What You Learned About How Planes Fly Was Probably Wrong(danielmiessler.com)
danielmiessler.com
What You Learned About How Planes Fly Was Probably Wrong
http://danielmiessler.com/blog/why-planes-fly-what-they-taught-you-in-school-was-wrong
10 comments
For anyone interested in this I recommend the really great "What makes airplanes fly" by Peter P Wegener.
http://www.amazon.com/What-Makes-Airplanes-Peter-Wegener/dp/...
This book was part of a course he taught for non aerospace engineers. His explanations of tough topics like boundary layer theory and airfoils are clear
http://www.amazon.com/What-Makes-Airplanes-Peter-Wegener/dp/...
This book was part of a course he taught for non aerospace engineers. His explanations of tough topics like boundary layer theory and airfoils are clear
Kind of expensive for anything short of consuming interest, though.
Here's more info on the author
http://opa.yale.edu/news/article.aspx?id=6066
Apparently he worked on the V2 rocket. But there are lots of scientists that can write technical books targeted to experts. This book is worth it because this guy is devoted a lot of his teaching career to explaining difficult topics to non-experts. I think he did a marvelous job, of mixing technical material and history. I think this book was part of a course he taught at yale univ.
It is an expensive book, but I'll go in and say it's worth it. Check your univ library.
http://opa.yale.edu/news/article.aspx?id=6066
Apparently he worked on the V2 rocket. But there are lots of scientists that can write technical books targeted to experts. This book is worth it because this guy is devoted a lot of his teaching career to explaining difficult topics to non-experts. I think he did a marvelous job, of mixing technical material and history. I think this book was part of a course he taught at yale univ.
It is an expensive book, but I'll go in and say it's worth it. Check your univ library.
The used copies of the older first edition are reasonably cheap.
http://www.amazon.com/gp/offer-listing/0387975136/
http://www.amazon.com/gp/offer-listing/0387975136/
$23.48 shipped with some of the sale going towards literacy charities.
http://www.betterworldbooks.com/What-Makes-Airplanes-Fly-id-...
http://www.betterworldbooks.com/What-Makes-Airplanes-Fly-id-...
Another good discussion about this is in the online pilot textbook "See How It Flies" at http://www.av8n.com/how/ (section 3). That author also disagrees with the Coanda effect conclusion.
The discussion in there is the best I've seen, it centers on the concept or circulation, the vorticity of the flow field around the wing. One example that's used is the fact that if you just take a wing by itself (try it with a long, flat piece of cardboard) it will "fly" while rotating in the air. It doesn't fly very well, admittedly, but it certainly exhibits a glide ratio a lot better than something with the same cross-sectional area that's not a wing. In that case, there's no top or bottom of the wing at all.
The discussion in there is the best I've seen, it centers on the concept or circulation, the vorticity of the flow field around the wing. One example that's used is the fact that if you just take a wing by itself (try it with a long, flat piece of cardboard) it will "fly" while rotating in the air. It doesn't fly very well, admittedly, but it certainly exhibits a glide ratio a lot better than something with the same cross-sectional area that's not a wing. In that case, there's no top or bottom of the wing at all.
Actually the whole av8n.com site is a great resource for anyone interested in learning more about physics. One of the reasons why I linked it ;).
Hm. a hundred years after first plane took off, we still argue how exactly do they fly...
Not because there's something especially mysterious about it; it's just an instance of the more general rule that (a) people misunderstand anything complicated and (b) this doesn't stop them from expressing opinions about it.
There is one thing that's slightly different in this case: the misleading coolness of diagrams of the Bernoulli effect. One can easily see how in the hands of a popular writer, that could get transformed from an interesting optimization for wing cross-sections to "how planes stay up." I was fooled by this till a later age than I'd like to admit. It never occurred to me to ask how planes fly upside down, if the shape of the airfoil is how they generate lift.
There is one thing that's slightly different in this case: the misleading coolness of diagrams of the Bernoulli effect. One can easily see how in the hands of a popular writer, that could get transformed from an interesting optimization for wing cross-sections to "how planes stay up." I was fooled by this till a later age than I'd like to admit. It never occurred to me to ask how planes fly upside down, if the shape of the airfoil is how they generate lift.
Well, but it is complicated. Hydrodynamics are not intuitive, and the fact is that none of these rules that are being thrown around are sufficient to solve the problem. They all just express different aspects of it.
You can't say, is it the Bernoulli effect or is it the fact that the wing makes the air behind it go down. Those are both different ways of expressing the same thing. You can't have one without the other. This is why I like the "See How it Flies" discussion.
