Will not solve quasi-static

mdsimulationgroupmdsimulationgroup
Hi,

I am attempting to wrap a film around a wire. I can solve the structure without quasi-static but once I try to time iterate with quasi-static, it can not converge.

I would like to set the node rotation value to "10*t" and attempt the wrapping. But just can't figure out why it won't converge.

Any ideas?

Thanks,
Frank

Comments

  • mdsimulationgroupmdsimulationgroup
    I believe I've been able to get it to solve by changing the time step to something small (0.01s) and maximum time step to 0.01s.
  • VictorVictor
    What I did:
    • Changed the tension force from nodes to faces to make it uniform.
    • Adjusted the X position of one side of the film's nodes so the elements are square, not parallelogram. This might not have been needed.
    • Ramped the tension force and the pressure.
    • Added CCX -> no extra contact step.

  • FatmacFatmac
    i played around with the model a little- did away with the pressure and attached the film to the wire - may be not a very elegant solution but it sort of worked. Once a full wrap is on i added contact from one side of the film to the other
  • FatmacFatmac
    video of out put
  • JohnMJohnM
    Fun problem. I increased the density, used symmetry, changed where the roll starts.
    I had trouble wrapping the wrap as it came full circle, I think this would take some clever time stepping.


  • mdsimulationgroupmdsimulationgroup
    JohnM, this looks really nice! Would you mind explaining how you used symmetry?
  • disladisla
    Not sure if it was @JohnM or Sergio who posted many years ago the idea of self-contact. It works very well in this case.
    One needs one contact for the initial Strip , Cilinder, and a second contact with itself once the first turn has roll. Don’t overlap master or slave surfaces.



  • mdsimulationgroupmdsimulationgroup
    Hi disla, this looks really nice! Would you mind posting your input?
  • JohnMJohnM
    By cutting the model in half and holding the cut surface in the X-direction (I think that was the direction), you essentially get a "mirror image" result. It cuts down on model size. Modeling your thin film with a single linear hex8 through the thickness is too stiff in bending. I will usually try using the c3d8i formulation (this can be changed in ccx options - also read the calculix manual about element types). But these and the 20-noded hexes don't always behave. So try to put at least 2 elements through the thickness.

    @disla - I'd like to see that model as well, please post, looks awesome!
  • disladisla
    https://we.tl/t-JKQM1dZVqb
  • mdsimulationgroupmdsimulationgroup
    @disla, I'm unable to access the site. Are you able to attach the liml file?
  • mdsimulationgroupmdsimulationgroup
    edited June 24
    Anyone got any ideas why my film is bouncing around like the wind is blowing on it?

    I found that if I suppress the "Film_to_Film" contact, the film does not flop around. So, I believe I may have the Film_to_Film contact setup wrong but I'm not sure in what way?
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register