Size effect in RC deep beams: Experimental investigation and STM verification

An experimental program consisting of three groups of 11 specimens is carried out to investigate the possible causes of size effect, typically represented by a reduction in shear strength with an increase in the height of deep beams. It is well-known that deep beams behave very differently from shallow beams as arch action rather than flexure dominates the behaviour, after diagonal cracking has occurred. However, causes of size effect in deep beams remain unresolved. It is postulated that the factors influencing the strength of a compression member, i.e. the strut geometry and boundary conditions, govern the size effect. The experimental program provides experimental evidence for the hypotheses from strut-and-tie models. By properly configuring the loading and support plate dimensions, size effect in ultimate shear strength is significantly mitigated, even for beams with high h/b ratios and plain concrete web. Evenly-distributed web reinforcement is found to mitigate the size effect to some extent. Effects of out-of-plane actions on beam strength are also investigated and discussed. The modified STM incorporating the causes of size effect outperforms several other methods in accurately and consistently predicting the ultimate shear strengths.