[Jinjolee]

I am trying to do the mesh convergence test by increasing the mesh refinement to see at what level will the drag force converge. I used simpleFoam as I want to solve for steady-state solution. However, I found that when the mesh refinement level increases, the residual values especially Ux, Uy and p keep increasing. Here is the residuals of the mesh (I will call this meshA) with refinement level = (edge = surface = region = 2; no layer):


And for mesh (I will call this meshB) with refinement level = (edge = surface = region = 4; boundary layer = relatvie 0.3; no. of layer = 3 (81.1% generated)):


Also, the forceCoeffs become unstable for high refinement level(meshB), while it is stable at low refinement level(A).

forceCoeffs of meshA:


forceCoeffs of meshB:


What may be the cause? Does this imply the solution maybe transient?
Thanks a lot!

FYI, here is my controlDict for both settings:

application simpleFoam;

startFrom latestTime;

startTime 0;

stopAt endTime;

endTime 600;

deltaT 1;

writeControl timeStep;

writeInterval 20;

purgeWrite 4;

writeFormat binary;

writePrecision 6;

writeCompression off;

timeFormat general;

timePrecision 6;

runTimeModifiable true;

functions
{
#include "streamLines"
#include "cuttingPlane"
#include "forceCoeffs"
}

fvSolution:

solvers
{
    p
    {
        solver          GAMG;
        smoother        GaussSeidel;
        tolerance       1e-7;
        relTol          0.01;
    }

    Phi
    {
        $p;
    }

    U
    {
        solver          smoothSolver;
        smoother        GaussSeidel;
        tolerance       1e-8;
        relTol          0.1;
        nSweeps         1;
    }

    k
    {
        solver          smoothSolver;
        smoother        GaussSeidel;
        tolerance       1e-8;
        relTol          0.1;
        nSweeps         1;
    }
    
    epsilon
    {
        solver          smoothSolver;
        smoother        GaussSeidel;
        tolerance       1e-8;
        relTol          0.1;
        nSweeps         1;
    }
}

SIMPLE
{
    nNonOrthogonalCorrectors 0;
    consistent yes;
    residualControl
    {
	U		1e-4;
	p		1e-4;
	"(k|epsilon)"	1e-4;
    }
}

potentialFlow
{
    nNonOrthogonalCorrectors 10;
}

relaxationFactors
{
    equations
    {
        U               0.9;
        k               0.7;
        epsilon       0.7;
    }
}

cache
{
    grad(U);
}

fvSchemes:

ddtSchemes
{
    default         steadyState;
}

gradSchemes
{
    default         Gauss linear;
    grad(U)         cellLimited Gauss linear 1;
}

divSchemes
{
    default         none;
    div(phi,U)      bounded Gauss linearUpwindV grad(U);
    div(phi,k)      bounded Gauss upwind;
    div(phi,epsilon)  bounded Gauss upwind;
    div((nuEff*dev2(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
    default         Gauss linear corrected;
}

interpolationSchemes
{
    default         linear;
}

snGradSchemes
{
    default         corrected;
}

wallDist
{
    method meshWave;
}

Thank you very much!!

[sjsivabharathy]

Hi Jinjolee,
What is the Reynolds number and what geometry are you computing the flow for?

[Jinjolee]

Originally Posted by sjsivabharathy

Hi Jinjolee,
What is the Reynolds number and what geometry are you computing the flow for?

Hi,
My geometry is a cyclist riding on a bike(as shown in the figure below).


And I am measuring for the air drag from the external flow with inlet velocity of 12.5m/s, so if I am correct(sorry for my poor fluid background), the Reynolds number should be:
(12.5*1.638)/(1.5*e-5) = 1365000
where 1.638 is the wheelbase length of the bike(I take it as characteristic length)

Thanks.

[igor.leo93]

Hello Jinjolee,
Im facing similar issue. Did you manage to solve your problem?