Ansys Products

Ansys Products

Discuss installation & licensing of our Ansys Teaching and Research products.

Challenge in getting hexa dominant meshing of a pipeline with a single defect

    • Martin Ileagu
      Subscriber

    • mjmiddle
      Ansys Employee

      Try smaller sizes such as a body sizing or face sizing. Also, hexa-dominant mesher may not be the best choice. It starts from a surface mesh and paves inward, and then just fills the inside with tetras and transition elements when the paving can't match from different directions converging. Because of this, the element quality in the center will be very bad, and may be hard to converge when you run the solution. Because of this method, hexa dominant is best for geometries with a good amount of bulk volume. In thin geometries, like you have, it can sometimes connect the opposite sides without the low quality elements in any bulk center. And sometimes it just fails. That is why I said try different element sizes, on the order of the thickness across the tube and smaller.

      But thin geometry is better meshed as a sweep method. Or try multizone which can be thought of as a more intelligent sweep which can handle some non-matching imprints on source and target faces. You can also slice off the non-sweepable section in the geometry modeler (looks like SpaceClaim), and connect with shared topology. Sweep-mesh the sweepable body and tetra mesh the non-sweepable body. Or you can slice the geometry in more complex ways so as to create all sweepable bodies.

Viewing 1 reply thread
  • The topic ‘Challenge in getting hexa dominant meshing of a pipeline with a single defect’ is closed to new replies.