PSI - Issue 64

Maria Teresa De Risi et al. / Procedia Structural Integrity 64 (2024) 959–967 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

960

2

1. Introduction About 67% of Italian existing RC buildings have been designed to sustain gravity loads only (GLD). Almost all of them lack stirrups in beam-column joints and have a minimum amount of transverse reinforcement in columns. These features make such buildings extremely prone to shear failures under seismic actions. Worldwide, the shear failure of beams/columns or joints, is assessed differently. This leads to very different seismic capacity for the same building depending on the code adopted for its assessment. The main aim of this work is quantifying such a difference with reference to RC GLD Italian case-study buildings with different number of stories located in sites with different seismic hazard, based on nonlinear static analyses. Code-based safety checks according to European (Eurocode 8-part 3 (CEN, 2005); Italian DM 2018) and American (ASCE-SEI 41, 2017) codes are carried out. The results show that capacity at Severe Damage (SD) Limit State (LS) is generally limited by joints failures. European codes rely on the limitation of the maximum tensile and compressive principal stresses in beam-column joints, generally strongly limiting the building capacity to the first joint shear cracking (tensile failure). On the contrary, ASCE-SEI 41 (2017) leads to significantly higher capacity at SD LS for the shortest buildings. Relevant design of seismic strengthening, thus, strongly depends on the adopted code approach, leading to generally higher strengthening costs when the current European approach is used. 2. Overview of code-based shear capacity models worldwide In this section, a description of the shear capacity models adopted by the European (EN 1998-3, 2005), Italian (DM 2018; Circolare C7/2019), and American (ASCE/SEI-41, 2017) codes, for beam/column elements and for joints, is provided. It is important to underline that all the mentioned codes provide for the application of reduction factors for the material strength values. The mean material strengths resulting from in-situ tests must be divided by the partial safety factor of the material and by the Confidence Factor (i.e., CF) related to the Knowledge Level attained during the survey campaign. In this work, these factors are used, and a comprehensive level of knowledge has been always assumed, corresponding to a value of CF=1. 2.1. Investigated beam/column shear capacity models According to EN 1998-3 (2005), the shear capacity V R of a beam-column element is obtained by means of the shear degrading model by Biskinis et al., 2004 (V R,BIS ), as the sum of three contributions (eq. 1): V N , the contribution due to the presence of compressive axial load N (eq. 2), V c , the contribution of the post-cracking resistance mechanisms of concrete (eq. 3), and V w , the contribution of transverse reinforcement (eq. 3). According to EN 1998 3, 2005, V R degrades with the ductility demand, μ Δ ( since the degradation coefficient k decreases increasing μ Δ ), going linearly from 1 (i.e., maximum shear resistance) to a minimum value of 0.75 (i.e., minimum shear resistance). The same capacity model is adopted by Italian technical code (DM 2018), which modifies it for low ductility demand using the Variable Inclination Truss (i.e., VIT) model (eq. 5, in case of stirrups). V R is the same provided by EN 1998-3 (2005) model when μ Δ ≥ 3 (i.e., V R,DM18 =V R,EC8 ). When μ Δ ≤ 2, V R is the maximum between the values provided by EN 1998-3 (2005) model and by the VIT model (i.e., V R,DM18 = max(V R,EC8 ; V R,TIV )). Lastly, when 2 ≤ μ Δ ≤ 3, the strength is obtained from a linear interpolation be tween the two models. According to VIT model (that is prescribed for not-seismic loadings by DM 2018 and EN 1998-1, 2004), the shear strength of the element is the minimum between the shear corresponding to the excessive compression failure of the strut, V Rc , and the shear corresponding to the tensile failure of the transverse reinforcement, V Rs . The model adopted by ASCE/SEI-41 2017 is that by Sezen and Moehle (2004), i.e. an additive degrading model based on two strength contributions (eq. 6): V c , due to the post-cracking mechanisms of the concrete and to N (eq. 7), and V w , the contribution of the transverse reinforcement. The degradation coefficient k is equal to a maximum of 1.0 for ductility demand not exceeding 2; whereas it reaches a minimum of 0.70 for ductility demand equal or greater than 6, varying linearly between the two extremes. This shear strength model predicts a greater strength degradation compared to that prescribed by EN 1998-3, 2005.