Evaluation of the performance of bridge steel pedestals under low seismic loads

Full-scale, quasi-static, reversed cyclic tests of a two-girder 40' bridge specimen rehabilitated with steel pedestals are conducted to characterize the behavior, assess the deformation and strength capacity, and experimentally evaluate any vulnerabilities of 19'' and 33½'' steel pedestals used by the Georgia Department of Transportation (GDOT). The six tests show varying deformation (1.75'' to 3.5'') and strength capacity (25.8 kips to 96.4 kips) of the pedestals for two loading directions (strong- and weak-axis) and two configurations for the placement of the anchor bolts. Tests from this study reveal the pedestals to be elastic, flexible components, where peak displacements reached were limited by a deformation mode rather than permanent deformation or instability of the pedestals themselves. The three modes of deformation observed are prying-action, bolts yielding, and concrete breakout. Prying-action of the anchor bolts embedded in the reinforced concrete bent cap is the predominant mode as the anchor bolts pullout from the concrete with increased cycling due to sliding (anchor bolts subjected to shear loading) and rocking of the pedestals (anchor bolts subjected to tensile loads). The force-displacement relationships capture the hysteretic behavior of the pedestals and indicate rigid body kinematics (sliding and rocking) of the system. From all six hysteresis loops, 7-17% of energy is dissipated through equivalent viscous damping of the pedestals, which is a necessary characteristic for the seismic performance of connection elements during an earthquake. The displacement ductility ratio, mu, ranges from 2.8 to 13 for all six tests, where ductility ratios from 4 to 6 may be expected in extreme seismic events. From the force-displacement hysteretic relationships, other response parameters such as the effective stiffness and equivalent viscous damping of the pedestals are computed and compared. The force-displacement hysteretic relationships, in turn, are used to calibrate an analytical bridge model to determine the displacement demands.; A GDOT candidate bridge with steel pedestals is idealized as a 3DOF system to determine the bridge displacement demands using two approaches: (1) by calculating the peak displacement from accelerations generated by uniform hazard spectra based on USGS (2002) maps for the range of structural periods from all six tests represented by the 3DOF system, and (2) by calculating the peak displacements from response spectra generated from a suite of site-specific synthetic ground motions developed by Fernandez and Rix (2006). For this simple 3DOF model of a candidate bridge in Georgia, the inelastic behavior defined for the pedestals shows satisfactory performance for low seismic loads where the deformation and strength capacity are adequate as long as a mode of deformation leading to a mechanism of failure does not occur. Best practices, design guidelines, and inspection and maintenance advice are recommended to GDOT based on this study.