Steel Plate Girder Diaphragm and Cross Bracing Loads

The wide spectrum of options available to designers for analyzing and determining cross-frame forces can be a source of problems because different options may not result in similar solutions. The main objective of this project was to develop a set of recommendations and procedures and instructions to address analysis, design, and construction issues related to braces in steel I-girder bridges. Different methods that can be used to calculate brace loads are categorized and discussed in detail to identify the strengths and limitations of each method. The traditional 2D-grid analyses that are often used by commercial software packages ( such as MDX and DESCUS) does not take into account the full warping stiffness, resulting in underestimation of torsional stiffness of the girders. Although improved 2D-grid analyses may result in an improved representation of the full warping stiffness, these models are generally applicable to a no-load fit condition for cross-frames or diaphragms. The procedure by which 2D-grid analyses can be used for calculating cross-frame forces and other structural responses of bridges detailed with dead load detailing methods (erected fit and final fit) are described. It has been found that performance of improved and traditional 2D-grid analyses also depends on the framing layout (contiguous or staggered). Improved 2D-grid analyses are preferred for calculating the cross-frame forces because of the satisfactory performance for most of the framing layouts. A relatively simple method of simulating lack-of-fit is introduced in this report which makes use of models using three-dimensional (3D) finite element method (FEM). Although past studies have recommended using initial strain to simulate lack-of-fit in the cross-frames, this method can be tedious and complex. The proposed method makes use of element birth and death techniques to activate or deactivate the cross-frame elements to obtain a measure of the force or deformation at specific desired stages. The element birth and death method is generally more simple than the initial-strain methods with the essentially the same level of accuracy. Finally, different options for framing layouts, detailing methods, cross-frame configurations, and design methods for sizing the cross-frame members are discussed.


  • English

Media Info

  • Media Type: Digital/other
  • Edition: Final Report
  • Features: Appendices; Figures; Photos; References; Tables;
  • Pagination: 149p

Subject/Index Terms

Filing Info

  • Accession Number: 01535987
  • Record Type: Publication
  • Contract Numbers: BDK80-977-20
  • Created Date: Aug 11 2014 4:47PM