The significance of displacement control mode in direct shear tests of rock joints

Servo-controlled direct shear tests of rough (RMS = 500 μm) and smooth (RMS = 7 μm) surfaces of Gabbro are performed under a constant normal stress of 5 MPa and a constant shear displacement rate of 0.01 mm/s. The tested interfaces are 10 cm long and 8 cm wide. Two servo-controlled shear displacement modes are studied: 1) Piston Displacement Control [PDC], and 2) Block Displacement Control [BDC]. In PDC mode the displacement output from the shear piston is used as the feedback signal for servo control whereas in BDC mode the displacement output from two horizontally oriented LVDT's positioned directly on the sheared interface are used for control. We find that when the surfaces are tested under BDC mode they exhibit a distinct peak shear stress followed by a stress drop to residual shear strength, whereas a continuous transition to “steady state” sliding is exhibited by the same surfaces when tested under PDC mode. This leads to the important observation that while peak shear strength is similar in both control modes the residual shear strength in PDC mode is distinctly higher. All tested interfaces exhibit stick-slip oscillations, the magnitude and frequency of which are strongly related to the control mode. We show that stick slip amplitude and frequency are artificially higher in BDC mode because of servo-control effects and that the correct stick slip behaviour can only be obtained in PDC mode. We find that the displacement control mode in direct shear tests drastically affects the obtained results and therefore investigators must consider the consequences before specifying which servo-control mode to use in direct shear tests.