CCEER-99-3: Seismic Response of Bridge Pier Walls in the Weak Direction

Abstract

The research presented in this report consisted of an experimental and an analytical study. The objective of the experimental study was to evaluate the out-of-plane seismic behavior of representative bridge pier walls that exist in the US. The analytical study had two objectives, the first was to develop and calibrate an analytical model to determine the seismic response of bridge pier walls, while the second was to develop an approach that relates the displacement ductility capacity to the amount of confinement steel.

A comprehensive bridge pier wall survey was conducted to collect information about existing typical pier walls in the US. The data were well distributed geographically and states with full range of seismicity were represented. A statistical analysis was performed on the collected data to select test parameters and specimens. Seven specimens were designed, built, and tested in the experimental study under slow cyclic loads. The failure mode of the wall specimens was either compression failure of the concrete or fracture of the vertical reinforcing bars due to low-cycle fatigue.

An analytical model was developed and calibrated. A computer program called "PIER" was written to implement the analytical model. Good agreement was found when comparing the calculated and measured responses of the pier wall specimens tested in the course of this study and at the University of California at Irvine (UC-Irvine). A parametric study was conducted to extend the seismic response study to bridge pier wall cases that were not tested experimentally using the computer program "PIER". The parameters were the ratio of the wall height to thickness, the vertical steel ratio, the confinement steel ratio, and the axial load index. Pier wall cases that need retrofit were identified based on the expected seismic response.

A new practical approach to relate the confinement reinforcement in the plastic hinge zones of bridge pier walls to the displacement ductility capacity was developed based on the results of the parametric study. The proposed approach is recommended for design because, unlike other available models that are based on testes of columns, the proposed model is calibrated against wall data.

The displacement ductility capacity of six typical pier walls that contained confinement steel designed using the available code provisions was calculated using the proposed approach. A comparison of the resulting ductilities was made to identify design provisions that lead to best level of performance. The displacement ductility capacity of pier walls 29 and 30 in Moribe Viaduct that was severely damaged during the 1995 Hanshin Awaji Earthquake, Japan, was calculated. The damage and the poor seismic performance of these walls, indicated that the actual ductility capacity was lower than those calculated. The likely reason for the poor seismic performance of the walls was discussed.