[ingarcia1703]

Hello everyone

At fisrt I do appreciate if someone could help me with this problem which is driving me crazy. I'm working on a simulation of an airfoil in 2D by using the Spalart-Allmaras turbulence model. I want to know if the simulation is right by comparing the Cl and Cd results written at other research by wind tunnel test, nevertheless, I'm not getting near values of the Cl and Cd coefficients. Would you help me please?

At first I want to check if I'm setting wright the parameters at the controlDict directory, under the functions, forces part, this is the case:
For an angle of attack= 10°, U= 18 m/s and a chord of c=0.25, I wrote:

CofR ( 0 0 0 ); // ¿What is this?
liftDir ( -0.174 0.985 0); //for x and y, -sin(10) and cos(10) respectively
dragDir ( 0.985 0.174 0); // for x and y, cos(10) and sin(10) respectively
pitchAxis ( 0 0 1) ;
magUinf 18.00;
lRef 0.25;
Aref 0.25;

Are these parameters right?

[nimasam]

which solver do you use?

remember pisoFoam or simpleFoam or pimpleFoam use p/rho instead of p, so it may the reason getting different result, however you should present your result here, then other users can help you to figure out what the problem is:
1- it may be due to mesh resolution
2-it may be due to numerical Scheme

[ingarcia1703]

Hello NimaSam

Thank you for answer!

At first as you can see in the following link I set up the domain with the airfoil with an angle of attack of 0°, so as I want to know the Cl and Cd coefficient values with an angle of attack of 10° instead of changing the domain I change the flow direction, is that ok? I'm getting as well a y+ mean value of 9.5, with a minimum of 0.6 and max. of 16.4.

On the other hand, I'm working with simpleFoam, so which parameters should I change in order to take into account the p/rho value?

The fvSchemes use is also shown in the given document. What do you think of that?

Thank you very much.

https://dl.dropboxusercontent.com/u/...%20results.pdf

[nimasam]

simpleFoam and etc use p/rho you may want to multiply ur result by rho it may cure your results

[ingarcia1703]

Well hopefully but it doesn't works with the Cp and Cl values :/
Another suggestion maybe?

[tomf]

Hi Luis,

Make sure your Aref is equal to lref*thickness of your cell in empty direction.
CofR is center of rotation for the calculation of your pitching moment. I would suggest to take the quarter chord point.

Your density should not matter in coefficient calculations. But you could add these two lines (assuming air) to check:

rhoName    rhoInf;
rhoInf         1.225;

Regards,
Tom

[ingarcia1703]

Hi Tom!

Thank's for your answer!

I change the CofR point and when you say "the quarter point of the chord" I guess I must put over there its coordinates, right? Also I change the Aref value , but checking the Cl and Cd equations it only influence in the Cm value, so I guess it won't work, but let's wait the results.

The rfo lines are already set in that part of the code, I just didn't write it here, sorry by that! XD

Maybe another suggestion?

Greeting from Colombia!

[ingarcia1703]

Hi Tom!

I've already finish two different simulations and they gave me better results than before! I am wondering about the reason why by changing the CofR and the Aref I got different results in the Cl and the Cd values, if according to their equations, they are not related at all.

Thanks for your help! I do appreciate it!

[tomf]

Well CofR is not related to cl or cd indeed. It is to cm. aRef is your reference area, which in 2D would be your chord times your span of 1 cell thickness. Since

cl=Lift/(0.5*rho*mag(U)^2*aRef)

aRef directly influences all.

Regards,
Tom

[ingarcia1703]

Yes Tom you're right! Thanks for your help!

I wish you the best!