LS Dyna

LS Dyna

Topics related to LS-DYNA, Autodyn, Explicit STR and more.

Composite Modelling

    • Raghav
      Subscriber

      Hello everyone, I am modelling composites for impact simulations but i am unable to validate existing literature. I have alligned the fibers into desired direction as well.

      I have given eroding surface to surface contact between projectile and composite layers and automatic surface to surface tiebreak between composite layers.

      After the impact, the elements of one composite layer are falling into other without maintaining the proper contact (image below). How can I solve this issue?  What other possible errors could be there for my simulation to not match the previously published results apart from the material properties. Would symmetry plays a major role in results. I have set 3 boundary conditions, fixed and x and z symmetry by restricting the degree of freedom for respective case.

      Thankyou

    • igandiko
      Ansys Employee

      Please check if the tiebreak contact is working correctly. Try setting IGNORE=1 in automatic tiebreak contact and make sure the segment normals are oriented correctly. It might be netter to use segment sets instead of part ids in the contact definitions. Look for warnings in d3hsp file regarding missing tied nodes. For solid elements, use type 1 with HG control type 2. 

       
    • Raghav
      Subscriber

      Thankyou for the reply, I set the IGNORE =1 in automatic surface to surface tiebreak and this problem is resolved. I had to align the normal in the direction of impact. 

       But still I am not able to properly validate the paper. It is a possibility that erosion might be taking place early so I am not getting comparable results. But there are high chances that my contact is also wrong.

       

      Eroding surface to surface

      Automatic surface to surface tiebreak.

      The erosion is set to be 0.04 for maximum prinsipal strain.

      I am getting very little velocity drop. Please help

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