[EZByK]

Hello everybody,

currently I am doing steadystate RAS external aerodynamics simulation for road cars e.g (DrivAer) for my thesis with OpenFOAM v1812.

On this thread

How to plot local residuals' field (for each cell) ? OF 5.x

I found that it is possible to view the local residuals (p Ux Uy Uz ..) for each cell in Paraview, and it works perfect for p but unfortunatly not for U , k, and Epsilon.

The feature is also described here:

https://www.openfoam.com/releases/op...esidual-fields

To use the utility I added the residualFieldDict to the systems Folder. It contains the following

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  v1812                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.com                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/


residualsField
{
    type            residuals;
    libs            ("libutilityFunctionObjects.so");
    writeFields     true;
    writeControl    outputTime;
    fields          (p U k epsilon omega);
}

residuals
{
    type            residuals;
    libs            ("libutilityFunctionObjects.so");
    writeFields     true;
    writeControl    outputTime;
    fields          (p U k epsilon omega);
}
// ************************************************************************* //

As far as I know there is a need to make a include in controlDict as well.

...
functions
{
    #include "residualFieldDict"
}

So as I already mentioned this works perfectly if you want to have a look at local pressure residual for each cell. It already helped me a lot to get more refinement to the right areas for my problem in Order to achieve lower residuals. But e.g. for Ux Uy and Uz there is no data written to the file.

Here you can see the diffrence between the created files in my folder for 3000 Iteration.

initialResidual:p contains a lot of field informaion

FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    location    "3000";
    object      initialResidual:p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 0 0 0 0 0 0];


internalField   nonuniform List<scalar> 
364811
(
-4.15537e-09
-9.58299e-10
-8.97218e-09
1.77756e-09
-1.5212e-09
8.05321e-09
-2.7556e-10
-3.48638e-09
-5.27514e-09
3.91361e-09
-6.64565e-10
1.52268e-09
-5.25854e-10
-4.30708e-09
7.58053e-09
...

but if you have a look at initialResidual:Uz the file contains only this without Field information

FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    location    "3000";
    object      initialResidual:Uz;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 0 0 0 0 0 0];


internalField   uniform 0;

boundaryField
{
    symmetryWall
    {
        type            symmetry;
    }
    sideWall
    {
        type            zeroGradient;
    }
    inlet
    {
        type            zeroGradient;
    }
    outlet
    {
        type            zeroGradient;
    }
    lowerWall
    {
        type            zeroGradient;
    }
    upperWall
    {
        type            zeroGradient;
    }
    wheel_FL_wheel
    {
        type            zeroGradient;
    }
    body_body_lights_front
    {
        type            zeroGradient;
    }

Is there anybody who can see if I did something wrong setting up this feature?
Would be nice if you can give me related informations to solve this Problem.

Thanks for your help,

Eric

[Andrea159357]

have you solved your problem? i would like to see cell's residual too.

[EZByK]

Hi Andrea,

I had no progress on this one so far, but maybe a more advanced user can have a look at this or provide another strategie to get these kind of information.

There are several options:

- I am using the tool in the wrong way
- I did a small typo somewhere
- It is a bug in the tool (but I dont think so)

So as I mentioned before, if you want to have a look at pressure residual it will be fine but not for U k epsilon ...

Did you have the same result Andrea? Would be nice if some can test this on their own calculation and share experience.

Have a nice day!

[marluc]

Hi,


I think the reason lies in the solver used for variables other than p, which is usually (and presumably in your case) solved with GAMG. If you solve "U, k, omega, ..." with PBiCGStab or PBiCG than the residual field of every can be stored using solverInfo:

    residuals
    {
        type            solverInfo;
        libs            ("libutilityFunctionObjects.so");
        writeResidualFields yes;
        writeControl    writeTime;
        fields          (".*");
    }

Best,
Luca

[tas38]

I was recently having some difficulty writing out the residual fields (to visualize in paraview), but was unable to do so.


In v1912, the "residuals" utility function object has been replaced by "solverInfo" as mentioned above. However, even with the "writeResidualFields" set to yes/true, the fields were not being written.


The issue appears to be with the choice of linear solver. I could only get the residual fields to be written using GAMG and NOT PBiCG or smoothSolver.



I do not have an explanation as to why the residual fields are only written for GAMG.

[HPE]

You can follow the discussions for the fix:

solverInfo functionObject: writeResidualFields only works for pressure and not for other fields