General Mechanical

General Mechanical

Convergence Issue for Large Deformation Static Structural Task (Bending Wire)

    • dusev
      Subscriber

      Trying to bend a wire into an arch shape. Using 2 pieces to do this ( a sheath and an arched push tool). I've fixed the sheath and apply a 1-DOF displacement using the arched push tool piece to push the wire into the sheath. I have 2 frictionless contact defined between bodies, and I've turned on large deformations.

      Seems like this should be pretty straight forward, but I keep failing to converge. What am I missing here?

      The errors generally mention that the wire has a node (then it mentions a single node) that has been displaced too far. I've tried to remesh the wire, and increase the number of steps in the step controls making the displacements small for each step, also tried quasi-static. Tried changing some of the nonlinear controls options too.

      Any suggestions? Here's the .wpbz file: wpbz_file_wire_bending

    • Erik Kostson
      Ansys Employee

      We can not download any files - perhaps someone else can.

       

      In general if you search the internet and this site for "convergence and issue" you will get lots of information that will be of help.

      Also the help manual has some help:

      https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v222/en/wb_sim/ds_contact_best_prac_converge.html

       

    • dusev
      Subscriber

      Ok, I retuned some parameters and increased my geometry sizes. I'm getting some solutions, but the wire overall is not moving:

      You can see the insert going in and bending the wire piece, however the wire ends are not moving. All my contacts are frictionless. The wire here is nitinol (austensitic so superelastic). I have the following settings under static structural constraints:

      • Displacement of the arched push piece - body - 10 steps at 1 second for 1 mm each
      • Fixed support - body - of thick 'sheath'
      • Displacement - of top face of wire (couldn't select body): X-Free, Y - 0 (to constrain motion in the Y), Z - Free

      I imagine theres something tricky going on with the supports/constraints that I'm missing. I expect the ends of the wire to get pulled in to align with the inner edges of the sheath due to the superelastic properties of the wire to start to bend, however there is only local bending occuring here.

      - Should I focus on doing this analysis under a transient structural setting instead as maybe the static structural doesn't like the large displacements? Why isn't it bending properly (only locally)?

    • dusev
      Subscriber

      With the transient solution it's the same thing. Here's an example of the failed output:

      I keep getting Elements that are highly distorted using both transient and static structural methods.

      Any suggestions here?

    • Erik Kostson
      Ansys Employee

       

       

       

      Again see these links and the help manual for more advice (also the mesh on the parts is large, and one would like to have a similar mesh between parts – see our free course on contact analysis – also make the large parts rigid, and mesh them as the wire with a finer mesh - finally make sure that the wire is not flying of before contact is made - use the contact tool to check the contact status)

      In general if you search the internet and this site for “convergence and issue” you will get lots of information that will be of help.

      Also the help manual has some help:

      https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v222/en/wb_sim/ds_contact_best_prac_converge.html

       

       

       

Viewing 4 reply threads
  • You must be logged in to reply to this topic.