CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Simulating flow past airfoil with different AOA (

Quarkz December 26, 2005 23:50

Simulating flow past airfoil with different AOA

I am now trying to simulate flow past airfoil with different AOA (angle of attack). I've used a software called MeshPilot to generate the structured grids and successfully simulated the flow at 0deg AOA using my FVM NS solver. Due to the limitation of the software, it is not possible to generate an airfoil with some other AOA.

Hence, I am wondering if it is possible to just change the AOA by changing my initial/boundary conditions ie instead of u=1,v=0 at inlet, u=0.707=v for 45deg AOA.

Will there be problems? Or is generating an airfoil at different AOA a more popular approach? Can someone recommend a suitable structured grid generator to do that?

Thanks alot

ag December 27, 2005 23:29

Re: Simulating flow past airfoil with different AO
Your approach will work fine assuming your freestream is steady. It's referred to as a Galilean transformation. Just remember that lift and drag coefficients are computed parallel and normal to the freestream flow, and are not the components of the force in the x-y coordinate system (except for aoa=0). If you want to change the angle of your airfoil, why not just rotate the grid? It's a simple transformation and is the approach I follow. Write yourself a simple program to apply the transformation.

AnotherCFDUser January 6, 2006 11:56

Re: Simulating flow past airfoil with different AO
There should be no problem with your approach (or a rotation of the grid) provided that the angle-of-attack variation is relatively small (certainly within the linear region). Even in the linear regime it would be better to use a grid with points that are aligned with the wake direction in the wake region (usually the wind direction to a good approximation).

Beyond the linear region you will expect to see changes in the flow topology (separation, separation bubbles etc.) which really should be tackled with grids generated to capture these features.

All times are GMT -4. The time now is 11:29.