Musings on Lift

[PT20J]

The March 2025 AOPA Pilot magazine got me thinking: I wonder why it seems so hard to understand how our wings produce lift?

Bernoulli? Newton? Both? Neither?

It's not as hard as we make it, but it is not intuitive and trying to make it match our intuition seems to me to be what trips up those that try to explain it in simple terms.

Consider Bernoulli. Yes, the air flowing over the top of the wing speeds up and produces a lower pressure. But why does it speed up? It is NOT because the air above the wing is funneled through some sort of half venturi. This is easily shown by the fact that a flat sheet will generate lift at a positive angle of attack, but, since there is no camber, there is no half venturi.

Consider Newton. Lift is not created by downwash. Downwash creates drag. Downwash is due to the tip vortices of finite span wings. 2D airfoils generate lift but have no downwash. Ground effect reduces downwash with the effect that drag decreases and lift increases. If downwash was responsible for lift, lift would decrease as we neared the ground and our Mooneys wouldn't float so much.

There are two problems that impede our understanding of lift. The first is that many of the drawings of streamlines flowing past an airfoil are incorrectly drawn. An airplane wing actually affects the airflow at a considerable distance ahead, above, below and behind the wing. The second problem is that we are used to simple systems that have discernable cause and effect. But fluid dynamics isn't like that. If the air speeds up, is it because the pressure changed? Or, does the pressure change because the air speeds up?

What is really happening is that the wing presents an obstacle that the air must go around. And, there are laws of nature that the air must obey in so doing. The air must change direction to get out of the way. Air has mass so changing its direction requires a force. Force on a fluid such as air is a pressure. So, as the air flows around the wing there will be pressure gradients. And pressure gradients cause the air to speed up or slow down. It's not correct to say that the pressure causes the direction and speed changes or that the direction and speed changes cause the pressure changes because all three parameters are part of one system. 

[Hank]

8 hours ago, PT20J said:

The March 2025 AOPA Pilot magazine got me thinking: I wonder why it seems so hard to understand how our wings produce lift?

Bernoulli? Newton? Both? Neither?

. . .

Consider Newton. Lift is not created by downwash. Downwash creates drag. Downwash is due to the tip vortices of finite span wings. 2D airfoils generate lift but have no downwash. Ground effect reduces downwash with the effect that drag decreases and lift increases. If downwash was responsible for lift, lift would decrease as we neared the ground and our Mooneys wouldn't float so much.

Yes, that is certainly "a" downwash, but as you pointed out, it also certainly does not create lift.

The Newtonian downwash theorized to produce lift is spanwise along the wing, not way out just at the tip. As the airfoil moves through air, it displaces air in a downward direction along its length, and by the 3rd Law, the downward moving air creates an upward force on the wing. This is why introductory aerodynamics discusses infinite length airfoils, to eliminate the special causes of drag created at the tips.

Then there is the Bernoulli Effect.

Our airplanes fly because of both the Newtonian upward force caused by pushing air down, and the Bernoulli effect cause by pressure drop atop the wing. Newton pushes up, Bernoulli pulls up, and away we fly.

Newtonian lift explains why a flat sheet (or your tilted hand stuck out the car window) generates lift. The lift can be increased by changing the plate / hand into an airfoil shape, and the airfoil can be optimized for a particular flight speed and air density; NACA generated A LOT of data covering this decades ago that designers still use.

[PT20J]

The spanwise vortex sheet is a consequence of the tip effects. Lift can be calculated by either accounting for the pressure differences above and below the wing or the momentum changes in the air as it moves around the wing. But the latter is not the same as ‘forcing air down” as the example of ground effect demonstrates. Aerodynamics text books first describe 2D airfoils in order to derive equations of lift and drag and then go on the describe 3D wings in order to account for the effects of planform and span. The NACA catalog of airfoils is 2D wind tunnel data because the wing models completely filled the test section from wall to wall so there were no tips.

[0TreeLemur]

One of my favorite views of lift generation in action.  Streaklines in a wind tunnel at an angle of attack near stall.

Newton's second law F=ma written perpendicular to the streamlines does  a better job of describing the physics that produces lift better than any other equation.  It clearly shows that streamline curvature produces a pressure gradient in the direction opposite of the radius of curvature.  The minus sign indicates that the pressure decreases in the direction pointing towards the center of curvature, the n direction on the figure.   The wind tunnel observation also shows why our stall detectors are placed where they are.

