added tutorial for foam-extend

[hoehnp]

I added a modified version of the quickstart tutorial for foam-extend with the purpose of testing the foam-extend adapter.

[MakisH]

Thanks for the contribution! A first quick review of the changes, without having tried to run the code.

We generally have alternatives for the perpendicular-flap tutorial (which is meant to be something to copy from), not for the Quickstart (which is meant to be a specific first case to run). With that said, I do understand that this is a minimal case for you.

Since the changes are very minimal, I wonder if we could make the original case compatible with foam-extend, even if we just applied a small patch automatically. Some of the changes for foam-extend should still work with OpenFOAM (e.g., the noSlip/zeroGradient or the path of the preCICE config) or not be that relevant (e.g., the forces function object).

---

List of differences compared to fluid-openfoam:

0/U:

- type            noSlip;
+ type            zeroGradient;

dynamicMeshDict:

- motionSolverLibs ("libfvMotionSolvers.so");
+ motionSolverLibs ("libfvMotionSolver.so");

+ diffusivity uniform;

controlDict:

  libs
  (
+    "libforces.so"
+    "liblduSolvers.so"
     "libpreciceAdapterFunctionObject.so"
  );

• forces function object configuration (do we actually need that?).

fvSchemes:

-     div((nuEff*dev2(T(grad(U))))) Gauss linear;
+     div((nuEff*dev(T(grad(U))))) Gauss linear;

fvSolution:

     p
     {
         solver           GAMG;
         tolerance        1e-6;
         relTol           1e-4;
         smoother         DICGaussSeidel;
+        agglomerator    faceAreaPair;
+        nCellsInCoarsestLevel 10;
+        mergeLevels 1;
     }

preciceDict:

- preciceConfig "../precice-config.xml";
+ preciceConfig ".."/"precice-config.xml";

[MakisH]

For easier maintenance, it would be good if this file followed the same order and formatting as the one in fluid-openfoam, so that we can easily take a diff.

[MakisH]

This should not be needed.

[MakisH]

Is this explicitly needed?

[hoehnp]

yes, without it, it was not running.

[hoehnp]

I start to wonder if this only applies to the debug version for some reason.

[MakisH]

But is this even a valid combination? If adjustTimeStep is disabled, what effect should maxCo have? We could still set one with a comment that it is not used and only there for compatibility reasons.