CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   HELP!!! (http://www.cfd-online.com/Forums/fluent/49930-help.html)

kostas November 30, 2008 10:18

HELP!!!
 
I try to use the k-e model on the airfoil and when i solute it, i don't take the right solution. Maybe, must i care the grid near the wall? How?

erkan November 30, 2008 18:28

Re: HELP!!!
 
It may be because of coarse edge meshes on airfoil surface. I may recommend you having at least eighty meshes on one surface.

Moreover use last length from LE to any point where the curvature is small and then use succesive ratio up to TE.

kostas December 1, 2008 07:28

Re: HELP!!!
 
What do you mean when you say that i must have at least 8 meshes on any section? I separate the section at 8 surface for the best control of mesh?

For operation of model i must follow a repeating prosedure, thus i come at right y+?

zongtwi December 2, 2008 12:33

Re: HELP!!!
 
You have to take care of the grid near the wall, the Y+ values need to be correct, and a corresponding near wall treatment should be used as well. You should be able to find more info on solving near wall flows in the FLUENT documentation.

Hope that helps.

kostas December 2, 2008 13:06

Re: HELP!!!
 
Thanks... Yes i must care the grid near the wall...but how??? The tutorial of fluent say that the first node must be at position (Y+>11) for operating the stadart wall function of k-e model. But how i find the y length from Y+. The problem is that the Y+ depend on flow characteristicks and thus i dont compute the y length (Y+=(y.U*.r)/n) where y= the real (dimentional length) r= density n=viscosity U*=friction velocity (U*=sqr(t/r)) where t=stress of wall Y+=non-dimentional length.

I look the fluent tutorial but it dont say anything!!!

zongtwi December 4, 2008 01:51

Re: HELP!!!
 
It's always tricky trying to get the correct Y+ value, I see it as a chicken and egg problem. You need to have the correct Y+ value for your simulation, and the only way to get the correct value is to run the simulation...hence chicken and egg!

The best way is to approximate the initial Y+ value, run your simulation, and in the post processing stage, plot your Y+ values on the wall. You will not have a single value, like you have realised the Y+ values depend on flow conditions, so you will get different values at different sections of the wall. This is where your judgement comes in, whether to be satisfied with the values or not. If the values are too high, then you have to go back to your model and recreate a new mesh with a smaller Y value (remember to always have the boundary layer elements in your mesh), rerun the simulation and check your Y+ values again.

There's also a mesh adaption criteria in FLUENT that you can use, which is based on the Y+ values but I personally don't like this feature, as the mesh adaption doesn't maintain the prismatic or boundary layer mesh near the wall, and also it is proned to result in a lot of left-handed elements (i.e. not good).

Hope that helps.

Kostas Pan December 4, 2008 07:37

Re: HELP!!!
 
Thanks for your apply.. ok Simply, i belived that there is a method for computation off Y+...


All times are GMT -4. The time now is 15:23.