ABSTRACT: A simple index test inspired by Brinell hardness measurement has been designed. The test procedure is described and validated. A correlation between the measured hardness and the uniaxial compressive strength is established from data of several reservoir sandstones. Various applications of the test to petroleum engineering practice are presented.
1 INTRODUCTION In the oil industry, a shortage of core material is frequent for both technical and economical reasons. Yet, some problems, such as sand production prediction, often require a precise mechanical description of tens of metres of heterogeneous formations (Santarelli et al., 1989). This need to establish a mechanical core log related to the uniaxial compressive strength added to the constraint of having little material available leads to the use of index tests. The point load test (Broch and Franklin, 1972), the Schmidt Hammer test (ISRM, 1981), the shore scleroscope test (ISRM, 1981) or the Vickers hardness test (Das, 1974) were all rejected on the basis that the volume of material tested was too small or too large or that the index was not directly related to the uniaxial compressive strength. Brinell or Rockwell hardness tests were considered unsuitable because of the brittle nature of rocks (ISRM, 1981). However, Van der Vlis (1970) proposed a method to perform Brinell hardness tests which gave good correlations with rock deformation moduli whilst Geerstma (1985) showed that this index correlated well with the pressure collapse of hollow cylinders. The possibility of using it as a basis for core mechanical description was investigated.
2 TEST PROCEDURE The standard procedure to measure Brinell hardness on metals (AFNOR, 1980) consists in applying a given load on a flat surface and creating a spherical print by plastic deformation (Figure 1). The print surface area is measured and the Brinell hardness computed. For rocks, the print surface is not always well defined. Furthermore, the application of the load results not only in the formation of a small print but also in the creation of a subvertical tensile crack on each side of the print because of the large value of the ratio of the compressive to the tensile strength (sc/st) for rocks. In order to avoid this, two steps were included in the testing procedure. First the rock specimen is casted in epoxy resin which applies passive confinement during the test and delays the formation of the tensile crack around the print. Furthermore, the test is performed by increasing linearly the load and the ball displacement is monitored throughout. The load vs displacement curve is analysed and the hardness is determined by measuring the slope of the curve in its linear part (Figure 2). Standard specimen preparation and testing procedure are as follows.
1.A more or less circular disc is cut from a core (6 to 8 cm diameter by 3cm thickness).
2. The specimen is coated with a layer of viscous araldite glue and allowed to dry in order to avoid any penetration of the epoxy resin in the pores of the rock.