PSI - Issue 44

Samuel Barattucci et al. / Procedia Structural Integrity 44 (2023) 426–433 Barattucci et al./ Structural Integrity Procedia 00 (2022) 000–000

430

5

100 120 140 160 180

100 120 140 160 180

0 20 40 60 80

0 20 40 60 80

F [kN]

F [kN]

0

20

40

60

80 100 120 140 160 180

-140 -120 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140

Displacement [mm]

Diagonal Displacement [mm]

difference of the absolute displacements of both the upper and lower beams. The drift cycles protocol was specifically designed taking into account the expected displacement capacity of the system. 3.3. Test results and observed failure modes Figure 3 reports the main experimental results in terms of cyclic force-displacement curve. The continuous line of the envelope curve represents the 1 st cycle envelope while the dotted line represents the 3 rd cycle envelope. From the results, it can be observed a rigid response of the system up to the peak force, then the damaging of the masonry infill induces a progressive strength degradation especially for repeated cycles. For large amplitude cycles the strength contribution is provided only by the RC since the masonry infill is completely damaged. Figure 4 describes the failure mode observed during the test at different drift level. The diagonal cracking activation and the corresponding damaging of the end portion of the RC column for a drift level of about 0.5% is depicted in Figure 4a. The full infill corner damage was achieved for a drift level of 2% with the consequent extension of the cracking field of the end portion of the RC column (Fig. 4b). The nonlinear response of the case study frame has been investigated by means of a 2D numerical model implemented in OpenSees (Mazzoni et al., 2003). The frame is assumed to be rigidly fixed to the ground. A member-by-member modelling is adopted for the analysed frames. In particular, the “beamWithHinges” element is used and beams and columns of the r.c. frame are modelled as members constituted by an elastic element with plastic hinges at their ends. The length of the plastic hinge L pl is equal to the depth of the cross section. A fibre cross section is assigned to each plastic hinge, where both concrete and steel components are considered. The concrete part of the cross section is subdivided into fibres having 5 mm depth and width equal to the width of the cross section. Single fibres are used to model rebars. The Kent-Scott-Park constitutive law (“Concrete01” uniaxial material represented in Figure 5a) is assigned to concrete fibres. An elasto-plastic with strain kinematic hardening constitutive law (“Steel02” uniaxial material, represented in Figure 5b) is assigned to steel fibres. Concentrated loads equal to 250 kN are applied to the top column cross sections. Infill panels are modelled by means of a pair of diagonal trusses without tension resistance. As proposed by Panagiotakos and Fardis (1996) and Celarec et al. (2012) the force-displacement relationship of the diagonal truss is calibrated to replicate the shear force-drift relationship of the infill panel. This relationship consists of four branches, as shown in Figure 5c. The first branch corresponds to the linear elastic behaviour up to the first cracking of the infill and it is defined by the elastic stiffness K el and the (b) Fig. 3. Experimental results: (a) horizontal load vs horizontal drift; (b) horizontal load vs diagonal displacement obtained with wire instruments. 4. Numerical analysis 4.1. Numerical model (a)