InterFace mismatch in solids4Foam toolkit for HronTurekFSI3
 

InterFace mismatch in solids4Foam toolkit for HronTurekFSI3

Hi...
a simple ./Allrun of the ...solids4foam/tutorials/fluidSolidInteraction/HronTurekFsi3/laminar shows unresolved interface. See figure => Attachment 71859.
However, the same case runs fine in fsiFoam solver. What is the cause of such issues?




I also facing such issues in fsiFoam as well for a different geometry (see figure Attachment 71860).

Please help on how to rectify that?

Hi Guarav,

Are you using solids4foam with foam-extend-4.0 or foam-extend-4.1?

Philip

I am running on foam-extend-4.0 for both solids4foam and fsiFoam.

Did you use the "Warp By Vector" filter in ParaView on the solid sub-domain to show its deformed position? It should be used with the total displacement vector field (D or pointD) and a Scale Factor of 1.

This is needed for any total Lagrangian style solid models which do not move the mesh. The "unsNonLinearGeometryTotalLagrangian" solid model is used by default in the HronTurekFSI3/laminar case.

Philip

Hi again Guarav,

Just to confirm, I have checked this case on the latest master branch of solids4foam and it works as expected.

If you still have issues, please give more details.

Philip

Hello Phillip,

Sorry for late response,
Yes using "Warp by vector" works well and good.
There are no issues for now.

Thank You.

Hello

I have problems with the tutorial: HronTurekFsi3. The following appears in the compilation..

"/home/angel/OpenFOAM/angel-6/FluidSolidInteraction/src/fluidSolidInteraction/lnInclude/constitutiveModel.H:52:10: fatal error: IOReferencer.H: Arquivo ou diretório inexistente
#include "IOReferencer.H"
^~~~~~~~~~~~~~~~
compilation terminated.
/opt/openfoam6/wmake/rules/General/transform:25: recipe for target 'Make/linux64GccDPInt32Opt/hronTurekReport.o' failed
make: *** [Make/linux64GccDPInt32Opt/hronTurekReport.o] Error 1
Running fsiFoam on /home/angel/OpenFOAM/angel-6/FluidSolidInteraction/run/fsiFoam/HronTurekFsi3/fluid"

Could someone guide me:

Thank you

Angel

Has the package compiled successfully?

I suggest you try solids4foam instead, as the extend bazaar FSI toolbox i sno longer maintained.

Kind regards,
Philip

No. No....Apply the procedure explained on the following page:

http://openfoamwiki.net/index.php/Ex...tutorial_cases


Try to execute the 3dTube case with the steps explained on the previous page. When I open for View the following error appears:

"angel@angel-virtual-machine:~/OpenFOAM/angel-6/FluidSolidInteraction/run/fsiFoam/3dTube/fluid$ paraFoam
Created temporary 'fluid.OpenFOAM'
I/O : uncollated


--> FOAM FATAL IO ERROR:
Essential entry 'value' missing

file: /home/angel/OpenFOAM/angel-6/FluidSolidInteraction/run/fsiFoam/3dTube/fluid/0/p.boundaryField.inlet from line 38 to line 40.

From function Foam::fvPatchField<Type>::fvPatchField(const Foam::fvPatch&, const Foam::DimensionedField<Type, Foam::volMesh>&, const Foam::dictionary&, bool) [with Type = double]
in file /home/ubuntu/OpenFOAM/OpenFOAM-6/src/finiteVolume/lnInclude/fvPatchField.C at line 131.

FOAM exiting

Segmentation fault (core dumped)"

I installed foam-extend-4.0 in OpenFoam 6.0

I need help:(

Thank you

Hi:),
What:eek:?!
foam-extend-4.0 is a separate distribution of openfoam, not just an extension for the current version that you have(namely v6).
Did you activate foam-extend-4.0 "alias" in your terminal before compilation of fsi toolkit? Note that this toolkit can not be compiled under openfoam 6. Finally, you can use solids4Foam rather than fsi toolkit (read this).

Best,
Ali

Hi all,

Recently, I installed solids4foam under foam-extend-4.1 (running Ubuntu 18.04). I only got to testing things today, and with the HronTurekFsi3/laminar tutorial, I get warnings/errors from blockMesh, setsToZones, etc:

Which makes sense, because the directory structure is different, with fluid and solid sub-directories in system. Apart from this, the solver seems to be running fine up to now. Should I be worried about these warnings/errors, and is there a way to get rid of them?

Many thanks,
Sita


P.S. I'm also getting warnings in the solver logfile:
What's this about? And again, should I be worried?

As you have already stated, those errors/warnings are completely harmless.
This is because foam-extend-4.1 now outputs fvSchemes and fvSolution dictionaries at the beginning of each simulation and due the different structure of fsi cases it fails to find those files in system directory. One workaround would be to simply copy fvSchemes and fvSolution to the system directory.

No need to worry about this either as updateMovingWallPressureAcceleration() function is disabled on purpose.

Ah, that's good to know, thanks a million!

Hello, Sita

I installed OpenFoam-16-ext on Ubuntu 14.04. It works excellently;). Then I installed or solver miniGeotechFoam that was developed by Tian Tang in 2013 for this extended version of OpenFoam. . I performed some simulations of porous media to verify Tersagui and Mandel. Now I am checking cases of soils with elasto-plastic behavior with this solver. In the future I will place some of my results in this forum ..

Thank you..

Atte .. Angel

Hi everyone,

When executing the Allrun script in the HronTurekFSI3 tutorial, it crashes at Time = 3.151, on a floating point exception (Courant number max. 222.16). I'm using solids4foam under foam-extend-4.1.

Also, when viewing the results in ParaView, it gives me errors about a duplicate entry D.

Did anyone else experience the same? Am I doing something wrong?

Many thanks,
Sita

Unfortunately, I think there seems to be a problem with the new algorithm implemented in foam-extend-4.1 or maybe the test case needs some modifications to be compatible. The HronTurekFSI3 test case runs fine on foam-extend-4.0, but again pimpleFluid fails after 3 seconds(simulation time).

I'm not receiving this error, how do you open the case?

Hi Daniel,

Thanks for getting back. Good to know that you're getting the same error with the HronTurekFSI3 tutorial.

The errors in ParaView may be due to an installation problem on my side; somehow I couldn't get ParaView installed correctly during the foam-extend-4.1 install, so I installed it separately and told foam-extend to use my system's ParaView, which in turn is giving trouble with the PVFoamReader, causing ParaView to use a VTK reader to open my foam-extend results (see also this thread). But anyway, good to know that you're not getting these errors, that means that I should look further into my paraFoam install.

Cheers,
Sita

update on HronTurekFSI3 divergence:

I have tried several cases with pisoFoam and pimpleFoam on foam-extend-4.1, and interestingly they all do diverge immediately. Even, foam-extend-4.1 pisoFoam tutorial pitzDaily diverges instantly.

I have also created a fluid only case based on HronTurekFSI3 test case and here are the results:

• foam-extend-4.1

1- pisoFoam: diverged
2- pimpleFoam: diverged, increasing nOuterCorrectors will prevent divergence but solution is wrong.

• foam-extend-4.0

1- pisoFoam: successful
2- pimpleFoam: does not diverge, but the vortex shedding is totally dampened out and it is identical to a stead-state solution.

So, this seems to be a bug with foam-extend itself.
Also found this thread describing the exact same problem.

Regards,
D. Khazaei

solids4foam query for setSet and parallelization.
 

Hi phil,

I noticed two following issues:

1.
In the latest dev/master branch for the FSI cases(say HronTurekFsi3), I see that "setSet -batch ..." has been avoided, but follows the command "setstoZones -noFlipMap" which logs to
I could not understand this if it is intentional?

2.
On running parallel, the makelinks file and command is missing. This causes the "solids" files to not generate along with fluid in different processor directory.
> On reconstructing only fluid part gets reconstructed, which causes non-compliant interface in paraview (not even by warp by vector) because the solid fields doesn't load.

Thank you in advance.