Results from 23 cyclic tests, including 18 cantilever-typed steel bridge piers and five beam-to-column connections, are presented to investigate their ductile fracture behavior as related to the seismic design of steel bridge structures, and based on shell and fiber models, two evaluation methods of ductile crack initiation are proposed. The effect of various parameters, including plate width-thickness and column slenderness ratios, cross-section shape, loading history, repeated earthquakes and initial weld defect is investigated experimentally. Among these parameters, width-thickness ratio, loading history and initial weld defect are shown to have significant influence on ductile fracture behavior. The test data suggest that for unstiffened box specimens, current seismic design provision limits on ultimate strain may not provide sufficient ductility for seismic design. On the other hand, based on the experimental results, two damage index-based evaluation methods respectively using shell model and fiber model are successfully employed to predict ductile fracture of steel bridge structures. Comparisons between experimental and analytical results show that they can predict ductile fracture behavior with good accuracy across the specimen geometries, steel types, loading histories and initial weld defects.
Effects of the Weld Electrode on the Fracture Behavior of Welded Steel Beam-to-Column Joints