[michael157]

Hello,
i have a problem in solving a case with chtMultiRegionFoam. I get an error message and I can not undertand what the problem is.
I have created an easy case with a simple geometry and this case is running without any problems.
Then i replaced the "easy" geometry with a complex one and the case started to run but crashes after some seconds. The following error message i got:

[...]
Solving for solid region platte
DICPCG: Solving for h, Initial residual = 1, Final residual = 3.102837e-07, No Iterations 5
Min/max T:min(T) [0 0 0 1 0 0 0] 708.3925 max(T) [0 0 0 1 0 0 0] 790.7728
ExecutionTime = 166.2 s ClockTime = 234 s

Region: air Courant Number mean: 0.5470849 max: 7.974256
Region: platte Diffusion Number mean: 0.0001257544 max: 0.0001329767
deltaT = 0.0002029867
Time = 0.00200155


Solving for fluid region air
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
DILUPBiCG: Solving for Ux, Initial residual = 0.2788012, Final residual = 8.959984e-07, No Iterations 37
DILUPBiCG: Solving for Uy, Initial residual = 0.1237281, Final residual = 8.137486e-07, No Iterations 36
DILUPBiCG: Solving for Uz, Initial residual = 0.2438061, Final residual = 8.851508e-07, No Iterations 36
DILUPBiCG: Solving for h, Initial residual = 0.2280842, Final residual = 7.831367e-07, No Iterations 36
Min/max T:86.47079 790.6948
GAMG: Solving for p_rgh, Initial residual = 0.09683618, Final residual = 0.0007611302, No Iterations 3
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors (air): sum local = 3.468532e-05, global = -4.339116e-06, cumulative = -1.073209e-05
[1] #0 Foam::error:rintStack(Foam::Ostream&) at ??:?
[1] #1 Foam::sigFpe::sigHandler(int) at ??:?
[1] #2 in "/lib/x86_64-linux-gnu/libc.so.6"
[1] #3 Foam::GAMGSolver::scale(Foam::Field<double>&, Foam::Field<double>&, Foam::lduMatrix const&, Foam::FieldField<Foam::Field, double> const&, Foam::UPtrList<Foam::lduInterfaceField const> const&, Foam::Field<double> const&, unsigned char) const at ??:?
[1] #4 Foam::GAMGSolver::Vcycle(Foam::PtrList<Foam::lduMa trix::smoother> const&, Foam::Field<double>&, Foam::Field<double> const&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::PtrList<Foam::Field<double> >&, Foam::PtrList<Foam::Field<double> >&, unsigned char) const at ??:?
[1] #5 Foam::GAMGSolver::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const at ??:?
[1] #6 Foam::fvMatrix<double>::solveSegregated(Foam::dict ionary const&) at ??:?
[1] #7 Foam::fvMatrix<double>::solve(Foam::dictionary const&) at ??:?
[1] #8
[1] at ??:?
[1] #9 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
[1] #10
[1] at ??:?
[TP-T430s-MD-LX:14212] *** Process received signal ***
[TP-T430s-MD-LX:14212] Signal: Floating point exception (8)
[TP-T430s-MD-LX:14212] Signal code: (-6)
[TP-T430s-MD-LX:14212] Failing at address: 0x3e800003784
[TP-T430s-MD-LX:14212] [ 0] /lib/x86_64-linux-gnu/libc.so.6(+0x36c30) [0x7fd317fc3c30]
[TP-T430s-MD-LX:14212] [ 1] /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x39) [0x7fd317fc3bb9]
[TP-T430s-MD-LX:14212] [ 2] /lib/x86_64-linux-gnu/libc.so.6(+0x36c30) [0x7fd317fc3c30]
[TP-T430s-MD-LX:14212] [ 3] /home/michael/OpenFOAM/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so(_ZNK4Foam10GAMGSolver5scaleERNS_5Fi eldIdEES3_RKNS_9lduMatrixERKNS_10FieldFieldIS1_dEE RKNS_8UPtrListIKNS_17lduInterfaceFieldEEERKS2_h+0x be) [0x7fd31910eb6e]
[TP-T430s-MD-LX:14212] [ 4] /home/michael/OpenFOAM/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so(_ZNK4Foam10GAMGSolver6VcycleERKNS_7 PtrListINS_9lduMatrix8smootherEEERNS_5FieldIdEERKS 8_S9_S9_S9_S9_S9_RNS1_IS8_EESD_h+0x222d) [0x7fd31911341d]
[TP-T430s-MD-LX:14212] [ 5] /home/michael/OpenFOAM/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so(_ZNK4Foam10GAMGSolver5solveERNS_5Fi eldIdEERKS2_h+0x4ae) [0x7fd319113ffe]
[TP-T430s-MD-LX:14212] [ 6] /home/michael/OpenFOAM/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libfiniteVolume.so(_ZN4Foam8fvMatrixIdE15solveSegr egatedERKNS_10dictionaryE+0x132) [0x7fd31ae00122]
[TP-T430s-MD-LX:14212] [ 7] chtMultiRegionFoam(_ZN4Foam8fvMatrixIdE5solveERKNS _10dictionaryE+0x14f) [0x48968f]
[TP-T430s-MD-LX:14212] [ 8] chtMultiRegionFoam() [0x431872]
[TP-T430s-MD-LX:14212] [ 9] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7fd317faeec5]
[TP-T430s-MD-LX:14212] [10] chtMultiRegionFoam() [0x434fee]
[TP-T430s-MD-LX:14212] *** End of error message ***
--------------------------------------------------------------------------
mpirun noticed that process rank 1 with PID 14212 on node TP-T430s-MD-LX exited on signal 8 (Floating point exception).
--------------------------------------------------------------------------

