Modeling Roughness on a channel

Does anyone know of a way to model roughness through the walls of a channel for both laminar and turbulent flow in ansys fluent? 

Comments

  • DrAmineDrAmine GermanyForum Coordinator
    edited January 2019
    This is pure research but generally laminar flows are not influenced by roughness as the streamlines will cover the roughness being stratified. Consequence is the friction factor is independent of roughness. For turbulent flows sticks to omega based models in Fluent and use standard wall resolution as if the wall is smooth.
  • DrAmineDrAmine GermanyForum Coordinator
    edited January 2019
    Sans grain height or something equivalent is tout roughness height.
  • Raef.KobeissiRaef.Kobeissi United KingdomMember
    edited January 2019

    Not sure if it is suitable:

  • kawad24kawad24 Member
    edited January 2019

    Is this method equitable to roughness heights by using an Ra value?

  • DrAmineDrAmine GermanyForum Coordinator
    edited January 2019
    Check paper of Adams for that to relate measured values to sand grain roughness. "A simple algorithm to relate measured
    surface roughness to equivalent sand-grain roughness"
  • DrAmineDrAmine GermanyForum Coordinator
    edited January 2019
    It should increase
  • DrAmineDrAmine GermanyForum Coordinator
    edited February 2019

    As ANSYS-Stuff we can only comment on attached pictures. 

  • DrAmineDrAmine GermanyForum Coordinator
    edited February 2019

    And what is now going wrong? Provide more information about boundary, conditions and your judgment of the result.

  • kawad24kawad24 Member
    edited February 2019

    The problem I am facing, is that when I add a roughness factor on the walls, the pressure differential is decreasing rather than increase. The center of the model is an air fluid region with a mass flow rate inlet and pressure outlet. The fluid is surrounded by aluminum walls with a constant heat flux on the top wall and insulated on the left and right wall as well as the bottom. I am not sure why this is the case? I was questioning my mesh being too course and possibly giving invalid results. Or a possible error in the settings:

    ·      Turbulent: Viscous-Standard K-Epsilon approximation. 

    ·      To model forced convection, the density was set to as a model approximation to assume incompressible ideal-gas

    ·      Operating pressure: 98,338.2 Pascal (14.2 PSI)

    ·      Gauge Pressure at outlet: -1112.3 Pascal (-.16 PSI)

  • DrAmineDrAmine GermanyForum Coordinator
    edited February 2019

    Which near wall method are you using? Do you exp. data or any correlations?

  • kawad24kawad24 Member
    edited February 2019

    I was using the scalable near wall function.  I am trying to simulate this and will be experimentally be verifying the results. I just don't understand why the wall height i place isn't following the predicted trend for pressure differentials

  • kawad24kawad24 Member
    edited February 2019

    can you tell me what settings were used for this turbulent model? For some reason, my pressure differential is decreasing as the roughness is increasing. I see the K-Epsilon model, realizable model was used, but is there something else I should be aware of?

  • DrAmineDrAmine GermanyForum Coordinator
    edited February 2019
    I do not understand your density setting. Regarding viscous model and near wall modelling I recommend using sst model.
  • DrAmineDrAmine GermanyForum Coordinator
    edited March 2019

    1/It is up to you to choose the most suitable meshing method. Just fulfill the mesh metric quality requirements: Aspect Ration, Cell Volume Change, Skewness and Ortho Quality.

    2/Create an iso-surface based on Mesh>X-Coordinate

     

  • regitasyaliregitasyali Member
    edited June 1

    Hi, can you provide reference for this statement please? Thanks.

Sign In or Register to comment.