Page 112 - Rock Mechanics For Underground Mining
P. 112

ROCK STRENGTH AND DEFORMABILITY

              Figure 4.8 Principle of closed-loop
              control (after Hudson et al., 1972b).































                                          The essential features of closed-loop servocontrol are illustrated in Figure 4.8.
                                        An experimental variable (a force, pressure, displacement or strain component) is
                                        programmed to vary in a predetermined manner, generally monotonically increasing
                                        with time. The measured and programmed values are compared electronically several
                                        thousands of times a second, and a servo valve adjusts the pressure within the actuator
                                        to produce the desired equivalence.
                                          Modern servocontrolled testing systems are used to conduct a wide variety of tests
                                        in rock mechanics laboratories. The key to the successful use of these systems is the
                                        choice of the control variable. The basic choice is between a force (or pressure) and a
                                        displacement (or strain) component. Figure 4.9 shows why it is not feasible to obtain
                                        the complete uniaxial force–displacement curve for a strain-softening specimen by
                                        programming the axial force to increase monotonically with time. When the peak
                                        strength of the specimen is reached, the program will attempt to continue to increase
                                        the axial force, but the load-carrying capacity can only decrease with further axial
                                        displacement. However, the test can be successfully controlled by programming the
                                        axial displacement to increase monotonically with time.
                                          The post-peak portions of the force–displacement curves obtained in compression
                                        tests on some rocks may be steeper than, or not as smooth as, those shown in Figures
                                        4.7 and 4.9. In these cases, better control can be obtained by using the circumferential
                                        displacement rather than the axial displacement as the control variable. Figure 4.10
                                        shows the complete axial stress (  a )–axial strain (ε a ) and circumferential (or radial)
                                        strain (ε r )–axial strain curves obtained in such a test on a 50 mm diameter by 100 mm
              Figure 4.9 Choice between force
                                        long specimen of an oolitic limestone (Portland stone) in which a wrap-around trans-
              and displacement as the programmed
                                        ducer was used to monitor circumferential displacement. Although the possibility of
              control variable (after Hudson et al.,
              1972a).                   extracting energy from the machine–specimen system offered by this technique is
                                        94
   107   108   109   110   111   112   113   114   115   116   117