Improved Peak Dilatancy Angle for Shear Strength Criteria of Sawtooth Rock Joints

The shear strength criterion for sawtooth joints, which considers only the shear contribution of inclination angle and ignores the effect of height difference, requires improvement. Direct shear tests based on Bahaddini’s numerical model were conducted on sawtooth joints with the same apparent inclination angle but different undulating heights. A comparative analysis of the measured peak shear strength with the predictions from typical shear strength criteria [Patton and Ladanyi and Archambault (L&A) criteria] revealed that these criteria actually estimate the shear strength of the sawtooth joint at the maximum undulating height corresponding to the apparent inclination angle. Then, an average slope angle ih was proposed to characterize the roughness differences between joints at the maximum undulating height and those below the maximum undulating height. Subtracting ih from the apparent inclination angle provided an improved peak dilatancy angle for the shear strength criterion of sawtooth rock joints. Validation results revealed that the average prediction error decreases from 145% to 21% for the Patton criterion and from 29% to 11% for the L&A criterion. To verify the revised criteria, 48 groups of sawtooth joints were tested in the laboratory. The results showed that the accuracy improved by 47.41% for the Patton criterion and 51.92% for the L&A criterion. Additionally, 22 published shear test data sets of sawtooth joints were collected to further verify the revised criteria. It was observed that the revised criteria have a better prediction performance, with root-mean-square errors of 2.036 and 0.927 for the revised Patton and L&A criteria, respectively.