CFD Online Logo CFD Online URL
Home > Forums > Main CFD Forum

An observation

Register Blogs Members List Search Today's Posts Mark Forums Read

LinkBack Thread Tools Display Modes
Old   March 11, 2004, 17:34
Default An observation
CFD Rookie
Posts: n/a
Hi everyone out there:

I have an observation I would like to share with you guys. I am modelling a 2D NASA airfoil at zero angle of attak (so that there is no separation!). (Mach = 0.05, Re = 1.44e6, chord = 4.089 ft). I have experimental data on this airfoil: Cl = 0.54 @ alpha = 0 deg.

Here is what I did:

I use RNG k-e model with standard wall function. According to the code, y+ range is within 35 - 350. I believe my mesh density is sufficient to capture all the flow structure. The procedure of my analysis is to gradually reduce the distance between the first node and the airfoil surface, while keeping the rest of the parameters constant (This means the overall node count in the entire model remains constant). y+ is gradually reducing, which makes sense, and the y+ values all around the airfoil for different cases still satisfy the 35 - 350 range. What I observed is my computed Cl deviates more (getting smaller) from the experimental data when the y+ values are gradually reduced.

I get the same trend when I try to refine the mesh around the airfoil.

In conclusion, computed Cl is closer to the experimental Cl if my y+ is bigger. My question is, is this the general trend? Normally when we do mesh dependency test, we expect the computed solution to get closer to the experimental result or exact solution (if available). However, this is not what I observed, which means a very serious thing, i.e. the notion of mesh dependency test guide me to a more incorrect computed solution. And the changes in computed result (in this case Cl computed) is big (not like a small change of .1% in Cl that I can ignore).

I appreciate any comment and advice. Thanks!
  Reply With Quote

Old   March 11, 2004, 20:55
Default Re: An observation
Jarmo Monttinen
Posts: n/a
In case a bug free code, you should get closer to the exact solution when refining the mesh.

But getting closer to the experimental solution is another issue:

- the experiment is never fully 2-D - how is the model machined, ie. is your computed shape "exactly" the same as the one in experiments - freestream turbulence in the experiments - was the AOA really 0 or maybe 0.1 degrees in experiments? - ....

I know these issues are far more severe a lower Re than what you are dealing with but might still be valid.

Do you have any information on the Cd?

Just some thoughts in the end of the day.

-- Jarmo
  Reply With Quote


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

Similar Threads
Thread Thread Starter Forum Replies Last Post
force mesurment aginor FLOW-3D 5 October 14, 2008 05:53
What is the difference between flex and foamFlex allenzhao OpenFOAM Installation 1 August 14, 2008 17:17
An observation in interFoam code liu OpenFOAM Running, Solving & CFD 1 May 28, 2005 02:18
An intresting observation about SIMPLE Scheme Abhijit Tilak Main CFD Forum 12 March 14, 2003 12:48
an interesting observation!!!! mahesh prakash Main CFD Forum 1 October 26, 1999 10:49

All times are GMT -4. The time now is 05:59.