We meassure lift and drag for 2D and 3D wing case,we use same airfoil at same AoA and same airflow speed for both case. Wind tunnel can change test section walls width.

2D wing test:

So first we put airfoil in wind tunnel so walls touch arifoil tips(2D wing),at AoA=15degrees and airflow speed=20m/s and meassure lift and drag..

3D wing test:

Now we increase walls width so airflow "act" like 3D wing in test section.We use same AoA=15 and same airflow speed (20m/s) over airfoil like in 2D wing- test. note: AoA=15 is angle between chord line and freestream airflow not angle between chord line and effective airflow at 3D wing

Question is,what will be drag of 3D wing-test compare to 2D wing-test,smaller ,equal or larger?

My logic is that 3D airflow leakage on tips will decrease overall pressure difference at wing ,so lift will be smaller than in 2D wing-test but drag will be also smaller than in 2D-test,because when you integrate smaller pressure difference over airfoil surface you will get smaller drag as well.

(BUT if we must have SAME LIFT IN BOTH TEST,than 3D wing test will have larger drag,because then we must increase AoA at 3D wing to compensate lift reduction caused by airflow leakage,so now when we have increased AoA we also have larger drag..)

3D wing will have less drag,drag decrease when AoA decrease,3D wing "feel" effective AoA,so it fly at smaller "AoA" than 2D wing.

2D vs 3D

enter image description here

how blue-lift can be larger than red-lift,wing at AoA=10 can not produce larger lift than wing at AoA=20 ??

enter image description here


2 Answers 2



At relatively high Reynolds number (>1 million) and in attached flow, where the boundary layer is thin, 3D wing should always have larger overall drag than a 2D airfoil (infinite span), at the same free-stream incidence angle.


The downwash generated from trailing vortices is actually very small compared to the free-stream airspeed; therefore, the reduction in effective AOA is also very small.

This results in three effects:

  1. Reduction in lift due to tilting. This effect is exceedingly small, since cosine of a small angle is essentially unity.

  2. Reduction in lift and pressure drag due to lower effective AOA. The reduction in lift is small but definitely non-negligible. For unseparated flow around an airfoil, the overall pressure drag is very small to begin with, so the reduction is even smaller.

  3. Creation of induced drag due to tilting of the lift vector. This is the main one, since sine of a small angle produces non-negligible linear factor. For example, if the local lift coefficient, $C_l$, is 0.4, and we have an effective decrease in AOA of 1.0deg, then $C_l\cos{\frac{1.0\pi}{180}}\approx 0.9998C_l\approx0.4$, while the local induced drag coefficient, $C_{d_i}$, is $C_l\sin{\frac{1.0\pi}{180}}\approx0.0175C_l=0.007$.

By the way, the drag polar you cited doesn't help your argument since it's for the whole wing (at aspect ratio of 6). What you really want to cite is the airfoil drag (from Airfoil Tools):

Clark-Y airfoil lift and drag

As you can see, the form drag (includes pressure drag) is a small portion of the overall 3D drag at aspect ratio of 6, in the linear lift range. The pressure drag increase is even smaller.


From your comments, it seems you have totally missed the point about induced drag. Induced drag is completely absent in 2D. In fact, if you assume that the flow is inviscid and without local supersonic flow, you have zero drag on any 2D airfoil. This is called d'Alembert's paradox.

As mentioned above, induced drag appears in 3D inviscid flow due to finite span and induced downwash.

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – Federico
    Commented May 4, 2020 at 7:38
  • $\begingroup$ @JZYL This is only true when compare for the same lift,not for same AoA.JanHudec explain that here :aviation.stackexchange.com/questions/80839/… $\endgroup$
    – user52248
    Commented Sep 6, 2020 at 8:46
  • $\begingroup$ @Сократ, no this answer is absolutely correct. This compares a 2-D airfoil with a 3-D one, but in the linked answer we are comparing the same 3-D airfoil with and without end-plates, which makes a big difference. $\endgroup$
    – Jan Hudec
    Commented Sep 6, 2020 at 9:18
  • $\begingroup$ @JanHudec What is difference in puting wall to wall 3D airfoil in wind tunnel or 3D airfoil in ductec fan? in both case you get 3D airfoil with infity big endplates What is definition of 2D airfoil?Just because he put airfoil inside wind tunel from wall to wall ,that doesnt mean he has "theoretical" 2D airfoil with finite lift and infity span=zero induced drag? "theoretical"2D airfoil can not be simulate in real life $\endgroup$
    – user52248
    Commented Sep 6, 2020 at 9:39
  • $\begingroup$ @Сократ a “2D” airfoil means infinite and for rather weird reasons that can produce even infinite lift with zero (not even just finite) induced drag. But it is a theoretical construct and that property comes out of the limits for increasing span. Endplates, no matter how big, will not make the wing behave as if it was infinite, only as if it was a little longer. $\endgroup$
    – Jan Hudec
    Commented Sep 6, 2020 at 11:17

3D wing has more drag as it considers vortices forming at the wing tip which creates more drag.But incase of 2D wing, the wing is assumed to be infinite wing. Aerodynamically, the effect of trailing vortices reduces the slope of the coefficient of lift vs. angle of attack curve. As a result,the coefficient of lift is higher in 2D wing than 3D wing for same AoA. enter image description here


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .