[Pyrokrates]

Hi, I´m new to fluent udfs. I tired to implement some input viscous resistance using porous media:

DEFINE_PROFILE(Darcy_JJ,t,i)
{
    cell_t c;

    real celli[ND_ND]; // Vector with dimensions of present case
    real z;

    real z_in = 8.107e-6; // Length of inlet in z
    real z_pm = 8.107e-6; // Length of porous media in z
    int Nz = 10; // Number of pm cells in z
    real dz = z_pm / Nz; // Length of single filter cell in z

    real Darcy = 3.2268e+13;

    begin_c_loop(c,t)
    {
        C_CENTROID(celli,c,t);
        z = celli[2];

        if(celli[0] == 0 && celli[1] == 0) // Just to check the calculation
        {
            Message("Results for Cell coords: %g %g %g \n",celli[0],celli[1],celli[2]);
            Message("Results for multiplier: %g \n",((z-z_pm-dz/2)/dz + 1));
            Message("Darcy: %g \n", Darcy * ((z-z_pm-dz/2)/dz + 1));
        }

        C_PROFILE(c,t,i) = Darcy * ((z-zPF-dzF/2)/dzF + 1);

        C_UDMI(c,t,0) = C_PROFILE(c,t,i);
    }
    end_c_loop(c,t)
}

And here is my problem... What I expect is that for 10 cells the first cell Center layer in z-direction gets Darcy*1, the second one Darcy*2,... until Darcy*10. What I get is from messages: Exactly what I want, but for the calculation and the output in cfd post a different result. The last cell is something like interpolated meaning I don´t reach the Darcy*10... The resistance starts to decrease over the final cell and not after it... Do you know where the problem is? The first cell is not being interpolated.


Thanks for any help in advance


Chris

[AlexanderZ]

Actually I think that's the problem of your contours.

make the picture inside the fluent, turn off "node values" option while making contour.

for the first cell you have gradient from 0 to value -> so picture may look as you expected, but the last cell has value to 0 gradient, so you may see some negative gradient. That's what I expect, didn't see our pictures actually