[laurentD]

Hello Michael,
can you tell us more about your timestep please ?
Do you define a maximal value for the Courant number or do you fix a value for the timestep, or maybe another thing ?
I am currently working with chtMultiRegionFoam and actually the value of the courant number is a key value.
Maybe a Courant number greater than 7 is too much in the case you are simulating.
In three words : tell us more.

Laurent

[michael157]

Hi Laurent,
i have specified a max. Courant number of 0.9 so this should not be the problem.

Here is what i have defined in the controlDict file:

application chtMultiRegionFoam;

startFrom latestTime;

startTime 0.001;

stopAt endTime;

endTime 600;

deltaT 0.001;

writeControl adjustableRunTime;

writeInterval 2;

purgeWrite 0;

writeFormat ascii;

writePrecision 7;

writeCompression off;

timeFormat general;

timePrecision 6;

runTimeModifiable true;

maxCo 0.9;

maxDi 10.0;

adjustTimeStep yes;

And to give more information about my case the attached picture is showing the case. Blue arrows are the inlet (minY), velocity BC with a speed of 15m/s
BC specified in the changeDict, the opposite side (maxY)outlet.
U
{
internalField uniform (0 0 0);

boundaryField
{
".*"
{
type fixedValue;
value uniform (0 0 0);
}
minY
{
type fixedValue;
value uniform (0 15 0);
}

maxY
{
type inletOutlet;
inletValue uniform (0 0 0);
value uniform (0 15 0);
}
}
The other boundary are wall BC.

[laurentD]

Hi Michael,

I understand your message but if you limit the value of the Courant number with maxCo = 0.9, you should not have Courant numbers equals to 7.9 as we can see on your results.
Your controlDict seems good but i think you have to put a lower value for the timestep. You put deltaT = 0.001 but i think in your case you should use deltaT = 1e-4. With it OpenFOAM will adjust the timestep to have Courant lower than 0.9. Thanks to the command :
adjustTimeStep true,
it should work.
Tell me what happens please, it's interesting for me too.
Laurent

[laurentD]

Another question :
At which iteration does the error appear ?
The error you copied in the first message, when did she occur ?
If this is not the first iteration, can you tell me what are the previous values of Courant and timesteps ?

