Effect of skew on the minimum support length requirements of single-span bridges with seat-type abutments

Skew bridges are known to be susceptible to girder unseating during earthquakes, and empirical equations for minimum support length are used in their design. To determine the degree of conservatism or un-conservatism in these equations, a rigorous model of a skewed bridge was developed in OpenSees and validated against a comprehensive dataset from shake table experiments. The validated model was then used to perform a parameter study on a series of single-span, simply supported, prototype bridges with seat-type abutments. Nonlinear response history analyses were performed that included impact and friction effects at the abutments and were repeated for both far-field and near-field motions. It was found that the additional support length required to prevent unseating is a linear function of skew angle over the range 0°–60° for both types of motions. This is contrary to common practice, which uses a quadratic function (American Association of State Highway and Transportation Officials, AASHTO) or the inverse of cosine of angle of skew (Federal Highway Administration (FHWA)). Consequently, common practice appears to be underestimating support length requirements by up to 50% at midrange skew angles of 30°–40°.