ABSTRACT: Fully instrumented laboratory overcoring tests have been made to compare the USBM borehole deformation gage and the Australian CSIRO hollow inclusion cell under simulated field conditions with known applied stresses. The errors are less than 10% for both in isotropic, homogeneous and continuous rocks, but neither cell works reliably in thinly bedded or highly fractured rock. Both are affected by bedded and jointed rock with the errors being roughly equal but in the opposite sense so that a combined reading was reasonably correct. An explanation is given for this behaviour. The CSIRO cell has a greater flexibility of use in poor conditions but care must be taken with bonding and temperature compensation. The USBM gage is simpler but one instrument cannot give the complete stress state.
1 INTRODUCTION The USBM borehole deformation gage (USBM gage) and the CSIRO hollow inclusion cell (CSIRO cell) are two commonly used in-situ stress measurement devices (Merrill, 1967; Worotnicki and Walton, 1976). Both of them utilize the overcoring stress relief principle but adopt different sensing techniques. The interpretation of data obtained with the devices is based on assumptions that the rock is continuous, isotropic, homogeneous and linear-elastic, which may result in the calculation of incorrect stress values in practical rock conditions. Frequent instrumentation failure has indicated that each technique is subject to certain limitations of rock types and conditions. The choice of the technique and the assessment of validity of the measurements usually rely on personal preference and experience because no objective information exists for the choice and the assessment. Although some comparative work has been conducted in the field to investigate the performance of different techniques in different conditions, such comparisons were made between unknowns because the real field stress is unknown, and that reduces the value of the comparison work. This paper presents a simple and practical approach to investigate the performance of overcoring stress measurement devices in different rock conditions, which allows an objective comparison to be made between the measured results and the known stress values. Some improvements and corrections have been made to increase the accuracy of the measurement with the USBM gage and the CSIRO cell. More than fifty overcoring tests have been completed with the two devices. The results of the tests are sufficient to make an assessment of the reliability and suitability of the two devices in various rock types and conditions. A list of recommendations are provided on an appropriate choice of devices and on how to adjust the measured results to allow for real rock conditions [Cai and Blackwood, 1987: Cai, 1990].
2 EXPERIMENTAL METHOD To simulate in situ stress, blocks of rock were subjected to biaxial pressure by two hydraulic rams (Figure 1). While pressure was applied, a 38 mm borehole was drilled and a stress device was installed and overcored with a 150 mm thin-walled bit. From the measured changes, stresses were calculated and compared with the actual applied values to evaluate the performance of the device.