Bagley Street Pedestrian Bridge

• Michigan's first cable stayed footbridge

• Unique asymmetric design with asymmetry in two major planes

• Erection engineering analysis to develop an erection manual

Detroit’s Mexicantown Bagley Street Pedestrian Bridge is the first cable-stayed bridge to be built in the state of Michigan. The bridge was designed by HNTB Corporation for Michigan Department of Transportation which appointed URS Corporation to act as its Construction Inspection engineer. Stress analysis, erection engineering and the preparation of a detailed erection manual outlining all stages for construction was carried out by specialist erection engineering consultant Genesis Structures for its client Walter Toebe Construction using LUSAS Bridge analysis software.

Overview

Constructed as part of the $230 million I-75 Gateway Project, Bagley Street Pedestrian Bridge is a two-span, asymmetric cable stayed structure that crosses 10 ramps and roadways, including both I-75 and I-96, to re-connect the east and west sides of Detroit’s Mexicantown community which has been divided since I-75 was built in the late 1970s. The bridge has a total length of 417ft made up of a 276ft main (west) span and a 141ft back (east) span. A single 155ft tall inclined and tapering concrete and steel pylon supports the main span via ten forestays set in a fan arrangement. The back span is fully supported by a deadman abutment and a pylon support strut. The bridge superstructure comprises a trapezoidal single-cell steel box girder of constant depth and varying width topped by a concrete slab that varies from 15ft-3in to 34ft in width over its length. Concrete barriers cast along the deck stiffen the deck and provide fixity for glazed pedestrian barriers. Five tuned mass dampers limit any vibrations due to wind or pedestrian traffic.

Staged erection modelling

Genesis Structures used LUSAS Bridge to model the bridge’s complex geometry, its members and their associated material properties . A detailed beam and shell element model represented all temporary and permanent structural components and allowed the 62 defined stages of the construction process to be accurately analysed.

Because of the unique geometry of the pylon base a separate 3D solid model was used to fine-tune the base stiffness and beam modelling method used for the main bridge model. For the main model, section properties for the pylon itself were obtained from a 3D CAD model. During its construction the pylon was temporarily restrained using four guy cables at two vertical levels and in addition to providing support during construction this also helped maintain alignment and restrict movement of the pylon prior to the installation of the stay cables and subsequent post-tensioning of the pylon. Temporary guys, cable load eccentricities, post-tensioning and time-dependent effects such as creep and shrinkage to CEB-FIP 1990 were all included and assessed at relevant stages of the modelling process.

The steel plates, stiffeners, composite concrete slab and concrete barriers of the superstructure were all modeled using 3D shells. Internal girder bracing was modelled with beam elements. Time-dependent (creep and shrinkage) effects on the concrete slab were also included. The concrete deck pours required temporary bearings to be used initially at the abutments (and on the model) to allow for beam rotation. Then, once the integral abutment connections were made, the modelled supports were updated accordingly. The eccentric cable loading on the box girder system produces torsion and lateral thrust in the girder. This is resisted by upward, downward, and lateral bearing supports at the abutments and by tension linkages and vertical and lateral bearings at the pylon support strut. Three falsework towers that supported the west span were modelled using compression-only supports that allow accurate modelling of the lift-off behaviour due to cable tensioning.

Cable installation and stressing were carried out in a balanced manner and to forces defined by the proposed erection sequence. Some stays required only a single jacking operation while others required two jacking operations at different construction stages. As the installation and stressing of the permanent stays progressed, the temporary pylon guys were removed. Progressive installation of cables and post-tensioning of the pylon caused a gradual decrease in reactions and eventual lift-off at the falsework supports. A 30-year creep analysis carried out at the end of the staged construction process evaluated long-term effects on the structure.

Design limitations

The contract documentation placed design limitations on the pylon restricting the horizontal displacement at its top to two inches and also restricted the maximum amount of bending moment about its weak axis at a specified level at its base to be 800 kip ft during its construction, unless any exceeded values could be proved safe. John Boschert, Structural Engineer at Genesis Structures explains: "The aim of limiting these values was to minimize the creep and shrinkage effect and achieve better geometry control." He continues, "By carrying out a detailed LUSAS analysis of the pylon at all stages of construction we were able to show that the effect on creep from the actual pylon deflection was acceptable, and that the overall moment capacity about the weak axis of the pylon was sufficient to resist the actual bending moments seen during construction."

Results obtained

From the analysis carried out with LUSAS, numerous graphs and diagrams were created to show time-history effects for key components of the structure. These included:

• Displacement time-history for the top of pylon

• Bending moment diagrams for the pylon weak axis during construction

• Cable tension time-histories for the temporary and permanent cables

• Reaction time-histories for all abutments, bearings and falsework supports

• Vertical displacement and stress plots for the top and bottom of the steel box for all stages of construction and for the beginning and end of service conditions

Beginning and end of service stresses for the cable stays and beginning of service stresses in the concrete deck slab were also obtained. Representative graphs produced from the results of LUSAS analyses are shown below:

Bending moment in pylon for a specific construction stage

Summary

From the staged erection analysis carried out with LUSAS Bridge, Genesis Structures was able to simulate and prove the proposed erection sequence and prepare a detailed erection manual and associated geometric control plan for its client and general contractor Walter Toebe Construction. Three-dimensional target coordinates and elevations were provided for key points on the structure, including at the pylon stay housing, at temporary shoring, at box girder splices, along the box girder deck and at all stay cable connection points. 

For more details on this project see Modern Steel Construction magazine July 2010.

"Modelling each stage of construction in LUSAS facilitated accurate geometry control and allowed us to confidently monitor the pylon bending demands."

John Boschert, Structural Engineer, Genesis Structures

Share this article