Studies of a full-scale horizontally curved steel I-girder bridge system under self-weight

This dissertation presents original contributions in two areas: the testing of individual round steel tubular members under concentric and eccentric loads and the behavior of curved systems during erection. The tubular member studies established member capacity limits and helped optimize instrumentation on cross frames within the experimental structure. The primary results of this part of the work were the determination that (a) the original 1/8" (3 mm) member thickness would not be sufficient and that thicker 1/4" (6 mm) members would be needed, and (b) that only four strain gages could be used to measure the bending and axial forces in the members.; The curved bridge system is the first realistic large-scale curved steel bridge to be tested anywhere in the world. The tests produced the first sets of laboratory data on fullscale curved steel I-girder bridge behavior during construction. Six variations of the final experimental structure, a simply-supported three-girder system with a 90 ft. (27 m) span and a 200 ft. (61 m) radius, were examined with differing combinations of intermediate shoring support. Results from the erection studies showed that adequate equilibrium could be obtained using proposed instrumentation schemes. In addition, the results showed that a companion analytical model produced good predictions of behavior. It was also shown that an approximate hand-calculation method used to analyze curved steel girder bridge systems, the V-Load Method, gave conservative approximations of exterior girder and cross frame member behavior but nonconservative predictions of behavior of the interior girder.; The work described in this dissertation is part of a much larger project aimed at developing new rational guidelines for the design of horizontally curved steel I-girder bridges. Because of the uniqueness of the tests and their scale and cost, insuring the reliability of the data obtained and documenting all aspects of the material and geometric properties of the structure were primary concerns. The development of a comprehensive test methodology, ranging from the careful selection of the instrumentation to the validation of the data reduction schemes, is another primary contribution of this dissertation. It is expected that the data generated from the full-scale tests will become the benchmark for future extensive analytical and parametric studies of curved steel bridges.