How do airplanes fly upside down?

How do military jets and stunt planes fly upside down?

I have only rudimentary understanding of the physics of how airplanes fly. From what I understand planes couldn’t possibly fly upside down. Empirical evidence would suggest otherwise however, so my understanding must be flawed or lacking (or both).

Wings produce lift, right? In right-side-up flight the lift is being produced in the opposite direction than gravity. Lift > gravity = ascent, lift = gravity = level flight, lift < gravity = descent. Right?

But upside-down, those same wings are still producing lift, right? However, the lift is now being produced in the **same ** direction as gravity. So…. What’s keeping the plane up?

I also understand that thrust plays a role in ascent and descent thusly: point the nose of the plane up and the combination of lift (from the wings) and being “pushed” up will make the plane climb faster than with just “wing-lift” along. Likewise point the nose down and you overcome some (all?) of the “wing-lift” because the entire plane is being “pushed” down. The extreme example of this is rockets like those used to launch satellites – they ascend completely from “push” (thrust) but no “wing-lift”.

So I can see that a plane could fly upside down if the “push” from the engines completely overcomes both gravity and the “wing-lift”, both of which are now pointed in the same direction (the ground). However, this makes me think that the nose of the plane would have to be pointed away from gravity/“wing-lift”/the ground at a very severe angle.

But I’ve seen film, pictures, etc. of military jets and stunt planes flying upside down with their wings apparently parallel to the ground. How??? What am I missing here?


There’s your answer. Inverted flight actually requires the nose to be slightly “up” (as in pointed toward the sky) to produce lift. Very inefficient, to be sure, but enough to sustain flight.

So, how do they fly “knife-edge” (wings vertical)?

They don’t. Not for very long anyway. It’s purely momentum, with very slight lift generated by the fuselage.

Unless you’ve got a high-performance jet with engine thrust exceeding the weight of the plane. Then you can fly (and accelerate) straight up.

Welcome to the boards! :slight_smile:

We covered this in a thread about a month ago. A search for “upside down” should be useful.

I think by knife-edge he meant flying horizontal with the wings in a vertical position, generating no lift. Unless you point the nose slightly up or get enough lift from the fuselage, no amount of thrust will keep that plane in the air for long.

if a plane flying normally lowers it flaps to decrease altitude, could a plane flying upside down lower it’s flaps to gain altitude? Is this ever done?

Lowering flaps does not equal “decrease altitude,” at least not directly. The purpose of the flaps in normal flight is to increase the lift at low speeds by increasing the apparent angle of attack. “Angle of attack,” if you didn’t know, is the angle between the wing and the incident air, so pitching up increases the angle of attack. The flaps also increase parasitic drag, so airspeed and altitude will decrease.
Lowering the flaps while inverted will have the opposite effect, i.e. it will decrease the angle of attack, while still increasing drag, so you’d lose a lot of lift. Doesn’t seem like a safe thing to do, in my opinion.

Planes don’t deply flaps to lower altitude…they deploy flaps to increase lift at a given speed. this does haf the effect of increasing drag, as well, thuse causing the plane to lose speed and altitude. But the flaps are there so that the plane can fly at lower speed without stalling. (and it does this by increasing lift)

Well, you wouldn’t be able to tell by the way I butchered the English on my post, but as a praciticing aerospace engineer, I am, in fact, qualified to answer the question.

Preview is your friend. Actually, simply proofreading is your friend.

What you understand is not the problem. The problem involves what you were taught. The vast majority of intro textbooks are wrong in their explanations of flight. You have to read the aerodynamics texts to encounter the right stuff.

The introductory texts contain what’s now known as the “equal transit time fallacy”, also called the “longer path fallacy.” They state that parcels of air are separated by the leading edge of the airfoil, and the two halves of the parcel must race to meet each other at the trailing edge. Therefore an asymmetrically-shaped airfoil will generate lift. Nope. Wrong.

