Lateral load distribution for steel beams supporting an FRP panel.

Abstract

Fiber Reinforced Polymer (FRP) is a relatively new material used in the field of civil engineering. FRP is composed of fibers, usually carbon or glass, bonded together using a polymer adhesive and formed into the desired structural shape. Recently, FRP deck panels have been viewed as an attractive alternative to concrete decks when replacing deteriorated bridges. The main advantages of an FRP deck are its weight (roughly 75% lighter than concrete), its high strength-to-weight ratio, and its resistance to deterioration. In bridge design, AASHTO provides load distributions to be used when determining how much load a longitudinal beam supporting a bridge deck should be designed to hold. Depending on the deck material along with other variables, a different design distribution will be used. Since FRP is a relatively new material used for bridge design, there are no provisions in the AASHTO code that provides a load distribution when designing beams supporting an FRP deck. FRP deck panels, measuring 6 ft x 8.5’, were loaded and analyzed at KSU over the past 4 years. The research conducted provides insight towards a conservative load distribution to assist engineers in future bridge designs with FRP decks. Two separate test periods produced data for this thesis. For the first test period, throughout the year of 2007, a continuous FRP panel was set up at the Civil Infrastructure Systems Laboratory at Kansas State University. This continuous panel measured 8.5 ft by 6 ft x 6 in. thick and was supported by 4 Grade A572 HP 10 x 42 steel beams. The beam spacing’s, along the 8.5 ft direction, were 2.5 ft-3.5 ft-2.5 ft. Stain gauges were mounted at mid-span of each beam to monitor the amount of load each beam was taking under a certain load. Linear variable distribution transformers (LVDT) were mounted at mid-span of each beam to measure deflection. Loads were placed at the center of the panel, with reference to the 6 ft direction and at several locations along the 8.5 ft direction. Strain and deflection readings were taken in order to determine the amount of load each beam resisted for each load location. The second period of testing started in the fall of 2010 and extended into January of 2011. This consisted of a simple-span/cantilever test set-up. The test set-up consisted of, in the 8.5 ft direction, a simply supported span of 6 ft with a 2.5 ft cantilever on one side. As done previously both beams had strain gauges along with LVDTs mounted at mid-span. There were also strain gauges were installed spaced at 1.5ft increments along one beam in order to analyze the beam behavior under certain loads. Loads were once again applied in the center of the 6 ft direction and strain and deflection readings were taken at several load locations along the 8.5 ft direction. The data was analyzed after all testing was completed. The readings from the strain gauges mounted in 1.5 ft increments along the steel beam on one side of the simple span test set-up were used to produce moment curves for the steel beam at various load locations. These moment curves were analyzed to determine how much of the panel was effectively acting on the beam when loads were placed at various distances away from the beam. Using these “effective lengths,” along with the strain taken from the mid-span of each beam, the loads each beam was resisting for different load locations were determined for both the continuously supported panel and the simply supported/cantilever panel data. Using these loads, conservative design factors were determined for FRP panels. These factors are S/5.05 for the simply supported panel and S/4.4 for the continuous panel, where “S” is the support beam spacing. Deflections measurements were used to validate the results. Percent errors, based on experimental and theoretical deflections, were found to be in the range of 10 percent to 40 percent depending on the load locations for the results in this thesis.