[michael157]

Hi Laurent,
I have copied all what i got from the log file. It seems for me that the error occur not in the first iteration.
I set the time step to 1e-5 and I have the same error at the same stage. It seems that the Temperature is the problem, see Min/max T:86.47079 790.6948 in the logfile. The Temperature at the case with time step 1e-5 is Min/Max -876 1968.
--------------------------------------------------------------------------------------
No finite volume options present

Region: air Courant Number mean: 0.02518538 max: 0.5
Region: platte Diffusion Number mean: 6.991947e-05 max: 7.393502e-05
deltaT = 0.001798561
Region: air Courant Number mean: 0.04529745 max: 0.8992806
Region: platte Diffusion Number mean: 0.0001257544 max: 0.0001329767
deltaT = 0.001798561
Time = 0.00179856


Solving for fluid region air
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
DILUPBiCG: Solving for Ux, Initial residual = 1, Final residual = 8.750894e-07, No Iterations 35
DILUPBiCG: Solving for Uy, Initial residual = 1, Final residual = 9.656731e-07, No Iterations 33
DILUPBiCG: Solving for Uz, Initial residual = 1, Final residual = 9.484505e-07, No Iterations 63
DILUPBiCG: Solving for h, Initial residual = 1, Final residual = 8.38986e-07, No Iterations 84
Min/max T:300 737.9726
GAMG: Solving for p_rgh, Initial residual = 0.9927845, Final residual = 0.0043827, No Iterations 6
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors (air): sum local = 0.0002022969, global = -9.427607e-06, cumulative = -9.427607e-06
GAMG: Solving for p_rgh, Initial residual = 0.09379622, Final residual = 0.000382585, No Iterations 4
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors (air): sum local = 0.0002579665, global = 3.034633e-06, cumulative = -6.392974e-06
DILUPBiCG: Solving for epsilon, Initial residual = 0.1579261, Final residual = 7.899828e-07, No Iterations 16
DILUPBiCG: Solving for k, Initial residual = 1, Final residual = 2.952635e-07, No Iterations 80

Solving for solid region platte
DICPCG: Solving for h, Initial residual = 1, Final residual = 3.102837e-07, No Iterations 5
Min/max T:min(T) [0 0 0 1 0 0 0] 708.3925 max(T) [0 0 0 1 0 0 0] 790.7728
ExecutionTime = 166.2 s ClockTime = 234 s

Region: air Courant Number mean: 0.5470849 max: 7.974256
Region: platte Diffusion Number mean: 0.0001257544 max: 0.0001329767
deltaT = 0.0002029867
Time = 0.00200155


Solving for fluid region air
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
DILUPBiCG: Solving for Ux, Initial residual = 0.2788012, Final residual = 8.959984e-07, No Iterations 37
DILUPBiCG: Solving for Uy, Initial residual = 0.1237281, Final residual = 8.137486e-07, No Iterations 36
DILUPBiCG: Solving for Uz, Initial residual = 0.2438061, Final residual = 8.851508e-07, No Iterations 36
DILUPBiCG: Solving for h, Initial residual = 0.2280842, Final residual = 7.831367e-07, No Iterations 36
Min/max T:86.47079 790.6948
GAMG: Solving for p_rgh, Initial residual = 0.09683618, Final residual = 0.0007611302, No Iterations 3
diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors (air): sum local = 3.468532e-05, global = -4.339116e-06, cumulative = -1.073209e-05
[1] #0 Foam::error:rintStack(Foam::Ostream&) at ??:?
[1] #1 Foam::sigFpe::sigHandler(int) at ??:?
[1] #2 in "/lib/x86_64-linux-gnu/libc.so.6"
[1] #3 Foam::GAMGSolver::scale(Foam::Field<double>&, Foam::Field<double>&, Foam::lduMatrix const&, Foam::FieldField<Foam::Field, double> const&, Foam::UPtrList<Foam::lduInterfaceField


Thank you.
Michael

[laurentD]

Yes i agree with you, the first iteration seems to be correct.
One thing is answering me :
you say in the controlDict that you want the job start from 0.001, but when i look the first iteration, it start from 0, doesn't it ?
Maybe the problem is not there but it seems strange...
Anyway the thing we have to understand is : Why does the Courant increase so much ? I am convinced that the error you have come from the Courant, it is too big for a transient job.
Laurent

[michael157]

I agree with you, the job starts from 0.
The strange thing is that another job, exactly same BC and Initial conditions, only difference is the geometry runs well.
I have no idea where to look for the error.
I changed the starting time and the job crashes as well.

