AN EVALUATION OF TEST DATA FOR SELECTING A YIELD CRITERION FOR SOILS WITH DISCUSSION

Abstract

The existing contradictory evidence supporting various failure criteria for soils is briefly summarized. The results of triaxial extension and compression tests are presented, in which an attempt is made to measure the variations of strain throughout the specimens. While the overall average behavior of the specimens appears to favor the Mohr-Coulomb criterion, more detailed investigation suggests that the extended von Mises, or possibly the extended Tresca, criteria are more relevant. Discussion: B. Ladamyi contends that there is a danger in making any conclusions based on the comparison between the two types of triaxial tests (drained compression tests and drained extension tests). The danger is that the two tests will differ from each other for the double value of the stress difference mobilized during isotropic consolidation. Ladamyi contends this phenomenon may be clearly identified in the results of some triaxial compression and extension tests published by R.H.G. Parry. W.B. Truesdale and R.W. Rusin state that in an effort to study the problem of strain distribution a gage was developed. It is used to measure small increments of displacement over a short gage length in a triaxial soil specimen undergoing deformation at a uniform rate. On the basis of the tests using the gage the following remakrs were made: (1) nonuniform straining is most prominent at low confining pressures; (2) strain varies both longitudinally and radically throughout the specimens; (3) in the Ottawa sand specimens, greatest straining occurred near the specimen center at mid-height; (4) in the ball clay specimens, the greatest staining occurred near the specimen ends.