CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   about the drag coefficient (http://www.cfd-online.com/Forums/main/8952-about-drag-coefficient.html)

maximus April 1, 2005 22:57

about the drag coefficient
 
I have a question about the drag coefficent of an airfoil. I use the RANS to compute the flow field of an airfoil,but I found the drag can't be well predicted. The drag induced by friction is very small compared with the drag induced by pressure even at the low attack angle. The drag coefficient is larger than the experimental data. Can anybody give me some suggestions? I'll appreciate any response! Thanks! Maximus

Jarmo Monttinen April 4, 2005 13:29

Re: about the drag coefficient
 
What is the Reynolds number in your simulation?

-- Jarmo

maximus April 5, 2005 09:32

Re: about the drag coefficient
 
Hi,Jarmo: Thanks for your response. The Reynolds number in my computation is about 2*e6, and the airfoil is NACA 23012. I use the S-A, K-e and K-w turbulence model to make a comparison, but I found they are not very different with each other. I compared the distribution of pressure coefficent with the experimental data,and the predicted results are in good consistency with experimental data except at the suction peak region. I don't know if I make a mistake when I integrate the force along the airfoil surface. Thanks!!

Jarmo Monttinen April 5, 2005 12:14

Re: about the drag coefficient
 
Hmmm.. so the RE is high enough that the fully turbulent approach should work and there is no separation.

Double check the calculation of viscous drag, if the Cp is close to the experimental values, you should get close values for Cd as well.

-- Jarmo

maximus April 7, 2005 00:13

Re: about the drag coefficient
 
Jarmo,thanks for your reply. I still have a question. The Cp predicted is well consistent with the experimental data, also the Cl. So I think the Cd induced by pressure may be well predicted. But the Cd integrated from the pressure is 0.012 which is larger than the total Cd of experiment(about 0.007). If the visous drag is considered the total Cd predicted must be larger than the experimetal data. I was confused.Are there some considerations to deal with the integration to get Cd from pressure? Thanks!! Maximus

Jarmo Monttinen April 7, 2005 12:33

Re: about the drag coefficient
 
Hmmm.... it is possible that you are taking the wrong component of the force at some points, or maybe have a sign error for some segments? Total force is pdA, in this case, dA is just 1*(cell face length), right? If you are calculating the drag this way using components of this force, it is possible that you have a sign error for the projection somewhere. Look at the individual segments on the airfoil, the pressure (or Cp) there and the shape of the airfoil. Positive Cp at LE should add to the drag, positive at TE (if so) should decrease it and so on...

-- Jarmo

Sandy April 11, 2005 02:14

About Cd calculation
 
Hi, I am trying to calculate the Cd for a truck body. I am getting the coefficient of drag as 11.35 The drag force in truck motion axis is 18800N. and the truck is moving at a velocity of 140Km/h with wind acting in opposite direction at 100 Km/h. The projected area of truck in X direction is 7.69m2 and the entire surface area of truck front is 22 m2. The air density assumed is 1.2kg/m3. I have taken the projected area for calculation and not the entire surface area. Pls tell me if its correct?? Also if I take a circular plate and a sphere for drag coeff. calculation, as I understand, I should take the projected area for calculation of 1/2 . rho . v2 . A But if the dia of both sphere and circular plate is same, then Cd for both should also be same. Or is there any other factor which I am missing out here?? Pls explain?

Jarmo Monttinen April 18, 2005 19:00

Re: About Cd calculation
 
You better look at the dimensional drag as the drag coefficient as the coefficient definition always uses some area, which for standard problems is defined, but for your problem probably not. You might want to look up some papers to see what the authors have used, perhaps wetted surface would be more appropriate?

-- Jarmo


All times are GMT -4. The time now is 05:39.