CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   enhanced wall treatment (http://www.cfd-online.com/Forums/fluent/42144-enhanced-wall-treatment.html)

saeed August 26, 2006 05:19

enhanced wall treatment
 
Dear users I'm simulating a closed cavity with Flunt 6 .. Realizable... but I dont know what wall function i should use for modeling it . my problem have 2 walls with heat flux and i'm validating with a paper written by Kenjeres ... so i should solve my problem with high accuracy .. I search help contents of Fluent but .. there is;nt any subject about enhanced wall treatment .. If its possible fou you please send me some information about it tanks...saeed


Evan Rosenbaum August 29, 2006 11:19

Re: enhanced wall treatment
 
The enhanced wall treatment is nothing more than stacking cells against the walls to resulve the boundary later. Put a bunch of cells near the walls and solve. Then check y+. If y+ is about 1, your OK. For y+ >> 1 add cells near the wall. For y+ << 1, reduce the number of cells.

Albert F. August 30, 2006 03:58

Re: enhanced wall treatment
 
I thought that if the y+ was <<1 it was ok, wasn't it? Lower y+ you have better the results will be. Is this false?

Evan Rosenbaum August 31, 2006 11:47

Re: enhanced wall treatment
 
Yeah, my mistake. You can't have too low a y+.

Albert F. September 5, 2006 04:44

Re: enhanced wall treatment
 
Do you know how the results are affected by imposing a wrong y+? I mean If what I want to know is the temperature distribution along a wall a wrong y+ can vary the results severly? Do you trust in an other viscosity model?

Evan Rosenbaum September 6, 2006 11:38

Re: enhanced wall treatment
 
You cannot have multiple layers of elements all using the log law. The wall function would be applied to the first one only, even through others will be in the boundary layer. Thus, the boundary layer thickness would be too low. Switch to enhanced wall treatment of the y+ is too small for standard wall functions.

It is an extreme generalization (with plenty of exceptions) that you probably should not use standard wall functions when wall flows are critical. So, for the flow in a large room, wall functions are usually OK. For flow between two closely spaced plates, the usually are not OK.


All times are GMT -4. The time now is 22:20.