• Geomechanics and Engineering
• /
• 제12권6호
• /
• pp.883-898
• /
• 2017

To eliminate the holding and gluing problems making the direct tensile strength test hard to be applied, a new method of testing specimens prepared using lathe machine to make the dog bone shape is assessed whether it could be applied to determine accurate direct tensile strength values of rock materials. A series of numerical modelling analyses was performed using finite element method to investigate the effect of different specimen and steel holder geometries. In addition to numerical modelling study, a series of direct tensile strength tests was performed on three different groups of rock materials and a rock-like cemented material to compare the results with those obtained from the finite element analyses. A proper physical property of the lathed specimens was suggested and ideal failure of the dog bone shaped specimens was determined according to the results obtained from this study.

• Geomechanics and Engineering
• /
• 제24권4호
• /
• pp.349-358
• /
• 2021

The Brazilian test has been widely used to determine the indirect tensile strength of rock, concrete and other brittle materials. The basic assumption for the calculation formula of Brazilian tensile strength is that the elastic moduli of rock are the same both in tension and compression. However, the fact is that the elastic moduli in tension and compression of most rocks are different. Thus, the formula of Brazilian tensile strength under the assumption of isotropy is unreasonable. In the present study, we conducted Brazilian tests on flat disk-shaped rock specimens and attached strain gauges at the center of the disc to measure the strains of rock. A tension-compression bi-modular model is proposed to interpret the data of the Brazilian test. The relations between the principal strains, principal stresses and the ratio of the compressive modulus to tensile modulus at the disc center are established. Thus, the tensile and compressive moduli as well as the correct tensile strength can be estimated simultaneously by the new formulas. It is found that the tensile and compressive moduli obtained using these formulas were in well agreement with the values obtained from the direct tension and compression tests. The formulas deduced from the Brazilian test based on the assumption of isotropy overestimated the tensile strength and tensile modulus and underestimated the compressive modulus. This work provides a new methodology to estimate tensile strength and moduli of rock simultaneously considering tension-compression bi-modularity.