[ajneil]

can anyone help me in my turbulent flow problem with sudden expansion pipe. i am using k-e model with standard wall function in fluent. i ve meshed the geometry in gambit and i didnt get t y+ values along the wall above 12.

regards
jose

[Aeronautics El. K.]

Maybe you could use the Enhanced Wall Treatment.

Originally Posted by ajneil

can anyone help me in my turbulent flow problem with sudden expansion pipe. i am using k-e model with standard wall function in fluent. i ve meshed the geometry in gambit and i didnt get t y+ values along the wall above 12.

regards
jose

[ajneil]

1) can u explain how enhanced wall treatment gives better results?
2) is it very herd to mesh in the flow separation region?

[Aeronautics El. K.]

Using standard wall functions means that you ignore the viscous sublayer and that's why you need a y+ >30 (you may want to take a look here: http://www.cfd-online.com/Wiki/Law_of_the_wall).
If y+ is lower than that, Fluent offers the choice to use the enhanced wall treatment whose details you can read here: http://cdlab2.fluid.tuwien.ac.at/LEH...ug/node514.htm

To mesh the flow separation region you need to have some indication as to where this region is. This can be deduced from experimental data or from a simulation on a coarse grid which you refine at the separation region.