Announcement

Collapse
No announcement yet.

how do we tackle excessive distortion

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • how do we tackle excessive distortion

    Dear all,

    I am working on constitutive modelling and my model deals with large deformations.

    I am using c3D8r elements. The problem that I face is 'excessive distortion' in the elements.

    The vumat code worked just fine with single c3d8r element, however, when i use it for multiple elements there is excessive distortion...

    How do I tackle it...any suggestions are greatly appreciated.


    thanks
    brunda

  • #2
    Originally posted by brunda View Post
    Dear all,

    I am working on constitutive modelling and my model deals with large deformations.

    I am using c3D8r elements. The problem that I face is 'excessive distortion' in the elements.

    The vumat code worked just fine with single c3d8r element, however, when i use it for multiple elements there is excessive distortion...

    How do I tackle it...any suggestions are greatly appreciated.


    thanks
    brunda

    What sort of a material model are you dealing with?

    Since your error may be due to lots of things, the best method is trying to debug the code through writing data in a text file "formatted outputs" and tracking the variables.

    Comment


    • #3
      I am using Anand and Ames constitutive model...

      I am applying the model for a dogbone sample of 65x10x0.22 (mm) dimensions.

      From the output file, I noticed that the F(3,3) (deformation gradient in the 0.22 mm direction) has instability..

      So I am not sure if the dimension in thickness direction is creating excessive distortion..

      B

      Comment


      • #4
        According to the given dimensions I understand that you have a sheet specimen (i.e. the thickness is considerably small compared to the other directions).

        In order to eliminate the possible modelling errors (I mean errors other than your material routine) I suggest you to run the test with a built in material model in ABAQUS.

        Another point is that, does the error appear at the very beginning of the analysis or in the middle? If at the beginning youre relatively luckier. In either case try to monitor your computed stresses. Since the strains or deformation gradients are automatically passed by ABAQUS excessive distortions are generally due to illposed material response in the form of misderived tangents or stresses.

        Comment


        • #5
          Hi Kumar,

          As you had suggested, I used a built-in vumat and the model worked fine.. I have looked at outputs of various quantities that are computed in my vumat too.. there seems to be a sudden in jump in deformation gradient (plastic part of F=Fe*Fp) which is causing a sudden change in stress and other values..

          The model uses inverse langevin function and I am doubting if this might be the cause for excessive distortion..

          Unfortunately the code breaks in the middle of the analysis :(...

          Comment

          Working...
          X