It is pointed out that an ellipsoidal yield surface in principal stress space applicable to materials such as soil that undergo plastic volume changes has already been proposed, and complete equations including the definition of the effective plastic strain increment for applying this yield surface to anisotropic as well as isotropic materials has been published. One of the advantages of such a yield function is that the algebraic form of the associated flow rule is identical to that for the Von Mises yield function; any computer program that has been written for the Von Mises yield function should be adaptable to the ellipsoidal yield function. Another advantage is that the ellipsoidal yield function combines the shear and volume characteristics of the material, including both dilation and compaction into one yield function. In the process, the need to artificially change the value of Poisson's ratio to 1/2, for undrained conditions is eliminated. The necessary parameters for ellipsoidal yield function are obtainable from the results of standard tests. It is pointed out that the authors of the original paper have allowed the inclusion of strain softening, which implies "instability in the small" and which is prohibited in the basic derivation of the normality rule of plasticity. It is suggested that the authors elaborate further on how the problem is circumvented. The opinion is expressed that segmented yield surfaces cause undesirable mathematical complexities, although in the case of certain interesting but elementary problems, they can lead to the availability of closed-form solutions. It is recommended that consideration be given to applying the ellipsoidal yield function to soil mechanics problems.

  • Discussers:
    • Merkle, J G
  • Publication Date: 1975-10

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00127648
  • Record Type: Publication
  • Report/Paper Numbers: ASCE#11599 Proceeding
  • Files: TRIS
  • Created Date: Mar 10 1976 12:00AM