General Mechanical

General Mechanical

Topics relate to Mechanical Enterprise, Motion, Additive Print and more

Non Convergence issue for Mohr colulomb Soil model for a cohesionless soil

    • Srinjoy Gupta

      Hi, I have developed a soil block of volume 8mx8mx8m in ansys workbench static structural with following engineering properties

      Now after Meshing and providing Boundary condition by restraining the block in the X and Y direction and applying a fixed support in the bottom (pic attached below), I am applying gravity load but because I want the soil to be already in a consolidated state hence it should not deform upon appplying gravity but since it is showing significant deformation. To prevent this i'm writing an insitate command snippet as following:

      ! SET AN INITIAL Overburden stress which INCREASES WITH DEPTH
      ! INISTATE, DEFINE, ID, EINT, , --, FuncName, C1, C2, ..., C12

      But once I apply this my model no longer converges but it does so for the Engineering Property where Cohesion value in Mohr-Coulomb property is significant. Please could you guide me how to solve the error. Also for a multilayered soil deposit the inistate comand snippet that I wrote wont suffice in that instance what procedure should I follow if want deformation to be nil.


    • John Doyle
      Ansys Employee

      If I understand the definition of cohesionless soil, it is like dry sand with no resistence to shear sliding.   Is that what you are trying to simulate?  This sounds like a recipe for non-convergence in a static structural run.  You will need enough cohesion for numerical stability, in a static run.  Would running this as a transient (with system damping) be an option or perhaps an explicit solver like LS-Dyna? 

      Perhaps there are additional tips in the references listed below (from Section of the Material Reference Guide):

      1. Simo, J. C., Kennedy, J. G., & Govindjee, S. (1988). Non-smooth multisurface plasticity and viscoplasticity. Loading/unloading conditions and numerical algorithms. International Journal for Numerical Methods in Engineering. 26(10), 2161-2185.

      2. Simo, J. C. & Hughes, T. J. R. (1998). Computational Inelasticity. New York: Springer.

      3. Roscoe, K. & Burland, J. B. (1968). On the generalised stress-strain behaviour of wet clay. Engineering Plasticity. 169(1), 535-609.

      4. Wood, D. M. (1990). Soil Behaviour and Critical State Soil Mechanics. Cambridge: Cambridge University Press.



    • Srinjoy Gupta

      Hi, I was able to model the cohesionless soil (i.e. dry soil as you correctly identified) in Static Structural. Coming on to my second query when I am applying gravity to my model it's deforming under its own weight because it's ramped up from 0 to 9.81m/s2. I don't want that to occur and already want the geostatic stresses that develop at the end of applying gravity to be already present in the soil (at time = 0s) as they exist in an already consolidated state as discussed in the Technology Showcase: Example Problems ANSYS, Inc. Release 2022 R2. For that, I need to incorporate an INISTATE command but I'm unable to understand how to apply it for the soil model. Please could you guide me regarding the same especially when we have a multilayered soil deposit with varying unit weight or density 

Viewing 2 reply threads
  • You must be logged in to reply to this topic.