Hallo again

I do agree that drawing an isoline would be helpful. However, the boudary layer thickness for airfoil cannot be defined as U_loc = 0.99*U_inf like in a flat plate. The main reason is, that flow accelerates significantly near the airfoil contour. This holds for inviscid and viscous flows especially when the airfoil thickness is large or when the angle of attack is not zero. In contrast, the inviscid velocity for a flat plate remains U_inf along the streamwise direction. Therefore, the definition of the BL thickness shall be V_total_loc = 0.99*V_total_loc_inviscid. In this sense, one should obtain the inviscid distribution outside the boundary layer but still in the vicinity of the wall. This is not a straightforward task. Therefore, plotting an isoline directly for U_loc = 0.99*U_inf will introduce some errors.

you can evaluate the thick of boundary layer by draw the velocity profile

on the airfoil at different position

Before that you must know the flow type ( laminar or turbulent or transition)

Have you already performed the simulation using a grid fine enough to resolve the BL??

After completing the calculation, you have several option, ploting the profile and or seeing the contour of velocity or temperature due to the kind of your evaluation(for velocity or temperature).

You need to add more details on your project. e.g. What kind of fluid regime? Turbulence? Which one of the turbulence models? Do you know how to generate the mesh due to your turbulence model? and etc.

Dear Roh Dargahi Filippo Maria Denaro Abbas J Jubear

1. I have done simulation before.

2. Fluid regime is transition.

3. I used K-Kl transition model.

More explain:

I want to put vortex generators on the upper surface of airfoil (Experimentally). I have been performed simulation for evaluate boundary layer thickness that I need for vortex generator's height.

by drawing velocity profile at different position on upper surface, I need many position which It is so difficult due to creating line manually. I want another way (simpler and more accurate) that evaluate B.L. thickness.

Best regards

how about your grid resolution?

Filippo Maria Denaro

grid resolution is fine (y+ is 0.2).

Have tried to plot the isolines of the streamwise velocity component =0.99*U_inf?

if the geometry of airfoil was symmetrical and the angle of attack is zero the vortex was not generated

Hi,

I guess a global method for calculating the airfoil B.L. thickness (or other integral thickness) is hardly achievable. Once I thought of a general coordinate transformation method in OpenFOAM to do the task but never got the time to try it!

The idea was to transform the airfoil grid coordinates in Cartesian space to a local normal-tangential coordinate system. By this way, I suppose it would be easy to consider the B.L. thickness based on 0.99*U_inf criterion. I hope this could serve as a hint.

Regards,

Ehsan

Hello,

To my opinion, you could do that through a script, you can use tecplot or any other post-processing software available for that purpose. The idea of the script is following:

1. You should determine the point of analysis, for example from 0 < x/c < 1 with step size of Delta_X/c.

2. Determine the analysis will be on the suction or pressure sides, it can be both but it has to be done sequentially.

3. Get the wall location (x,y,z) of the airfoil at your desired location, and get the normal vector of the wall.

4. Follow your normal vector line, and interpolate (get) the velocity components at that area for a specific step, for example Delta_n, for a number of iterations.

5. Then set a break point of the script, if total_velocity_(n) = 0.99 * total_velocity_(n+1), then iterations stop; move to a next observation point x/c = x/c_i + Delta_X/c.

I hope this code logic is helpful to you.

Then you've done your simulation, right? ok, so make sure you have decent Y+ near the wall. It's up to yourself.

As Filippo Maria Denaro pointed out, why not simply plot the isolines of the streamwise velocity/temp component =0.99*U_inf or 0.99*T_inf? or drawing the contour of velocity/temp. for U>0.99*U_inf or T=0.99*T_inf?

Hallo again

I do agree that drawing an isoline would be helpful. However, the boudary layer thickness for airfoil cannot be defined as U_loc = 0.99*U_inf like in a flat plate. The main reason is, that flow accelerates significantly near the airfoil contour. This holds for inviscid and viscous flows especially when the airfoil thickness is large or when the angle of attack is not zero. In contrast, the inviscid velocity for a flat plate remains U_inf along the streamwise direction. Therefore, the definition of the BL thickness shall be V_total_loc = 0.99*V_total_loc_inviscid. In this sense, one should obtain the inviscid distribution outside the boundary layer but still in the vicinity of the wall. This is not a straightforward task. Therefore, plotting an isoline directly for U_loc = 0.99*U_inf will introduce some errors.

Yes, a rigorous approach is to work along the curvilinear direction s and the normal direction n. Then the velocity streamwise component must be referred as to the tangential and normal vector unit.

However, if the profile has low thickness I would try a simple approach to see if the field is well resolved

In fluent all tangential, axial, X and Y velocity components are available.

Dear @https://www.researchgate.net/profile/Galih_Bangga @https://www.researchgate.net/profile/Alireza_Beiki

One can compare velocity profile with a velocity profile obtained at another calculation at different Reynolds number in account of freestream pressure/dencity. Inviscid parts of dimensionless (or even dimensional?) velocity profiles should nearly coicide with each other.

Thanks for all the people who participated in this question.

I used velocity "isoline" by 0.99*V_inf.

I got relative acceptable results, but I think as Ehsan Asgari and Filippo Maria Denaro said, if we transform coordinate, we can get more accurate results.

I did not transform coordinate yet because isoline had rough estimation that it was what I need now.

Best Regards