Strength And Fatigue Of Three Glass Fiber Reinforced Composite Bridge Decks With Mechanical Deck To Stringer Connections

Replacement of the steel grating deck on the lift span of the Morrison Bridge in Portland, OR, will utilize glass fiber reinforced polymer (FRP) panels to address ongoing maintenance issues of the deteriorated existing deck, improve driver safety and introduce bridge water runoff treatment. This report outlines the testing methods and results of an experimental program aimed primarily at evaluating a new open cell deck. While most FRP panels are connected via shear studs that are grouted within isolated pockets, the panels in this case were bolted directly to the steel stringers. Two different FRP deck options were evaluated for comparison: one with open cells and the other with more conventional closed box extrusions. The objective was to evaluate the strength of the FRP to steel stringer connection with individual bolt connection tests, the strength and fatigue resistance of the FRP decks themselves, and the relative lateral stiffness contribution of the panels. Additional related tests were also included to complement the research effort such as the inclusion of tests on a closed box deck removed from the Broadway Bridge in Portland, OR, and strength tests of a retrofit attachment option of FRP deck to stringer using bolted clamps. While the monotonic, flexural, and shear strength of the deck exceeded the design values, the associated failure mode of the open cell panels was consistently via shear flow through the stem near the top flange. The residual displacement of failed FRP decks was found to be minimal, which would make visual identification of failed panels without applied load difficult in the field. Fatigue strength evaluation was conducted with two different protocols, where one was run to over 6 million cycles based on AASHTO defined loading and the other to 2 million cycles with higher than AASHTO defined loading. Fatigue failure was observed in only one specimen that was subjected to the higher loading condition, providing a sense of fatigue life of this material. Fatigue failure mode initiated in flexural fiber rupture, which was different to monotonic tests under the same loading configurations. Bolted deck to steel stringer connection tests indicated failure modes in the FRP with strength values that were in certain configurations well below the strength of the bolts. For cases where the bolted FRP deck was counted on to provide lateral stiffness, such as the case in the raised configuration of the bascule span, the closed cell was found to have approximately twice the stiffness. The results of these tests provide valuable data that can be applied to FRP bridge deck designs that utilize bolted connections and open and closed cell deck configurations under high traffic volumes.