CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   LES non-isotropic grid (http://www.cfd-online.com/Forums/main/108125-les-non-isotropic-grid.html)

martyn88 October 15, 2012 10:34

LES non-isotropic grid
 
Hi everyone,

I was wondering if it is possible to run LES on a non-isotropic grid? What are the grid requirements for an LES?

Thankyou,

Hugh

FMDenaro October 15, 2012 11:03

Quote:

Originally Posted by martyn88 (Post 386697)
Hi everyone,

I was wondering if it is possible to run LES on a non-isotropic grid? What are the grid requirements for an LES?

Thankyou,

Hugh

Yes, the grid is related to the filter distribution. It is common to use non-homogeneous filter in direction normal to wall for example. LES on unstructured grid is also a further example.
The grid requirement is that the large energetic structures must be resolved

vonboett November 26, 2012 09:47

There is a nice current paper of C.J. Keylock et al. in Geomorphology 179 (2012) "The application of computational fluid dynamics to natural river channels:
Eddy resolving versus mean flow approaches" which gives some upper limits for applying LES in dependency to the wall shear stress and flow depth. Using wall functions, it is said that vertical to the wall one can estimate that the first cell center has to be in the range of 30 to 300 wall units close to the wall, without wall functions you have to be below 4 wall units close with your first cell center when applying LES. Furthermore, he suggest to have the diameter of the smallest vertice you want to resolve represented by six cells, and that close to the surface (for example using interFoam) you can become coarser but should keep smallert than 5% of flow depth with your cells.
By the way: To my knowledge Foamers calculate the wall unit for example as y+ = (shear stress at wall / density of fluid)^0.5 * ((distance to wall, for example your cell center) /kinematic viscosity)


All times are GMT -4. The time now is 09:57.