Facing a floating point error when reducing time step in a unsteady simulation

withwolf · April 6, 2016, 12:55

Hi, everyone,

I'm running a unsteady simulation of a valve closure process using DFBI and overset mesh. The simulation runs well with time step 2ms, 1ms, 0.5ms. But an error occurs when I tried time step 0.25ms.

The error message is "A floating point exception has occurred: floating point exception [Invalid operation]. The specific cause cannot be identified. Please refer to the troubleshooting section of the User's Guide.
Context: star.keturb.KeTurbSolver
Command: RunSimulation
error: Server Error"

I tried increase the maximum inner iteration, but didn't work.

I am using implicit unsteady, 2nd-order temporal discretization, K-E turbulence model, reliazable K-E two layer. The mesh continua is trimmed mesher and prism layer.

Can anyone give some explanation or solutions?　　　Thank you.

withwolf · April 11, 2016, 21:09

Can any one give some suggestions?

I'm at my wits' end.

sarahtuvia · April 18, 2016, 02:06

Try to decrease the under relaxation factor of the KE turbulence in solvers

withwolf · April 23, 2016, 15:01

Quote:

Originally Posted by sarahtuvia

Try to decrease the under relaxation factor of the KE turbulence in solvers

Thank you for your reply.

I've already tried that. I decreased the KE turbulence relaxation factor from defaulted 0.8 to 0.2. But got the same error again.

nakkasujith@gmail.com · November 9, 2018, 13:40

Hey, I am getting same error. Did you get any answer ?

Maddin · November 10, 2018, 17:11

Stop the simulation before the error came up.
Make a Threshold for some physical values

and then look for the too high values...
Then decide is it a mesh issue or a iteration/time step issue.

acalado · November 14, 2018, 16:33

Yes, it might be a mesh issue. You can try plotting different residuals to see where the highest are or some unrealistic values and pin point which cells might need to be fixed.

Yongbo · November 14, 2018, 19:38

Hi, wolf

I use starccm+ to solve shiphydrodynamics. I found the 2nd order sometimes cannot work. please try 1first-order temporal discretization.

Yongbo

April 6, 2016, 12:55	floating point error when reducing time step in an unsteady simulation	#1
withwolf New Member lailai Join Date: Jul 2015 Posts: 6 Rep Power: 10	Hi, everyone, I'm running a unsteady simulation of a valve closure process using DFBI and overset mesh. The simulation runs well with time step 2ms, 1ms, 0.5ms. But an error occurs when I tried time step 0.25ms. The error message is "A floating point exception has occurred: floating point exception [Invalid operation]. The specific cause cannot be identified. Please refer to the troubleshooting section of the User's Guide. Context: star.keturb.KeTurbSolver Command: RunSimulation error: Server Error" I tried increase the maximum inner iteration, but didn't work. I am using implicit unsteady, 2nd-order temporal discretization, K-E turbulence model, reliazable K-E two layer. The mesh continua is trimmed mesher and prism layer. Can anyone give some explanation or solutions?　　　Thank you. Last edited by withwolf; April 7, 2016 at 16:02.

November 14, 2018, 16:33		#7
acalado Senior Member André Join Date: Mar 2016 Posts: 133 Rep Power: 10	Yes, it might be a mesh issue. You can try plotting different residuals to see where the highest are or some unrealistic values and pin point which cells might need to be fixed. __________________ Sapere aude!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Other] Contribution a new utility: refine wall layer mesh based on yPlus field	lakeat	OpenFOAM Community Contributions	58	December 23, 2021 02:36
Stuck in a Rut- interDyMFoam!	xoitx	OpenFOAM Running, Solving & CFD	14	March 25, 2016 07:09
AMI interDyMFoam for mixer nu problem	danny123	OpenFOAM Programming & Development	8	September 6, 2013 02:34
mixerVesselAMI2D's mass is not balancing	sharonyue	OpenFOAM Running, Solving & CFD	6	June 10, 2013 09:34
pisoFoam with k-epsilon turb blows up - Some questions	Heroic	OpenFOAM Running, Solving & CFD	26	December 17, 2012 03:34

April 11, 2016, 21:09		#2
withwolf New Member lailai Join Date: Jul 2015 Posts: 6 Rep Power: 10	Can any one give some suggestions? I'm at my wits' end.

April 18, 2016, 02:06		#3
sarahtuvia Member tuvia milshtein Join Date: Jun 2014 Location: Israel Posts: 42 Rep Power: 11	Try to decrease the under relaxation factor of the KE turbulence in solvers

November 9, 2018, 13:40		#5
nakkasujith@gmail.com New Member Join Date: Sep 2018 Posts: 4 Rep Power: 7	Hey, I am getting same error. Did you get any answer ?

November 10, 2018, 17:11		#6
Maddin Senior Member Join Date: Mar 2009 Posts: 260 Rep Power: 18	Stop the simulation before the error came up. Make a Threshold for some physical values and then look for the too high values... Then decide is it a mesh issue or a iteration/time step issue.

November 14, 2018, 19:38		#8
Yongbo New Member Yongbo Chen Join Date: Jun 2017 Location: China Posts: 2 Rep Power: 0	Hi, wolf I use starccm+ to solve shiphydrodynamics. I found the 2nd order sometimes cannot work. please try 1first-order temporal discretization. Yongbo