SEISMIC BEHAVIOUR AND RETROFITTING OF URM BUILDINGS

Abstract

Earthquakes are natural hazards under which disasters are mainly caused by damage to or collapse of buildings and other man-made structures. The unreinforced masonry (URM) buildings have proved to be the most vulnerable to earthquake forces and have suffered maximum damage during the past earthquakes worldwide. Unreinforced masonry is an old age traditional method used for most of the low to medium rise buildings in many countries, including India. In most cases, neither seismic loads are considered while designing nor earthquake resistant features are incorporated in masonry buildings leading to their excessive damage during earthquakes. Understanding the failure mechanism of these buildings subjected to seismic loads will help in improving their performance. Thus, the various failure mechanism of different masonry buildings subjected to seismic loads has been presented herein. To improve the seismic performance of masonry buildings, a number of techniques have been adopted to strengthen the existing masonry buildings. A review of the existing strengthening technology used for strengthening masonry has been comprehensively discussed in this thesis. In addition, a detailed review of existing codal recommendations and guidelines have also been presented. Numerical modeling which serves as a powerful tool has also been reviewed with respect to masonry modeling. The procedure adopted for the strengthening of URM recommended in Indian standard code of practice IS 13935: 2009, using welded wire mesh (WWM) and coarse cement sand mortar has been adopted and presented in detail for all considered test specimens. The conventional masonry panels were constructed and tested to obtain the material properties and to develop a nonlinear material model for finite element (FE) modeling. Concrete Damaged Plasticity (CDP) constitutive model has been used in this study to simulate the non-linear behavior of masonry. Experiments have been conducted to evaluate the in-plane and out-of-plane behavior of both reinforced and unreinforced masonry panels. Two half-scale masonry models have also been tested on shake table whose sequential construction details are also presented herein. The experimental results of URM and reinforced masonry (RM) panels, as well as the two half-scale masonry models, have been compared with numerical simulations. In the first phase, masonry panels of size 500 mm x 500 mm x 230 mm have been tested for diagonal compression as per ASTM E519 to study the in-plane behavior of URM panels. The URM specimens were strengthened using WWM (1 inch, 1.5 inch, and 2 inch spacing) and 1:3 coarse sand mortar as per IS 13935: 2009. The behavior of both strengthened and URM panels ii have been compared in terms of strength, stiffness and ductility. The incorporation of WWM reinforcement on URM masonry panels resulted in an average increase of strength ranging from 1.88, 2.35 and 2.42 times more compared to control specimen of 1:4 mortar masonry samples. In case of 1:6 mortar masonry samples increase in strength varied from 11.51, 12.24, 13.00 times the controlled specimen in 1 inch, 1.5 inch and 2 inch spacing WWM respectively. The numerical results simulated with CDP model were compared with the experimental findings in terms of damage pattern and shear stress-strain plots. In the second phase, the URM panels were investigated for out-of-plane behavior as per ASTM E518. Masonry panels of size 1000 mm x 500 mm x 230 mm were tested under four-point loading condition to study the out-of-plane behavior of both URM and RM panels. The enhancement in flexural strength of RM specimen compared to URM specimen was investigated in terms of load carrying capacity, displacement, and ductility. The flexural load carrying capacity of masonry has significantly increased at an order of four in case of strengthened specimen compared to that of conventional URM panels. The experimental test results were numerically validated in terms of load-deflection plots and damage pattern. In the third phase, two half-scale masonry models, one URM and the other strengthened with 1.5 inch spacing WWM and 1:3 coarse sand mortar has been evaluated using shake table test facility in the Department of Earthquake Engineering, IIT Roorkee. Initially, free vibration test was carried out on both the models to compute the time period and natural frequency of the URM and RM models. The models have been tested on a shake table for a series of artificially generated acceleration time history compatible with Indian standard response spectra for seismic zone V on hard soil. These ground motions were applied at the base of the model and response has been recorded at the base and at the top of these models. The URM model experienced extensive damage confirming that URM buildings are highly unsafe during the earthquake and require retrofitting/ strengthening. Acceleration at the top of the models was observed and recorded during testing. The RM model was able to withstand three times more intense load (ground motion) compared to the URM model without any sign of distress. The modes of failure were observed and roof acceleration was recorded. The experimental results were validated numerically using finite element analyses. The results obtained from the numerical simulation were found to be in good agreement with the damage pattern and peak ground acceleration obtained from the experimental results. The RM model performed well during the dynamic testing confirming that the adopted technique can be effectively used for strengthening/ retrofitting of existing masonry structures.