InterFoam - time step

atlan · January 26, 2015, 13:07

Hallo,
Can somebody help me with the following problem? I am trying to run InterFoam simulation on a general geometry (large wessel filled with water). The mesh is made of tetrahedral elements and the minimum elements size is over 20 mm. The inlet velocity is about 7m/s. If I calculate the time step for this configuration with Courrant number under 0.1 this is about 3e-4s. But if I run the simulation the solver decreases the time step under 1e-10, which is for 30s calculation impossible to calculate. Can somebody advise me what to do to keep the time step on some realistic level?
Regards
Atlan

alexeym · January 26, 2015, 14:10

Hi,

Just to clarify a little bit: Co = u*dt/dx: Co = 0.1, u = 7, dx = 20e-3 -> dt ~ 3e-4.

But really, show your:

1. checkMesh output
2. fvSchemes & fvSolution (maybe also controlDict to check your initial deltaT)
3. IC & BC (i.e. archive of your 0 folder), if you set initial distribution of alpha1 with setFields, post a screenshot of your initial distribution of alpha1.

Phicau · January 27, 2015, 05:24

Hi atlan,

from my experience is better to forget about using tetrahedral meshes with interFoam, as the spurious velocities at the interface are enormous and advection of the phases looks horrible. Try a hex-dominant mesh instead.

Best,

Pablo

atlan · March 13, 2015, 16:09

Dear Pablo,

you were completely right. I prepared completely hexa mesh and there is no problem with the time step anymore.

Atlan

January 26, 2015, 13:07	InterFoam - time step	#1
atlan New Member Join Date: Jul 2014 Posts: 21 Rep Power: 11	Hallo, Can somebody help me with the following problem? I am trying to run InterFoam simulation on a general geometry (large wessel filled with water). The mesh is made of tetrahedral elements and the minimum elements size is over 20 mm. The inlet velocity is about 7m/s. If I calculate the time step for this configuration with Courrant number under 0.1 this is about 3e-4s. But if I run the simulation the solver decreases the time step under 1e-10, which is for 30s calculation impossible to calculate. Can somebody advise me what to do to keep the time step on some realistic level? Regards Atlan

March 13, 2015, 16:09	Time step problems	#4
atlan New Member Join Date: Jul 2014 Posts: 21 Rep Power: 11	Dear Pablo, you were completely right. I prepared completely hexa mesh and there is no problem with the time step anymore. Atlan

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Sudden jump in Courant number	NJG	OpenFOAM Running, Solving & CFD	7	May 15, 2014 13:52
AMI interDyMFoam for mixer nu problem	danny123	OpenFOAM Programming & Development	8	September 6, 2013 02:34
same geometry,structured and unstructured mesh,different behaviour.	sharonyue	OpenFOAM Running, Solving & CFD	13	January 2, 2013 22:40
Full pipe 3D using icoFoam	cyberbrain	OpenFOAM	4	March 16, 2011 09:20
calculation diverge after continue to run	zhajingjing	OpenFOAM	0	April 28, 2010 04:35

January 26, 2015, 14:10		#2
alexeym Senior Member Alexey Matveichev Join Date: Aug 2011 Location: Nancy, France Posts: 1,930 Rep Power: 38	Hi, Just to clarify a little bit: Co = u*dt/dx: Co = 0.1, u = 7, dx = 20e-3 -> dt ~ 3e-4. But really, show your: 1. checkMesh output 2. fvSchemes & fvSolution (maybe also controlDict to check your initial deltaT) 3. IC & BC (i.e. archive of your 0 folder), if you set initial distribution of alpha1 with setFields, post a screenshot of your initial distribution of alpha1.

January 27, 2015, 05:24		#3
Phicau Senior Member Pablo Higuera Join Date: Jan 2011 Location: Auckland Posts: 627 Rep Power: 19	Hi atlan, from my experience is better to forget about using tetrahedral meshes with interFoam, as the spurious velocities at the interface are enormous and advection of the phases looks horrible. Try a hex-dominant mesh instead. Best, Pablo