icoFoam floating point exception (8)

leizhao512 · September 6, 2015, 18:15

Hi, everyone. I was quite new to openFoam. I am simulating a cylindrical channel with a spiral spacer. It is a laminar flow since the inlet flow velocity is 2e-5m/s and the Re is 1.2. I initially was using Fluent and it the results were good. Then I want to use icoFoam. The mesh I was using in openFoam was converted from a Fluent mesh, and by checking mesh, it said OK. I am using the time step of 10s, which is exactly same with what I was using in Fluent. However, the solution always diverge and gave me an error of "floating point exception (8)". Besides, the Courant number kept increasing. Can anybody help me with this?
Thank you.

The error message is like this:

Quote:

--------------------------------------------------------------------------
in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
[1] #1 Foam::sigFpe::sigHandler(int) in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
[1] #2 in "/lib64/libc.so.6"
[1] #3 Foam::symGaussSeidelSmoother::smooth(Foam::word const&, Foam::Field<double>&, Foam::lduMatrix const&, Foam::Field<double> const&, Foam::FieldField<Foam::Field, double> const&, Foam::UPtrList<Foam::lduInterfaceField const> const&, unsigned char, int) in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
[1] #4 Foam::symGaussSeidelSmoother::smooth(Foam::Field<d ouble>&, Foam::Field<double> const&, unsigned char, int) const in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
[1] #5 Foam::smoothSolver::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so"
[1] #6
[1] in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/bin/icoFoam"
[1] #7
[1] in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/bin/icoFoam"
[1] #8
[1] in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/bin/icoFoam"
[1] #9
[1] in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/bin/icoFoam"
[1] #10 __libc_start_main in "/lib64/libc.so.6"
[1] #11
[1] in "/opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/bin/icoFoam"
[hs122:14272] *** Process received signal ***
[hs122:14272] Signal: Floating point exception (8)
[hs122:14272] Signal code: (-6)
[hs122:14272] Failing at address: 0x1ba912000037c0
[hs122:14272] [ 0] /lib64/libc.so.6(+0x32920) [0x7f1aef331920]
[hs122:14272] [ 1] /lib64/libc.so.6(gsignal+0x35) [0x7f1aef3318a5]
[hs122:14272] [ 2] /lib64/libc.so.6(+0x32920) [0x7f1aef331920]
[hs122:14272] [ 3] /opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so(_ZN4Foam22symGaussSeidelSmoother6sm oothERKNS_4wordERNS_5FieldIdEERKNS_9lduMatrixERKS5 _RKNS_10FieldFieldIS4_dEERKNS_8UPtrListIKNS_17lduI nterfaceFieldEEEhi+0x512) [0x7f1af03928d2]
[hs122:14272] [ 4] /opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so(_ZNK4Foam22symGaussSeidelSmoother6s moothERNS_5FieldIdEERKS2_hi+0x2d) [0x7f1af039303d]
[hs122:14272] [ 5] /opt/apps/gcc4_5/openmpi1_6/OpenFOAM/2.3.0/OpenFOAM-2.3.0/platforms/linux64GccDPOpt/lib/libOpenFOAM.so(_ZNK4Foam12smoothSolver5solveERNS_5 FieldIdEERKS2_h+0x412) [0x7f1af0388b32]
[hs122:14272] [ 6] icoFoam() [0x43bfa5]
[hs122:14272] [ 7] icoFoam() [0x43e0aa]
[hs122:14272] [ 8] icoFoam() [0x43e3e0]
[hs122:14272] [ 9] icoFoam() [0x41d97c]
[hs122:14272] [10] /lib64/libc.so.6(__libc_start_main+0xfd) [0x7f1aef31dcdd]
[hs122:14272] [11] icoFoam() [0x4196b9]
[hs122:14272] *** End of error message ***
--------------------------------------------------------------------------
mpirun noticed that process rank 1 with PID 14272 on node hs122 exited on signal 8 (Floating point exception).
--------------------------------------------------------------------------
usage: kill [ -s signal | -p ] [ -a ] pid ...
kill -l [ signal ]

leizhao512 · September 6, 2015, 18:18

I also uploaded my case file without mesh file as the mesh file is too large.
The log file is:

Quote:

