti ff ness in the major and minor panel direction of CLT (Gu¨lzow, 2008; Giaccu et al., 2019; Llana et al., 2022). Additionally, modelling approaches, such as Finite Element Analysis (FEA) or composite theory, can be utilized to estimate the panel properties based on the lamellae properties, which can be obtained using established non-destructive grading methods. Within Service Limit State (SLS) design, sti ff ness properties play an important role. SLS addresses building func tionality and limitation of non-structural damage during regular usage, as well as extreme events such as earthquakes. Since SLS verification is governed by displacement criteria, accurate prediction of global sti ff ness becomes essential. This research presents an experimental-numerical study on the determination of out-of-plane mechanical properties of hybrid CLT panels. Eigenmodal analysis, using a grid of accelerometers, was employed to obtain the longitudinal, transverse and torsional shear modulus of 5-layer hybrid CLT panels. Before panel fabrication, the lamellae were tested using modal analysis. An FEA model of the panels was generated and di ff erent statistical approaches were considered for the selection of the lamellae and layer properties to determine the influence of made assumptions. Di ff erent wood species were used to assemble 5-layered hybrid CLT panels, namely Dutch-grown Ash ( Fraxinus excelsior ), 50 year old reclaimed Spruce (likely Picea abies , formerly used as internal cladding) and Dutch Douglas Fir ( Pseudotsuga menziesii ). The individual pieces were finger-jointed to a length of 6 meter using a structural melamine-urea-formaldehyde (MUF) adhesive. The lamellae were subsequently planed to a specific cross-section (width ( w ) and thickness ( t )), which is given in Table 1. Before panel fabrication, the longitudinal sti ff ness ( E x ) of all lamellae was evaluated using a Mobile Timber Grader ( MTG 960 , Brookhuis , the Netherlands). The MTG 960 me chanically excites the beam and measures the vibrational response to find the eigenfrequency. Following EN 14081-2 (CEN, 2019), the eigenfrequency, combined with the density and moisture content ( MC ), is used to find E x . The density was determined by gravimetrically obtaining the mass and the dimensions, and the MC was obtained using a handheld capacitive moisture meter ( FMW , Brookhuis , the Netherlands). The density ( ρ ), average E x , including coef ficient of variation ( CoV ), and number of tested lamellae ( n ) of each specie are given in Table 1. Ash has the highest average sti ff ness, and the reclaimed Spruce is surprisingly higher than the virgin Douglas-Fir. Hence the CLT panel layup given in Table 1. A schematic representation of hybrid CLT panels is shown in Figure 1 (a). A total of ten 5-layer hybrid CLT panels ( L × W × H = 2980 × 1200 × 138 mm) were assembled with Ash as outer layers (layers 1 & 5), the reclaimed Spruce as cross layers (layers 2 & 4), and the Douglas-Fir as the central layer (layer 3). This layup was chosen in order to maximize the bending sti ff ness of the hybrid CLT, based on the lamellae results (Table 1). The hybrid CLT were made from single edge-glued layers that were formed prior to the face-gluing process. Due to the limited availability of the reclaimed Spruce thicknesses, an asymmetrical layout had to be made with layers 2 and 4 having di ff erent thicknesses. The lamellae was placed randomly within each layer and the location of each lamella was registered. A structural MUF 2. Materials and Methods 2.1. Lamellae 2.2. Hybrid CLT