External edge restraint in reinforced concrete (RC) elements arises when naturally
occurring free strain deformations are prevented from occurring due to the presence of a
more mature, stiffer restraining member, upon which the new element is cast. The restraint of the free strain (namely shrinkage and thermal) has the potential to generate detrimental cracking in the restrained element, if the tensile capacity of the concrete is exceeded. This may occur at early age, following large thermal strain generated during hydration, or later in the structure lifetime, as shrinkage in the element is prevented. The importance of a comprehensive understanding of external edge restraint in RC walls is two-fold. First, a comprehensive understanding of the restraint mechanisms will prevent excessive and/or unexpected cracking occurring, which will likely lead to deterioration of the structure.
Secondly, without comprehensive understanding of both material and structural
behaviour, design priorities, such as targeted reinforcement and reduced material usage,
for improved sustainability, cannot be implemented to their potential. The need for further research in this area is clearly demonstrated by a critical review of
the literature. Crack width predictions from widely used guidance documents (BSEN
1992 and CIRIA C766) do not align with what is observed experimentally or in practise.
The literature shows that the current understanding of external edge restraint
fundamentally does not reflect the true behaviour. Recently, it has been proposed that the role of the restraining member is critical in the interactions occurring in the edge restraint system – a previously ignored influence when partial scale tests and steel restraining members were used to develop external edge restraint understanding. This research project furthers the understanding of the restraint mechanisms present in external edge restrained RC walls, through a two stage experimental investigation.
Full-scale RC walls cast onto mature RC base slabs experimentally quantified the restraint
of imposed strains present in RC walls. The experimental test set up encouraged the
development of external edge restraint, and the reinforcement was varied parametrically.
The results illustrated the importance of considering a ‘global experimental restraint
factor’ to represent the true restraint occurring in the RC walls. The results showed the
current design guidance assumption that the restraint is solely a function of the relative
stiffness ratio between the restrained and restraining members is an oversimplified
concept. The external edge restraint magnitude is significantly affected by the movement
induced in the base slab due to the casting of the wall. Experimental monitoring of the
restraining member highlights the two way nature of the interaction across the joint, with
the direction of the interaction primarily being a function of the age of the base slab at the time of wall casting.
The experimental cracking patterns in the edge restrained walls highlighted the
importance of considering the internal restraint of the shrinkage, due to the reinforcement. An experimental programme was conducted to quantify the contribution of reinforcement restrained shrinkage with reinforcement diameter and cover distance. Findings demonstrated a single 20mm diameter reinforcing bar provides up to 62% restraint to the free shrinkage at a 30mm cover distance. When a closely spaced reinforcement grid is included in a larger structural element, results showed that (a) the reinforcement restraint between two reinforcement bars is superimposed, and (b) the restraining influence produced by a reinforcement grid is distributed throughout the element and observed as a global effect. With reinforcement restrained shrinkage offering a significant contribution to the interactions occurring in external edge restrained walls, the influence this has on the tensile capacity of the element to resist external edge restraint, and how the restraint induced cracking is affected must be better considered by the guidance documents. Again, the importance of considering a ‘global experimental restraint factor’ is shown.
