[HMQ]

Hi!
I'm very new about OpenFoam. I followed the damBreak tutorial and wanted to change it to jet flow in 3D. I meshed and set the boundary condition with Pointwise, when I checkMesh, it said mesh OK, then I setFields, it looked correctly in paraview. But when I started to compute, the max alpha.water was rapidly becoming much more than larger 1 rapidly, and I cant find the problem.
Here's the error code:



PIMPLE: iteration 1

smoothSolver: Solving for alpha.water, Initial residual = 0. 00144282, Final residual = 5.6773e+102, NO Iterations 1000

Phase-1 volume fraction = 8.32042e+103 Min(alpha .water) = -1.27119e-09 Max(alpha .water) = 4.70168e+110

MULES: Correcting alpha. water

#0 Foam: :error: rintStack (Foam: :Ostream&) at ??:?
#1 Foam: :sigFpe: :sigHandler (int) at ??:?
#2 ? in "/lib/x86_ 64- Linux-gnu/libc.so.6"
#3 Foam: : tmp<Foam : :GeometricField<double, Foam: : fvsPatchField, Foam: : surfaceMesh> > Foam: :mag<Foam: :Vector<double>, Foam: : fvsPatchField, Foam: : surfaceMesh> (Foam: :GeometricField<Foam: :Vector<double>, Foam: : fvsPatchField, Foam: : surfaceMesh>const&) at ??:?
#4 Foam: : interfaceProperties: :calculateK() at ??:?
#5 ? at ??:?
#6 _libc_start_main in "/lib/x86_ 64-linux-gnu/libc. so.6"
#7 ? at ??:?
[1] 98269 floating point exception (core dumped) interFoam |
98270 done tee log


Thanks for your help!

[gkarlsen]

Your simulation appears to be blowing up. This could be caused by any number of reasons, but from your description I would guess that you have increased the velocity without reducing the time step accordingly (high Courant number). This could cause stability issues. You might need to provide some more information if this does not help you. Please see:

How to give enough info to get help

[HMQ]

Thank you for your guidance!
I used Pointwise to mesh and set BC, than exported it into OF. I set the initial conditions accordingly like the tut damBreak and setFields, and I didn't change other factor like fvschemes or others, but it just can't work out. Before I use blockMesh to make the same mesh the same as this and it ran correctly with the value of max/min alpha.water was 1/0. I didn't know if I ignored some important step when I export the mesh?

[ChiaraViola]

Hi! Did you solve your problem? I'm running interFoam and I want to simulate a propagation of waves against a dike but I had a similar problem with high value of max (alpha.water) , say 20, and the min value is around -0.06. I had also the floating point exception error and I really don't know how to solve it.

Also, can you explain me please if you understood the meaning of a high value of alpha?I thought that the max value has to be around 1 since it means a cell full of water in the VoF method..
The wave theories I tried are StokesI and StokesII and both gave me errors after 3 seconds of simulation, instead the solitary wave gave me a first overtopping but then the solver was blocked at running the same time step for other hours... Do you have any suggestion?


Best regards,

Chiara

[luoyang]

Originally Posted by HMQ

Thank you for your guidance!
I used Pointwise to mesh and set BC, than exported it into OF. I set the initial conditions accordingly like the tut damBreak and setFields, and I didn't change other factor like fvschemes or others, but it just can't work out. Before I use blockMesh to make the same mesh the same as this and it ran correctly with the value of max/min alpha.water was 1/0. I didn't know if I ignored some important step when I export the mesh?

Hi, HMQ
I used the mesh generated from Pointwise to calculate a two-phase flow problem before. Divergence happened but I dont why. Using the blockMesh tool with the same case is ok. So I guess maybe sth is not right in the mesh exported from pointwise and I dont use it anymore...

[miotto]

I had this issue too. The problem was also caused by the mesh.
Eventhough

checkMesh -allGeometry -allTopology

was all OK, the mesh was somehow broken. I generated it again changing some minor details and the exploding alpha vanished.
I still don't know what was wrong with the first mesh.

[thanosniotis]

Hey Guys,

I have had the same problem with the interFOAM and overInterDyMFOAM.

I am trying to simulate the flooding process in the internal of the ship after a breach on her hull. The geometry is indeed complicated in the internal of the vessel.

I have generated three meshes coarse, medium and fine , and in the three cases a am facing the same problem. Form the initial time step the alpha.water value is unbounded. Order or magnitude +-1 for coarse and medium grids and +47 (!!!) for the fine one.

As far as I am concerned until now the problem exists due to the mesh. I am trying to change the geometry and to regenerate the meshes to see what is going on.

Best regards,
Thanos

[mostafa R]

Originally Posted by thanosniotis

Hey Guys,

I have had the same problem with the interFOAM and overInterDyMFOAM.

I am trying to simulate the flooding process in the internal of the ship after a breach on her hull. The geometry is indeed complicated in the internal of the vessel.

I have generated three meshes coarse, medium and fine , and in the three cases a am facing the same problem. Form the initial time step the alpha.water value is unbounded. Order or magnitude +-1 for coarse and medium grids and +47 (!!!) for the fine one.

As far as I am concerned until now the problem exists due to the mesh. I am trying to change the geometry and to regenerate the meshes to see what is going on.

Best regards,
Thanos

Hi,
Did you find the source of this problem with mesh?
I have this alpha increases with dynamicMeshDict when i want to refine the mesh at boundary o two phase.

[thanosniotis]

Originally Posted by mostafa R

Hi,
Did you find the source of this problem with mesh?
I have this alpha increases with dynamicMeshDict when I want to refine the mesh at boundary o two-phase.

Hi Mostafa,

Finally, it was not the problem with the mesh or the geometry. The problem was on the overset interpolation method. More specifically, I had least square method when I had the problem, and when I changed to inverse distance, everything worked fine.
I don t know if this helps.

Best regards,
Thanos

[mostafa R]

Originally Posted by thanosniotis

Hi Mostafa,

Finally, it was not the problem with the mesh or the geometry. The problem was on the overset interpolation method. More specifically, I had least square method when I had the problem, and when I changed to inverse distance, everything worked fine.
I don t know if this helps.

Best regards,
Thanos

Hi Thanos,

Thank you for the quick reply. I will use your advice for overset interpolation in the next step.

But before it, After investigation of all parametersin CFD, In my case, If it might be useful for anybody else, just remember to check the mesh first of all (by paraFoam etc.). I had defaultFaces in file ''constant/polyMesh/boundary'' due to inappropriate defined patches.
look at this thread:
using blockMesh - defaultFaces

However, the error is still remained and gradually increases (almost maxalpha=1.0004)...

[binyang]

Originally Posted by mostafa R

Hi Thanos,

Thank you for the quick reply. I will use your advice for overset interpolation in the next step.

But before it, After investigation of all parametersin CFD, In my case, If it might be useful for anybody else, just remember to check the mesh first of all (by paraFoam etc.). I had defaultFaces in file ''constant/polyMesh/boundary'' due to inappropriate defined patches.
look at this thread:
using blockMesh - defaultFaces

However, the error is still remained and gradually increases (almost maxalpha=1.0004)...

Hi, Do you have a solution now? I also face the same problem. Thanks

[5tern]

Hey,
I had the same problem, using Pointwise, but then remembered I forgot to renumberMesh -overwrite.
This is strongly recommended after exporting to OpenFOAM and even simple cases may fail if not done.