CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   How set experimental points values as wind tunnel boundary conditions (http://www.cfd-online.com/Forums/fluent/110850-how-set-experimental-points-values-wind-tunnel-boundary-conditions.html)

vidade December 22, 2012 14:47

How set experimental points values as wind tunnel boundary conditions
 
Hi everyone, my first thread..
I had to test car performances using a k-epsilon realizable model in a slotted walls WT and I need to impose some experimental point values (pressure) as wind tunnel wall boundary conditions. How can I impose a wall pressure conditions like this? Is it possible?
thank you all, if you need some extra info please ask!

LuckyTran December 23, 2012 02:05

Without doing some magic, you cannot specify the pressure on the wall.

You must follow the physical mathematics of what each boundary condition is. For a solid wall, this means that the wall-normal velocity is equal to the wall velocity (kinematics). For the no-slip condition, the tangential velocity is equal to the wall tangential velocity. If there is a slip-condition then tangential velocity is related to the wall tangential velocity by some relation. Note, there is no pressure!

Think of it this way. You cannot specify the pressure at the wall in your wind tunnel either. You can try by various methods to adjust the tunnel conditions to obtain a certain pressure at the wall but by no means are you specifying the pressure at the wall, you are just changing the inlet and outlet conditions of your wind tunnel.

With that said, I have seen cases where variables are "fixed" over an entire zone. Not sure if you can do specific locations (possibly with a UDF). If you set the pressure in this manner, you will end up with inconsistencies and your solution most likely will have trouble converging / converging to an inaccurate result.

vidade December 23, 2012 11:56

Quote:

Originally Posted by LuckyTran (Post 398736)
Without doing some magic, you cannot specify the pressure on the wall.

You must follow the physical mathematics of what each boundary condition is. For a solid wall, this means that the wall-normal velocity is equal to the wall velocity (kinematics). For the no-slip condition, the tangential velocity is equal to the wall tangential velocity. If there is a slip-condition then tangential velocity is related to the wall tangential velocity by some relation. Note, there is no pressure!

Think of it this way. You cannot specify the pressure at the wall in your wind tunnel either. You can try by various methods to adjust the tunnel conditions to obtain a certain pressure at the wall but by no means are you specifying the pressure at the wall, you are just changing the inlet and outlet conditions of your wind tunnel.

With that said, I have seen cases where variables are "fixed" over an entire zone. Not sure if you can do specific locations (possibly with a UDF). If you set the pressure in this manner, you will end up with inconsistencies and your solution most likely will have trouble converging / converging to an inaccurate result.

thanks, I know about Navier Stokes formulation actually, and that's the problem. It need a magic trick as you said.. I thought to modify the wall function equation imposing not directly the pressure but the tangential stress (with no slip boundary condition) and iteratively impose the desired pressure. Another way could be an energetic way playing with kinetics energy and its dissipation to impose right experimental pressure condition. I am not a fluent expert so these stuff, easy on paper, could be impossible to do on CFD, if anyone has some other ideas i'd be happy to know.:)

LuckyTran December 23, 2012 17:02

I recommend visiting the Fluent Manual section 7.2.4 on Fixing the Values of Variables. I have used it before for fixing y,z velocity in an entire zone. You can specify pressure if you are using the pressure-based solver. Like all the default properties in Fluent, it is designed for you to fix variables throughout an entire zone and not a single point.

You can specify non-constant profiles by specifying the profile or using a UDF. If you construct the correct profile/UDF then I imagine you can get it to specify the pressure at a specific location.

Sorry I can't give a complete walkthrough after that as I avoid these options (complicated CFD is not very useful, IMO). But it is possible and fairly straightforward (if you can get the profile/UDF). You do not need to change the solver, wall functions, or mess with the details of the turbulence modeling.


All times are GMT -4. The time now is 12:12.