Author: Renxiang Lu

Publisher:

Published: 2020

Total Pages: 321

ISBN-13:

Skewed highway bridges are commonplace, but the behavior of a skewed bridge can be significantly more complex compared to a straight bridge. As a consequence, the theoretical equations currently used to determine the load rating of a skewed bridge can be inaccurate for a large skew angle because the location of the girder and the moment (positive or negative) under consideration, the arrangement and position of cross-frames and parapets, and the location of loading applied on the bridge are neglected. In this study, an accurate load rating method is proposed that accounts for these effects. In the proposed method, a field test is used to determine the safe load-carrying capacity of the bridge, and a two-dimensional grid with longitudinal and transverse beam elements rigidly connected at intersecting nodes and vertical loads applied at the nodes (“grillage”) model is used to calculate the contribution due to skew. The field test is conducted using a calibrated truck driven over the bridge in successive runs that traverse the entire width of the roadway, and the measured strain response is used to calculate internal moments and experimental live load effects. To compare the behaviors between the actual bridge and the analytical estimation, the ratio of the bridge experimental and analytical load ratings is obtained. Additionally, contributors to the bridge live load are decomposed and compared to the analytical load rating to obtain the bridge deaggregated ratio of load ratings. The effect of critical span adjustment, longitudinal and lateral load distribution, unintended composite action between the slab and the girders, slab flexure, and additional stiffness in the system on the ratio of load ratings are qualified and quantified. The contribution of additional stiffness in the system to the actual load rating in comparison to the analytical is further discretized into the contribution to stiffness due to skew and the contribution to stiffness due to curbs and railings. The contribution to stiffness due to skew is calculated as the ratio of the statical moment in the skewed bridge grillage model to the corresponding statical moment in the equivalent straight bridge. The proposed method is illustrated using a case study of a steel girder highway bridge with a 43-degree skew angle. The tests indicate that the experimental load rating is 156% greater than the analytical load rating at the positive moment location, and 11% greater at the negative moment location. The increased load rating is primarily due to unintended composite action. Since unintended composite action is not reliable for load-levels above the linear-elastic region, the effect of unintended composite action is removed, resulting in a reliable experimental rating load rating that is 46% greater than the analytical load rating at the positive moment location. The grillage model indicates that additional stiffness due to skew contributes to a 19% increase in the positive moment load rating. The case study demonstrates that the proposed method provides a more accurate determination of the skew effect, compared to the theoretical equations currently in use, because the location of the girder and moment under consideration, the arrangement and position of cross-frames and parapets, and the location of loading applied on the bridge are explicitly incorporated. Therefore, field testing coupled with a grillage model has the potential to effectively load rate skewed highway bridges.