Problem with airflow over naca0012 and angle o f attack

Lucas · February 17, 2011, 14:27

Hello,

I'm working on a comparison between FA and CFX with some airfoils. I'm working with laminar flow Reynolds number of 360,000 for naca 0012. CFX simulates the data as followed. Now I'm trying to simulate in the OF, but the values of Cd and Cl are very different from those obtained in CFX and literature, and an angle of attack of 2 degrees the pressure field is completely irregular, I'm changing the angle of attack from speed change. Attached is the photo of the pressure field and cases.

naca0012_0.jpg

naca0012_1.jpg

naca0012_2.jpg

CFX:
CFX.jpg

Openfoam cases:

http://www.mediafire.com/?bgdmx03updfgckd

Thank you.

FelixL · February 17, 2011, 14:50

Hello, Lucas,

with a Reynolds number of 360,000 the flow is expected to be turbulent, not laminar! A turbulence model would be appropriate to improve the numerical results, especially drag.

Another thing: the freestream boundaries are much too close to the airfoil surface. The distance between airfoil and farfield boundaries should be at least 20 cord lengths! Otherwise the flow close to the airfoil surface is distorted by the farfield boundary conditions and the results are negatively affected.

Greetings,
Felix.

Lucas · February 18, 2011, 14:13

Hello FelixL,
thanks for your reply.

I was wondering, why this problem did not show up in CFX, OF is much more sensitive than CFX? Because the pressure field for an 3º is completely different from the 2º.

Other thing i am not sure if my case is set up correctly, now i am trying to
use the airfoild2d tutorial case, can you help me with the reference length and area for that airfoil.

Thanks,
Lucas Vieira.

February 17, 2011, 14:27	Problem with airflow over naca0012 and angle o f attack	#1
Lucas New Member Lucas Vieira Join Date: Nov 2010 Posts: 6 Rep Power: 16	Hello, I'm working on a comparison between FA and CFX with some airfoils. I'm working with laminar flow Reynolds number of 360,000 for naca 0012. CFX simulates the data as followed. Now I'm trying to simulate in the OF, but the values of Cd and Cl are very different from those obtained in CFX and literature, and an angle of attack of 2 degrees the pressure field is completely irregular, I'm changing the angle of attack from speed change. Attached is the photo of the pressure field and cases. naca0012_0.jpg naca0012_1.jpg naca0012_2.jpg CFX: CFX.jpg Openfoam cases: http://www.mediafire.com/?bgdmx03updfgckd Thank you.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
when the angle of attack =?, the total lift coefficient begin to decrease suddenly	liujmljm	Main CFD Forum	6	August 3, 2012 06:46
Unreal vortices in NACA0012 at high angle of attack	parvez2683	FLUENT	1	September 18, 2009 13:27
Cp for NACA0012, angle of attack=14degree	liujmljm	Main CFD Forum	1	October 29, 2008 23:12

February 17, 2011, 14:50		#2
FelixL Senior Member Felix L. Join Date: Feb 2010 Location: Hamburg Posts: 165 Rep Power: 18	Hello, Lucas, with a Reynolds number of 360,000 the flow is expected to be turbulent, not laminar! A turbulence model would be appropriate to improve the numerical results, especially drag. Another thing: the freestream boundaries are much too close to the airfoil surface. The distance between airfoil and farfield boundaries should be at least 20 cord lengths! Otherwise the flow close to the airfoil surface is distorted by the farfield boundary conditions and the results are negatively affected. Greetings, Felix.

February 18, 2011, 14:13		#3
Lucas New Member Lucas Vieira Join Date: Nov 2010 Posts: 6 Rep Power: 16	Hello FelixL, thanks for your reply. I was wondering, why this problem did not show up in CFX, OF is much more sensitive than CFX? Because the pressure field for an 3º is completely different from the 2º. Other thing i am not sure if my case is set up correctly, now i am trying to use the airfoild2d tutorial case, can you help me with the reference length and area for that airfoil. Thanks, Lucas Vieira.