[Roxxor91]

Hello Foamers.

I am running pimpleFoam on a domain composed of an unmeshed box (called box) surrounded by 6 patches (namely inlet, outlet, top, bottom, emptyFaces). I have access to the pressure field on the patch as a surfaceScalarField on the box patch.

I am also running an in-house OpenFOAM solid solver on a box domain (also called box) that has the same cartesian coordinates than the box in the first case. I want to map the pressure field from the first case to that second case.

As a first attempt, I wrote a coded functionObject that writes the pressure values on the box patch of the first case (it is a surfaceScalarField even if it is saved as a simple scalarField). The problem is the box patch from the first case has a different number of points than the box patch from the second case, and most likely the points are organised in a different way.

                                label patchID = mesh().boundaryMesh().findPatchID("box");

                                IOField<scalar> pOut
                                (
                                        IOobject
                                        (
                                                "pBox",
                                                mesh().time().timeName(),
                                                mesh(),
                                                IOobject::NO_READ,
                                                IOobject::NO_WRITE
                                        ),
                                        p.boundaryField()[ patchID ]
                                );

So my second idea is to use mapFields to perform a non-consistent mapping. The dictionnary system/mapFieldsDict looks like this

/*--------------------------------*- C++ -*----------------------------------*\
=========                 |
\\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
 \\    /   O peration     | Website:  https://openfoam.org
  \\  /    A nd           | Version:  7
   \\/     M anipulation  |
\*---------------------------------------------------------------------------*/
FoamFile
{
version     2.0;
format      ascii;
class       dictionary;
location    "system";
object      mapFieldsDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

patchMap        ( box box);

cuttingPatches  ();


// ************************************************************************* //

and I use the command

mapFields ../Fluid -case . -sourceTime 0.1

The log file is

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Source: "/mnt/datasim/OpenFOAM/solvers/test" "Fluid"
Target: "/mnt/datasim/OpenFOAM/solvers/test" "Solid"

Create databases as time

Source time: 0.1
Target time: 0.25
Create meshes

Source mesh size: 2400  Target mesh size: 300


Mapping fields for time 0.1

    interpolating p

End

and when I look at the directory 0.1/ I can see that no new p field has been created. Moreover I noticed that the source time is not the target time.

Note: For testing purpose, I made both cases run until 0.25 which is probably my target is 0.25. It is not my main concern at the moment because eventually, I want to perform this mapping at every (written) timestep.

Do you have an idea on how to properly use mapFields ?

EDIT: I use OpenFOAM-6 on Debian.