You can't say, is it the Bernoulli effect or is it the fact that the wing makes the air behind it go down. Those are both different ways of expressing the same thing. You can't have one without the other. This is why I like the "See How it Flies" discussion.
> You can't say, is it the Bernoulli effect or is it the fact that the wing makes the air behind it go down. Those are both different ways of expressing the same thing. You can't have one without the other.
Actually, you can. It's fairly straight-forward to construct an example where you do get the Bernoulli effect and there is no downwash. What is true is that when you fly you have both, and both contribute.
What's more true is that there are many effects, each of which contributes, many of which are inter-dependent, and all of which are simple in isolation, and complex in interaction and action.
Actually, you can. It's fairly straight-forward to construct an example where you do get the Bernoulli effect and there is no downwash. What is true is that when you fly you have both, and both contribute.
What's more true is that there are many effects, each of which contributes, many of which are inter-dependent, and all of which are simple in isolation, and complex in interaction and action.
Well, yeah, I meant in the explanation of why an airplane flies. Sure, in a venturi you have Bernoulli effect with no downwash, but the fluid is instead changing speed. The effect is still there, it's just operating in a different direction. Any time you have a pressure change, you must have an acceleration.
In the venturi, the pressure is lower where the speed is higher, that's a fact. But is the pressure lower because the speed is higher, or is the speed higher because the pressure is lower? That question makes no sense, because it depends on how you think about it.
On the one hand, you can say: mass conservation dictates that the fluid must go faster in the narrow part of the tube. If the fluid is to go faster, it must accelerate, so there must be a pressure gradient. Hence, the pressure in the narrow part must be lower.
On the other hand, you can say: Since the fluid goes faster in the narrow part of the tube, the pressure is lower there. Since the pressure is lower, there's a pressure gradient, and that's what causes the fluid to speed up.
Neither of these explanations make sense, because there is no cause and effect in the problem, it's just that one state globally obeys all constraints on the fluid and that's the state with higher speed and lower pressure in the narrow part.
In the venturi, the pressure is lower where the speed is higher, that's a fact. But is the pressure lower because the speed is higher, or is the speed higher because the pressure is lower? That question makes no sense, because it depends on how you think about it.
On the one hand, you can say: mass conservation dictates that the fluid must go faster in the narrow part of the tube. If the fluid is to go faster, it must accelerate, so there must be a pressure gradient. Hence, the pressure in the narrow part must be lower.
On the other hand, you can say: Since the fluid goes faster in the narrow part of the tube, the pressure is lower there. Since the pressure is lower, there's a pressure gradient, and that's what causes the fluid to speed up.
Neither of these explanations make sense, because there is no cause and effect in the problem, it's just that one state globally obeys all constraints on the fluid and that's the state with higher speed and lower pressure in the narrow part.
An interesting video, that reinforces everything I saw when I explain these effects - I've taken a copy and must get in touch with the owner to talk about using it.
The point about Bernoulli is that it only applies in its naive form in fluid flow where it's effectively a closed system. If you introduce airstreams of varying speeds then all bets are off.
Effectively Bernoulli works because the velocity changes are being caused by the pressure differences. Take a very, very long plate with a hump in small part of it:
As the fluid approaches that faster flowing area, it must accelerate, and the only thing to accelerate it is a pressure gradient. In the video he is using other means to accelerate the fluid flow, so it's different.
At the end of the plate there is no downwash, so the only effect is Bernoulli, and in this experiment you do get a pressure difference between the sides of the plate, and hence "lift".
In the case of the flow around the "elbow" there is no acceleration of the air, hence no Bernoulli effect. Quite the opposite, I would expect a Bernoulli effect to push the "elbow" to the left. However, the air is being sucked around the plate, so the "downwash" effect dominates.
There are some really, really bad explanations in the literature and on the net, often written by people who do one experiment without separating the effects. They go on to teach, and unsurprisingly people get confused.
The point about Bernoulli is that it only applies in its naive form in fluid flow where it's effectively a closed system. If you introduce airstreams of varying speeds then all bets are off.
Effectively Bernoulli works because the velocity changes are being caused by the pressure differences. Take a very, very long plate with a hump in small part of it:
____________XXXX____________
As the fluid approaches that faster flowing area, it must accelerate, and the only thing to accelerate it is a pressure gradient. In the video he is using other means to accelerate the fluid flow, so it's different.
At the end of the plate there is no downwash, so the only effect is Bernoulli, and in this experiment you do get a pressure difference between the sides of the plate, and hence "lift".
In the case of the flow around the "elbow" there is no acceleration of the air, hence no Bernoulli effect. Quite the opposite, I would expect a Bernoulli effect to push the "elbow" to the left. However, the air is being sucked around the plate, so the "downwash" effect dominates.
