PSI - Issue 64

Austin Martins-Robalino et al. / Procedia Structural Integrity 64 (2024) 418–425 Martins-Robalino and Palermo / Structural Integrity Procedia 00 (2019) 000 – 000

423

6

rupture or buckling of longitudinal steel reinforcement and instead predicted failure of the concrete which led to sliding of the wall at a height of 50 mm above the base.

a) b) Fig. 4. Lateral load-displacement response of (a) Wall SWS V1; (b) Wall SWN V1.

3.2. Tension Stiffening with Local Fracture and Reduction of Bar Area Two phenomena were investigated to improve the prediction capabilities: local fracture of bars due to tension stiffening and bar area reduction due to strain gauge installation. These changes were considered as they reflect observed phenomena during experimental testing. During experimental testing it is common to observe a lower rupture strain of reinforcement bars in concrete compared to the rupture strain obtained from bare bars such as those in tension tests. This phenomenon is a result of tension stiffening and cracks causing strain localization at exposed sections of reinforcement instead of the more uniform strain development along the length of a bare bar present during tension tests. This local fracture is also reflected in the inelastic rotation demands of seismic provisions in design codes such as CSA A23.3-14 (2014); which assumes a maximum reinforcement strain of 0.05, significantly below the rupture strain of steel. The “Modified Bentz with Local Fracture” tension stiffening model in VecTor2 considers this phenomenon by modifying the rupture strain of embedded reinforcement based on the work of Zhang (2020). For Walls SWS and SWN, this resulted in embedded rupture strains of 72 mε and 68 mε, respectively, for the 10M reinforcing bars, lower than bare bar rupture strains reported in Table 1 and used in the preliminary models. During the process of stain gauge installation, it is necessary to provide a smooth even surface to attach the strain gauge, which is accomplished by sanding of the bar to remove the ribs over the gauge length and a portion of the bar diameter. Although the process attempts to remove the least amount of material, there was a reduction in the bar area which led to fracture and buckling of bars at these reduced sections. This pattern of fracture and buckling at the location of the strain gauges was observed after removal of damaged concrete in both Wall SWS and Wall SWN (Fig. 5). Note that SE-SMA bars did not require removal of material beyond the surface finish to attach the strain gauges and there was no observed fracture or buckling of these bars. To incorporate this, the truss elements at the locations corresponding to strain gauge on the 10M bars had their area reduced by 2.75%.

a) b) Fig. 5. Web and boundary region bars after completion of testing in (a) Wall SWS and (b) Wall SWN (Morcos & Palermo, 2019).