| description abstract | The behavior of skewed steel I-girder bridges under live load is complex and challenging to interpret due to sparse existing field data; predictions used in the design process can include oversimplifications regarding superstructure load distribution and girder lateral response. To further understand the static and dynamic live load response of skewed steel I-girder bridges with different abutment conditions, two continuous-span steel I-girder bridges, one of skew 41° with stub abutments and the other of skew 45° with integral abutments, were evaluated in the field under isolated (single truck) live load as full and temporary half (during staged construction) bridges. Tests were conducted at 3, 32, and 56 km/h (2, 20, and 35 mi/h, respectively) for multiple load paths; data acquisition systems measured field response at 20 Hz. Girders and cross-frames were instrumented with strain gauges, girder end rotations were measured with tiltmeters, and bridge end movements were monitored with displacement transducers. Field instrumentation provides insight into superstructure behavior and data for validation and refinement of three-dimensional numerical simulation approaches (e.g., representation of abutment and bearing conditions, proper inclusion of structural components, and modeling of connections). The simulation approaches were then adopted for assessing bridge live load behavior beyond the range of field measurements, which can also be used in other research with limited access to field information. The live load distribution factor for girder strong-axis bending in standard line girder analysis was observed to be conservative, especially for interior girders when considering loading from the isolated test truck. Girder bottom flange lateral bending stress along the bridge span (in-span) was more significant for interior than exterior girders under the live load tests, whereas the lateral bending stress near bridge supports was mostly uniform for all girders (and of smaller magnitude than the in-span stress). Truck live load generally induced the most critical response on adjacent girders and cross-frames, except when loading close to an exterior girder—considerable bottom flange lateral bending response was observed on multiple adjacent interior girders for the exterior loading, with peak stresses away from the directly loaded bridge obtuse corners. Dynamic live load effects were evaluated by calculating dynamic load allowance (DLA) based on field measurements; the DLA used in standard practice for strength limit states was generally acceptable but slightly unconservative for an integral abutment bridge exterior girder. | |
