CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   boundary layer thickness (http://www.cfd-online.com/Forums/main/94608-boundary-layer-thickness.html)

rfar0028 November 21, 2011 06:10

boundary layer thickness
 
When modelling a flat plate, how can one obtain a plot of the development of the boundary layer thickness over a flat plate?

Thanks

Caluca November 21, 2011 06:22

Well I am not sure if I really understand your question. Do you mean a plot of time versus boundary layer thickness? If that is the case you could select a section on the flat plate (for example x = 0.5L) and store the boundary layer thickness at that section within each time step. At the end you plot a time array against a boundary layer thickness array.

rfar0028 November 21, 2011 06:34

No I need a plot of boundary layer thickness (delta) vs distance from the leading edge of the flat plate. The objective is to compare with the analytical solution and find the transtion point from a laminar to turbulent b.l

Thanks for your help

Caluca November 21, 2011 07:05

You could consider a 3D surface plot, with x and t on the horizontal axes and delta on the vertical axis.

You would then need to store all the delta's in a 2 dimensional array as you would need all the delta's along the entire plate for every time step.

swetkyz November 22, 2011 12:58

Quote:

Originally Posted by rfar0028 (Post 332904)
No I need a plot of boundary layer thickness (delta) vs distance from the leading edge of the flat plate. The objective is to compare with the analytical solution and find the transtion point from a laminar to turbulent b.l

This is steady state, correct? What package are you using for post processing? Depending on the package, you may need to plot velocity vs vertical distance at several discrete locations on the plate for a fast visual representation of BL thickness.

A more elegant solution might be to write some code to look through your data and determine when the velocity has become close enough to nominal and store that information to be viewed by excel or some other plotting program.

I'm assuming you are talking about a momentum boundary layer and not thermal--either way, you can replace temperature with velocity in the above statements, and they still apply.

rfar0028 November 22, 2011 17:42

Yes its steady state and momentum b.l. I'm using the CFD-POST via Ansys Workbench. I don't have any experience with code but I was hoping that there is some function on CFD-POST.

Thanks

mettler November 23, 2011 10:46

have you tried a contour plot of a plane? - but that would only be a 2-d plot. I am not sure if I understand as to whether or not you need a 3d plot or 2d. You can construct a plane perpendicular to the plate and produce a contour plot of the velocities in that plane. From that you should be able to see the boundary layer development.

rfar0028 November 23, 2011 13:00

Yes this is a simple 2D problem..the one you find in the cornell and penn statefleunt tutorials. Does the method you mentioned apply for the 2D case? I.e how can construct a perpendicular plane?

Thanks for your time

mettler November 23, 2011 13:03

yes, you can definitely do it in 2d. Fluent has an option 'construct a plane'..you'll have to find it and define the plane etc. Also, you can export that plane's velocities to Excel as a matrix and find the boundary layer by looking thru the velocity numbers. A contour plot would help you locate it, but the exact location might be take a little more digging.

rfar0028 November 24, 2011 08:15

Just another question..my next step is to increase the free stream velocity over the flat plate such that at some point the boundary layer changes form laminar to tubulent. From the CFD model I would like to obtain the point/region where the transition from laminar to turbulent b.l start occuring.
Therefore which turbulence model is adequate to obtain a detailed analysis of the development of such a boundary layer?
Thanks

cfdnewbie November 24, 2011 10:43

Quote:

Originally Posted by rfar0028 (Post 333417)
Therefore which turbulence model is adequate to obtain a detailed analysis of the development of such a boundary layer?
Thanks

None, really. If you want to have a detailed analysis of the transition process and the exact location, all current turbulence models will give you more or less wrong results.

If you are limited by your solver, then 2-equation models generally do an ok job.

aero_suresh November 25, 2011 21:30

Quote:

Originally Posted by rfar0028 (Post 333417)
Just another question..my next step is to increase the free stream velocity over the flat plate such that at some point the boundary layer changes form laminar to tubulent. From the CFD model I would like to obtain the point/region where the transition from laminar to turbulent b.l start occuring.
Therefore which turbulence model is adequate to obtain a detailed analysis of the development of such a boundary layer?
Thanks

The transition point w.r.t to distance and boundary layer thickness is purely depends on the type and quality of mesh near to the bounday along with type of turbulence used.
Most of my flat plate case i have used Kepsilon tubulence model and mostly matches with theoretical valve well.


All times are GMT -4. The time now is 23:04.