full text is available online.
|Author:||Melanie Marie Bragdon|
|Title:||Behavior and Design of FRP-Reinforced Longitudinal Glulam Deck Bridges|
|Committee Chair:||Habib J. Dagher, Professor of Civil Engineering|
|Committee Members:||William G. Davids, Assistant Professor of Civil Engineering; Roberto Lopez-Anido, Assistant Professor of Civil Engineering|
|Date of Defense:||2002|
In 1977, the Weyerhaeuser Company developed a system for short-span timber bridges. The girder-free system consisted of longitudinal, vertically-laminated glulam panels joined by below-deck Transverse Stiffener Beams (TSB). This project addresses two potential areas of improvement in the construction and design of these bridges: a reinforced deck panel and an improved method for TSB design. This project has two objectives: (1) To evaluate the behavior and advantages of longitudinal glulam deck panels reinforced with Fiber-Reinforced Polymers (FRP) and (2) To evaluate existing AASHTO empirical TSB design criteria. The tension-reinforced deck panels can alleviate reliance on high grade wood laminations and allow longer spans and lighter decks. The new panels have the middle two-thlrds of the tension side reinforced with longitudinal E-glass FRP. The research addressed the selection of the FRP material system, the manufacturing process used for applying the reinforcement to the panels, the structural and economic benefits of FRPglulam panels, and the durability of the FRP. The approach included design, laboratory manufacture, and construction of a municipal pier in Milbridge, Maine. Wet-impregnated unidirectional E-glass fabrics were used to reinforce the 164. wide, 167-ft. long, 7-span vehicular pier. A crosssection reinforcement ratio of one percent was used, increasing panel stiffness by six percent. The pier showed the FRP-glulam deck as cost competitive with a prestressed concrete deck. The pier was load tested and performed as predicted under full design live load. The FRP has performed well after two years of harsh marine exposure. To evaluate AASHTO designs of the TSB, a parametric study was performed using a finite element model developed for this study. The model was validated against full-scale laboratory tests conducted at The University of Maine and Iowa State University. The finite element model incorporated orthotropic plate elements for deck panels, offset beam elements for TSB, nonlinear models for deck-to-TSB connections, elements to allow pretensioning of the connections, and elements to model bearing between the deck and TSB. The parametric study focused on shear and bending response of the TSB and the relative movement between adjacent panels. Over 140 analyses were conducted on 43 southern pine bridges designed according to current AASHTO criteria, using 50 load cases. Results showed that the empirical AASHTO design criteria for the TSB may be unconse~ative. In the most critical cases under AASHTO HS20 loading, TSB designed according to AASHTO criteria may experience maximums of either 68% more shear stress than allowable or 61% more bending stress than allowable. In addition, relative panel deflection may exceed the 0.1-inch asphalt serviceability criteria by 79%. Based on the parametric study performed on curb-free bridges, the following design criteria are recommended to replace the current AASHTO TSB design criteria. "In lieu of a more accurate analysis, the transverse stiffener beam shall be designed for the following bending moment and shear values: Shear = 0.45*wheel load and Bending Moment = (3.5 inches) *wheel load, as the wheel load represents the maximum wheel load for HS & H vehicles and 1.75*maximum wheel load for alternate military loading."
Bragdon, Melanie Marie, University of Maine, CET2002-001