Full-Scale Laboratory Evaluation of Hybrid Composite Beams for Implementation in a Virginia Bridge

This research project studied a steel-reinforced concrete and fiber-reinforced polymer (FRP) structural element called the Hybrid-Composite Beam (HCB). The beam was used in a skewed simple span superstructure replacement project over the Tides Mill Stream in Colonial Beach, Virginia. For typical HCB construction, each beam is transported to site as a lightweight FRP beam shell. Self-consolidating concrete is pumped into the shell interior arch form, and when the concrete hardens, it stiffens and strengthens the beam so that it can act as falsework to carry the weight of a cast-in-place concrete bridge deck. Unstressed prestressing strands are embedded in the FRP shell bottom flange during the resin placement, and these strands equilibrate thrust in the arch and stiffen the beam to meet service deflection criteria. After the deck is placed, the HCB system performs as a longitudinal flexural member, with the bridge deck resisting compression and prestressing strands and the FRP bottom flange resisting tension. The primary research goal was to document the HCB as a structural component and as a bridge system, with the outcome being validation of key assumptions that can be applied to future designs such as, for example, strain compatibility between the FRP shell and steel strands. The research was conducted in five phases. In Phase 1, the HCB flexural rigidity and through-depth strain distributions were quantified considering just the FRP shell with unstressed strands. These tests confirmed flexural rigidity estimated by hand calculations and strain compatibility under uniform loads. Phase 2 evaluated flexural behavior of the HCB FRP shell and poured concrete arch. Phase 3 testing was performed after three HCBs were made integral with cast-in-place concrete end diaphragms and a reinforced concrete bridge deck. Point loads, to simulate an HL-93 design truck as specified in American Association of State Highway and Transportation Officials (AASHTO) LFRD Bridge Design Specifications, were applied to the bridge deck to maximize shear, flexure, and torsion in the skewed bridge. Live load distribution between the three girders was approximately equal and the assumption of strain compatibility between the bridge deck, FRP shell, and steel strand was confirmed. Stresses in bottom flange FRP strands and the top of deck concrete were less than 30% of material limits under service level live loads. The concrete arch fell below the composite neutral axis, placing it in tension along the span. After the live load system tests, a more detailed investigation was performed in Phase 4 to explore transverse deck behavior. Transverse flexural demands were approximately 20% of the design capacity and standard truss bars, as specified by the Virginia Department of Transportation, are not necessary because of the small clear span of the slab between beams. In Phase 5, the bridge system was saw-cut longitudinally to separate it into three individual HCB composite beams. Two beams were load tested to failure at the Structures Laboratory at Virginia Tech. For one of the two beams tested at Virginia Tech, 14 out of a total 22 strands were cut at mid-span to simulate strand deterioration and for comparison the other beam remained undamaged prior to testing. The observed beam failure modes were mid-span concrete crushing for the undamaged beam and mid-span strand-FRP bond failure for the damaged beam. In support of Phase 5, a three-dimensional (3D) finite element model was developed to explore flexural and shear force distributions along the span, which led to a shear design procedure in which shear force is distributed based on the relative moments of inertia of the FRP shell and arch. Shear resistance is provided by the FRP shell webs and the concrete arch and fin.