Modelling of Shear Behaviour of Reinforced Concrete Members without Shear Reinforcement

Abstract

Reinforced concrete members without shear reinforcement are still commonly encountered in civil engineering practice, particularly in existing structures and in elements where shear reinforcement is difficult or impractical to arrange. In such members, shear failure occurs in a brittle manner, with limited deformation capacity and little warning. Although many models have been proposed in the literature for the shear behaviour of reinforced concrete members without shear reinforcement, a generally accepted mechanical model for brittle flexural-shear failure has not yet been established.
A major difficulty is that flexural-shear failure is not governed by isolated mechanisms, but by the interaction among several complex mechanisms, including crack propagation, aggregate interlock, dowel action, residual tensile strength and shear stress transfer through the uncracked compression zone. Existing models remain limited in two respects. First, many models do not provide a comprehensive framework in which these mechanisms and their interactions are consistently represented. Instead, they often select one mechanism as the governing mechanism and introduce a calibrated failure criterion based on experimental databases. Second, the available mechanical models for individual shear transfer mechanisms are not always suitable for direct integration into such a comprehensive framework, because they may rely on simplified assumptions or may not be formulated in terms of crack kinematics. As a result, the reliability of these models becomes uncertain when they are applied beyond the range covered by the databases on which they were developed.
The main objectives of this dissertation are twofold. The first objective is to refine the models of selected shear transfer mechanisms to enable their consistent integration into a discrete crack-based mechanical framework. The second objective is to develop a calibration-free mechanical framework that provides a rational basis for assessing shear capacity beyond the current experimental range. In this framework, the critical shear crack is explicitly represented, and the contributions of different shear transfer mechanisms are evaluated along this crack.
The selected shear transfer mechanisms re-examined in this dissertation are dowel action, aggregate interlock and contribution from the uncracked compression zone. A new dowel action model is developed by combining the Beam on Elastic Foundation Theory with concrete fracture mechanics. This model can predict a complete force-displacement relationship for dowel action, whereas most models in the literature are mainly limited to predicting the ultimate dowel capacity. A 3D scanning-based extension of the Two-Phase Model is developed for aggregate interlock by incorporating the real morphology of fractured crack surfaces. This extension provides a systematic framework for quantifying the influence of crack surface roughness on aggregate interlock, whereas currently available models generally rely on empirical factors to account for this influence. The evaluation of the contribution of the uncracked compression zone is further refined by adopting a revised shear stress distribution along the beam depth, motivated by the aggregate interlock stress distribution obtained in the cracked region.
The dissertation further proposes and validates a simplified shear crack model based on comprehensive experimental observations of crack patterns using the author’s database and data reported in the literature. Unlike existing simplified shear crack models, the proposed model considers the propagation of secondary cracks in the compression zone and treats their length as an unknown variable. This crack model, together with the proposed dowel action model, is then integrated into the final mechanical framework for flexural-shear failure. In the proposed framework, moment equilibrium and transverse force equilibrium are first satisfied. The shear capacity is then obtained by minimising the residual in the longitudinal force equilibrium. In this manner, the shear capacity is determined through a conventional limit-state analysis that satisfies three predefined limit-state conditions. Without empirical calibration factors, the calculated shear capacities show good agreement with an extensive shear beam database. Based on these results, a simplified method using the critical shear displacement is further developed to improve the applicability of the proposed model.
Overall, this dissertation contributes to a more rational mechanical understanding of brittle flexural-shear failure in reinforced concrete members without shear reinforcement through the following aspects:
• Development of a mechanical model for dowel action that provides the complete force-displacement relationship and accounts for the development of splitting cracking along the longitudinal reinforcement. (Chapter 3)
• Extension of the Two-Phase Model that incorporates the real morphology of fractured crack surfaces from 3D scanning data, which provides a basis for applying aggregate interlock modelling to concretes with different fracture characteristics. (Chapter 4)
• Development and validation of a simplified shear crack model that captures the main geometric and kinematic features of flexural-shear cracks based on experimental observations. (Chapter 5)
• Development of a calibration-free mechanical model, together with a simplified method based on the critical shear displacement, for predicting the shear capacity of reinforced concrete beams without shear reinforcement. (Chapter 6)