 

[varlajo]

7 hours ago, Hank said:

Newton pushes up, Bernoulli pulls up, and away we fly.

Well said Sir, well said!! 

[Joshua Blackh4t]

I remember a comic from a flying magazine. It had graphical illustrations of a quiz.

What makes a wing fly? Is it:

A: Bernoulli

B: Newton

C: Coanda effect

D: Bucket loads of cash

[AH-1 Cobra Pilot]

I have not yet read that issue of AOPA Pilot, but the conversation above makes me think of three things:

As nobody above has mentioned it, does the article talk about compressible vs. incompressible flow?

Pressure is simply Newton's Third Law in gas.

Most of these things are really obvious if you take some time to play with a wind tunnel that has smoke generators and a rotatable airfoil.

[N201MKTurbo]

19 hours ago, PT20J said:

 

What is really happening is that the wing presents an obstacle that the air must go around.

So, not really. The air is stationary (sort of) and the wing must pass through it. But your analogy does not make the analysis incorrect. 

[N201MKTurbo]

So, the disturbances ahead of the wing travel forward from the wing at the speed of sound. When the wing is moving faster than these disturbances you hit the sound barrier. This drastically changes the airflow over the wing. But we don’t have worry about these things….

[PT20J]

A lot of YouTube videos are just wrong. Here’s a good one by a retired Boeing engineer (his book goes into more detail)

This one is good also.

 

[Utah20Gflyer]

Does a propeller create thrust by creating lower pressure in front of it or by moving the air from in front of the propeller to behind it!  A propeller is a miniature wing.  
 

Lift is created by forcing air downward.  The significance of the lower air pressure above the wing is that the air is being sucked downward and therefore is creating extra lift through the direction change downward.   The process of changing the direction of the air creates drag.  
 

This explains ground effect perfectly since the downward moving air is now hitting a solid surface and the air can’t move as freely as it does at altitude.  
 

Without forcing air downward you can’t push a plane up.  It really is just that simple.  

[MikeOH]

Not sure what the point of this thread really is....starts out with "it's not as hard as we make it" and the next thing you know we've got the triple-integral of Biot-Savart!

@Utah20Gflyer get's it, IMHO 

[EricJ]

47 minutes ago, Utah20Gflyer said:

Does a propeller create thrust by creating lower pressure in front of it or by moving the air from in front of the propeller to behind it!  A propeller is a miniature wing.  

They do both.   Standing behind a running prop demonstrates the thrust from the high pressure pretty readily.   Sucking up rocks during a static stationary run demonstrates the low pressure created in front of the prop.   Wings are the same.   When the sun is low it's pretty easy to see the skin on top of the wing puffing up in between the ribs and spars, like it's being inflated, but it's just the low pressure on top of the wing pulling the skin up.  It's kind of cool to realize that that force pulling up the skin is a lot of what's holding the airplane up.

[Utah20Gflyer]

1 hour ago, EricJ said:

They do both.   Standing behind a running prop demonstrates the thrust from the high pressure pretty readily.   Sucking up rocks during a static stationary run demonstrates the low pressure created in front of the prop.   Wings are the same.   When the sun is low it's pretty easy to see the skin on top of the wing puffing up in between the ribs and spars, like it's being inflated, but it's just the low pressure on top of the wing pulling the skin up.  It's kind of cool to realize that that force pulling up the skin is a lot of what's holding the airplane up.

I agree both are happening.  Because you are flying through a homogeneous air mass and you want to create a zone of high pressure under the wing you necessarily have to create a low pressure area above the wing.  You can’t have one without the other.  
 

For every action there is an opposite and equal reaction.  So to me it makes more sense that it’s the downward high pressure that is pushing the plane up against gravity rather than an equal and non opposite (aligned) force that is pulling it up.  
 

I think this is more intuitive with propellers and helicopters than our wings because wings are so big and slow moving.  They only have to displace the planes weight worth of air to stay aloft and because that is spread over such a large area it’s not obvious to us how much air they are pushing down.  The tip of a propeller blade is approaching the speed of sound which is way faster than our wings move through the air.   Therefore the effect is exaggerated in comparison.  

