General Mechanical

General Mechanical

POLAR symmetry not working

    • javat33489
      Subscriber

      Hi all. I

      have one segment. In total, there are 6 of them in the product. First, I bend the segment in the calculation using the APDL code with the ekill command (highlighted in red) and pull it onto the edge. Then I deform the segment by moving. I am having the following problems.

      1. If I do symmetry with POLAR and divide by the number of segments (6), I get 60 degree symmetry, everything is ok. Cyclic symmetry + cartesian doesn't work. Further, when calculating, the boundaries between the segments (highlighted in yellow) are not fastened together and are separated during deformation. Why? how to keep the binding in these places?
      2.  If I don't use symmetry, but use frictionless support (in places highlighted in yellow) I get a DOF error. Why? Using frictionless support I want to simulate symmetry.

    • Aniket
      Ansys Employee

       

      If you use Frictionless supports only on the faces highlighted in yellow, the part is free to move in its axial direction. The Frictionless supports won’t constrain that, and thus you are getting the DOF error.

      Cyclic symmetry won’t work with cartesian; as it needs a cylindrical coordinate system. Can you please explain issue #1 by posting the result images?

      -Aniket

      How to access Ansys help links

      Guidelines for Posting on Ansys Learning Forum

       

      • javat33489
        Subscriber

        Step 1

        Step 2

        Step 3

        Displacemnet and Fixed support.

        The top moves. Internal fixed.

        Step 4

        At the junctions of symmetry, detachment occurs. Why? How to avoid it?

      • javat33489
        Subscriber

        Help please

    • javat33489
      Subscriber

       Help Help Help

    • Aniket
      Ansys Employee

       

      Can you please check the initial node locations and final node locations at the separation? You can use a single node for this.

      As you are using a beta feature, in this case, it may have issues, mostly graphical. Beta features do not go through thorough testing as other release features and hence can give this type of issue.

      -Aniket

      How to access Ansys help links

      Guidelines for Posting on Ansys Learning Forum

       

      • javat33489
        Subscriber

        Symmetry has been around for a very long time, I don't know why it's BETA. Cyclic symmetry doesn't allow it, so I followed the advice from this forum to use POLAR. But the problem is that the symmetry parts don't stick together, there are only 4 parts.

        • Aniket
          Ansys Employee

          The symmetry feature is NOT beta, showing graphical expansions using Symmetry details is beta. 

          If you have applied frictionless support, or symmetry on the parts that are separating, a good check would be checking node locations at the final step and comparing them with their initial locations.

          Most likely this would be a graphical artifact.

          -Aniket

          How to access Ansys help links

          Guidelines for Posting on Ansys Learning Forum

    • javat33489
      Subscriber

      This is not an artifact, parts of the symmetry are being pulled apart. Although symmetry implies gluing elements together

    • Aniket
      Ansys Employee

       

      Kindly post all the images of your setup. Make sure to include all BCs, details, and graphics. Ansys employees can not exchange any files, but can only guide looking at the inline images.

      Also, have you tried replacing cyclic symmetry with frictionless support? does that work well? In the image above, you are checking the total deformation at a node, which can be increasing. Check directional deformations, in the tangential direction and see if those are zero or near zero.

      -Aniket

      How to access Ansys help links

      Guidelines for Posting on Ansys Learning Forum

       

      • javat33489
        Subscriber

        >>Also, have you tried replacing cyclic symmetry with frictionless support? does that work well?

        Yes, it works well. But sometimes symmetry is needed.

        >>n the image above, you are checking the total deformation at a node, which can be increasing.

        There is not a general deformation, but a deformation in the direction of one of the coordinate axes

         

         

         

         

        • Aniket
          Ansys Employee

          Sorry, but no, this is not directional deformation. As per the image you have uploaded the legend on top left mentions that this is a Total Deformation:

          -Aniket

          How to access Ansys help links

          Guidelines for Posting on Ansys Learning Forum

        • javat33489
          Subscriber

          All right. I first chose TOTAL, but in the settings I did not specify ALL, but chose the axis.

Viewing 5 reply threads
  • You must be logged in to reply to this topic.