Size effect and boundary conditions in the Brazilian test: Experimental verification

Abstract

The effect of both size specimen and boundary conditions on the splitting tensile strength, determined from the Brazilian test, were studied experimentally. A total of 110 splitting tests of granite and mortar specimens were performed, using cylindrical and prismatic specimens of sizes between 17 mm and 300 mm. To analyze the effect of the boundary conditions, the specimens were tested with different widths of load bearing strip in the range of size recommended by the standards. The influence of the rupture mode (stable or unstable crack propagation) on the splitting tensile strength was also explored. The results of the tests were compared with the theoretical predictions obtained from a closed form analytical expression based on the cohesive crack model. The validity of the classical limit strength theory for larger size specimens was also analyzed. The results indicate that the splitting tensile strength depends strongly on specimen size and on the boundary conditions of the test. As the size of the specimen increases and the relative width of the bearing strip decreases, the splitting strength tends asymptotically to the minimum value coincident with the tensile strength. The dependence of the Brazilian test on the specimen size and boundary conditions closely follows the predictions of the cohesive crack model.