Cavitation – Pressure Coefficient

    • KostasI

      I am dealing with cavitation around a 2-D hydrofoil. I am using the Zwart-Gerber-Belamri cavitation model. The code runs fine and I get back reasonable results. What I need for my project is the pressure coefficient 

      Pressure coefficient :    Cp = (p - p_infty) / (0.5 * rho * U^2)

      p - p_infty = gauge pressure

      rho = density of the liquid

      U  = inflow velocity

      When I get the pressure coefficient from fluent, the pressure coefficient is constant underneath the cavity. Specifically the pressure coefficient underneath the cavity is what you get when the pressure is equal to the vapor pressure.

      However when I get the pressure contours from fluent, the pressure is not constant underneath the cavity. If I print the pressure around the hydrofoil and non-dimensionalize it myself using the above definition ,I get a different pressure coefficient  along the foil. Indeed the pressure coefficient that I obtain this way is not constant underneath the cavity.

      It's easy to prove that the pressure coefficient should not be constant underneath the cavity, where there are sources/sinks of velocity (div(u) =/=0). However based on different models (e.g. barotropic model, potential formulation etc) the pressure cannot get lower than the cavitation pressure.

      It seems that Fluent allows the pressure to have any values when calculating the flow but enforces the condition of (pressure >= vapor pressure) when calculating the pressure coefficient. Is this true? Is there any other reason why I get to different results for the pressure coefficient?


    • DrAmine
      Ansys Employee

      Your plot is not really helpful,, sorry. But yes there is a cosmetic clipping of pressure for post-processing things. I am pushing behind removing that clipping.

    • KostasI

      Thank you for your comment. Is there anywhere I could find more about this cosmetic clipping  and how or when Fluent applies it?

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