What means "bounding epsilon"?

gara1988 · October 25, 2012, 05:15

Hello, I have an easy question. I don't understand what means "bounding epsilon" or k, during running of simpleFoam. Is there someone that can tell me what it means?

Thanks a lot.
Best Regards.

Mattia

batta31 · October 25, 2012, 09:51

Hi Mattia,
I have experienced the same problems. What I can suggest you is to evaluate carefully the initial/boundary values for your turbulent variables, and maybe you could have a look to your fvSchemes and fvSolution files.

More precisely:
- in fvSchemes you could use upwind scheme for the terms involving your turbulent variables;
- in fvSolution have a look to the tolerances you use for your turbulent variables and to the relaxation factors (maybe you could be a little more conservative and use a lower value of relaxation factor)

With these modifications my simulation has not diverged any more.

Best regards
Simone

gonpe · October 26, 2012, 08:10

Hi Mattia

The solver is reporting high or low values of a turbulence variable. It is typically indicative of some instability within the solution.

As the post above mentions there are a couple of ways to deal with it. In my experience it often comes down to mesh quality. Maybe lack of refinement in certain areas with high gradients etc.

Sometimes these bounded values go away as the solution converges.

Goncalo

gara1988 · October 29, 2012, 08:59

Thanks batta31 and gonpe for the replies.

Best regards.

October 25, 2012, 05:15	What means "bounding epsilon"?	#1
gara1988 New Member Mattia Join Date: Oct 2012 Location: Milan Posts: 28 Rep Power: 13	Hello, I have an easy question. I don't understand what means "bounding epsilon" or k, during running of simpleFoam. Is there someone that can tell me what it means? Thanks a lot. Best Regards. Mattia

October 26, 2012, 08:10	Bounding Turbulence variables	#3
gonpe Member Goncalo Pedro Join Date: Nov 2009 Location: Victoria, British Columbia Posts: 30 Rep Power: 16	Hi Mattia The solver is reporting high or low values of a turbulence variable. It is typically indicative of some instability within the solution. As the post above mentions there are a couple of ways to deal with it. In my experience it often comes down to mesh quality. Maybe lack of refinement in certain areas with high gradients etc. Sometimes these bounded values go away as the solution converges. Goncalo batta31 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
error means?	hanid	Main CFD Forum	0	September 8, 2008 13:25
BC evaluation by means of external lists allocated in the solver	shrina	OpenFOAM Running, Solving & CFD	2	January 18, 2007 11:29
What does the floating point error means?	jon william	FLUENT	1	October 8, 2006 08:57
plot order points... what does it means ?	Cyril	FLUENT	0	June 7, 2006 05:56
What means % in Fortran?	What means % in Fortran	Main CFD Forum	2	April 20, 2004 14:34

October 25, 2012, 09:51		#2
batta31 Member Simone Join Date: Sep 2012 Posts: 95 Rep Power: 13	Hi Mattia, I have experienced the same problems. What I can suggest you is to evaluate carefully the initial/boundary values for your turbulent variables, and maybe you could have a look to your fvSchemes and fvSolution files. More precisely: - in fvSchemes you could use upwind scheme for the terms involving your turbulent variables; - in fvSolution have a look to the tolerances you use for your turbulent variables and to the relaxation factors (maybe you could be a little more conservative and use a lower value of relaxation factor) With these modifications my simulation has not diverged any more. Best regards Simone

October 29, 2012, 08:59		#4
gara1988 New Member Mattia Join Date: Oct 2012 Location: Milan Posts: 28 Rep Power: 13	Thanks batta31 and gonpe for the replies. Best regards.