PSI - Issue 78

Beatrice Travasoni et al. / Procedia Structural Integrity 78 (2026) 1111–1118

1114

(b)

(a)

Fig. 3. Calibration of the wall model based on EN 1998-3 formulations. The dashed line represents the lateral force-displacement response, with ultimate displacement d u indicated by the dotted line. The schematic on the right shows the wall geometry and boundary conditions.

The walls were modeled as nonlinear plate elements governed by the Drucker–Prager failure criterion. Compression-only springs were included at the base to simulate rocking within the plane of the wall, which was identified as the prevailing failure mode. The material parameters for the Drucker–Prager model were derived from the values listed in Table 1. The cohesion was set equal to f v0 , the initial shear strength for regular masonry, while the friction angle Φ was obtained from the friction coefficient μ using the standard relation Φ = tan -1 ( μ ). The base spring stiffness was computed as a function of the elastic modulus and wall thickness, expressed in MPa/m. The shear–displacement response obtained from the FEM model showed good agreement with the analytical predictions, with less than 10% deviation in ultimate shear strength. The drift limit d u , as defined by the code for compression bending failure, was also respected, as shown in Fig. 3. The shear response of the connections was defined as follows: for Connection Types 1 and 2, the shear stiffness was assigned proportionally to the ratio between the elastic moduli along and across the grain direction of the timber; for Connection Type 3, the shear behavior was calibrated based on experimental data derived from Foraboschi (2022). The calibrated material parameters and modelling assumptions are considered valid for the entire structural model and suitable for future applications involving similar masonry configurations. 2.2. Calibration of wall-to-diaphragm connections The timber floors, both original and retrofitted, are connected to the masonry walls through three types of nonlinear connectors, each defined by a specific force–displacement relationship, as illustrated in Fig. 4. The first system, based on Mirra (2021), consists of a steel angle bracket fixed below the main timber beam and anchored to the masonry wall with mechanical fasteners. Its nonlinear behavior was implemented using the analytical law proposed in the same study. The second system involves a CFS angle chemically anchored into the masonry and connected to a retrofit plywood panel through SDS screws. In this case, the global response is governed by failure at the plywood interface, with experimental data derived from Karki et. al (2022) and the mechanical behavior of individual screws determined in accordance with EN 1995-1. The third system, used exclusively in the presence of a reinforced concrete topping, includes chemical anchors connecting the steel amgle bracket to the masonry. The connection between the bracket and the slab is assumed rigid. The nonlinear behavior was calibrated using pull-out test results on threaded rods embedded in masonry, as reported by Tselios et al. (2023).