Same Courant number on the entire mesh

KateEisenhower · February 16, 2017, 04:57

Dear OpenFOAM-Community,

currently I am calculating the flow around an airfoil in 2D. As expected the cells with the highest Courant number are in the direct vicinity of the airfoil while the cells in the farfiled of course have very low values.

So the question is:
Is it advisable to aim for more or less the same Courant number over the whole domain?

Of course this would lead to unbelievable huge cells in the farfield.

I would appreciate to read your opinion on this.

Best regards,

Kate

matejfor · February 16, 2017, 06:31

There is no way around this. To lower the Courant number you would need to coarsen the mesh close to airfoil which would destroy your simulation and simply you cannot do that, or you can refine the mesh in the far-field resulting in mesh overresolved where it is not needed. So to fight with the broad range of CFL where the maximum value limits your time step size, you do not do magic with the mesh, but - if your case is steady state or pseudo-transient you may apply local time stepping. The other solution is to solve your problem in other space than the good old Euklidean space.

happy foaming,

Matej

February 16, 2017, 04:57	Same Courant number on the entire mesh	#1
KateEisenhower Senior Member Join Date: Mar 2015 Posts: 250 Rep Power: 12	Dear OpenFOAM-Community, currently I am calculating the flow around an airfoil in 2D. As expected the cells with the highest Courant number are in the direct vicinity of the airfoil while the cells in the farfiled of course have very low values. So the question is: Is it advisable to aim for more or less the same Courant number over the whole domain? Of course this would lead to unbelievable huge cells in the farfield. I would appreciate to read your opinion on this. Best regards, Kate

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Foam::error::PrintStack	almir	OpenFOAM Running, Solving & CFD	91	December 21, 2022 04:50
[OpenFOAM.org] OF2.3.1 + OS13.2 - Trying to use the dummy Pstream library	aylalisa	OpenFOAM Installation	23	June 15, 2015 14:49
SigFpe when running ANY application in parallel	Pj.	OpenFOAM Running, Solving & CFD	3	April 23, 2015 14:53
dynamic Mesh is faster than MRF????	sharonyue	OpenFOAM Running, Solving & CFD	14	August 26, 2013 07:47
[snappyHexMesh] external flow with snappyHexMesh	chelvistero	OpenFOAM Meshing & Mesh Conversion	11	January 15, 2010 19:43

February 16, 2017, 06:31		#2
matejfor Senior Member matej forman Join Date: Mar 2009 Location: Brno, Czech Republic Posts: 182 Rep Power: 17	There is no way around this. To lower the Courant number you would need to coarsen the mesh close to airfoil which would destroy your simulation and simply you cannot do that, or you can refine the mesh in the far-field resulting in mesh overresolved where it is not needed. So to fight with the broad range of CFL where the maximum value limits your time step size, you do not do magic with the mesh, but - if your case is steady state or pseudo-transient you may apply local time stepping. The other solution is to solve your problem in other space than the good old Euklidean space. happy foaming, Matej