WheelSeperation--viscous models and wall functions

Mark G

Hiya, hopefully I can get some bloody useful answers because I am stumped!

My project is to investigate endplate design and the interation with the wheel (modelled as static is this case)

As I have just self taught in a short period of time, I do not know how to create 3d boundary layers in gambit, so used size functions to create a coarse mesh in fluent and adapt both "velocity gradients" and Wall Plus values.

With speeds of 50m/s (reynolds numbers of 2e6+) I was expecting the flow to separate on the wheel (with 1,000,000 grid meshes), however it doesnt look like it does seperate from post anylsis. (coeffiecents of drag are similar to the invicid case)

What viscous model is best for high pressure gradient bluff bodies, with seperated flows. I have wall plus values to within 300 (which is recommended for resolving BL)...so it should work :S ????

ANY advice, much appreciated!!

Mark