Experimental and numerical investigation of shear strength at dapped end beams having different shear span and recess corner length

Abstract

The geometric properties of the support region of dapped end beams (DEBs) cause a significant change in the ratio of the distance between the support and suspension reinforcement (a) to the effective depth (d). This ratio is effective in the shear capacity of DEB. In addition, in the ratio of the distance of the load on the beam from the support (av) to the depth of the beam (deff) is also effective on the behavior of the beam under vertical loads. For the first time in the literature, the combined effects of these two ratios on the shear capacity of DEBs was investigated by experimental and numerical methods in this study. Firstly, an experimental study was carried out having 1/1 cross-sectional scale six different DEBs. In the experimental study, while the a/d ratio was kept constant as 1.8, the av/deff ratio was varied between 0.75 and 3. The experiments results indicate that towards increasing the av/deff ratio, the DEBs failed by flexure. The strength of DEBs decreases with an increase in the av/deff ratio. With smaller av/deff, the DEBs show greater stiffness and ultimate load. Numeric finite element models by using ABAQUS were verified by the load–displacement relationships and damage modes obtained from these experiments. After that, at the parametric stage of the study numeric models were also constructed in order to further investigate the effect of different a/d ratios and av/deff on the strength of DEBs. According to all test and numeric results, it is recommended that DEBs should be placed on the supporting member by considering the minimum a/d ratio. Otherwise, the capacity of the DEBs will be significantly reduced since the damage in the thinned zone will occur at lower load levels. At the end of study, considering the combined effects of a/d and av/deff, a shear capacity reduction coefficient based on these parameters has been proposed. The shear capacity, which is reduced using the proposed coefficient, approaches the numerical results, which converge to a great extent with the experimental data by 92.46%.