This project sought to increase the seismic safety of the state’s bridges by improving the connections among bridge components. A typical Washington state concrete bridge bent consists of cast-in-place piers, precast, pre-stressed girders, and a cap beam. Successful interaction among all three components must be achieved to transfer induced loads effectively and provide adequate resistance to seismic shaking.
The cap beam comprises a precast crossbeam and a cast-in-place diaphragm, flush with the girders. The method of creating the tension connection between the bottom girder flange and the cap beam that WSDOT uses is to extend some of the bottom steel strands into the cast-in-place diaphragm, where they are anchored with strand vices and bearing plates. However, that hardware configuration can cause space conflicts and difficulties during construction. Therefore, the goal of this project was to create a reliable, effective, and practically applicable way of anchoring strands extended from the girder into the cap beam.
To develop the girder-diaphragm seismic connection, the researchers first established the adequacy of the smallest possible strand anchor that would still lead to strand ductile failure due to yielding rather than strand anchor failure due to crushing of the concrete. Through laboratory testing, they determined that a barrel anchor alone, with no backing plate, is easily sufficient.
Next, the researchers investigated the possibility of a group of strands breaking out from the diaphragm. They conducted tests on strands, anchored by strand vices, grouped in various patterns and embedded at various depths in concrete blocks representing a diaphragm. They found that the results were closely predicted by the Concrete Capacity Design method, which they concluded can be used to predict the minimum embedment length required to ensure that strand fracture occurs before break-out failure.
Finally, the researchers evaluated the distribution of girder bending moments across the bridge deck. They determined that the important parameters are the flexural properties of the girders and the torsional properties of the cap beam. The outcome focused on the girder end moments when the column reaches its full flexural capacity.
One conclusion of the study was that for bridges constructed with a two-stage cap-beam, which WSDOT uses extensively, the number of strands that should be extended from each girder may be reduced from the number suggested by the existing design method.
Authors:
Kristina Tsvetanova
John F. Stanton
Marc O. Eberhard
UW Department of Civil and Environmental Engineering
Sponsor: WSDOT
WSDOT Technical Monitor: Bijan Khaleghi
WSDOT Project Manager: Lu Saechao