ABSTRACT: Experiments performed with a set of brittle sandstones indicate that failure envelopes determined by conventional triaxial and Multiple Failure State tests are only slightly different. Acoustic wave velocity measurements may be a valuable tool for detecting onset of failure in the multiple failure state test.
RESUME. Des experiences faites sur un ensemble de grès fraglles indiquent que les enveloppes de rupture determinees par des testes triaxiales classiques et par des testes multiple failure state sont très proches. Nous montrons que les mesures de vitesse des ondes acoustiques sont très utiles pour la detection du commencement de rupture dans les testes multiple failure state.
ZUSAMMENFASSUNG: Experimente an einer Serie von spröden Sandsteinen zeigen, daß sich die durch konventionelle Triaxialversuche und multiple failure state Versuche besti_ten Grenzfastigkeitskurven nur geringfuegig voneinander unterscheiden. Schallgeschwindigkeitsmessungen haben sich als eine wertvolles Mittel zur Feststellung des Bruchbeginns in multiple failure state Versuchen erwiesen.
1. INTRODUCTION Rock mechanical laboratory analyses form a basis for the petroleum industry in order to decide strategies for well completion and well stimulation. The amount of core material available from a specific zone in a reservoir may, however, often be very limited. In particular, in weak zones, core recovery may be poor. Expenses in coring and testing programmes may also be substantial if a complete failure envelope shall be established from a series of triaxial tests. About 20 core plugs are required for establishing, within statistical confidence, a complete failure envelope from traditional triaxial tests.
ISRM has proposed (Kovari et al., 1983) an alternative technique,the which allows the use of one single core plug or determination of an entire failure envelope. In this test procedure, which is described in more detail below (Section 2), the sample is brought close to failure repeatedly by increasing the confining pressure stepwise. We shall subsequently (Section 3) present results of experiments performed to compare the two methods in the case of relatively weak sandstones from reservoirs and from outcrops. Acoustic velocity measurements have been performed during the tests to guide the interpretations. In Section 4, the results are discussed, with focus on the damage caused by inducing failure in one of the steps of the test, and the possible consequences this may have for the remaining parts of the test. The conclusions are then drawn in Section 5.
2. EXPERIMENTAL TEST PROCEDURE A cylindrically shaped core plug (here: 1½" diameter; α 3" length) is placed in a standard triaxial cell. The sample is embedded within a plastic sleeve and surrounded by a confining oil. The confining pressure is applied through the oil, and can be controlled independently from the axial load. An example of a multiple failure test is shown in Figure 1. The test consists of a series of segments. In the first segment, the stress state is first brought to a predetermined hydrostatic level (in this case 2.5 MPa). Then the axial load is increased, until the sample yields (i.e. the stress-strain curve starts bending over).