CCEER 01-6: Experimental Study and Analysis of Retrofitted Flexure and Shear Dominated Circular Reinforced Concrete Bridge Columns Subjected to Shake Table Excitation

Abstract

This study presents the experimental and analytical study of four 1/3rdscale circular flexural dominated and two 1/3rd-scale circular shear dominated 1971 Caltrans detailed bridge columns subjected to shake table loading. The benchmark as-built flexural column reached it�s design capacity but failed as expected due to severe lap-splice failure at very low ductility. A second as-built flexural column was retrofitted with a steel jacket and subjected to multiple El Centro earthquake motions. The column performed well but the lap-splice slipped before the full flexural capacity was reached. A third as-built flexural column was retrofitted with a steel jacket and subjected to a large initial El Centro motion with subsequent aftershocks. The column performed well but failed to reach post yielding strains in the lap-splice region. A fourth as-built column was retrofitted with a carbon fiber wrap and subjected to multiple El Centro amplitudes until failure. The carbon wrap performed significantly better than the steel jacket retrofits, providing higher lap-splice bar strains and larger displacement ductility. Measured stiffness for all flexural columns were less than calculated using standard moment curvature procedures. Non-linear time history analysis was performed to compare the calculated and measured response, predicting force levels well but under-predicting the higher displacement responses. The two shear-dominated columns were tested in double curvature using a unique dual-link system. A severe shear failure followed by total collapse occurred at 3.25 x El Centro for the first column subjected to incremental earthquake motions. The second column ii subjected to a large initial motion also failed at 3.25 El Centro but was not followed by a total collapse. Current shear capacity equations were used to compare measured versus calculated capacity. The Caltrans equation including the effect of strain rate closely matched the measured shear capacity. Current shear stiffness calculations matched well with measured stiffness. Non-linear time history analysis was performed using hysteretic models created from the shear capacity equations. The non-linear dynamic models failed to capture the ultimate conditions of the shear columns.