Experimental Assessment of Controlled Rocking Masonry Shear Walls without Post-tensioning

Abstract

Several research studies have recently evaluated the seismic response of controlled rocking masonry walls (CRMWs) with unbonded post-tensioning (PT) tendons. These studies demonstrated that such walls typically have both low damage and the ability to self-center, thus presenting an enhanced seismic response compared to conventional fixed-base shear walls. However, practical difficulties related to PT implementation during construction, coupled with the problem of PT losses that subsequently affect the wall’s self-centering ability, point to an opportunity to develop an alternative approach to control rocking walls. In response, the current study introduces controlled rocking masonry shear walls without PT tendons and with an energy dissipation (ED) device of embedded unbonded axial yielding dog-bone bars, named as ED-CRMWs. Experimental results are presented herein from six half-scale two-story fully grouted ED-CRMWs that were tested under displacement-controlled cyclic loading. All six walls were designed to have the same lateral resistance to facilitate investigating the influence of different design parameters, including toe confinement strategies through steel plates or boundary elements, axial compressive stress levels, ED device locations, and horizontal reinforcement ratios. The experimental results are presented in terms of the failure modes and damage levels, force-displacement response, residual drifts, and ductility capacities. The results show that even with no PT, all ED-CRMWs preserved the intended self-centering behavior with a flag-shaped hysteretic response, having a maximum residual drift ratio of 0.15%, except for one unconfined wall. In addition, the strategy of using end-confined boundary elements produced the most effective performance of the system pertaining to strength degradation, self-centering, and displacement ductility with a drift ratio of 2.35% being reached before strength degradation. In general, all walls exhibited limited and localized damage at the wall toes, thus demonstrating the promising concept of relying on gravity loads and ED devices in CRMWs, without the need for unbonded PT tendons.