CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Help! lift/drag coefficients do not match experimental data! (http://www.cfd-online.com/Forums/fluent/101940-help-lift-drag-coefficients-do-not-match-experimental-data.html)

ali.qasim May 17, 2012 12:52

Help! lift/drag coefficients do not match experimental data!
 
Hello

I came across this paper which contains lift and drag coefficients for a few airfoils at various Reynolds numbers.

I'm trying to get these results in FLUENT, but my values are way off.

The airfoil I've chosen is NACA 0012. Reynolds number is 80,000. The airfoil model used in the experiment had a chord length of 15.24 cm. I got the coordinates online, and scaled them to this length.

Now, assuming values for density (1.225 kg/m^3) and dynamic viscosity (1.7894e-5 kg/m-s), I arrived at a free-stream velocity of about 7.668 m/s. I used this as my inlet boundary condition.

For the reference values in FLUENT, I set the reference length = chord length, depth = 1m, and therefore area = (chord length) * 1.

I chose the Inviscid model. I know this is not perfect, but for now I just want to see if I get a rough approximate for the lift coefficient.

Here's a link to a spreadsheet showing the values I got, and the percentage difference from the experimental values. As you can see, they are not even close. I get a constant difference of around 40% (I expect it to be within 5% at least for the first 8 degrees), which tells me I could be missing a factor here.

Can anyone see what I'm missing? Please help me out, as I've been stuck on this for many days!

Thanks

Ali

cfd seeker May 17, 2012 14:40

your lift results are ok and drag results are bad?? isnt it the case with you?

ali.qasim May 17, 2012 14:44

Quote:

Originally Posted by cfd seeker (Post 361719)
your lift results are ok and drag results are bad?? isnt it the case with you?

No, both are.

I'm not concerned about drag right now, as I'm using the inviscid model.

The problem is, the lift coefficients are not what they should be. The approximation of inviscid should not really affect that very much.

ali.qasim May 19, 2012 11:11

Problem solved.. but doesn't make sense!
 
I'm now getting pretty decent results for the lift coefficient. But the weird thing is, it was because of the symmetry boundary condition.

I am using a C-mesh for the airfoil, and had defined the arc of the semicircle as the inlet, and the side edges (on top and bottom) as symmetry. This time, I removed that symmetry BC, and made these side edges also as part of the inlet.

I have absolutely no idea how that would make such a big difference. The percentage deviation from wind tunnel data is now within 5% compared to the 40% deviation I had before. Can someone please explain why this happened?

Far May 19, 2012 11:32

Are you sure? Please post this problem on ANSYS meshing forum to get detailed discussion

ali.qasim May 19, 2012 11:57

Quote:

Originally Posted by Far (Post 361974)
Are you sure? Please post this problem on ANSYS meshing forum to get detailed discussion

I'm pretty sure about that.

Started a thread in ANSYS meshing.


All times are GMT -4. The time now is 21:33.