hai everybody....

I would like to ask very simple question... what factor could effect an accuracy of CFD results for external flow aerodynamic...

appreciate... if I get a reply...from you all...

best regard.. ERIK

Actually there are plenty of factors:

1. Differencing scheme. High order schemes usually give better results. Usually...

2. Location of boundaries - the far you move the boundaries the "better" the results (i.e. your model flower is closer to real unbounded flow)

3. Turbulence modeling. The way it influence the results is pretty complex: more accurate turbulence modeling gives you more accurate friction. Friction drag is quiet low comparing to pressure drag but accurate prediction of friction gives you more accurate prediction of separation regions - their location and size. And this influence much at pressure drag and general flow structure.

4. Convergence monitoring. If your residuals have droped below the prescribed value (0.001 in Star) this doesn't meen that your solution is converged. You have to wait untill the residuals will stop dropping. And it will be better if you'll look not at the residuals but at "enginering values" - drag and/or lift. Ususally they keep changing long after the residuals stop droping.

5. 2D vs 3D and transient vs steady-state modelling. Actually all REAL separational flows are 3-dimensional transient...

Hope somebody will remember something else...

One additional important thing is the mesh quality and density. One needs to use fine enough grids in order to achieve good results.