The effect of surface roughness and a steel anchor on the direct-shear bond between existing concrete and new concrete

Abstract

The strengthening of concrete structures must consider the transfer of various forces at the joint, especially the direct-shear bond between existing concrete and new concrete, which is dependent on several variables. This research focused on the investigation of the direct-shear bond at the joint of concrete samples having a cylindrical shape with a diameter of 150 mm and a length of 300 mm. The joint of concrete was in the middle of the sample and was prepared by casting new concrete on the existing concrete. A steel anchor was installed in the middle of the joint with an embedded length of 75 mm. The studied variables included the surface roughness of the existing concrete, the size of the anchor, and the use of steel fibers in the new concrete. It was found from the tests that the preparation of the existing concrete surface to be rougher, which was less than the minimum roughness specified by various codes, could increase the direct-shear bond by 210–260 % and the displacement of concrete samples by 140–200 % compared to an as-cast surface. The larger the size of the anchor, the greater the cohesion bond and the first-cracking bond, but these bonds would not follow the mechanical effect of the larger cross-sectional area. The use of steel fibers in the new concrete hardly increased the direct-shear bond and displacement. Furthermore, the results of statistical data analysis showed that the variables significantly influencing the direct-shear bond at a 95 % confidence level included the roughness of the existing concrete surface and the installation of a steel anchor. Very good agreement was observed between the test results and the predicted equations for cohesion bond and first-cracking bond, with R2 values of 0.9284 and 0.8948, respectively. When comparing the equations provided by the various codes for determining interface shear resistance, the values obtained from ASSTHO and Eurocode 2 codes were closer to test results than those of the ACI 318 code because both codes have considered both cohesion and friction at the interface.