ABSTRACT
This paper investigates the effect of varying stress levels on the determination of deformation moduli and examines the sensitivity of the borehole dilatometer in measuring rock mass scale features and its ability to represent rock mass characteristics. Cubical blocks of sandstone, limestone and granite were tested in the laboratory under triaxial stress conditions. Field programs were conducted in both granite and sandstone rock masses surrounding underground openings. The dilatometer's zone of influence was determined by conducting tests within an instrumented concrete cylinder and by performing a finite element analysis. Laboratory and field results suggest that the deformation modulus of a rock mass can be substantially affected by the stress field and that the borehole dilatometer is sensitive enough to profile deformation moduli as dictated by the in situ rock mass characteristics.
INTRODUCTION
Several authors, for example Boughton (1968), Howarth (1984) and Patrick et al. (1985) have indicated that the stress field influences the deformation moduli obtained during load testing programs in rock. These observations were also noted by the authors in previous modulus testing projects conducted in various rock types (Koopmans, 1983; Koopmans and Hughes, 1985). It would be reasonable to assume that an increase in stress would result in a more compact rock mass of higher stiffness due to the closure of microcracks and discontinuities. On the contrary, the opposite effect would be encountered at lower stress levels. This study investigates the effect of varying stress levels on the determination of deformation moduli in rock. It also examines the sensitivity of the borehole dilatometer in measuring rock mass scale features and its ability to truly represent rock mass characteristics. It has been reported that borehole devices, such as dilatometers are relatively insensitive to rock mass scale features and provide results less characteristic of the true rock mass modulus because of the small volume of rock influenced in each measurement (Schrauf and Pratt, 1979). On the contrary, Heuze (1980) suggests that borehole devices such as the NX borehole jack influence the rock mass some seven borehole diameters from the loading plate. This represents a rock mass volume which is more than 500 times greater than that of a standard NX core sample. Laboratory dilatometer tests were conducted in cubical blocks of sandstone, limestone and granite under triaxial stress conditions. Field programs were undertaken in both granite and sandstone. In Manitoba, testing was carried out in the granitic rock mass surrounding the access shaft of an underground research laboratory. In Nova Scotia, testing was conducted within a sandstone rock mass located between two parallel circular tunnels. A computerized borehole dilatometer system was used for all testing programs.
COMPUTERIZED BOREHOLE DILATOMETER SYSTEM
The computerized borehole dilatometer system (Figure 1) is highlighted by an intensifier which is operated by an electrically-driven hydraulic pump to allow controlled pressure loading and unloading of the dilatometer. The intensifier has a 6:1 piston differential ratio between the oil and water chambers, respectively. This allows lower oil pressures to be administered from the hydraulic pump through flexible hoses to the intensifier where the pressure is increased six times to allow water inflation of the dilatometer.