Fluids

Fluids

Slow at iteration 1 and then diverges

Tagged: 

    • atulsingh92
      Subscriber

      Hello ,

      So, I wanted to know if there would be any general reasons to fix if my 1st 10-15 iterations take a lot of time to come, and then my simulation diverges at 100-200 iterations later. The ones from lets say 30 to 100 are quick enough on a 6 core pc with 300k cells (small size), although it diverges anyways.

      The question is why would iterations 1 to 10 take so much time, compared to others, where ideally, the solver runs quickly.

      I have mass-flow-inlet at the blue region, attached to the 4 tubes and a symmetric design at the other end.

    • DrAmine
      Ansys Employee
      First of all check if the mesh of proper quality in Fluent: Orthogonal quality, Skewness as well as aspect ratio
    • atulsingh92
      Subscriber
      Min orthognal quality comes at around 0.2
      And I think it should be less than 0.1
      But as I am using the "outline tab" for meshing in fluent meshing using batch submission with the help of TUI commands (2019R2), I am unsure of how can I get this under 0.1 ?
      Any suggestions?
    • DrAmine
      Ansys Employee
      Orth quality should be as large as possible: values near 1 are ideal (1 will imply pure hex). So no need to try to reduce it...
    • Rob
      Ansys Employee
      Did you define "zero temperature" or leave it as the default condition? What is the inlet velocity?
    • atulsingh92
      Subscriber
      Yes, the rectangular region is a wall with zero temperature, while the tubes are at 300k, with fluid at 373K. The blue area is mass flow inlet at 0.0014 kg/s. The whole domain should have the bounding box of (700 X 100 X 40 )mm.

    • Rob
      Ansys Employee
      You set zero temperature? Which units are you using?
    • atulsingh92
      Subscriber
      With energy on, I have set thermal BC-> Temperature-zero to the recatangular region as follows

      However, I have also tried with the heat-flux as zero for the same case, where it diverged!
      Units are SI
Viewing 7 reply threads
  • You must be logged in to reply to this topic.