Bridges, Tunnels and other Transportation Structures
Single Abstract
Seaska Perez-Aviles, EIT
Graduate Research Assistant
University of Florida
Gary Consolazio
Professor
University of Florida
H.R. (Trey) Hamilton
Associate Professor
University of Florida
Prestressed concrete systems, especially multi-strand post-tensioned systems, are widely used in bridge construction and are considered to be cost-effective, structurally efficient, and durable. While occurrences have been rare, issues arising from poor grouting practice or poor material performance have led to occasional tendon durability problems. In response to these issues, future post-tensioned bridges in Florida will incorporate flexible fillers into some or all of their multi-strand tendons. Use of flexible fillers will typically integrate combinations of unbonded tendons and bonded prestress and/or mild steel reinforcement (mixed reinforcement conditions), which presents constructability and structural implications. Although this design approach is new in the United States, it has been widely used for major construction projects in Europe.
The flexural behavior of prestressed beams is affected by the presence of unbonded tendons. For instance, strains in unbonded tendons are not compatible with the adjacent concrete strains, which invalidates the use of strain compatibility. The tendon stress is dependent on deformation of the entire member and on assumptions concerning friction and tendon contact within the member. Also, the ultimate stress for unbonded tendons is lower than for bonded tendons. Although current AASHTO design guidance includes separate provisions for bonded and unbonded systems, it does not provide a rational approach for prestressed concrete members with mixed reinforcement conditions. To address this issue, this study involves development of guidelines for the analytical evaluation of ultimate strength of concrete girders with mixed reinforcement. Non-linear finite element analyses are conducted in order to evaluate the behavior of this type of structure. Girder models are validated against data from experimental tests conducted on full-scale I-girder beams with internal parabolic tendons and externally deviated tendons. A parametric study varying the ratios of unbonded and bonded reinforcement on prestressed beams is conducted and results are presented. Guidelines that identify specific analytical features that are needed to accurately represent the ultimate strength behavior of concrete girders with mixed reinforcement are also presented.