The Effect of Long Term Loading on the Behaviour of Steel-Concrete Continuous Composite Beams Prestressed by Propping

It is now accepted practice to construct composite steel concrete bridges as continuous structures. Whilst the general adoption of continuous systems should bring useful economies, there is a technical disadvantage associated with the concrete deck in the hogging regions ever the internal supports. Here the deck is in tension and detracts from the improved stiffness expected from the use of continuous beam systems.
This loss of stiffness may be compensated for in one of several ways. In this study the technique of pre-cambering the steel beam prior to casting the deck slab and the subsequent removal wholly or in part of the pre-camber to induce compression in an otherwise tension zone is investigated with regard to long term effects.
Date collected over many years shows that concrete is a time dependent material. It is well known that concrete strength improves with age but, equally, when under permanent load, decays in elastic stiffness; concrete also shrinks.
In composite beams the two inter-related time dependent phenomena creep (elastic decay) and shrinkage are important. Creep is revealed as a loss in composite inertia, with a consequent change in stress patterns. Free shrinkage of the concrete deck slab is inhibited by virtue of its physical connection to the steel beam beneath it together with the restraint given by embedded reinforcement and is therefore stress inducing. Additional shrinkage restraint t is induced by virtue of continuity.
From this study of continuous composite beams prestressed by jacking to induce negative camber at intermediate supports, both creep and shrinkage are shown to be responsible for changes in reactions, deflections and stress configurations. Of these two time dependent phenomena shrinkage is shown to be dominant.
Eight continuous composite beans were studied with four resulting concrete strengths. Four arrangements of continuity reinforcement were employed. Jacking and de-jacking displacements were constant. The test period for all beams was 400 days.
Supporting data on the strength and elasticity of concrete employed and the influence of creep and -shrinkage is also provided. Data relating to the strength and elasticity of steel and the push out connector capacities is given. Concluding tests to destruction provide data on the ultimate load characteristics.