CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   Why 3D solid-pore geometry showing diverged solution? (http://www.cfd-online.com/Forums/openfoam/110086-why-3d-solid-pore-geometry-showing-diverged-solution.html)

Sargam05 December 3, 2012 16:45

Why 3D solid-pore geometry showing diverged solution?
 
Hello everyone,

I am using solidDisplacement solver for 3D solid-pore cube geometry. Following are displacement D boundary conditions:

boundaryField
{
electron_minX
{
type tractionDisplacement;
traction uniform ( 0 0 0 );
pressure uniform 0;
value uniform (0 0 0);


}

electron_maxX
{
type tractionDisplacement;
traction uniform ( 0 0 0 );
pressure uniform 0;
value uniform (0 0 0);

}
electron_minY
{
type tractionDisplacement;
traction uniform ( 0 10 0 );
pressure uniform 0;
value uniform (0 0 0);



}
electron_maxY
{
type tractionDisplacement;
traction uniform ( 0 -10 0 );
pressure uniform 0;
value uniform (0 0 0);


}
electron_minZ
{
type tractionDisplacement;
traction uniform ( 0 0 0);
pressure uniform 0;
value uniform (0 0 0);


}
electron_maxZ
{
type tractionDisplacement;
traction uniform ( 0 0 0 );
pressure uniform 0;
value uniform (0 0 0);

}
electron_to_pore
{
type tractionDisplacement;
traction uniform ( 0 0 0 );
pressure uniform 0;
value uniform (0 0 0);

}
electron_to_npore
{
type zeroGradient;
}



}


OpenFOAM simulation is giving diverge solution. Following is part of log file:
GAMG: Solving for Dx, Initial residual = 0.0407966, Final residual = 0.00378547, No Iterations 1
GAMG: Solving for Dy, Initial residual = 0.0521521, Final residual = 0.00500495, No Iterations 2
GAMG: Solving for Dz, Initial residual = 0.0856927, Final residual = 0.00255286, No Iterations 2
Min/Max/Ave DEqn:(-7.00897e+65 -2.76576e+77 -1.2423e+65) (7.19343e+65 -2.76576e+77 1.17129e+65) (1.36111e+64 -2.76576e+77 -2.05129e+64)
GAMG: Solving for Dx, Initial residual = 0.030862, Final residual = 0.0010546, No Iterations 2
GAMG: Solving for Dy, Initial residual = 0.0446989, Final residual = 0.00435236, No Iterations 3
GAMG: Solving for Dz, Initial residual = 0.069533, Final residual = 0.00225699, No Iterations 2
Min/Max/Ave DEqn:(-6.97373e+65 -2.76576e+77 -1.2305e+65) (7.17125e+65 -2.76576e+77 1.07291e+65) (1.47816e+64 -2.76576e+77 -2.09031e+64)
GAMG: Solving for Dx, Initial residual = 0.0282657, Final residual = 0.000732531, No Iterations 2
GAMG: Solving for Dy, Initial residual = 0.0351312, Final residual = nan, No Iterations 1000
GAMG: Solving for Dz, Initial residual = 0.0653158, Final residual = 0.00194052, No Iterations 2
Min/Max/Ave DEqn:(-6.92125e+65 nan -1.19559e+65) (7.11257e+65 nan 1.03401e+65) (1.54778e+64 nan -2.09798e+64)


I have done similar 3D cube geometry problem without pore in which got converge solution. But I do not know why similar boundary conditions are not giving converge solution for solid-pore 3D cube geometry. I have also tried OpenFOAM-ext but similar problem is coming. When I am using three symmetry planes I get converge solution but when I am taking all surfaces as solution domain I get diverge solution. How I can get converge solution with considering all surfaces of the 3D cube (solid-pore phase) geometry?

Does someone has any idea why this problem is coming? How I can solve this problem?

Best regards,
Sangeeta


All times are GMT -4. The time now is 05:31.