Design and response of lower-level beam-column joints in ductile reinforced-concrete double-deck bridge frames

A research program of field and laboratory tests to study the seismic response of reinforced concrete double-deck freeway structures has been concluded at the University of California, Berkeley. The program had three phases of testing. The first phase of testing was conducted in situ immediately after the 1989 Loma Prieta Earthquake on a section of the Cypress Street Viaduct that had not collapsed. The purpose of these tests was to gain an understanding of the possible causes for collapse and to investigate temporary retrofit measures. The second phase of testing was the proof test of a comprehensive retrofit concept for a bent of the Alemany Freeway. This retrofit consisted of replacing the non-ductile frame elements with ductile ones. The test results confirmed that the retrofit design achieved a high level of strength and ductility capacity. The third phase of testing focused on design criteria for beam-column connections in new ductile double-deck freeways. This report describes the research findings of the third phase of the program.; Because current bridge design practices result in column yielding, the applicability of design criteria for beam-column joints based on yielding in beams has been questioned. Two one-third scale test specimens representing the lower-level beam-column connection in a double-deck bridge structure were built and tested in the laboratory. The difference in strength of the structural members of the two test specimens resulted in different peak demands on the beam-column joint. The tests were used to determine the response of the joints when subjected to unidirectional and bidirectional load cycles and compare the response to the different levels of maximum joint shear stress demands. The test specimens were able to sustain the design forces and deformations without significant distress in the joint. Most of the deformation and damage concentrated at the column ends.; The first test specimen was designed to impose a lower demand on the beam column joint. The joint was able to sustain the large deformations demands and provided the proper bond conditions necessary to develop the full strength of the columns. The second specimen had a joint shear stress demand in excess of that prescribed by the current design recommendations of the ACI-ASCE Committee 352. Even though the joint did not sustain significant visible damage, it did not provide satisfactory anchorage conditions to develop the full flexural strength of the columns.; The scope of the program consisted of analyzing the measured response and comparing it with currently available data and design methodologies. A force transfer mechanism was developed on the basis of the measured data and on currently accepted models. This force-transfer mechanism was used to model the response of the test structure. Because of the basic assumptions of this design model, it can be expanded to model beam-column connections of various geometric and detailing characteristics.