The shear strength parameters for a particular soil can be determined by means of laboratory tests on specimens taken from representative samples of the in-situ soil.
Laboratory determination of shear parameters – Direct shear test
The shear strength parameters for a particular soil can be determined by means of laboratory tests on specimens taken from representative samples of the in-situ soil. Great care and judgment are required in the sampling operation and in the storage and handling of samples prior to testing, especially in the case of undisturbed samples where the object is to preserve the in-situ structure and water content of the soil. In the case of clays, test specimens may be obtained from tube or block samples, the latter normally being subjected to the least disturbance. Swelling of a clay specimen will occur due to the release of the in-situ total stresses. Shear strength test procedure is detailed in BS 1377 (Parts 7 and 8) .
The specimen is confined in a metal box (known as the shear box) of square or circular cross-section split horizontally at mid-height, a small clearance being maintained between the two halves of the box. Porous plates are placed below and on top of the specimen if it is fully or partially saturated to allow free drainage: if the specimen is dry, solid metal plates may be used. The essential features of the apparatus are shown diagrammatically in Figure. A vertical force (N) is applied to the specimen through a loading plate and shear stress is gradually applied on a horizontal plane by causing the two halves of the box to move relative to each other, the shear force (T) being measured together with the corresponding shear displacement (l). Normally, the change in thickness (h) of the specimen is also measured. If the initial thickness of the specimen is h0 then the shear strain can be represented by l/hr and the volumetric strain (v) by h/h. A number of specimens of the soil are tested, each under a different vertical force, and the value of shear stress at failure is plotted against the normal stress for each test. The shear strength parameters are then obtained from the best line fitting the plotted points.
The test suffers from several disadvantages, the main one being that drainage conditions cannot be controlled. As pore water pressure cannot be measured, only the total normal stress can be determined, although this is equal to the effective normal stress if the pore water pressure is zero. Only an approximation to the state of pure shear is produced in the specimen and shear stress on the failure plane is not uniform, failure occurring progressively from the edges towards the centre of the specimen. The area under the shear and vertical loads does not remain constant throughout the test. The advantages of the test are its simplicity and, in the case of sands, the ease of specimen preparation.