Create time

Create mesh for time = 0

Reading transportProperties

Reading field p

Reading field U

Reading/calculating face flux field phi

Starting time loop

Time = 10

Courant Number mean: 0.000107184 max: 0.266238
smoothSolver: Solving for Ux, Initial residual = 0, Final residual = 0, No Iterations 0
smoothSolver: Solving for Uy, Initial residual = 1, Final residual = 9.04463e-09, No Iterations 147
smoothSolver: Solving for Uz, Initial residual = 0, Final residual = 0, No Iterations 0
DICPCG: Solving for p, Initial residual = 1, Final residual = 25.9149, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.0840296, Final residual = 0.0010788, No Iterations 1001
time step continuity errors : sum local = 0.00167782, global = -0.000131827, cumulative = -0.000131827
DICPCG: Solving for p, Initial residual = 0.0516744, Final residual = 9.79542e-05, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.0153046, Final residual = 6.87564e-05, No Iterations 1001
time step continuity errors : sum local = 0.000188136, global = -3.53417e-06, cumulative = -0.000135361
ExecutionTime = 122.56 s ClockTime = 123 s

Time = 20

Courant Number mean: 0.721766 max: 6.47043
smoothSolver: Solving for Ux, Initial residual = 0.694741, Final residual = 9.63998e-09, No Iterations 239
smoothSolver: Solving for Uy, Initial residual = 0.608072, Final residual = 9.55807e-09, No Iterations 254
smoothSolver: Solving for Uz, Initial residual = 0.698027, Final residual = 9.68723e-09, No Iterations 244
DICPCG: Solving for p, Initial residual = 0.120674, Final residual = 0.0317097, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.136847, Final residual = 0.00306148, No Iterations 1001
time step continuity errors : sum local = 0.042808, global = 0.0014139, cumulative = 0.00127853
DICPCG: Solving for p, Initial residual = 0.0569325, Final residual = 0.000320575, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.0201781, Final residual = 0.000587244, No Iterations 1001
time step continuity errors : sum local = 0.0180552, global = 0.000196775, cumulative = 0.00147531
ExecutionTime = 250.23 s ClockTime = 251 s

Time = 30

Courant Number mean: 2.44691 max: 124.162
smoothSolver: Solving for Ux, Initial residual = 0.75502, Final residual = 9.61463e-09, No Iterations 248
smoothSolver: Solving for Uy, Initial residual = 0.634171, Final residual = 9.97674e-09, No Iterations 266
smoothSolver: Solving for Uz, Initial residual = 0.759349, Final residual = 9.81736e-09, No Iterations 248
DICPCG: Solving for p, Initial residual = 0.135442, Final residual = 0.0361863, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.154305, Final residual = 0.00439654, No Iterations 1001
time step continuity errors : sum local = 0.757218, global = -0.0137298, cumulative = -0.0122545
DICPCG: Solving for p, Initial residual = 0.0564087, Final residual = 0.000233005, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.0271797, Final residual = 0.000498899, No Iterations 1001
time step continuity errors : sum local = 0.194416, global = 0.00252102, cumulative = -0.00973351
ExecutionTime = 384.17 s ClockTime = 385 s

Time = 40

Courant Number mean: 28.8262 max: 1431.65
smoothSolver: Solving for Ux, Initial residual = 0.660078, Final residual = 9.66353e-09, No Iterations 308
smoothSolver: Solving for Uy, Initial residual = 0.55, Final residual = 9.62978e-09, No Iterations 357
smoothSolver: Solving for Uz, Initial residual = 0.663699, Final residual = 9.85606e-09, No Iterations 298
DICPCG: Solving for p, Initial residual = 0.174997, Final residual = 0.0255697, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.113939, Final residual = 0.00314605, No Iterations 1001
time step continuity errors : sum local = 7.08733, global = 0.208589, cumulative = 0.198855
DICPCG: Solving for p, Initial residual = 0.0402492, Final residual = 0.000105796, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.0286548, Final residual = 0.000240382, No Iterations 1001
time step continuity errors : sum local = 1.11688, global = 0.00996845, cumulative = 0.208824
ExecutionTime = 525.73 s ClockTime = 527 s

Time = 50

Courant Number mean: 340.81 max: 12310.1
smoothSolver: Solving for Ux, Initial residual = 0.595993, Final residual = 9.93393e-09, No Iterations 827
smoothSolver: Solving for Uy, Initial residual = 0.605277, Final residual = 1.84456e-08, No Iterations 1000
smoothSolver: Solving for Uz, Initial residual = 0.596082, Final residual = 9.99107e-09, No Iterations 824
DICPCG: Solving for p, Initial residual = 0.147091, Final residual = 0.01536, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.180302, Final residual = 0.00397467, No Iterations 1001
time step continuity errors : sum local = 41.5979, global = -1.37837, cumulative = -1.16955
DICPCG: Solving for p, Initial residual = 0.0424793, Final residual = 0.000639773, No Iterations 1001
DICPCG: Solving for p, Initial residual = 0.0471108, Final residual = 0.00111893, No Iterations 1001
time step continuity errors : sum local = 34.9303, global = -0.74496, cumulative = -1.91451
ExecutionTime = 728.05 s ClockTime = 730 s

Time = 60

Courant Number mean: 3151.15 max: 124948

Cyp · September 6, 2015, 23:28

what if you reduce your time step?

leizhao512 · September 7, 2015, 09:40

Quote:

Originally Posted by Cyp

what if you reduce your time step?

