Preprocessing

Preprocessing

Different mesh for different solids

    • cmoreno98
      Subscriber

      Hello everybody,


      I'm new to ANSYS and I'm trying to generate a mesh on a T-junction. Using DesignModeler I divided the geometry in 4 different bodies and I want to mesh the central solid in a different way:



      In the 2 inlets and 1 outlet I want a coarse simple mesh:



      • Solver Preference: Fluent

      • Element Size: 0.1 mm


      In the central solid I want a fine mesh:



      • Solver Preference: Fluent

      • Element size: 0.05 mm

      • Inflation: 8 layers with max. thickness 0.1 mm

      • Multizone


      How can I do this? I'm using ANSYS Workbench Academic Student 2019 R3.


       


      Thank you in advance,


      Carlos

    • peteroznewman
      Subscriber

      I don't expect Mulitzone to work for this geometry. Try to add a Mesh Control Method of Multizone and pick the center body, then Generate Mesh. See if you get Hex elements. If you get Tet elements, then delete the Multizone method.


      Under Mesh, type 0.1 for the Element Size. This will be the Global element size.


      Add a Sizing Mesh Control and pick the center body and set the Element Size to 0.05 mm.


      I recommend you use inflation on all walls. Why do you think you can leave it out of the other walls?


      Pick the outside faces of the body (or bodies) and type N to create a Named Selection, call it Inflate.


      Under Mesh, use the Inflation settings shown below.



       

    • cmoreno98
      Subscriber

      Hi peteroznewman,


      I tried to add the Multizone Method and I got Tet elements, so I deleted it.


      Yes, I also think that Inflation on all walls would be better, but just the central body has around 420k elements and it takes around 22h to simulate (using OpenFOAM). If I add Inflation on every solid I obtain a mesh of 624k elements (it will take almost 2 days to simulate). If I only add Inflation on the central body I get a total of 474k (faster simulation). Nonetheless, I will perform trials with both meshes to see the differences.


      Thank you very much, that's what I wanted



       


      PS: Just for curiosity, this is what simulations look like, air entering through the horizontal inlet and water through the vertical one:


Viewing 2 reply threads
  • You must be logged in to reply to this topic.