CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Lift co-efs (http://www.cfd-online.com/Forums/fluent/28046-lift-co-efs.html)

F.P. mie November 23, 2000 06:41

Lift co-efs
 
Hi I'm trying to model (2D) flow around a NACA0009 foil, to compare with experimental results. I've drawn the foil from x/y co-ords, so I think its fairly accurate. However, when I try to compute lift and drag forces on the foil, they just oscillate like crazy, even after 10000 iterations! My flow speed is 10 m/sec, the attack angle is 10deg, and my fluid is water. I've specified a trapezoid area around the foil, with 3 inflows and an outflow. Any suggestions what might be wrong? In particular, should I have gravity on/off, and will it matter what operating pressure I use? Cheers FP

Trac November 23, 2000 18:18

Re: Lift co-efs
 
At 10deg, you could be getting unsteady flow from the separation, and you might need to switch to a transient analysis.

Gravity & operating pressure shouldn't matter, but you will need to make sure the reference values (under report) are correct, or your Cd and Cl will be calculated wrongly.

You also need to make sure the boundaries are far enough away (at least 10c) and you should probably check your y+ values since (assuming air and c=1) your flow is turbulent. You might want to consider a two-layer zonal model to pick up the separation.

It might be a good idea to run the model with a lower incidence, just to make sure you have everything working okay.

Tracie.

jangho.,lee November 27, 2000 00:13

Re: Lift co-efs
 
hi i got the same problem to compute the NACA00xx series. in my experience, NACA 0009 series oscillate at 14 degree angle of attat at Ma=0.085.

i use fluent5.1 and tubluent model is k-e model.coule implicit method.

i think that you should check your "courant number" and your time in the solver is steady..

if you want to any detail about the datum, write the mail to me...

ps.. fluent can not predict the exact stall angle...

F.P. mie November 27, 2000 07:12

Re: Lift co-efs
 
Guys Thanks for the help, I changed my turbulence model, and my lift co-efs are fine now. New problem- all my drag values are negative!! How is this possible-if Cd comes from 1/2(rho)(v^2)(Area), where is the minus coming from? Any ideas? FP

Peter Attar November 27, 2000 10:22

Re: Lift co-efs
 
Check your definition of the drag coefficient...you are missing the numerator.

FP Mie November 27, 2000 12:18

Re: Lift co-efs
 
Peter- Yeh, its Cd=D/0.5(rho)(v^2)(area), but my query remains, providing the drag force cant be -ve (which it can't, can it??), how are the Cd's coming out -ve? I'm new to Fluent, so forgive me if my questions seem simple! FP

Peter Attar November 27, 2000 12:25

Re: Lift co-efs
 
I'm not a fluent user at all actually..: ) just like to read the boards.I am not sure how things are defined but it would make sense to me that since the Drag opposes the motion of the airfoil that it's sign would be negative..this would give you your negative Cd..I would assume that in most aero books this definition is assummed and that is why you never see Cd defined as negative.


jangho,Lee November 27, 2000 21:30

Re: Lift co-efs
 
hi. guy. i think your drag problem based on your drag corrinate. first, check your force to drag.. i think that you write positive x cordinate at force. x cordinate should negative.

check your force...

P.venkataraman December 13, 2000 12:09

Re: Lift co-efs
 
Just accessed the group

By definition Drag and lift coefficients are defined parallel and perpendicular to flow. You have to transform those values using the angle of attack

Venkat


All times are GMT -4. The time now is 16:03.