[sharonyue]

I just made a testify, add this code into icoFoam:

fvVectorMatrix UEqn
        (
            fvm::ddt(U)
          + fvm::div(phi, U)
          - fvm::laplacian(nu, U)
        );

Info<<"UEqn:" << UEqn << endl;
Info<<"UEqn.A():" << UEqn.A()<< endl;
Info<<"UEqn.H():" << UEqn.H()<< endl;        
        //solve(UEqn == -fvc::grad(p));

        // --- PISO loop

Then I run a 9 nodes cavity case. It output this in the first timestep:

Create time

Create mesh for time = 0

Reading transportProperties

Reading field p

Reading field U

Reading/calculating face flux field phi


Starting time loop

Time = 0.005

Courant Number mean: 0 max: 0
UEqn:Lower triangle = 
12
(
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
)


diagonal = 9(0.00012 0.00015 0.00012 0.00015 0.00018 0.00015 0.00012 0.00015 0.00012)

Upper triangle = 
12
(
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
-3e-05
)



[0 4 -2 0 0 0 0]
9{(0 0 0)}

3
(
3((6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05))
9((6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05) (6e-05 6e-05 6e-05))
0()
)


3
(
3((6e-05 0 0) (6e-05 0 0) (6e-05 0 0))
9{(0 0 0)}
0()
)


UEqn.A():dimensions      [0 0 -1 0 0 0 0];

internalField   nonuniform List<scalar> 9(800 700 800 700 600 700 800 700 800);

boundaryField
{
    movingWall
    {
        type            zeroGradient;
    }
    fixedWalls
    {
        type            zeroGradient;
    }
    frontAndBack
    {
        type            empty;
    }
}

UEqn.H():dimensions      [0 1 -2 0 0 0 0];

internalField   nonuniform List<vector> 9((0 0 0) (0 0 0) (0 0 0) (0 0 0) (0 0 0) (0 0 0) (200 0 0) (200 0 0) (200 0 0));

boundaryField
{
    movingWall
    {
        type            zeroGradient;
    }
    fixedWalls
    {
        type            zeroGradient;
    }
    frontAndBack
    {
        type            empty;
    }
}

So in this log, Why Ueqn.A()'s internal field is not equal to the Ueqn's diag. term? I mark this two in the red.

Thanks in advance,

[sharonyue]

If A()=diag()/mesh.V(),
this is my mesh:

convertToMeters 0.03;

vertices
(
    (0 0 0)
    (1 0 0)
    (1 1 0)
    (0 1 0)
    (0 0 0.1)
    (1 0 0.1)
    (1 1 0.1)
    (0 1 0.1)
);

blocks
(
    hex (0 1 2 3 4 5 6 7) (3 3 1) simpleGrading (1 1 1)
);

So V is 0.003*0.01*0.01=0.000003.
but 0.00012/0.000003 does not equal to 800.

[Martin80]

Hi,

Ueqn.A()=diag()/mesh.V() holds true only for cells, which do not contact the boundary. E.g. in your example:

600=0.00018/(0.01*0.01*0.003)

[and be careful mesh.V()=3e-7 (not 3e-6)].

At the moment I'm also trying to clarify details about the "Ueqn.A()".

B.r. Martin.

[sharonyue]

Originally Posted by Martin80

Hi,

Ueqn.A()=diag()/mesh.V() holds true only for cells, which do not contact the boundary. E.g. in your example:

600=0.00018/(0.01*0.01*0.003)

[and be careful mesh.V()=3e-7 (not 3e-6)].

At the moment I'm also trying to clarify details about the "Ueqn.A()".

B.r. Martin.

So this is only true for inner cells, not boundary cells. Thats clear! Thx very much!