There are some really, really bad explanations in the literature and on the net, often written by people who do one experiment without separating the effects. They go on to teach, and unsurprisingly people get confused.
The point about Bernoulli is that it only applies in its naive form in fluid flow where it's effectively a closed system. If you introduce airstreams of varying speeds then all bets are off.
It applies to fluid elements that have the same internal energy. In the case of the airplane wing, the air ahead of the wing is effectively unaffected by the wing's presence, so you can apply the Bernoulli effect to comparing different fluid elements.
Sure, if you compare the air coming out of a hair dryer with air that doesn't, you'll get in trouble, but that has nothing to do with an airplane wing.
It applies to fluid elements that have the same internal energy. In the case of the airplane wing, the air ahead of the wing is effectively unaffected by the wing's presence, so you can apply the Bernoulli effect to comparing different fluid elements.
Sure, if you compare the air coming out of a hair dryer with air that doesn't, you'll get in trouble, but that has nothing to do with an airplane wing.
At the end of the plate there is no downwash, so the only effect is Bernoulli, and in this experiment you do get a pressure difference between the sides of the plate, and hence "lift".
Impossible, that would violate momentum conservation.
A pressure difference implies a force on the plate, so upward momentum is transferred to the plate. That momentum has to come from somewhere and it can only come from the air, which must move downward.
Impossible, that would violate momentum conservation.
A pressure difference implies a force on the plate, so upward momentum is transferred to the plate. That momentum has to come from somewhere and it can only come from the air, which must move downward.
If you mount the plate on a spring, and measure the steady state, the spring extends, showing that there is a force on the plate. There is no violation of conservation of momentum.
Look, this is pointless. Static states can isolate the individual effects, and then they all get combined into a dynamic state in varying amounts, and it becomes horribly complicated. People insist on trying to produce and explain overly simplistic models, and others insist on misunderstanding them. I've actually physically done these experiments and I know that what I say is true.
I'm not going to reproduce all the nitpicking tiny details in this forum because it's hard, inappropriate, and people will continually try to pick holes in it. It's the Monty Hall problem all over, and I'm just too tired to care.
There's every chance that your understanding is right in the cases you're considering, but I can't be bothered finding out where our experimental models differ.
I wish I hadn't bothered.
Look, this is pointless. Static states can isolate the individual effects, and then they all get combined into a dynamic state in varying amounts, and it becomes horribly complicated. People insist on trying to produce and explain overly simplistic models, and others insist on misunderstanding them. I've actually physically done these experiments and I know that what I say is true.
I'm not going to reproduce all the nitpicking tiny details in this forum because it's hard, inappropriate, and people will continually try to pick holes in it. It's the Monty Hall problem all over, and I'm just too tired to care.
There's every chance that your understanding is right in the cases you're considering, but I can't be bothered finding out where our experimental models differ.
I wish I hadn't bothered.
I'm ashamed to admit I didn't think about inverted flight until I read this article. Shit.
Once I realize something like this it makes me want to go back and do an integration / cleaning pass on my web of knowledge. Not sure how to do that in practice, though.
Once I realize something like this it makes me want to go back and do an integration / cleaning pass on my web of knowledge. Not sure how to do that in practice, though.
For upside-down flight, though, the shape of the wing isn't irrelevant. Many aerobatic aircraft have symmetric wing profiles, since the curvature of the wing reduces performance when inverted. (Higher speeds are required to maintain the required lifting force, and drag is increased).
Some of the popular explanations of aerodynamic lift are outright wrong, but most of them are correct about part of the story. Our CFD equations are obviously correct, otherwise our simulation software would give bad results. Maybe it's just hard to explain the results of these equations in an unambiguous way.
It is true that "air goes down, wing goes up", but this explanation is only one "why" closer to the heart of the matter.
Some of the popular explanations of aerodynamic lift are outright wrong, but most of them are correct about part of the story. Our CFD equations are obviously correct, otherwise our simulation software would give bad results. Maybe it's just hard to explain the results of these equations in an unambiguous way.
It is true that "air goes down, wing goes up", but this explanation is only one "why" closer to the heart of the matter.
There's that, and the fact that both Bernoulli and Newton's laws offer a thorough explanation of flight.
From everything I've read, it seems there is debate on exactly how to describe/explain what generates lift. Most of the explanations are correct, but insufficient by themselves.
I built a fair sized (5m, 16') windmill and this was one of the most surprising things I learned while designing the blades.
Windmill blades and airplane wings have a lot in common.
I wrote some python software for it to model the curvature and get an idea of how to get the maximum effect out of a blade cut from a given blank.
The neat thing was that without the software being interactive we'd have never clued in to some of the possibilities.