[kmefun]

You can have a try to turn off adjustableTimeStep. Let timeStep fix to like 1*e-5. To see when Time = 0.00200155 is the case blow up or not.

[michael157]

I think the problem is something with the mesh. The solid domain consists of geometry with e.g. a plate with 3mm thickness and the fluid sourrounding is 0.6 m.
I tried to do with simple plate of 3mm thickness and got the same error. I tried to do with a plate of 10mm thickness and the case runs well.

[michael157]

Hello everybody,
i did last nigth a lot of trials and i have no idea what the problem is.
The complex geometry doesn´t run - crash after first iteration
i exchanged the complex geometry with a plate, 3mm thickness and the job with exactly same parameters run.
The time step of 0.0001s should be small enougth for a courant number of 1.
Has anybody any idea what i else can do?

[kmefun]

There are several reasons might cause it blow up such as improper boundary conditions setup etc. If you can upload your test case, it might be easiler to find out where the promblem is.

Have a nice weekend

[michael157]

I have uploaded the whole case to a dropbox folder.
See link below:
https://www.dropbox.com/sh/1z89bylbg...lwaXk4vFa?dl=0

There are two files, one with only the case and Preprocessing (Filename: M01_TESTAUSSCH_3_0-7_poly0-8_7_Co0-9_k-e-on.tar.gz)
and the other with the crashed simulation (Filenmae: M01_TESTAUSSCH_3_0-7_poly0-8_7_Co0-9_k-e-on_including-logfiles.tar.gz) (logfiles included).

I hope this help to solve the problem

[laurentD]

Hello Michael,
i can see that others persons have come to discuss this problem. Good thing.
You say that the job runs when you use another geometry.
Have you tried the command checkMesh with the complex geometry? Maybe it can help you to see where are the problems...
Laurent

[michael157]

HI Laurent,
yes i did the checkMesh and OF says the Mesh is OK.

[kmefun]

Hi,

Could you briefly discribe your test case? Is the geometry an enclosure? In your test case, I look at all patch type for air zone are all wall type. But in 0/air/U dictionary like

minY
{
type fixedValue;
value uniform ( 0 15 0 );
}
maxY
{
type inletOutlet;
value uniform ( 0 15 0 );
inletValue uniform ( 0 0 0 );
}

you specified a velocity for walls? Does that mean walls move in your case?

[michael157]

The test case is a geometry which is surrounded by air. The air should move towards the geometry with an air-velocity of 15m/s.
In the picture you can see the case set-up visualized. The blue arrow should show the "inlet" velocity, on the opposite site is the outlet. The others are walls.

You mentioned that every boundary is defined as wall. Yes this is right. In the blockMeshDict I defined everything initially as wall because I think this has no influence on the simulation. Later I do changeDictionaryDict and here I define the right BC.

Regards michael

[Mohammad Jam]

Originally Posted by michael157

The test case is a geometry which is surrounded by air. The air should move towards the geometry with an air-velocity of 15m/s.
In the picture you can see the case set-up visualized. The blue arrow should show the "inlet" velocity, on the opposite site is the outlet. The others are walls.

You mentioned that every boundary is defined as wall. Yes this is right. In the blockMeshDict I defined everything initially as wall because I think this has no influence on the simulation. Later I do changeDictionaryDict and here I define the right BC.

Regards michael

Hi Michael,

I know this thread is a bit old!
Could you find the solution? I have same problem.

Thanks in advance,
Regards, Mohammad