Topics relate to Fluent, CFX, Turbogrid and more

2D 6DOF Rigid Body Dynamic Mesh Solution Stabilization

    • piggh18

      Hello, I am currently trying to run a transient 2D multiphase open-channel wave simulation with a 6DOF rigid body in the domain. The rigid body is defined as SDOF rotation with a surrounding deforming dynamic mesh zone that uses smoothing and remeshing. I am running with the PISO scheme with NITA. When I ran the first time, all went well for approximately 6.5 seconds of simulation time when I got the message "Negative K in 1 cells after linear solve." After this, the dynamic meshing failed with the error "Error at host: Update-Dynamic-Mesh failed. Negative cell volume detected." There were 953 cells with non-positive volume as well. I figured there were two ways to go about fixing this:

      1) Use implicit update settings under dynamic mesh options. I've tried this and I think it leads to better convergence but it takes far too long to be realistically used. It also doesn't work with NITA.

      2) Use solution stabilization for the rigid body zones I have defined. I'm not sure if this will work as I am currently using it along with NITA and have not had problems yet. With this option, I wonder what value I should use for the scale factor and whether I should use volume or coefficient based stabilization.

      Thank you very much in advance!!!

    • Prashanth
      Ansys Employee


      Did you try to address the negative volume mesh issue? They are due to higly skewed degenerate cells from the remeshing. Have you tried with overset mesh?

Viewing 1 reply thread
  • You must be logged in to reply to this topic.