I accept I could be wrong on this and many other things, but this is how I think about it.  It’s a bit of a chicken and egg situation.  
 

 

[AH-1 Cobra Pilot]

45 minutes ago, Utah20Gflyer said:

I agree both are happening.  Because you are flying through a homogeneous air mass and you want to create a zone of high pressure under the wing you necessarily have to create a low pressure area above the wing.  You can’t have one without the other. 

Not quite correct. 

1.  Think of a symmetrical airfoil at zero angle of attack.  Both sides will have high pressures in equal amounts up to a portion of the chord, and you have no lift or moment, but you do have drag.

2.  How would that explain a stall?

[Utah20Gflyer]

1 hour ago, AH-1 Cobra Pilot said:

Not quite correct. 

1.  Think of a symmetrical airfoil at zero angle of attack.  Both sides will have high pressures in equal amounts up to a portion of the chord, and you have no lift or moment, but you do have drag.

2.  How would that explain a stall?

Approximately half of the air being forced downward is coming from the top side of the wing.  When the air detaches from the top surface it is no longer being forced downward and you lose a lot of your lift. 
 

When you look at the side section of an airfoil you’ll see that the air from the top of the wing is actually being forced downward at a greater angle than the bottom of the wing.  You gain a lot of efficiency from that top surface smoothly dropping down toward the rear.  Sucking that air downward.  
 

I guess I would think about this like a leaf blower.  It’s a device that performs work by creating air pressure differentials.  At the intake there is low pressure and at the nozzle there is high pressure.  The point of the device is to move the leaves with the high pressure air and I would propose to you that an airplane is also a device that uses high pressure air to do its work.  Low pressure is just a required side effect of creating the high pressure that is actually performing the work.  
 

But you can’t create high pressure without also creating low pressure so you could say that a plane flies by creating low pressure and that would technically true, but I think it’s more true to say it’s the high pressure that is making the plane fly.   Plane go up, air go down! 

[PT20J]

For those that believe that lift is caused by air being forced downward, consider these questions:

1. How do you explain lift generated by a 2D airfoil (like a wind tunnel model) that has no net downwash? (If you look at @0TreeLemur's wind tunnel picture, you'll notice that the downwash behind the airfoil is offset by an upwash ahead of the airfoil. So, the net downwash is zero.)

2. If lift is caused by forcing air downward, reducing downwash should reduce lift. So, how do you explain the fact that lift is increased slightly in ground effect while downwash is reduced (because the air cannot continue moving downwards due to the presence of the ground plane).

 

[M20F]

Me push throttle, me get lift, me no engineer. 

[PT20J]

5 hours ago, MikeOH said:

Not sure what the point of this thread really is....starts out with "it's not as hard as we make it" and the next thing you know we've got the triple-integral of Biot-Savart!

The point is that there are many fallacies about lift that keep getting repeated and now are so ingrained that we don't give them much thought. If you throw out the incorrect drawings and move beyond a rigid cause-effect mindset, then it's pretty easy to understand what is happening based on simple physics. It only gets complicated when you want to calculate the exact amount of lift, but that's what computational fluid dynamics programs are for.

[Utah20Gflyer]

18 minutes ago, PT20J said:

For those that believe that lift is caused by air being forced downward, consider these questions:

1. How do you explain lift generated by a 2D airfoil (like a wind tunnel model) that has no net downwash? (If you look at @0TreeLemur's wind tunnel picture, you'll notice that the downwash behind the airfoil is offset by an upwash ahead of the airfoil. So, the net downwash is zero.)

2. If lift is caused by forcing air downward, reducing downwash should reduce lift. So, how do you explain the fact that lift is increased slightly in ground effect while downwash is reduced (because the air cannot continue moving downwards due to the presence of the ground plane).

 

I fully concede that this topic probably exceeds my expertise. 
 

But for those that don’t believe the wings push air down.  Why does it get so windy when I start my planes engine?   What’s up with that rotor wash from Helicopters?  
 

