Skip to content

Cuboid Geometry-Drag and Lift

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #6145
    sahilbhapkar30
    Participant

    Hello,

    I would like to ask if I could calculate drag and lift forces on my cuboid geometry. Is it possible to calculate it? I tried implementing the Drag setup as per the cylinder2d/3d setup but I am getting infinite value for my cuboid geometry. Is it only that drag and lift can be calculated for spherical bodies?

    Thankyou

    #6146
    Adrian
    Keymaster

    No, the functors used to calculate lift and drag in e.g. cylinder(2,3)d support arbitrary shapes.

    However, I am unsure what you mean by “drag of cuboid geometry”. The CuboidGeometry class describes the block decomposition of the simulation domain and not the obstacle geometry of which you want to calculate flow coefficients.

    #6148
    sahilbhapkar30
    Participant

    Hello Adrian,

    Thank-you for your prompt reply.

    When I tried implementing the drag and lift for my inlet and outlet which are in cuboid shape,

    SuperLatticePhysDrag3D<T,DESCRIPTOR> drag( sLattice, superGeometry, 3, converter ); // for inlet
    SuperLatticePhysDrag3D<T,DESCRIPTOR> drag2( sLattice, superGeometry, 4, converter ); // for outlet

    T dragA[3];
    int input1[0];
    drag( dragA, input1 );
    clout << “; drag_inlet=” << dragA[0] << “; lift_inlet=” << dragA[1] << endl;

    T dragB[3];
    int input2[0];
    drag2( dragB, input2 );
    clout << “; drag_outlet=” << dragB[0] << “; lift_outlet=” << dragB[1] << endl;

    I am getting inf value for drag_inlet and drag_outlet and same for lift as well. Could you tell me the reason for this inf value?

    I did not understand the meaning of the sentence you said in your comment:

    The CuboidGeometry class describes the block decomposition of the simulation domain and not the obstacle geometry of which you want to calculate flow coefficients.

    Thank-you

    #6158
    Adrian
    Keymaster

    The drag functor assumes that the given material number / indicator describes the boundary of the object for which the drag and lift coefficients are to be computed. You can see this in action e.g. in our cylinder(2,3)d examples.

    What I meant by my other comment is that the cuboid geometry and material numbers describe different concepts. Asking for flow coefficients of a CuboidGeometry doesn’t make sense in OpenLB terminology. However, rereading your initial comment: Maybe you did not mean the CuboidGeometry class but the flow around cuboid shapes?

    #6161
    sahilbhapkar30
    Participant

    Hello Adrian,

    Thank-you for the reply.

    Yes,actually I mean the flow around the cuboid shapes.

    As per my above comment, I have described my drag definition at the inlet and outlet. However, I am getting infinite value at the inlet and outlet.

    I tried setting the drag definition even for bstep2d/3d (Cuboid Shape geometry) and found similar infinite value for drag coefficient and lift. That is the reason I asked whether Drag force is calculated using Stokes Law. If it is calculating by Stokes Law, the Radius of Curvature in the cuboid geometry would be taken to be infinite.

    So, could you please help me to imply the drag and lift definition for cuboid shapes?

    Thank-you

    #6163
    Adrian
    Keymaster

    If you mean the flow around the cuboid shape that you have to provide the material number of the cuboid shapes (an this number must only describe the cuboids). I repeat that the drag functor will not work when provided in / outflow material numbers in your approach.

    For these kinds of questions I also kindly refer you to our upcoming Spring School. In the practical sessions there we also offer support for getting started with your own cases.

Viewing 6 posts - 1 through 6 (of 6 total)
  • You must be logged in to reply to this topic.