[Anna Tian]

Hi,

Currently I need to model the 3D wing for lift and drag coefficient calculations. I need some 3D wind tunnel experimental data for the CFD validation so that I can feel more comfortable about my simulation data. Where could I find these 3D experimental data?

[MK1000]

i am not sure of any models available but i would advice you to look up the aerofoil you are using. ie the NACA number. in the abbott von doenhoff theory of wing section book. it has Cl, Cd graphs Vs AOA and many other relevant material to compare your simulation data to.

[Anna Tian]

Originally Posted by MK1000

i am not sure of any models available but i would advice you to look up the aerofoil you are using. ie the NACA number. in the abbott von doenhoff theory of wing section book. it has Cl, Cd graphs Vs AOA and many other relevant material to compare your simulation data to.

But 2D and 3D are very different things. The resolve of the secondary flow needs also to be validated. Otherwise how could induced drag prediction be validated?

[MK1000]

im assuming you used solid works to create your 2D model that could be exported into ansys. assuming you did, you can also create a 3D model from your 2D model in solidworks. then repeat the same procedures you did for 2D onto the 3D model. ie create a mesh etc then set the conditions you want i.e AOA's, velocities etc then compare results to abbott von doenhoff

[Anna Tian]

Originally Posted by MK1000

im assuming you used solid works to create your 2D model that could be exported into ansys. assuming you did, you can also create a 3D model from your 2D model in solidworks. then repeat the same procedures you did for 2D onto the 3D model. ie create a mesh etc then set the conditions you want i.e AOA's, velocities etc then compare results to abbott von doenhoff

But does the book 'Theory of Wing Sections: Including a Summary of Airfoil Data (Dover Books on Aeronautical Engineering)' also include 3D data? I thought it only has 2D airfoil data.

I know how to do 3D CFD. I just need 3D experimental data.

[MK1000]

this is only what i would suggest use 2D results to validate your simulation, reason: What is expected is that all the 2D values should match the experimental results for both CL and CD at the different angles of attack. The reason as to why 2D geometry is used rather than that of 3D is because in 3D geometry will not give accurate results as there are wing tip vortices which in turn will give lower values than expected.

[Anna Tian]

Originally Posted by MK1000

this is only what i would suggest use 2D results to validate your simulation, reason: What is expected is that all the 2D values should match the experimental results for both CL and CD at the different angles of attack. The reason as to why 2D geometry is used rather than that of 3D is because in 3D geometry will not give accurate results as there are wing tip vortices which in turn will give lower values than expected.

I don't get it. Does that mean CFD is unable to predict accurately the 3D wing aerodynamics performance like Cd?

[Anna Tian]

Originally Posted by MK1000

this is only what i would suggest use 2D results to validate your simulation, reason: What is expected is that all the 2D values should match the experimental results for both CL and CD at the different angles of attack. The reason as to why 2D geometry is used rather than that of 3D is because in 3D geometry will not give accurate results as there are wing tip vortices which in turn will give lower values than expected.

Any way we have 3D wing with wing tip vortices in real life. It makes very little sense to validate the 2D, right?

[MK1000]

what im trying to say is this: your errors from 3D simulation may occur due to the fact that is may be difficult to get the order of accuracy in the codes such that good drag predictions (for arguments sake) can be achieved

[Anna Tian]

Originally Posted by MK1000

what im trying to say is this: your errors from 3D simulation may occur due to the fact that is may be difficult to get the order of accuracy in the codes such that good drag predictions (for arguments sake) can be achieved

So people do 2D validation to feel a little more comfortable about the 3D prediction results? And where to find the 3D experimental data to have a clue about how far is 3D CFD from experiment?

[MK1000]

you could say that.

not sure exactly where to find 3D results. however wish you the best

[slug]

The Onera wing is a well known test-case for CFD validation (transonic flow). See,

http://www.grc.nasa.gov/WWW/wind/val...ng/m6wing.html

plus many other web references. The original data is published in an AGARD report (http://ftp.rta.nato.int/public/PubFu...ARD-AR-138.pdf).

I don't think they report CL or CD, but the pressure data can be used to get CL (estimate). Drag will vary to some extent for any given wind-tunnel/flight due to differing transition locations. If you use the authors names as a search term (V.Schmitt and F.Charpin), you should be able to find many subsequent papers (by different authors) that compare to this data and may provide you CL/CD etc.. computed with differing CFD codes and methods.

You should also be able to take 2D aerofoil data, and get a reasonable prediction of the 3D CL using your wing's aspect ratio. Most design/aero texts should have the relevant equations.
Hope this helps.

[Anna Tian]

Originally Posted by slug

The Onera wing is a well known test-case for CFD validation (transonic flow). See,

http://www.grc.nasa.gov/WWW/wind/val...ng/m6wing.html

plus many other web references. The original data is published in an AGARD report (http://ftp.rta.nato.int/public/PubFu...ARD-AR-138.pdf).

