The behavior of ductile reinforced concrete bridge columns subj ected to cyclic loading has been the subject of extensive research. It is now possible to predict with fair accuracy the strength and deformation capacities of those columns given a variety of parameters that include the section geometry, amounts of longitudinal and transverse reinforcement, slenderness ratio, axial load and material strengths. Nonetheless, some areas of uncertainty remain, including the response of such columns when subjected to one or more components of intense earthquake ground motions, and the ability of current analytical models to predict performance.
As such, an integrated series of experimental and analytical studies were undertaken. This report describes the dynamic testing of four circular reinforc ed concrete bridge columns on the earthquake simulator of the Earthquake Engineering Research Center (EERC) of the University of California at Berkeley. The specimens were divided into two pairs, with each pair subjected to a different ground motion. Within each pair, one specimen was subjected to one component of the ground motion, while the other specimen was subjected to two components.
The four columns exhibited stable ductile behavior under several repetitions of the targeted ground motions. The bidirectionally loaded columns behaved similarly to the unidirectionally loaded columns under the design earthquake, and were able to sustain more repetitions of loading before failure was reached.
A number of elastic and inelastic analytical models were evaluated in terms of their ability to predict the local and global behavior of the tested columns. While simple models such as the stiff- ness degrading Clough/Takeda model give satisfactory results, the use of refined fiber elements results in better prediction of bot h global and local forces and de formations. Use of bilinear
hysteretic models proved inadequate. Properly calibrated elastic models were able to provide good predictions of maximum displacements.
Finally, two large analytical studies were carried out on a wide array of column heights,
diameters, and axial load intensities. The columns were subjected to large suites of ground motions scaled to match on average the design response spectrum. Results indicate that columns detailed according to modern seismic criteria generally behave satisfactorily. The first portion of the para-metric investigations examined the effect of three alternative design methodologies on
performance under unidirectional loading. These studies demonstrated that certain design approaches used today might lead to excessive displacement demands for columns with short periods, and that some columns might be susceptible to low-cycle fatigue failure. The second phase of the parametric studies examined the effects of bidirectional motions. It was found that peak bidirectional response was similar to that predicted unidirectionally, but that increased demands might be observed in the short period range. Bidirectional loading also tended to increase residual displacements. Elastic analysis methods were able to provide adequate predictions of displacement demands for moderate and long period structures, but often substantially underestimated demands in the short period range.
Full List of PEER Reports: click here.
