Search

A new form of light steel composite beam has been developed that uses C sections acting in tension with shear connectors in the form of screws or bolts or perforations in the web of the C section. The shear and bending resistance is also increased by using side C sections to the beams. Bending tests on point-loaded beams of 0.8, 1.1 and 1.7 m span showed that for the short span beams, the longitudinal shear bond strength of the base C sections is 1.4 N/mm2 for plain C sections and 2.3 N/mm2 for perforated C sections when expressed over the horizontal plane. Plain side C sections added 80% and perforated side Cs added 130% to the load-bearing capacity of the composite beams with base Cs. The theory is extended to cover elastic design taking account of partial shear connection in which the shear stiffness of the perforated web of the C section is approximately 10 N/mm3 (per unit web area) and that of the plain web is approximately 3 N/mm3.

This paper describes a sequence of experiments on a long-span asymmetric composite cellular beam. This type of beam has become very popular, combining the composite action between the steel and concrete with the increased section depth, compared with more commonly used solid-web I sections. Openings in the steel web also reduce the self-weight and can accommodate the passage of service ducts. Eurocode 4 recommends a high degree of shear connection for asymmetric composite beams despite the practical difficulties in achieving this. Recent research suggests that the required degree of shear connection could be reduced, particularly for beams that are unpropped during construction. However, little test data exists to verify the behaviour of unpropped composite cellular beams. Therefore two series of tests were conducted on a 15.26 m long asymmetric composite cellular beam with regular circular openings and an elongated opening at the mid-span. The degree of shear connection was 36%, less than half of that recommended in Eurocode 4, and the beam was unpropped during construction. The beam was subjected to uniformly distributed loading and shear load during the tests. The end-slip, mid-span vertical deflection, shear connector capacity and strain distribution were examined. The beam failed at an applied uniform load of 17.2 kN/m 2 (3.4 × design working load 5.0 kN/m 2 ). The member withstood an applied shear load that was 45% higher than predicted, and exhibited a Vierendeel mechanism at the elongated opening. Overall, these tests demonstrated the potential of unpropped composite cellular beams with low degrees of shear connection.

The elasto-plastic analysis of composite beams is important when considering the increase in bending resistance of the beam and the end slip between the steel and concrete at higher strains. This paper provides a simplified method of elasto-plastic analysis by considering equilibrium of the composite cross-section as a function of its strain profile. A parabolic–rectangular stress block for concrete is used in this model with a declining concrete strength at strains exceeding 0.0035. The bending resistance of the composite beam is expressed as a function of the bottom flange strain, and is compared to fully plastic design to EN 1994-1-1: Eurocode 4 and the AISC LRFD Code. The effect of various parameters on the development of the plastic bending resistance of composite beams is investigated, such as asymmetry of the section, the steel strength, the influence of propped or un-propped construction, strain hardening in the steel and reducing concrete strength at high strains, interface slip, and the effect of openings in the web of the beams. It was found that a moment of 95% of the plastic bending resistance of a composite beam (0.95 M pl ) is reached at a flange strain of 2 to 4 × yield strain for propped beams and 5 to 10 × yield strain for un-propped beams. When strain hardening in the steel is included in the analysis, bottom flange strains at a moment of 0.95 M pl are reduced by up to 30% relative to the case without strain hardening.