In fact, parcels of air which are split at the leading edge do not recombine at the trailing edge. They never have, and it’s no suprise to the aerodynamics physicists (actually, one of the half-parcels vastly outraces the other, and they never meet again.) Only the intro textbooks and pilot-training textbooks use the false fact to explain where lift comes from. And because of the fallacy, they wrongly ascribe the cause of the lifting force to the airfoil asymmetry.


Airfoil misconception in textbooks
Articles about the controversy:

NASA GRC articles: debunking incorrect explanations
So how DO wings work? Here’s my attempt (note the animation)


A physical description of flight

See how it flys: airfoils

Note that the shape of the wing is important in avoiding the “stall” effect …but for generating lift, the only important thing is the angle of the trailing edge of the wing. If the trailing edge does not deflect air downwards, then the wing will not generate a lifting force.

Not quite. When lift > gravity, the aircraft accelerates upward. In a steady climb, lift = gravity, just as in level flight or in a descent.

Actually, it’s a bit more subtle than this - in a climb, what opposes gravity is a combination of lift and thrust. This could lead to a long discussion not especially germane to the question about inverted flight (very well answered by bbeaty, who is something of a crusader for the truth on this subject).

As well as Jan-Olov Newborg!

Anyone who wants more online messages than they can possibly read can find ten years of threads about this controversy. Search on keywords:

Here’s the guy who started it all with a journal article in AJP back in 1987:

Even Einstein was taken in by the misconceptions, and during WWI in Germany tried to design an improved airfoil based on the textbook mistakes. Didn’t work. See “Albert Einstein’s Wing” in this article:

bbeaty has it right, from what I understand.

I, as everyone else, was taught that airplanes generate lift from bernoulli’s principle - essentially, that the air molecules on the top surface have to flow faster then those on the bottom, and therefore create left. It turns out that this accounts for only a small portion of the lift required for airplanes to fly.

A couple years ago I bought a book that had a more realistic explanation of how things work, and it’s very much in line with what bbeaty seems to be saying. The explanation could be boiled down to this - airplanes fly at a slightly nose-up attitude. This causes a portion of the underside of the wing to be exposed to oncoming air. The oncoming air hits the underside of the wing, and because of forward air speed, this causes a region of high pressure underneath the wing. It’s this pressure that ‘pushes’ the plane up and allows it to fly.

I had a hard time believing it at first - afterall I went through high school physics and a chemical engineering program and was taught that lift is generated by air molucules moving faster over hte wing then underneath it. It was tough to let a $15 book contradict a multi-thousand dollar education (which I’m still paying for!). But what settled it for me was this - take any sort of flat, light, long rectangular object. Say a piece of cardboard 4 inches wide, and 3 feet long. Align it parallel with the ground and move it through the air. No lift. Now, raise the leading edge up slightly, and move it through the air. All of a sudden, it takes off and heads into the air! At the right speed and pitch attitude, it will ‘skip’ across the air and stay level.

The interesting thing about this is that the cardboard has absolutely no airfoil whatsoever… I’m still searching for an explanation of why an airfoil is necessary (other than decreasing drag).

So anyway, back to your question - how can an airplane fly upside down? By maintaining slightly nose up attitude. The airfoil lift is only a portion of that necessary to fly, and is counteracted by the region of high pressure generated underneath the wing.

And in answer to another question - how to planes fly knife edge. There is some lift generated by a fuselage, but it also helps to have a large surface area rudder, and you’ll usually notice that in knifeedge flight, the rudder is fully deflected in the relative ‘up’ position (that is, if a plane is flying with it’s left wing down, the rudder is set to full right). In this case, the rudder acts in the same way the elevator does in straight and level flight.

I don’t think this is much of an improvement over the “Bernoulli” explanation - bbeaty’s links don’t argue for this view of lift.

For a really simple intuitive look at the issue, consider a hovering helicopter. If you’ve ever stood underneath one, or seen pictures of one hovering over water, you know that it is pushing a massive amount of air downward with its rotor. Which makes sense - if you want to keep a giant object like that in the air, you need to generate a force equal to its weight, and the only way to do that is to mash a lot of air downward.

But (as is apparent when you look at one standing still) a helicopter rotor blade is nothing more than a long and narrow wing that travels in a circular path. A fixed wing traveling in a straight line is doing exactly the same thing - deflecting air downward.

No, No, No, airplanes fly because of magic. The engines and propellers and turbine vanes and high by pass ratios fan are all for cooling the pilot who is working the magic. Evidence of the sweat produced when the engines all stop is proof of what I say…

I could swear this was covered in a Cecil column some time ago. :slight_smile:

The omniscient one didn’t exactly cover himself with glory here. For example he states: “increasing your angle of attack also increases your aerodynamic drag. Too much drag = stall = plane drops like rock. For that reason stunt planes need low-drag wings”

It’s misleading to imply that a stall is caused by high drag, and wrong to say that “stunt” planes need low-drag wings (in fact, their wings have rather high drag, which can be useful).

Another way to test the high-pressure-on-bottom-of-wing idea is to do what I did (I feel proud that I figured that bit out intuitively :)). Get your friend to drive a car on the highway and stick your arm out the window (open it, first). Then you can use your flattened hand as a wing, and feel it get lifted upwards, pushed down, or stay resonably level. It’s fun. Only do it when there’s nothing you could hit your arm on, however.

Oh no! Cecil is wrong! He’s presenting the “60% 40% fallacy,” as if Bernoulli’s Equation wasn’t based on Newton’s laws.

In fact, 100% of the lifting force is predicted by the Bernoulli Equation. After all, the only way a force can appear on an airfoil is because of surface pressure. The air above a wing really does flow significantly faster. If you use a tilted barn door as your wing, the air above the barn door does flow significantly faster. See

But also 100% of the lifting force can be found by measuring the acceleration of each parcel of air near the wing, applying F=mA to calculate the net downwards force, and applying Newton’s 3rd to find the equal and opposite upwards force on the wing.

The shape of an airfoil is important in order avoid the “stall” effect. But the airfoil shape does not create the lifting force, instead the whole process hinges on the angle of the trailing edge of the airfoil. If the trailing edge does not deflect air downwards then there’s no “circulation”, no net deflection of air, and no lift.

Here’s an asymmetrical (cambered) airfoil adjusted to give zero lift (it’s the top image of the three.) Note the angle of the trailing edge:

Lots of confusion arises because the trailing edge of a cambered airfoil is angled downwards when the airfoil as a whole is not tilted. For decades people have been focusing on the humped-up center of the airfoil as the cause of lift. But it’s the downward-angled trailing edge that does it. In other words, lift is ALWAYS created by “angle of attack,” it’s just that only the rear part of the airfoil determines how the air moves, not the “angle of attack” of the airfoil as a whole.

Yarg! Wrong!

Here’s the GIF I wanted.

These are from “See How It Flys,”  3  Airfoils and Airflow

I agree.

This especially bugs me. High AOA (ie nose “up” tail “down”) while inverted would mean a rapid impact with terra firma. AOA doesn’t care if you are inverted, right side up or in a 90-deg bank. It’s strictly relevant to the mean chord of the wing and the incident air. A high AOA while inverted means you are pointed toward the ground.

Well, it is inefficient but not because you are increasing your AOA. It’s inefficient because you are using your elevator to produce a negative AOA (on a wing with conventional design), but still keeping the thing airborne.

Flying inverted induces -1G on an aircraft (Plus or minus small variances depending on the ham-fistedness of the pilot). Nothing horrendous, and I’d wager that any aircraft certified by the FAA can withstand this stress for a VERY long time. With regards to the engine, the most powerful prop (and jet) engine in the world will “conk out” while inverted if it doesn’t have boost pumps for the fuel and pressurized oil sumps for lubrication. Similarly, almost any engine can be made “inverted-proof” with a few modifications. “Stunt plane” engines are powerful for many reasons, but power has nothing to do with “conking out” while upside down.

P.S. Has this already been discussed in “Comments About…”?