FSI floating point error on stressModel displacement iteration

Kyle.Davis · July 22, 2016, 10:41

Hi

I am trying to run a simple case with foam-extend 3.1 using the FSI application. I have imported a simple mesh created in Netgen 4.9.13 in OpenFoam format as I will need to import more complicated meshes at a later stage for the solid structure. I used checkMesh and everything is fine. I then used snappyHexMesh to create a fluid mesh around the solid, which checkMesh also showed is fine.

When running fsiFoam, there is not a problem with the fluid-to-solid interpolators or solid-to-fluid. The simulation runs for the first fluid iteration, however I get a floating point error when trying to solve the stress model. The forces on the fluid and solid are calculated and displayed, however no iterations are displayed for actually solving the displacements of the solid.

Total force (fluid) = (-0.00382077 -1.38906 -1.60812)
Total force (solid) = (0.00382077 1.38906 1.60812)
Evolving stress model
-0.0166856, 1.00047
Floating point exception (core dumped)

I believe I could be having a problem with the stressFoam solver trying to use a mesh that was not created with an OpenFoam utility, however further meshes with be too complicated to use blockMesh, hence me trying to import a mesh of a simple case. I am also not sure how to determine where the problem could be in the source code of the solid solver, and any help on explaining how to do so will be appreciated.

I have added a link below to the zip file with the already run case. The log.fsiFoam is included with the error shown. The solid and fluid mesh have already been created and only fsiFoam needs to be run.

https://www.dropbox.com/s/pa6cz7l3uj...lock2.zip?dl=0

Kind Regards
Kyle

BiomechanicsBath · October 2, 2018, 05:51

Hi

I am in the exact same situation you were. Have you figured out what happened? I have been playing around with the stress Model in StressProperties (convergence tolerance and nCorrectors increased), but that seems a bit like a quick-and-dirty fix. Any better Idea on how to fix this issue?
Thanks in advance!

ssyadav · October 14, 2018, 10:19

Hello All

I am trying to run fsiFOAM in parallel, but I am getting segmentation fault error while the solver is solving for solid deformation. The error is shown below:

Evolving solid solver: unsTotalLagrangianSolid
[cfd2:20298] *** Process received signal ***
[cfd2:20298] Signal: Segmentation fault (11)
[cfd2:20298] Signal code: (-6)
[cfd2:20298] Failing at address: 0x3e800004f4a

The serial version is working perfectly fine.

My question to the community is that how to decompose the case?

Currently I am using simple method and " globalFaceZones (SOLIDZone_to_FLUIDZone-zone) " in the solid decomposeParDict file and similarly for the fluid side patch in fluid decomposeParDict file.

Thanks for the help in advance.

Regards

Shyam Sunder

BiomechanicsBath · January 15, 2019, 16:17

Hi Kyle
Have you ever solved your problem. I am at the exact same problem now. If i am using air as the fluid I have no problems, just when using water or any other fluid.
Did you experience the same issue?

July 22, 2016, 10:41	FSI floating point error on stressModel displacement iteration	#1
Kyle.Davis New Member Kyle Davis Join Date: Feb 2016 Posts: 1 Rep Power: 0	Hi I am trying to run a simple case with foam-extend 3.1 using the FSI application. I have imported a simple mesh created in Netgen 4.9.13 in OpenFoam format as I will need to import more complicated meshes at a later stage for the solid structure. I used checkMesh and everything is fine. I then used snappyHexMesh to create a fluid mesh around the solid, which checkMesh also showed is fine. When running fsiFoam, there is not a problem with the fluid-to-solid interpolators or solid-to-fluid. The simulation runs for the first fluid iteration, however I get a floating point error when trying to solve the stress model. The forces on the fluid and solid are calculated and displayed, however no iterations are displayed for actually solving the displacements of the solid. Total force (fluid) = (-0.00382077 -1.38906 -1.60812) Total force (solid) = (0.00382077 1.38906 1.60812) Evolving stress model -0.0166856, 1.00047 Floating point exception (core dumped) I believe I could be having a problem with the stressFoam solver trying to use a mesh that was not created with an OpenFoam utility, however further meshes with be too complicated to use blockMesh, hence me trying to import a mesh of a simple case. I am also not sure how to determine where the problem could be in the source code of the solid solver, and any help on explaining how to do so will be appreciated. I have added a link below to the zip file with the already run case. The log.fsiFoam is included with the error shown. The solid and fluid mesh have already been created and only fsiFoam needs to be run. https://www.dropbox.com/s/pa6cz7l3uj...lock2.zip?dl=0 Kind Regards Kyle

October 2, 2018, 05:51	Same PRoblem	#2
BiomechanicsBath New Member Evelyn Join Date: Nov 2017 Posts: 7 Rep Power: 8	Hi I am in the exact same situation you were. Have you figured out what happened? I have been playing around with the stress Model in StressProperties (convergence tolerance and nCorrectors increased), but that seems a bit like a quick-and-dirty fix. Any better Idea on how to fix this issue? Thanks in advance!

October 14, 2018, 10:19	FSI floating point error	#3
ssyadav New Member Shyam Sunder Join Date: Sep 2015 Posts: 27 Rep Power: 10	Hello All I am trying to run fsiFOAM in parallel, but I am getting segmentation fault error while the solver is solving for solid deformation. The error is shown below: Evolving solid solver: unsTotalLagrangianSolid [cfd2:20298] * Process received signal * [cfd2:20298] Signal: Segmentation fault (11) [cfd2:20298] Signal code: (-6) [cfd2:20298] Failing at address: 0x3e800004f4a The serial version is working perfectly fine. My question to the community is that how to decompose the case? Currently I am using simple method and " globalFaceZones (SOLIDZone_to_FLUIDZone-zone) " in the solid decomposeParDict file and similarly for the fluid side patch in fluid decomposeParDict file. Thanks for the help in advance. Regards Shyam Sunder

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
icoFoam - Floating point exception (core dumped)	File	OpenFOAM Running, Solving & CFD	26	March 6, 2021 04:26
A floating point exception - SEM Model	yansheng	STAR-CCM+	1	April 4, 2016 04:57
fluent divergence for no reason	sufjanst	FLUENT	2	March 23, 2016 16:08
rhoPimpleFoam floating point error	dancfd	OpenFOAM Running, Solving & CFD	6	January 5, 2014 20:57
a reconstructPar issue	immortality	OpenFOAM Post-Processing	8	June 16, 2013 11:25

January 15, 2019, 16:17		#4
BiomechanicsBath New Member Evelyn Join Date: Nov 2017 Posts: 7 Rep Power: 8	Hi Kyle Have you ever solved your problem. I am at the exact same problem now. If i am using air as the fluid I have no problems, just when using water or any other fluid. Did you experience the same issue?