Long Term Behaviour of a Fully Instrumented Test Room in Potash

INTRODUCTION In 1982 and 1983, as part of the mine design program for a New Brunswick potash mine, series of laboratory tests were conducted on potash and salt samples obtained during the mine exploration program. Concurrently, two rock mechanics monitoring programs were initiated in-situ the short term test at site A, and the long term test at site B (figures 1 and 2). SHORT TERM TEST (FIGURE 1) The purpose of this test was to obtain a quick confirmation of laboratory data and comparison of the rock response to excavation with the data available from Saskatchewan potash mines. Instrumentation in the main drift was installed on Day 4 after excavation, and the pressure cells were pressurized on Day 33. The packer type cells in the east crosscut were pressurized on Day 25. Excavation of the west cross-cut was begun on Day 71. This "mine by" test has provided important information required for calibration of laboratory tests, as will be shown later. The test data recorded are shown in a series of diagrams: a. initial room closures, main drift-figure 3; b. convergence graphs, main drift-figure 4; c. stresses recorded, main drift-figures 5 to 8 and, d. stresses recorded, east crosscut - figure15 Three types of instruments were used in this test. In the main drift. Saskatchewan Research Council water type hydraulic inclusions. encapsulated in potash, were used to measure lateral and vertical stresses. In the east crosscut a home-made packer type hydraulic inclusion was used to record radial stresses. The convergence stations consisted of end-grouted bars of reinforcing steel. LONG TERM TEST (FIGURE 2) The purpose of the long term test was to study the steady-state behaviour of a single opening in potash. Site B was placed in a remote area of the mine, well away from any anticipated early mining activities. To assure homogeneity of rock mass, the site was placed within a 30 meter thick vertical limb of a syncline-shaped ore body contained within a 360 metre thick vertical section of salt rocks. It was excavated mechanically as close as possible to the designed square cross-section for the entire length of the room. The square cross- section was chosen in order to achieve a comparable geometry in vertical and horizontal directions; that is, to eliminate any undesired effects of geometry interaction with lithostratigraphy which might have generated non-homogeneous behaviour. The site was extensively instrumented. As in site A, the convergence stations consisted of end-grouted bars of reinforcing steel. However, in order to avoid any measurement inaccuracies which would be associated with the brittle yield zone in the opening walls, bars were placed to the same depth in opposite sides of the opening. This allowed the convergence measurement to be obtained as the absolute convergence between two points, rather than the convergence of a single point with respect to a potentially unstable reference. For stress measurements the Saskatchewan Research Council cells were placed vertically, laterally and radially. Unfortunately, due to cost considerations, it was decided to use a compacted sand, rather than grout, to encapsulate the cells in the boreholes. Due to the extremely small strain rates, only radial ly placed cells could be adequately pressurized. The rest of the cells did not perform adequately, and are currently being replaced.