Mesh size for turbulence modeling

villager · July 16, 2012, 15:17

Good day everyone!
I try to simulate fully developed turbulent flow (k-e RNG model, enhanced wall function because of 1<y+<5 in FLUENT y+ contours). But I have only two-three finite-volume cells at boundary layer, instead of 10-12. Is the last requirement strict and I've got wrong simulation results?

flotus1 · July 16, 2012, 16:19

With y+ around 1, it is highly recommended to have a sufficient number of prism layers to resolve the boundary layer.
If you only have 3 prism layers and y+ is around 1, then the prism cells only cover the viscous sublayer of the bondary layer.
The rest of the boundary layer will be poorly resolved and the results will suffer from the increased numerical diffusion in an unstructured mesh (assuming your mesh is not of the stuctured hexahedral type)

2-3 Cells might be ok when using scalable wall functions with y+ around 60, but not in this case.

July 16, 2012, 15:17	Mesh size for turbulence modeling	#1
villager Member vlg Join Date: Jul 2011 Location: My home :) Posts: 81 Rep Power: 17	Good day everyone! I try to simulate fully developed turbulent flow (k-e RNG model, enhanced wall function because of 1<y+<5 in FLUENT y+ contours). But I have only two-three finite-volume cells at boundary layer, instead of 10-12. Is the last requirement strict and I've got wrong simulation results?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] Help with fixing imported IGES model	siw	ANSYS Meshing & Geometry	24	August 24, 2010 11:22
OF 1.6 \| Ubuntu 9.10 (64bit) \| GLIBCXX_3.4.11 not found	piprus	OpenFOAM Installation	22	February 25, 2010 13:43
Gambit Unstructed (Tet) Mesh - size control	newbie	FLUENT	3	August 26, 2008 11:38
fluent add additional zones for the mesh file	SSL	FLUENT	2	January 26, 2008 11:55
Convergence moving mesh	lr103476	OpenFOAM Running, Solving & CFD	30	November 19, 2007 14:09

July 16, 2012, 16:19		#2
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,399 Rep Power: 46	With y+ around 1, it is highly recommended to have a sufficient number of prism layers to resolve the boundary layer. If you only have 3 prism layers and y+ is around 1, then the prism cells only cover the viscous sublayer of the bondary layer. The rest of the boundary layer will be poorly resolved and the results will suffer from the increased numerical diffusion in an unstructured mesh (assuming your mesh is not of the stuctured hexahedral type) 2-3 Cells might be ok when using scalable wall functions with y+ around 60, but not in this case.