Investigation of the performance of a skewed steel bridge with and without cross-frames
Date
2023
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
In steel girder bridges, the girders are typically connected by cross-frames at different locations along their length. Cross-frames provide lateral-load resistance and stability to the steel girders during construction, reducing the unbraced length of the compression flanges of steel girders. Moreover, cross-frames are thought to have a beneficial impact on girder load distribution and, as a result, contribute to increased bridge system capacity. Therefore, once a simply supported composite bridge is constructed, the only theoretical purpose of the cross-frames is live load distribution. However, prior studies show that the role of cross-frames in distributing the load may be insignificant. In addition, cross-frames are introducing lateral forces that were not taken into account during the design stage. These localized forces induce many issues in steel bridges. For example, out-of-plane distortion, which is one of the main sources of fatigue cracking, occurs at the connection of longitudinal girders and transverse structural members such as cross-frames. Other studies emphasize that warping torsion occurs in the web of the girder cross-section. The maximum warping stresses are located at the mid height of the web near the cross-frame locations. In addition to the issues that these cross-frames cause on steel bridges, they also contribute a large percentage of the overall bridge cost due to fabrication complexities and construction fit-up issues. As the fabrication of the cross-frames can be as much as five times more than the fabrication costs of the steel I-girders, based on the modern steel construction. ☐ For these reasons, the advantages of reducing or eliminating cross-frames from steel bridges have been discussed in many studies. Thus, in this study, the opportunity for destructively testing a representative highly skewed steel I-girder bridge was leveraged to evaluate the role of cross-frames by considering an extreme scenario of removing all the cross-frames (but not the end diaphragms) from the structure prior to testing. The results of this test show that the bridge demonstrates a high load capacity. The equivalent of 15 design trucks were applied, which is 3 times the theoretical capacity and 7.5 times the number of trucks physically possible to be simultaneously traveling across the one-lane structure. Only local yielding occurred at the midspan of the girder’s bottom flange and the web of the most heavily loaded girder. No other yielding was noticed on any other girder. Load distribution, composite action, rotational and axial fixity, and direction of the transverse path were calculated throughout the loading range during this test. The results demonstrated that the load was distributed effective without the cross-frames, through the deck of the bridge. Moreover, the bridge kept handling the load even though the deck was severally cracked. Another finding was that the end I-beam diaphragms are a predominant source of fixity at the end of the bridge’s girder. ☐ In addition to this test, another test was previously conducted for the same bridge with the presence of the intermediate cross-frame, which gave an opportunity to compare the bridge's performance with and without cross-frames. The difference in the distribution factor between the bridge with the cross-frame and the bridge without them was less than or equal to 5%, according to the results of the comparison. Field data shows that the bridge with the cross-frames is subjected to higher lateral and warping stresses than the bridge without them. In addition to the field tests, a finite element analysis (FEA) was conducted, and compared to the field data. The FEA predictions closely match the field data. In addition, The FEA showed that at the inelastic load level, the cross-frames did not contribute to redistributing the load, as these cross-frames yielded much earlier than the heavily loaded girder. Finally, a unique finding in this study was that cross-frames induce concentrated major-axis bending moments that change the bridge's expected behavior.
Description
Keywords
Cross-frames, Skewed bridges, Steel bridges, Lateral forces