LAUSR

Abstract A total of 16 full-scale in-situ tests were carried out to investigate the performances of unreinforced clay brick masonry walls subjected to vented gas explosions. The pressure-time histories of vented gas explosions, displacement-time histories and failure modes of wall specimens in each test were recorded and analyzed. A detailed micro model for masonry wall was developed in the finite element software LS_DYNA 971. The numerical model was validated with the test data and intensive numerical simulations were conducted to explore the influences of boundary condition, bonding pattern and thickness of masonry walls on their performances against vented gas explosions. The results show that the classical yield-line modes dominate failure patterns of masonry walls subjected to vented gas explosions and the dynamic responses of masonry walls are determined by the peak value of overpressure. It is found that boundary condition and wall thickness have great influence on the performance of masonry walls, while bonding pattern has relatively limited influence on its performance. In addition, the recorded gas explosion loads were compared with the predictions from the conventional TNT equivalency method and the methods specified in NFPA 68 and EN 14994. The results indicate both NFPA 68 and EN 14994 approaches give very conservative predictions of peak pressure from gas explosions. The TNT equivalency method also overestimates the peak pressure and predicts a load profile which is significantly different from the test data.