47 In-situ shear tests
47.1 General principles of the direct shear test
The direct shear test should be designed to measure the peak shear strength of the in-situ material as a function of the stress normal to the sheared plane (see ISRM suggested method ).
NOTE 1 More than one test is normally required to obtain a sensible design value. The measurement of residual shear strength presents major practical problems in arranging for a sufficiently long travel, but a useful indication of residual strength can be obtained by continuing the test to the limits of travel of the apparatus. In certain applications, the test may be designed to establish the strength of the interface between concrete and rock or soil.
For this test, a sample of ground, prepared and tested in situ, should be subjected to direct shear, using a stress system similar to that of the laboratory shear box tests (see Figure 16). The samples should be selected to include one or more discontinuities, if this is what is to be tested. The orientation of the discontinuities should be selected as relevant to the stress conditions being considered. The maximum sample size is often limited by practical considerations of loading and accessibility.
The orientation of the sample and the forces applied to it should be governed by the direction of the forces that become effective during and on completion of the works, but should be modified to take account of the orientation of significant discontinuities. In many cases, however, to facilitate the setting-up of the test, the sample should be prepared with the shear plane horizontal. The normal and shearing stresses should be imposed as forces applied normally and along the shear plane. However, an inclined shear force passing through the centre of the shear plane may be used, as this tends to produce a more uniform distribution of stress on the shear surface (see ISRM suggested method ).
NOTE 2 Field shear testing of intact soil might be necessary sometimes. Although it is theoretically possible to carry these out in the consolidated, unconsolidated drained or undrained state, in practice it is not usually possible to prevent some drainage. Such testing might be used on weaker rocks.
|1 Reaction system||N1 Normal deflection gauges|
|2 Jack||S1 Shear deflection gauges|
|3 Load distributor||L1 Lateral deflection gauges|
|4 Low friction rollers||P Pressure gauges|
|5 Shear box or other load distributing system||T Shearing force|
|6 Spherical seat||N Normal force|
|7 Datum reference||M Load maintenance|
47.2 Limitations of in-situ shear test
COMMENTARY ON 47.2.1
Samples are normally prepared at the bottom of pits or trenches in soil. Adits are more common for rock testing. The excavation permits access to the material at the zone of interest and, in many cases, provides a suitable means of setting up the reaction for the applied forces.
As a rough guide, the sample dimension should be at least ten times that of the largest particle; in rock, the sample size should reflect the roughness of the rock discontinuity being tested. For stronger rocks, the sample can be rendered with suitably strong cement and reinforced concrete to ensure adequate load distribution. The equipment should be of robust construction. Samples between 600 mm2 and 1 500 mm2 have been used for testing soil and weak rocks. Larger samples might be required in ground containing boulders or in compacted fill material.
Great care should be exercised in preserving the environmental conditions when carrying out the excavation. Excavation techniques that give rise to crumbling, fracturing or excessive dynamic shock loading, which would affect the discontinuities in the sample test area, should be avoided. Hand sawing, cutting and diamond drilling should be used to prepare and trim the sample. Adequate protection from the weather should be provided. Final exposure and trimming of the sample to fit the loading frame and the testing should all be completed with the minimum of delay to avoid possible significant changes in the moisture and stress conditions of the sample. Where tests are carried out below the water table, precautions should be taken to avoid the effects of water pressure and seepage.
Where it is intended to test one discontinuity only, care should be taken to avoid disturbance to the surface of the discontinuity and to prepare the sample so that the forces are applied in the plane of the discontinuity in the manner intended. The spatial orientation of the discontinuity should be defined by its dip and strike.
Where drained conditions are required, suitable drainage layers should be inserted around the sample and on the loaded upper surface.
47.2.2 Test arrangement
A typical test layout for determining the peak shear strength is shown in Figure 16. In addition, a porous top plate or other suitable medium should be used to distribute the load where drained conditions are required. The alignment of the force should be maintained during the test.
If a constant normal load is required for this type or test, a suitable reduction should be made from the applied normal load during testing, to compensate for the increase in the vertical component with increasing shear force. The shear force application should be developed by similar means to the normal loading. In both cases, care should be taken to ensure that the ground reaction does not extend to the sample. The reaction system can frequently be provided by the excavation sidewalls. In certain cases, it might be necessary to provide the shear force by traction on a system anchored by piles or anchored cable. Sufficient travel should be provided to run the complete test.
47.2.3 Method of carrying out the test
The forces used for the testing programme should be in the range of the working stresses to be applied by the structure. The peak force and an estimate of the residual direct shear strengths should also be determined to establish a factor of safety.
When testing a single fissure or joint, care should be taken to establish the initial slope of the fitted line at the lower normal stresses.
NOTE Sometimes, low values occur at lower normal stresses until the asperities on the joint surface interlock. Photographs of the shear surface form a useful record.
Extrapolation of results obtained on single joints should not be attempted without due confirmation that the joint surface tested is representative of the planeness and roughness of the joints in the mass (see Section 6). If not, appropriate calculations for extrapolation can be used (see Patton, 1966 , Ladanyi and Archambault, 1970  and Barton, 1971 ).
Where failure occurs in a plane that dips at an angle to the applied shearing force, the analysis given in Bishop and Little, 1967  may be used.
On completion of the test, full identification of the material and that immediately surrounding the sample should be carried out by sampling, visual examination and laboratory testing. Tests on relatively small samples of single joints might give useful values of angle of shearing resistance, but the cohesion parameter tends to be size dependent.