The response of reinforced concrete slabs to concentrated loading

Abstract

The analysis and design of reinforced concrete slabs subjected to concentrated loading has been based on the assumption that the response of such structures is dominated by flexural actions. Theoretical considerations and experimental evidence show that this assumption may not be valid in many real situations and thus that such slabs may possess load capacities significantly different to those predicted by usual methods. The aim of this investigation was to improve our understanding of the behaviour of such slabs and thus to facilitate the development of improved methods of analysis and design. Slab specimens instrumented to allow the measurement of reinforcing bar forces and concrete strains at a crack were tested in a laboratory situation, with the boundary conditions of the specimens being the major parameter under scrutiny. Associated simple beam specimens subjected to combined axial and bending loads assisted in the interpretation of the results of the slab testing. It was found that both the elastic and ultimate load capacities of a slab specimen restrained against inplane movements at the supports were far greater than those predicted by usual methods. The testing of specimens with no restraint to inplane movements showed that the enhancement of capacity was partly due to the presence of the restraint, but that equally significant was the existence i-n a far 'flatter' profile of force response in uniformly cracked reinforced concrete slabs subjected to concentrated loading than is predicted by thin plate theory. A further finding was that concrete tensile stresses can reduce steel forces at cracks by a much greater degree than is usually recognised. The use of the finite element method to provide an analytical technique capable of including aspects of realistic reinforced concrete slab behaviour ignored by usual methods of analysis and design was investigated. The performance of a particular model able to include the effects of cracking and the associated coupling of bending and inplane displacements was examined, and a series of analyses of slabs similar to the experimental slab specimens carried out. It was found that in simple slab situations the finite element idealisation and solution techniques used were not capable of reproducing with sufficient accuracy the complete response predicted by the theoretical model for slab behaviour embodied in the finite element model. The introduction of a more refined finite element, derived for the purpose, enabled the analyses corresponding to the experimental specimens to be carried out with marginally acceptable accuracy at 'pre-yield' load levels. Several limitations in the theoretical model for slab behaviour became apparent, including an inadequate model for conditions at a crack and a lack of reversing material models. The measured behavour of the experimental specimens was not able to be reproduced with great accuracy, although the predictions were an improvement on those given by usual methods. However, the analyses did illustrate several important aspects of the response of reinforced concrete slabs to concentrated loading.