[Nick R]

Hi guys,

I've simulated the flow over an airfoil and have a more fundamental question to ask. Can the boundary layer, say over an airfoil, become turbulent thru natural transition without flow separation? I think the answer is positive. If you could shed more light on this, it would be great. Also, how would one verify the boundary layer characteristics, as in whether it's laminar or turbulent. Is checking contours of intermittency the only way? Or should you check the boundary layer velocity profile (mean) against one for a flat plate? But here's there's a pressure gradient so is this comparison fair?

Thanks
Nick R.

[davidwilcox]

you could look at the shape factor which, dare i say characterizes the fullness of the profile. the shape factor is just the ratio of your displacement thickness and the momentum thickness. Blasius laminar shape factor is around 2.5. this number will decrease as you increase Reynolds number.
One other characteristic ( more of a necessary but not sufficient condition) of instabilities in the flow is the inflection in the velocity profile.

[FMDenaro]

what do you mean exactly? natural transition to turbulence (immagine the flow over a flat plate) leads to create unsteady vortical structures, very small near to wall, so that by definition you have separation. Maybe I did not understand your question...

[Nick R]

Originally Posted by FMDenaro

what do you mean exactly? natural transition to turbulence (immagine the flow over a flat plate) leads to create unsteady vortical structures, very small near to wall, so that by definition you have separation. Maybe I did not understand your question...

But in that case there's no inflection in the mean velocity profile (flat plate) is there? So there's no boundary layer separation.

[Nick R]

Originally Posted by davidwilcox

you could look at the shape factor which, dare i say characterizes the fullness of the profile. the shape factor is just the ratio of your displacement thickness and the momentum thickness. Blasius laminar shape factor is around 2.5. this number will decrease as you increase Reynolds number.
One other characteristic ( more of a necessary but not sufficient condition) of instabilities in the flow is the inflection in the velocity profile.

...
Thanks. For calculating the boundary layer thickness, is the edge of the boundary layer in the case of flow over a foil, say over the suction side, still 99percent of the free-stream speed? Or is it different since the flow speeds up locally?

[FMDenaro]

Originally Posted by Nick R

But in that case there's no inflection in the mean velocity profile (flat plate) is there? So there's no boundary layer separation.

This is the issue, you are talking about the statistically averaged velocity profile with an inflection?

[Nick R]

Originally Posted by FMDenaro

This is the issue, you are talking about the statistically averaged velocity profile with an inflection?

....
Yes I think I wrote "mean".

[FMDenaro]

I think that in regular operating condition (low angle of attack) the mean velocity profile does not show inflection