hexagon logo
Register Sign in
  • State Not Answered
  • Replies 9 replies
  • Subscribers 8 subscribers
  • Views 111 views
  • Users 0 members are here
Options
Related

Hello guys, I am trying to study the flow around NACA0012 airfoil at zero angle of attack. However, when I try to refine the octree parameter around the airfoil and generate the mesh, every single time I get different values for Lift force! Any reasons??

hossam
Hello guys, I am trying to study the flow around NACA0012 airfoil at zero angle of attack. However, when I try to refine the octree parameter around the airfoil and generate the mesh, every single time I get different values for Lift force! Any reasons??
  • 0
    NACA0012 has a symmetrical airfoil.
    The lift force is zero when the angle of attack is 0 degrees.
     
    Will it be the values(floating) whose lift force cannot be considered zero?
    And does that the values(floating) change depending on the octree parameter?
    I think it is natural to get different values(floating) for different octree setting.
     
    Anyway, I used scFLOW to get the lift coefficient for NACA0012.
    It was a very good result with even a small number of meshes.
    Here are the results of the lift coefficient and examples of the octree and prism settings for your reference.
    The chord length is 1.0 meter.  The mesh is a pseudo-2D with one element in depth.
    CL
    Octree
    Prism
  • 0
    Hello Chuel,
    Thanks so much for your reply and dedicated help.
    I totally agree with you that the lift force is to be zero at 0 angle of attack.
    Actually, I am not sure what it is the float values refer to, but yeah, at different octree settings, I get different values for lift. Even more, I get lift coefficient sometimes at order of magnitude 1e-4 and 1e-3 and that of drag is 1e-2.
    Is this normal to happen? Is this numerically acceptable?
    Do I need to input the thickness of first layer that capture the boundary layers to get more accurate result?
  • 0
    Hello hassam,
     
    I think it is very difficult to get an exact value of zero as a solution for numerical calculations.
    Whether it can be regarded as zero or not does not depend on its absolute value, but on the problem.
     
    How about comparing it with the case of angle of attack, which is not zero as the lift coefficient?
    I will attach the results of my test below.
    0deg_vs_2deg
    In the case of 0 degree angle of attack, CL values are depending on the octant.
    In some cases, even the signs are different.
    On the other hand, for 2 degrees angle of attack, the dependence on the octant is small and  CL values are close to the experimental value.
    CL values at 0 degrees angle of attack are sufficiently small compared to CL values at 2 degrees angle of attack that they can be regarded as zero.
     
    By the way, regarding the boundary layer, I think it is essential to set up the boundary layer element in order to obtain highly accurate results.
    This is especially effective for problems where the viscous stress acting on the wall surface has a significant effect on the calculation results.
    For airfoil type problems, I recommend inserting about 10 layers of boundary layers.
  • 0
    Hello Chuel,
     
    Apologies for my late reply.
    Ok, now I get it. Thanks for enlightening me regarding this issue.
    I have one last question about this topic. When I enter an angle of attack, as velocity components in x and y direction, I don't get correct values for lift and drag.
    For example, I tried to replicate the same value in your excel screenshot with same octree size(=1mm). At AOA(=zero degree), I obtained a similar value as yours for lift coefficient and at the same order of magnitude, however, at AOA(=2degrees), I obtained value of ~0.052! That's not correct, I think I misunderstand something in the setup, so could you please help me?
    Hint: I used static pressure outflow as an outlet BC. for your consideration.
    inflow setup
    Force-y = Lift
  • 0
    Hello hassan,
     
    > at AOA(=2degrees), I obtained value of ~0.052! That's not correct,
     
    I guess that you didn't apply the boundary conditions at the bottom and top areas.
    If you didn't , Please try it !
     
    BC
  • 0
    Hello Chuel,
     
    It worked! I put wrong BCs. on top and bottom of the domain.
    Is there a way to get the value of the drag only? I need to understand why the value here in X is by negative. The lift value is fine for me, however the value in X is suspicious for me. Does it relate to the kind of surface forces I need to compute during the setup?
    Capture2
  • 0
    Is the attached data the result of an angle of attack of about 18 degrees?
    When the angle of attack is taken into account in the inflow velocity component, the force in the X direction on the wing parallel to the X axis becomes negative as the angle of attack increases.
    Please check the following for your reference.
    Drag_Lift
  • 0
    Yup it worked,
    I understand how all of it works now. Thanks so much Chuel.
    Have a good day!
  • 0
    Hello hossam,
    I'm sorry I made a mistake in your name.
    Anyway, I'm glad that you understand how all of it works.
    Have a good day !
© 2025 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement