[7islands]

Too bad it didn't work for you, guessing from our results that could well be the case. Things are too picky...!

T

[Thriceguy]

Hey Takuy,

that is indeed pretty worrying. I will test it with my icoFsiFoam-solver and report, if I find the same problem. I actually built it from scratch, referring to Karl Jacob Maus manual. I guess it yet is exactly the same as jasaks solver.

@matteo, I am trying very hard to run icoFsiFoam parallel. However, I am very new to all the parallelization stuff in openFoam and hence find it hard to start.

If I decompose both the fluid and the structure to the same numbers of processors from the .fluid-folder, the solver starts, but stops immediately before having finished the first time step.

Courant Number mean: 0.01309 max: 0.12 velocity magnitude: 4
Time = 0.0003

Setting pressure
Total pressure force = (0 0 0)
PCG:  Solving for Ux, Initial residual = 0, Final residual = 0, No Iterations 0
PCG:  Solving for Uy, Initial residual = 0, Final residual = 0, No Iterations 0
*** Process received signal ***
Signal: Segmentation fault (11)
 Signal code:  (-6)
Failing at address: 0x3ea0000081a
[ 0] /lib/libc.so.6(+0x33af0) [0x7f7a7dbf2af0]
[ 1] /lib/libc.so.6(gsignal+0x35) [0x7f7a7dbf2a75]
[ 2] /lib/libc.so.6(+0x33af0) [0x7f7a7dbf2af0]
[ 3] /lib/libc.so.6(+0x7b726) [0x7f7a7dc3a726]
[ 4] /lib/libc.so.6(__libc_malloc+0x6e) [0x7f7a7dc3b58e]
[ 5] /usr/lib/libstdc++.so.6(_Znwm+0x1d) [0x7f7a7e4a726d]
[ 6] /usr/lib/libstdc++.so.6(_Znam+0x9) [0x7f7a7e4a7389]
[ 7] /home/cfd/OpenFOAM/OpenFOAM-1.5-dev/lib/linux64GccDPOpt/libfaceDecompFiniteElement.so(_ZN4Foam4ListIcE7setSizeEi+0x33) [0x7f7a7f6b62d3]
[ 8] /home/cfd/OpenFOAM/OpenFOAM-1.5-dev/lib/linux64GccDPOpt/libOpenFOAM.so(_ZN4Foam8OPstreamC1ENS_7Pstream10commsTypesEiiNS_8IOstream12streamFormatENS3_13versionNumberE+0xcc) [0x7f7a7ead33fc]
[bionik205:02074] [ 9] /home/cfd/OpenFOAM/OpenFOAM-1.5-dev/lib/linux64GccDPOpt/libOpenFOAM.so(_ZN4Foam18processorPolyPatch12initGeometryEv+0xb7) [0x7f7a7eb9a427]
[10] /home/cfd/OpenFOAM/OpenFOAM-1.5-dev/lib/linux64GccDPOpt/libOpenFOAM.so(_ZN4Foam16polyBoundaryMesh10movePointsERKNS_5FieldINS_6VectorIdEEEE+0x109) [0x7f7a7ebc2209]
[11] /home/cfd/OpenFOAM/OpenFOAM-1.5-dev/lib/linux64GccDPOpt/libOpenFOAM.so(_ZN4Foam8polyMesh10movePointsERKNS_5FieldINS_6VectorIdEEEE+0x243) [0x7f7a7ebfadf3]
[bionik205:02074] [12] /home/cfd/OpenFOAM/OpenFOAM-1.5-dev/lib/linux64GccDPOpt/libfiniteVolume.so(_ZN4Foam6fvMesh10movePointsERKNS_5FieldINS_6VectorIdEEEE+0x141) [0x7f7a804e39d1]
[13] icoFsiFoam() [0x43c2c6]
[14] /lib/libc.so.6(__libc_start_main+0xfd) [0x7f7a7dbddc4d]
[15] icoFsiFoam() [0x425669]
*** End of error message ***

Although not having done very much, this is about what I am able to do, withoug getting any hints. I do not even have an idea where to continue. Any hints would be highly appreciated.

You mentioned that the patch-to-patch interpolation is the problem. However, googling did not result in any useful information, so if you have got some suggestions, that'd be great!

Kind regards.
*D

[7islands]

Hi Thriceguy,
In fact we're finding the problem is not so simple and straightforward as I've reported (fixing the surface-to-surface interpolation caused another problem).

In any case 1.6-ext has just been released; let's see how the "Updated fluid-structure interaction solver with parallelisation support" works

Takuya

[Thriceguy]

Hey Takuya,

that's good news. I will run some simulations with it, too. We'll see and hope the best ;-)!

David

[deepblue17]

Hi,

are there any updates with the parallelization of the icoFsiFoam solver? I tried the last version from the 1.6-ext git and the flappingConsole still does not run in parallel.

I tested where the error occured and think this is already in the volPointInterpolation (cpi) of the setMotion.H.

The patchToPatch interpolation also does not work in parallel, but we could switch over to a directMappedBased boundary condition (like the turbulentTemperatureCoupledBaffleMixed which is used in CHTMultiRegionFoam - this one works also in parallel) in order to couple fluid and solid patches.

Any idea how to get this solver working in parallel?

[PeterII]

Hi All
I'm struggling with some problems running the tutorials for icoFsiFoam solver (flappingConsoleSmall) on openfoam-1.6-ext.
I recive this message int the terminal window:
>/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall$ ./Allrun
>Running blockMesh on/home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/ flappingConsoleSmall/fluid
>Running blockMesh on /home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/ flappingConsoleSmall/solid
/home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall
>./Allrun: 23: makeFsiCaseLinks: not found
>Running icoFsiFoam on /home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall/fluid
>/Allrun: 27: makeFsiResultsLinks: not found
and log files are:
1) log.blockMesh
Default patch type set to empty
--> FOAM Warning :
From function polyMesh:olyMesh(... construct from shapes...)
in file meshes/polyMesh/polyMeshFromShapeMesh.C at line 583
Found 10 undefined faces in mesh; adding to default patch.


2) log.icofsiFoam
--> FOAM FATAL IO ERROR:
keyword divSchemes is undefined in dictionary "/home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall/fluid/system/solid/fvSchemes"
file: /home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall/fluid/system/solid/fvSchemes
From function dictionary::subDict(const word& keyword) const
in file db/dictionary/dictionary.C at line 457.

FOAM exiting

• can anyone gave me help?

[mvoss]

looks like blockMesh came up with the error in the first place.

[ZKM]

@ PeterII

You need to take the latest RunFunctions from
http://openfoam-extend.svn.sourcefor...ext/bin/tools/
and replace it with your RunFunctions in bin/tools/
Moreover comment out these lines in Allrun

#cd constant
#ln -s ../../solid/constant solid
#cd ..
#cd 0
#ln -s ../../solid/0 solid
#cd ..

[PeterII]

Thank you (ZKM) so much for this help. However the blockMesh log file show the this error:
Creating block mesh topology
Default patch type set to empty
--> FOAM Warning :
From function polyMesh:olyMesh(... construct from shapes...)
in file meshes/polyMesh/polyMeshFromShapeMesh.C at line 583
Found 10 undefined faces in mesh; adding to default patch.
Check block mesh topology


Now I want to model the deformation of the blood vessel under a pusatil pressure and anny helpe will be appreciate.

[matz]

Hi everybody,

I'm new to this Forum, but have some experience with OpenFOAM.

I'm interested in Fluid Structure Interaction and am currently testing the icoFsiFoam-Solver from the OF-1.6-ext version. In serial mode the tutorial and several other cases are working fine, but if it goes to parallel the results do not match the serial ones. The deformation of the solid in parallel (in some cases) is far less than expected.

The problem occurs when the fluid/solid patch is on 2 or more processors. If the fluid/solid patch is only on one processor the parallel and serial cases match perfectly.

My questions are, if somebody else has the same problem and if there is any solution?

Thanks in advance!

Best regards,
Daniel

PS: The Problem can easily be reproduced with the standard tutorial for icoFsiFoam (flappingConsoleSmall).
For the decomposeParDict the following can be used (it is only important that the fluid/solid patch is split!):


numberOfSubdomains 4;

method simple;

simpleCoeffs
{
n (1 4 1 );
delta 0.001;
}

[charbel]

Hi everyone,

You can find here a new solver that I am trying to make for fluid-structure interaction.
I am working on version 1.6-ext.

This solver included:
1/ pimpleDyMFoam in which we have ALE and turbulence models can be used
2/ outer fixed point adaptive relaxation by using Aitken's method for strong coupling
3/ solve transport equations for heat and mass transfer
4/ solid solver for large deformattion (2nd Piola-Krichhoff stress tensor...)
5/ computing traction in setPressure.H but it is not taken into account in tractionDisplacementFvPatchVectorField.C can you give some hint please

Can you try this solver on the attached testCase and give me remarks because I am sure there will be some mistakes especially for:
1/ Aitken's method
2/ solid solver and its part in tractionDisplacementFvPatchVectorField.C
3/ setting traction in tractionDisplacementFvPatchVectorField.C

Thank you in advance.
Charbel.

[gascortado]

Originally Posted by PeterII

Hi All
I'm struggling with some problems running the tutorials for icoFsiFoam solver (flappingConsoleSmall) on openfoam-1.6-ext.
I recive this message int the terminal window:
>/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall$ ./Allrun
>Running blockMesh on/home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/ flappingConsoleSmall/fluid
>Running blockMesh on /home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/ flappingConsoleSmall/solid
/home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall
>./Allrun: 23: makeFsiCaseLinks: not found
>Running icoFsiFoam on /home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall/fluid
>/Allrun: 27: makeFsiResultsLinks: not found
and log files are:
1) log.blockMesh
Default patch type set to empty
--> FOAM Warning :
From function polyMesh:olyMesh(... construct from shapes...)
in file meshes/polyMesh/polyMeshFromShapeMesh.C at line 583
Found 10 undefined faces in mesh; adding to default patch.


2) log.icofsiFoam
--> FOAM FATAL IO ERROR:
keyword divSchemes is undefined in dictionary "/home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall/fluid/system/solid/fvSchemes"
file: /home/khaled/OpenFOAM/of16ext/stressAnalysis/icoFsiFoam/flappingConsoleSmall/fluid/system/solid/fvSchemes
From function dictionary::subDict(const word& keyword) const
in file db/dictionary/dictionary.C at line 457.

FOAM exiting

• can anyone gave me help?

I have the exact same problem. Does anyone knows how to solve this one?? I will really appreciate any help. Thanks

[soheil nazmdeh]

Hi ,
I try to solve my case with icoFsiFOAM solver , but in my geometry this eroor accur when try to solve it :

.....
Moving mesh time step continuity errors : sum local = -92209, global = 1.98903, cumulative = 2.86326
CG: Solving for p, Initial residual = 0.999957, Final residual = 7.0708e-07, No Iterations 526
CG: Solving for p, Initial residual = 0.999109, Final residual = 0.000723223, No Iterations 1000
Moving mesh time step continuity errors : sum local = 2.07627e+14, global = 2.26112e+10, cumulative = 2.26112e+10
Floating point exception (core dumped)

please help me to solve it...!

[Maimouna]

Dear All,

could you help me please where can I find icoFsiFoam solver in OpenFOAM 2.2.1. I searched many times but unfortunately I didn't get any result. Also, I tried to make my FSI directory but the same it is not working.

Help me please to find the answer.

Regards

[msuaeronautics]

Originally Posted by Maimouna

Dear All,

could you help me please where can I find icoFsiFoam solver in OpenFOAM 2.2.1. I searched many times but unfortunately I didn't get any result. Also, I tried to make my FSI directory but the same it is not working.

Help me please to find the answer.

Regards

You will have to download OF1.6-ext in order to use icoFsiFoam. The solver isn't officially available in any other iteration.

[Maimouna]

Many thanks for your reply. I have some questions regarding your answer
1. Have I install OpenFOAM 1.6-ext inside OpenFOAM 2.2.1 that I had installed before or install it separately?

2. During installation I got some error, how can I re-install it again?

3. I got locked folders, OpenFOAM 1.6-ext locked, why?


4. How to work with fluid and solid in parallel in cases such that flappingConsoleSmall?

Many thanks for your helping.

[arashfluid]

Hi
Iam using pimpleFsiFoam solver for pitching motion of naca0012 airfoil in incompressible and turbulence flow.My problem is weak coupling. PimpleFsiFoam is a solver based on FEM. So for moving the airfoil by pointDisplacement, I converted it to FVM. I create 2 meshes in gambit, one for the fluid and another one with the airfoil only as my solidMesh. Then convert them using fluentmeshToFoam. Various mesh for y + <1 is tried and boundary conditions for U and P are:
U:

dimensions      [0 1 -1 0 0 0 0];
  internalField   uniform (13.7873 2.431 0);
  boundaryField
  {   
      outlet
      {
          type            zeroGradient;
       }
      inlet
      {
          type            fixedValue;
          value           $internalField;
      }
      airfoil
      {
         type            movingWallVelocity;
          value           uniform (0 0 0);
      }

P:

dimensions      [0 2 -2 0 0 0 0];
  internalField   uniform 0;
  boundaryField
  {   
       outlet
        {
            type            fixedValue;
            value           uniform 0;
        }
       inlet       
        {
          type            zeroGradient;
        }
       airfoil        
        {
          type            zeroGradient;
        }

When I run the case I get the Floating point error.
Do Solver wrong acts or the case is problem?
Geometry with coarse mesh and without motion and laminar flow was ran only by icoFsiFoam openfoam tutorial solver. For turbulent flow,I convert icoFsiFoam to pisoFsiFoam. But again, I fail.
How do I apply this motion in the FEM solver?

[codder]

Seeing a similar error PeterII experienced, and ZKM suggested a solution for.

Different situation - in that I'm adapting the icoFsiFoam solver to use a different solid domain solver. The error I see is:

--> FOAM FATAL IO ERROR: 
keyword grad(U) is undefined in dictionary "/home/eric/foam/eric-3.0/run/       applications/solvers/icoFsiElasticAcpSolidFoam/tutorial/fluid/system/fvSchemes::gradSchemes"

... where grad(U) needs to be defined in the solid but not fluid fvSchemes.

So could anyone shed some light on the underlying reason for ZKM's proposed solution:

You need to take the latest RunFunctions from
http://openfoam-extend.svn.sourcefor...ext/bin/tools/
and replace it with your RunFunctions in bin/tools/
Moreover comment out these lines in Allrun

#cd constant
#ln -s ../../solid/constant solid
#cd ..
#cd 0
#ln -s ../../solid/0 solid
#cd ..

... I'm guessing this is a linking issue?

[hassan1201]

I solved HronTurekFsi case in OF- Extend 3.1
for many time steps but i stoped the solver to show it in paraFoam.
It appears fluid domain only without solid

[arashfluid]

Hi friends,
I could impose motion on solid patch by applying oscillatingFixedValue BC in motionU file and solid moves properly.But unlike the solid, fluid patch moves in the opposite direction of solid patch. Solver is icoFsiElasticNonLin... in foam-extend-3.1. My test case is uploaded on:
http://www.uploadbaz.com/f4xg5rf6b2rv

Where is the problem? Can someone help me?

Christoph, I have a specific question from you. I want to impose a forced oscillatory motion on solid,According to your thread in post #10. Can you tell me how did you apply that motion?