No announcement yet.

Dynamic Rolling Process FE Simulation

  • Filter
  • Time
  • Show
Clear All
new posts

  • Dynamic Rolling Process FE Simulation

    Dear All,

    I hope all feel so fine with your simulations.

    Recently, I am simulation a rolling process with a very high dynamic speed (around 5 m/s). Hence, the model is quite dynamic and not Quasi-static.

    I use Abaqus Explicit for this simulation. The structure is composite material including several layers with different material properties. In this case, I am using ''Cohesive Surface'' definition as the surface interaction between different layers.

    I have a big problem and there is an error during the simulation process so called:

    ''the ratio of deformation speed to wave speed exceeds 1.0000 in at least one element''.

    I tried to find a solution for this problem such as:

    Larger mesh . But the larger mesh decreases the solution accuracy.

    Have you ever seen any solution for this case? In this respect, may I imagine to let me know please if you have already any hints or solution for this problem.

    Thanks for your attention.

    Kind regards,


  • #2
    That error message often indicates that you simulation is becoming unstable. Do you expect some instability (for example, failure) to occur?

    Jorgen Bergstrom, Ph.D. PolymerFEM Administrator


    • #3
      Dear Jorgen,

      It is 100% correct. Actually, the model includes some layers and there are delamination between layers during the rolling process.

      For delamination: I could use two methods.

      1- With Cohesive Zone Element (CZM) ------- >> the model works perfectly. Here the delamination would be implemented in terms of material modelling.

      2- With Cohesive Surface ------- >> the model was crashed due to the failure of cohesive bonds between the layers.!!! Here the delamination would be implemented in contact definition and surface interaction.
      I can increase the mass scaling and the problem was solved but I think it is not completely precise since the resulting ALLAE is then so high due to mass scaling effects.

      then, how can I make that stable for second method (Cohesive surface) without mass scaling?

      I would be happy to let me know if you have further ideas.

      thanks in advance,