VARIATION OF ACOUSTIC EMISSION AND SHEAR WAVE VELOCITY OF LAVASAN GRANITE UNDER UNIAXIAL COMPRESSION TEST

In this paper the results of the study on variations of shear wave velocity (Vs) and acoustic emission (AE) before failure of rock samples is presented. Some laboratory tests were carried out on Lavasan granite specimens (45 mm diameter and 100 mm height) using a servo–controlled uniaxial compression device with a constant rate of displacement. Four transducers were used for monitoring of acoustic emissions and two other transducers for measuring Vs during loading. Shear wave velocity changes were monitored in directions perpendicular to the applied stress, and acoustic emission event-count data were recorded continuously until the failure of the rock. The data processing showed that Vs is increasing gradually during the loading up to about 85 percents of the ultimate compression rock strength and then decreases sharply or quit completely in some cases (no signal at the receiver). It seems that the decrease of Vs shows the development of micro-cracks in rock samples and the interruption of the Vs recording occurs because of the development a macro-crack. During loading phase the acoustic emission events have been also recorded continuously. Plotting of accumulated count and accumulated absolute energy versus time show the absolute energy that is a more effective way than the count for the prediction of catastrophic failure of specimens. Indeed, the AE count rate continuously rises during a test but the cumulative absolute energy sharply rises near the Vs decreasing point. A practical implication of the results is that an increased AE activity (count per second) should not necessarily indicate that the tested specimen is reaching to a critical state (the ultimate strength). The increase of AE count could be related to other AE sources such as dislocations and noise which can be activated during loading. Finally it is shown in this paper that the use of AE techniques and simultaneous measurements of the elastic wave changes could be a useful approach for predicting rock failure that could be used for earthquake prediction.