BEHAVIOUR OF OVER-REINFORCED SELF-COMPACTING CONCRETE BEAMS

Abstract

This thesis presents the results of a comprehensive investigation of self-compacting concrete (SCC) beams containing steel reinforcement significantly in excess of the requirements for balanced strain conditions. Because of its superior workability compared to that of conventionally vibrated concrete (CVC), SCC is ideally suited for effective concreting of the congested over-reinforced beams. The superior workability of SCC is attributed to its unique mixture characteristics. At the first instance, the mechanical properties of SCC in terms of the stress-strain characteristics have been investigated for a wide range of concrete strengths and it is shown that selected constitutive models for CVC were unable to give accurate predictions for the complete stress-strain response of SCC, especially for the descending branch. A new constitutive model has been proposed for unconfined SCC and validated with the help of the experimental results of SCCmixtures of other investigators reported in the literature. The proposed constitutive model for unconfined SCC has been used to develop equivalent rectangular stress block parameters for the flexural design of SCC beams. The flexural capacity predictions of the equivalent rectangular stress blocks were found to be in good agreement with the experimentally obtained moment capacities of simply supported SCC beams tested under four-point loading configuration. On the other hand, current design codes gave overly conservative moment capacity predictions. Confinement of the compression concrete is inevitable if effective use is to be made of large amounts of tension reinforcement which maybe well in excess of the requirements for balanced strain conditions. A unified constitutive model valid for a wide range of concrete strengths has been proposed for helically confined SCC by modifying the relevant parameter inAssa 's (Assa etal, 2001) constitutive model for confined CVC. In the absence of experimental results for confined SCC in the literature, validation of the proposed model for confined SCC could not be carried out. As an aid to flexural design of helically confined SCC beams, the proposed constitutive model for confined SCC has been used to develop equivalent rectangular stress block parameters expressed in terms of the degree of confinement. The flexural capacity predictions of the proposed stress block parameters for confined SCC were found to agree well with experimental results where as in accurate predictions were obtained on the basis of stress-strain models for confined CVC. Due to the effect of confinement, an extended post-peak response was obtained in the conventionally over-reinforced SCC beams whose load-deformation response has been 11 characterised in terms of the parameter 'ductility ratio', lir, which is so calibrated as to yield a value of zero for brittle post-peak behavior and a value of 1.0 for an ideal elasto-plastic response. Though none of the confined SCC beams had a lir value of 1.0, the beams with high degrees of confinement had lir values in the vicinity of 0.75. The limiting percentage of tension reinforcement in the helically confined SCC beams has been shown to be a function of the degree of confinement. It has been shown that theoretically, any amount of tension reinforcement consistent with detailing requirements can be used in the helically confined SCC beams. The serviceability behavior of the relatively slender SCC beams containing large amounts of tension reinforcement has been examined in terms of the midspan deflections and maximum surface crack widths at service loads. All the experimental beams were seen to have deflections less than span/250. Compared to the British code, BS8110, more accurate predictions of the mid-span deflections were obtained by using the recommendations of the Indian code IS 456:2000. The maximum surface crack widths of the SCC beams were measured to be less than the limiting value of 0.3 mm for normal condition of exposure. Accurate crack width predictions for the SCC beams of this investigation and those of Sonebi et al. (2003) were obtained by incorporating a modification parameter in the crack width equation given in BS8110. Because of the relatively lower coarse aggregate content in SCC vis-a-vis CVC, there is concern that the shear strength of the former may be significantly lower than the latter. Shear strength studies were carried out by testing simply supported longitudinally reinforced SCC beams under a four-point loading configuration. The parameters in the shear strength experimentation were : (i) grade of concrete (ii) amount of tension reinforcement and (iii) shear span-to-effective depth ratio, a/d. Elaborate instrumentation was used to distinguish between the ultimate shear load and the shear at diagonal cracking. An evaluation of the experimental results indicate that the shear strength equations in the ACI as well as the IS code give overly conservative predictions for the SCC beams though better accuracies in predictions were obtained from the shear strength equation of Zsutty (1971). In the absence of comparable data in the literature, the limited experimental shear data of this investigation has been used to propose equations for the cracking and ultimate shear strengths of the SCC beams in terms of the parameters fc', p, and a/d ratio. The proposed equations have been used for a predictive assessment of the shear capacities of the SCC beams reported by Haassan et al. (2008) A poor match between the experimental and predicted values of the cracking shear strength was observed though the ultimate shear strengths of the relatively lighter beams of Haasan were predicted with greater accuracy. in