I don't think they report CL or CD, but the pressure data can be used to get CL (estimate). Drag will vary to some extent for any given wind-tunnel/flight due to differing transition locations. If you use the authors names as a search term (V.Schmitt and F.Charpin), you should be able to find many subsequent papers (by different authors) that compare to this data and may provide you CL/CD etc.. computed with differing CFD codes and methods.

You should also be able to take 2D aerofoil data, and get a reasonable prediction of the 3D CL using your wing's aspect ratio. Most design/aero texts should have the relevant equations.
Hope this helps.

Thank you very much! This will be really helpful. Btw, my simulation will be more around Ma=0.15 which is much lower than this test case. Could I still use this test data for the CFD validation for my case? How reasonable is it to use this as the CFD validation for me low Ma case? This test data is for Ma=0.85. Any experimental data for Ma around 0.15?

[slug]

If your code is compressible and can reproduce the Onera data, then I would trust it at lower Mach numbers. However, if your code has a separate "incompressible option/code-path", then this must be tested and shown to work (need not be a wing case though). However, there is plenty of other data out there (even if you don't have journal access) that maybe simpler for you to model than the Onera wing.

e.g. http://aerospace.illinois.edu/m-seli...-2012-3026.pdf (Low RE (<100,000), incompressible, low AR, rectangular wing. Paper provides CL,CD,CM vs AoA and some surface flow visualisation)

[Anna Tian]

Originally Posted by slug

If your code is compressible and can reproduce the Onera data, then I would trust it at lower Mach numbers. However, if your code has a separate "incompressible option/code-path", then this must be tested and shown to work (need not be a wing case though). However, there is plenty of other data out there (even if you don't have journal access) that maybe simpler for you to model than the Onera wing.

e.g. http://aerospace.illinois.edu/m-seli...-2012-3026.pdf (Low RE (<100,000), incompressible, low AR, rectangular wing. Paper provides CL,CD,CM vs AoA and some surface flow visualisation)

Thank you very much. That is very helpful. Would you like to have a look at this wing scaling question?

http://www.cfd-online.com/Forums/mai...raft-wing.html

[Anna Tian]

Originally Posted by MK1000

i am not sure of any models available but i would advice you to look up the aerofoil you are using. ie the NACA number. in the abbott von doenhoff theory of wing section book. it has Cl, Cd graphs Vs AOA and many other relevant material to compare your simulation data to.

Since the lift and drag coefficient also depends on the Mach number beisdes Re number, I'd like to know which page of this book tells the Mach number in its experiment for NACA 2412. I looked for this condition for quite a long time in that book but didn't find out that value. Could you tell at which page could I find the Mach value of its experiment?

[adrin]

Here's a report we used to benchmark our CFD technology development effort (it's in the Mach number of relevance to you):

McAlister, K. W. and Takahashi, R. K., “NACA0015 Wing Pressure and Trailing Vortex Measurements,” Tech. Rep. NASA TP-3151, 1991.

Word of caution: if you are using a traditional low-order CFD method, good luck getting an accurate tip vortex using moderate number of grid points. You'd need high-order and/or adaptive grid methods to capture the wake accurately (because of numerical diffusion in traditional grid-based methods).

Here's our paper benchmarking against the above NACA0015 data; it should serve as a good reference as to what you should expect to get in terms of level of accuracy.

M. J. Stock, A. Gharakhani, and C. P. Stone, "Modeling Rotor Wakes with A Hybrid OVERFLOW-Vortex Method on A GPU Cluster," Accepted for presentation at the 28th AIAA Applied Aerodynamics Conference, Chicago, IL, July 2010.

Adrin

[Anna Tian]

Originally Posted by adrin

Here's a report we used to benchmark our CFD technology development effort (it's in the Mach number of relevance to you):

McAlister, K. W. and Takahashi, R. K., “NACA0015 Wing Pressure and Trailing Vortex Measurements,” Tech. Rep. NASA TP-3151, 1991.

Word of caution: if you are using a traditional low-order CFD method, good luck getting an accurate tip vortex using moderate number of grid points. You'd need high-order and/or adaptive grid methods to capture the wake accurately (because of numerical diffusion in traditional grid-based methods).

Here's our paper benchmarking against the above NACA0015 data; it should serve as a good reference as to what you should expect to get in terms of level of accuracy.

M. J. Stock, A. Gharakhani, and C. P. Stone, "Modeling Rotor Wakes with A Hybrid OVERFLOW-Vortex Method on A GPU Cluster," Accepted for presentation at the 28th AIAA Applied Aerodynamics Conference, Chicago, IL, July 2010.

Adrin

Thank you very much for your answer. It's very helpful!

Btw, what is your experience on tip vortex CFD, like induced drag CFD calculation? What is the accuracy you achieved compared with experimental data?

[lzlz]

hi,im meeting the same problem,do u solve it? can u help me to get some 3D wind tunnel airfoil(wing) experimental data? im really worry now,thx!