Preprocessing

Preprocessing

What is the best way to mesh this geometry? (cones, cylinders and a disc with holes)

    • Doba
      Subscriber

      Geometry can be seen here - https://forum.ansys.com/forums/topic/some-of-the-residuals-increasing-sharply-after-20-or-so-iterations/


      I am trying to follow this tutorial -


      but it's not working out for my geometry. I am getting like 7 different errors at the same time.


       

    • Rob
      Ansys Employee

      I'd make sure share topology had been done and that the mesh controls were set for CFD-Fluent. Set a small minimum size & sensible larger size and tet the whole model.  Ideally move to 2019R2 and use the new meshing workflow shown here 

        

    • Doba
      Subscriber

      I'm assuming you want me to use the Watertight Geometry meshing? I have a basic question about that. In the video it says that "Watertight refers to that models that don't require much cleanup". What does this mean exactly? How do I if my geometry requires much cleanup?

    • peteroznewman
      Subscriber

      If you want to study the airflow over a car body, and you were given all the panels, the underbody and the glass, there would be a large number of gaps. In years past, a lot of work would go into plugging all the gaps until the body was "watertight" then it could be subtracted from the air and meshing of the air could be done.  Now ANSYS has some automation to drape a surface over the body and fill the gaps automatically.


      Contrast that to a simple solid body representation of a pipe, with a flange face around the inlet and outlet. By simply identifying the two flanges, a watertight fluid cavity is easily identified. That requires no cleanup, it is considered "watertight" geometry.

    • Doba
      Subscriber

      Thanks for the answer! That new feature looks interesting but for now I'll stick with the non-fluent mesh module.


      @rwoolhou I tried to do what you suggested but the mesh quality is quite bad (same as this link). Can you help me out with this? I'm really lost atm


       

    • Rob
      Ansys Employee

      Now you've got the graph working zoom in on the poor quality cells. I suspect the decomposition you did to make the O-grid is giving some very small gaps/faces on the filter. If you merge all the fluid zones together and remesh how does it look? Note, keep zone separate if you need porous media. 

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