In Situ Measurements Of Stress Change Induced By Thermal Load: A Case History In Granitic Rock

Vibrating wire stressmeters (VWS's) and borehole deformation gages (BDG) were deployed in two in-situ heater experiments at Stripa, Sweden to determine the thermally induced stress changes in the rock mass. The heater experiments ran for 18 months, including a six- month cool-down period after the heaters were turned off. Gages were installed in both horizontal and vertical boreholes and were subjected to temperatures ranging from 10°C to 120°C, depending upon the distance of a gage from the central heater. Most gages operated in a moist environment because groundwater flowed into many of the instrument boreholes during the experiments Of the 36 VWS's installed, six failed. Failures were attributed to corrosion of the vibrating wire due to moisture entering the gage body. Of the 30 BDG's installed at the beginning of the experiments, 22 failed, also as the result of water entering the gage body. Modification and reinstallation of the gages eliminated the problem of water infiltration, but much data from early stages of the experiments was lost due to gage failures. Calibration of the VWS in blocks of Stripa granite showed a lack of repeatability. Variability in calibration results are attributed to seating effects, to variations in the elastic modulus of the rock, and to changes in temperature. These three effects combined to produce an error judged to be approximately ±33% of the stress changes measured at Stripa. In contrast, the major source of error in determining stress with the BDG appears to be the uncertainty in rock modulus and thermal expansion coefficient. Sample calculations based on data acquired near the end of the experiments (during cool-down)