Moment redistribution and post-peak behaviour of steel fibre reinforced concrete flexural members

Abstract

This thesis presents a study of the moment redistribution capability and post-peak behaviour of conventionally reinforced steel fibre reinforced concrete (R-SFRC) continuous members designed for moment redistribution. Because of the lack of research in this area, limitations are placed in design standards on the application of steel fibres in RC continuous members designed for moment redistribution and no guidelines are available in the standards on the requirement for minimum reinforcement for which R-SFRC flexural members will show sufficient level of ductility. Consequently, two sets of experiments were designed to investigate the moment redistribution capability and post-peak behaviour of R-SFRC continuous members.

In the first set of experiments, six full-scale two-span continuous RC beams with and without fibres were designed for 30% of positive and negative moment redistribution with respect to the linear-elastic condition. Dramix 5D steel fibres with nominal dosages of 30 and 60 kg/m3 were used and the tensile reinforcement ratios varied between 0.69% and 1.38%. The second set of experiments comprised of six full-scale two-span continuous one-way RC slabs with and without fibres, which were designed for 0 to 30% of positive moment redistribution with respect to the linear-elastic condition by varying the tensile reinforcement ratios between 0.21% and 0.42%. Dramix 3D steel fibres with a nominal dosage of 60 kg/m3 were used.

The test results showed that in all R-SFRC beam and slab tests, two plastic hinges fully formed to develop the failure mechanism indicating the ability of R-SFRC continuous members to achieve the theoretical (elastic) design moment redistribution. The R-SFRC beams having tensile reinforcement ratios more than 0.5% maintained their capacity up to a displacement of 50 mm whereas the R-SFRC slabs having tensile reinforcement ratios less than 0.5% showed a shorter displacement length over which hardening occurred before the peak load was reached, followed by a period of gentle softening. A comparison of ductility based on displacement and work done indicates that all the specimens had a good level of ductility, however, the ductility decreased with increasing moment redistribution.

Finally, the post-peak behaviour of R-SFRC flexural members was investigated using finite element (FE) models, with the models validated using the test data collected in this study. Parametric studies were undertaken to investigate the influence of volume and degree of hardening of tensile reinforcement, and dosage and softening slope of SFRC on the post-peak behaviour of R-SFRC flexural members. From this study, a model was proposed for defining the conditions needed to achieve a defined level of ductility for R-SFRC flexural members and based on the model the relationships for minimum tensile reinforcement for which R-SFRC flexural members show sufficient ductility were developed and verified against the tests undertaken in this study.