Fluids

Fluids

Topics relate to Fluent, CFX, Turbogrid and more

Overset meshing – No valve displacement

    • Nicolas Le Barbier
      Subscriber

      Hello,

      I am setting up a CFD simulation of a hydromechanical valve (a movable valve with a permanent fuel passage through an orifice at its center, whether the valve is open or closed). In this simulation, I aim to capture the dynamics of the valve (opening, equilibrium position, etc.) using a dynamic mesh approach with the Overset method. Despite creating the overset interface and configuring the dynamic mesh with 6 DOF (1 DOF translation) along the valve's directional axis, no displacement is observed in the simulation (obtaining a hydrodynamic force of +3.6 N as opposed to a preloaded spring force of -0.3 N should lead to valve opening). A contact detection is performed in the calculation at the guiding zones of the valve. Do you have any insights into why no valve displacement is observed?

      Best regards,

      Nicolas

    • SRP
      Ansys Employee

      Hi,

      Ensure that your mesh quality is good and well-suited for dynamic simulations. Poor mesh quality can lead to stability and convergence issues.Double-check the Overset interface setup. Make sure the interfaces are correctly defined, and the overlapping grids match appropriately.

      Thank you.

    • Nicolas Le Barbier
      Subscriber

      Hello,

      Thank you for your response and advice.

      Regarding the mesh quality, here are some details:

      • Background mesh: 2.5 million nodes / 14 million elements / Skewness: maximum 0.83, average 0.23 / Orthogonal quality: maximum 0.997, average 0.77
      • Component mesh: 1 million nodes / 2 million elements / Skewness: maximum 0.58, average 0.17 / Orthogonal quality: maximum 0.999, average 0.83

      Based on the trends of the residue curves, it doesn't appear that stability and convergence of the calculation are the source of the issue. 

      Below, please find an enlarged section of the velocity field's opening area and the overset mesh (a deliberate preliminary opening was created) for your reference. A portion of the component mesh can be seen extending beyond the domain (inactive cells). The force balance indicates a positive axial component that should ideally lead to the valve opening.

      The simulation notes "3256 overset orphan cells in interface over." Could the presence of orphan cells potentially hinder the displacement?

      Furthermore, I'd like to remind you that the lateral guiding zones of the valve are modeled with gap filling. Could this potentially hinder movement?

      I thank you in advance for your assistance and remain at your disposal for any additional information you may require.

      Best regards,

      Nicolas

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