[0TreeLemur]

For incompressible flow (M<0.3), streamline curvature explains >95% of the pressure distribution on an airfoil, with viscosity (vorticity) playing a minor role.   Downwash has very little to do with lift produced by a wing.  Lift results from either angle of attack and/or asymmetry in the airfoil itself.  An asymmetric airfoil like the one  on our favorite airplane in normal cruise has 0-degree angle of attack!  How much downwash do you think that is that causing?   As Skip pointed out in the wind tunnel photo above, which is a symmetrical airfoil at a high angle of attack,  the net vertical velocity component around it very nearly zero.

In the case of a symmetric airfoil at 0-degree angle of attack, the lift is zero because the streamline curvature is the same on both upper and lower surfaces.  But as the wind tunnel photo shows, angle of attack greatly modifies the flow field and results in a lot more streamline curvature above the airfoil than below it.   That's all there is to it.  No magic.

Once you get into the compressible flow regime, then the magic starts to appear along with a few demons.

[Utah20Gflyer]

9 minutes ago, 0TreeLemur said:

For incompressible flow (M<0.3), streamline curvature explains >95% of the pressure distribution on an airfoil, with viscosity (vorticity) playing a minor role.   Downwash has very little to do with lift produced by a wing.  Lift results from either angle of attack and/or asymmetry in the airfoil itself.  An asymmetric airfoil like the one  on our favorite airplane in normal cruise has 0-degree angle of attack!  How much downwash do you think that is that causing?   As Skip pointed out in the wind tunnel photo above, which is a symmetrical airfoil at a high angle of attack,  the net vertical velocity component around it very nearly zero.

In the case of a symmetric airfoil at 0-degree angle of attack, the lift is zero because the streamline curvature is the same on both upper and lower surfaces.  But as the wind tunnel photo shows, angle of attack greatly modifies the flow field and results in a lot more streamline curvature above the airfoil than below it.   That's all there is to it.  No magic.

Once you get into the compressible flow regime, then the magic starts to appear along with a few demons.

Looking at the picture of the airfoil in the wind tunnel the air to the rear of the wing is lower than the air in front of the wing.   What’s up with that? 

[PT20J]

33 minutes ago, Utah20Gflyer said:

Looking at the picture of the airfoil in the wind tunnel the air to the rear of the wing is lower than the air in front of the wing.   What’s up with that? 

The airfoil is at a positive angle of attack so the trailing edge is lower than the leading edge.

[Hank]

30 minutes ago, 0TreeLemur said:

An asymmetric airfoil like the one  on our favorite airplane in normal cruise has 0-degree angle of attack!  How much downwash do you think that is that causing?

No, we don't fly at 0° angle of attack. The angle is measured from the center of curvature on the leading edges to the pointy tip on the trailing edge.

There is probably a single weight and a single speed that will approach 0°, but my plane rarely weighs the same for very long, or on very many flights. Lift also depends on temperature and air density, both also highly variable. So angle of attack is different for each flight, and for each moment of each flight. 

We do maintain level flight when the weight of the airplane is balanced by the lift the wing is producing. Some flights start heavier, and require more lift, so more angle of attack; some are in thinner air (due to higher temp or lower barometric pressure, or both) and require more angle of attack; some are at higher speed, and require lower angle of attack, because lift is proportional to speed.

To laymen like us, "downwash" is the body of air that the wing forces down. To an aerodynamicist, it's the much smaller portion of air that spills off the wingtip. The former creates lift. As the wing moves through air, air molecules must move out of the way--some accelerate upwards, creating a low pressure area that lifts the wing; some accelerate downward, creating a high pressure area that lifts the wing. Some spill off the wingtip, creating drag, which slows the airplane (and lower speed creates less lift, requiring higher angle of attack; round and round the mulberry bush we go!).

So part of the issue is simply terminology--there are two different concepts both called "downwash" by two different groups of people, with different effects. Sadly, both are correct . . . .

[Hank]

1 minute ago, PT20J said:

The airfoil is at a positive angle of attack so the trailing edge is lower than the leading edge.

And the airstream impacting the bottom of the wing pushes it upwards, according to Sir Isaac Newton.

Oncoming air is split--some goes upward, moving without physical limitations and creates low pressure due to increased flow speed; some is pushed downwards, and because air is compressible, it creates an area of high pressure as flow velocity is reduced by the change in direction. But air from the top surface is also deflected downwards simply because the flow direction on top of the wing is downwards for most of its length.