I was using time step of 10 in Fluent, and the solution converged quickly. I also tried to shrink the time step to 1s in openFoam, but it didn't work. Is there any problem with my case setup, the fvscheme?

Cyp · September 7, 2015, 12:27

What if you reduce the time step even lower ? I guess in Fluent you have an adaptative time step, something you don't have with icoFoam, so probably when you say 10s time step in Fluent, it is not exactly the time stepd used during simulation.

If the error persists, I will say something is wrong in your boundary condition set up.

leizhao512 · September 10, 2015, 11:36

Quote:

Originally Posted by Cyp

What if you reduce the time step even lower ? I guess in Fluent you have an adaptative time step, something you don't have with icoFoam, so probably when you say 10s time step in Fluent, it is not exactly the time stepd used during simulation.

If the error persists, I will say something is wrong in your boundary condition set up.

I was using fixed time step in Fluent.
Then I tried to use simpleFoam to simulate the case because I have proved that it was steady state case using Fluent. The case could be converged in the first 400 iterations. However, if I just copy the solution of simpleFoam to icoFoam as the initial condition, and the schemes and linear solvers are same, the icoFoam case will still be divergent in 100 time steps.

tas38 · September 10, 2015, 15:33

You need to restrict the Courant number in icoFoam. You can do this as suggested by decreasing the time step manually, or you can implement an adjustable time step with a maximum specified Courant number. You can add the following to the controlDict file...

Code:

adjustTimeStep true;
maxCo 1.0;

Depending on your case, you may have to drop maxCo even lower.

itsme_kit · November 1, 2018, 11:43

Hi Lei

Have you figured out what kind of problems could trigger "floating point exception (8)"?

I had the same problem recently.

I don't think it's about dividing by 0.

Many thanks

Quote:

Originally Posted by leizhao512

Hi, everyone. I was quite new to openFoam. I am simulating a cylindrical channel with a spiral spacer. It is a laminar flow since the inlet flow velocity is 2e-5m/s and the Re is 1.2. I initially was using Fluent and it the results were good. Then I want to use icoFoam. The mesh I was using in openFoam was converted from a Fluent mesh, and by checking mesh, it said OK. I am using the time step of 10s, which is exactly same with what I was using in Fluent. However, the solution always diverge and gave me an error of "floating point exception (8)". Besides, the Courant number kept increasing. Can anybody help me with this?
Thank you.

The error message is like this:

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Floating point exception (core dumped) for GAMG solver	yuhou1989	OpenFOAM Running, Solving & CFD	2	March 24, 2015 19:28
Inlet Velocity Profile BC - Floating Point exception during solution initialization	Janshi	STAR-CCM+	4	March 14, 2012 10:21
simpleFoam Floating point exception error -help	sudhasran	OpenFOAM Running, Solving & CFD	3	March 12, 2012 16:23
Pipe flow in settlingFoam floating point exception	jochemvandenbosch	OpenFOAM Running, Solving & CFD	4	February 16, 2012 03:24
block-structured mesh for t-junction	Robert@cfd	ANSYS Meshing & Geometry	20	November 11, 2011 04:59

September 6, 2015, 23:28		#3
Cyp Senior Member Cyprien Join Date: Feb 2010 Location: Stanford University Posts: 299 Rep Power: 18	what if you reduce your time step?

September 7, 2015, 12:27		#5
Cyp Senior Member Cyprien Join Date: Feb 2010 Location: Stanford University Posts: 299 Rep Power: 18	What if you reduce the time step even lower ? I guess in Fluent you have an adaptative time step, something you don't have with icoFoam, so probably when you say 10s time step in Fluent, it is not exactly the time stepd used during simulation. If the error persists, I will say something is wrong in your boundary condition set up.

September 10, 2015, 15:33		#7
tas38 Senior Member Troy Snyder Join Date: Jul 2009 Location: Akron, OH Posts: 219 Rep Power: 18	You need to restrict the Courant number in icoFoam. You can do this as suggested by decreasing the time step manually, or you can implement an adjustable time step with a maximum specified Courant number. You can add the following to the controlDict file... Code: adjustTimeStep true; maxCo 1.0; Depending on your case, you may have to drop maxCo even lower.