In case anybody is interested here is a snapshot of the python program:
http://pics.ww.com/v/jacques/renewables/windmill/snapshot3.p...
Windmill blades and airplane wings have a lot in common.
I wrote some python software for it to model the curvature and get an idea of how to get the maximum effect out of a blade cut from a given blank.
The neat thing was that without the software being interactive we'd have never clued in to some of the possibilities.
In case anybody is interested here is a snapshot of the python program:
http://pics.ww.com/v/jacques/renewables/windmill/snapshot3.p...
Cool. Have you ever used the Manipulate[] function in Mathematica? It blew my mind the other day. You can just wrap any expression (including plots), with something like Manipulate[expression, {k, 0, 10}], and you get the regular output with a slider for k in the range [0,10]. If expression is a plot, it'll redraw as you drag the slider.
Sure, a propeller, or a windmill, or a sail, are all airfoils that experience lift. They just don't "fly", ie keep something suspended.
do you have pictures of the windmill?
sure: http://pics.ww.com/v/jacques/renewables/windmill/
That documents the whole building from start to finish, including the 3d computer controlled router/plasmacutter we built in order to fabricate all the parts.
The whole thing including making the tools took about a year and half.
A single picture of the completed machine is here:
http://pics.ww.com/v/jacques/renewables/windmill/firstrun.jp...
It's a variable pitch 3 blader with a 'drum' type rotor that holds 18 2x1x.5" neos. Total power about 2000 Watts, design power was 2.5 KW so not perfect but still pretty good.
That documents the whole building from start to finish, including the 3d computer controlled router/plasmacutter we built in order to fabricate all the parts.
The whole thing including making the tools took about a year and half.
A single picture of the completed machine is here:
http://pics.ww.com/v/jacques/renewables/windmill/firstrun.jp...
It's a variable pitch 3 blader with a 'drum' type rotor that holds 18 2x1x.5" neos. Total power about 2000 Watts, design power was 2.5 KW so not perfect but still pretty good.
A better more scientific and mathematically detailed explanation can be found if you google "how wings work" - http://www.iop.org/EJ/article/0031-9120/38/6/001/pe3_6_001.p...
Thank you to the OP for pointing out the flaw in the popular explanation !
Thank you to the OP for pointing out the flaw in the popular explanation !
Yes, I linked to a similar post at the bottom of mine--as well as a few other resources. Many get put off by a four-page explanation, however, and move on to other things. I like to try and give the full punch in a single page and then provide resources to go into more detail if so desired.
I thought everyone knew this j/k. Really when I studied the Bernoulli principle in highschool it didn't make a lot of sense to me at the time. If the curvature of a wing was so important then how did objects with no curvature at all gain any lift? For example gliders.
Bernoulli lift only really comes into effect once you have level flight. Then the curvature alone can provide enough lift to support the weight of the plane without inducing drag. The Koanda effect is where ailerons and flaps come into play. The produce additional lift but also contribute more drag. As the angle of attack increases laminar flow drops. Using the authors example, placing a glass in a stream of water redirects the flow, but you'll notice there is a bit that sort of fans out. That is the equivalent of spoiled air. Too much of it and you've lost all laminar airflow, bye bye lift.
Anyways, I'm getting out of my area of expertise. It's been ages since I studied aerospace and my AIAA bible is in my parents' garage.
Bernoulli lift only really comes into effect once you have level flight. Then the curvature alone can provide enough lift to support the weight of the plane without inducing drag. The Koanda effect is where ailerons and flaps come into play. The produce additional lift but also contribute more drag. As the angle of attack increases laminar flow drops. Using the authors example, placing a glass in a stream of water redirects the flow, but you'll notice there is a bit that sort of fans out. That is the equivalent of spoiled air. Too much of it and you've lost all laminar airflow, bye bye lift.
Anyways, I'm getting out of my area of expertise. It's been ages since I studied aerospace and my AIAA bible is in my parents' garage.
Bernoulli's principle also doesn't explain how planes can fly upside-down.
Wow. I had completely forgotten how cool it was playing with a hand outside the car window as a kid. It really was an aha moment.
A bit dangerous, I know.
A bit dangerous, I know.
Another perspective how wings are vacuum lifted: vacuum air flow has a low mass, but very high velocity. I work with vacuum packaging equipment and we often see a few 90 turns in the hoses hits the flow like a brick wall at each turn. The speed of the vacuum near 1 torr carries tremendous energy and is enough to rip silicone caulking from inside the joints. Air at double the atmospheric pressure won't see the high velocities to do any damage. I can imagine the high speed downward air flow giving tremendous push.
From the article:
> Ask yourself why planes can hang tons of massive crap (engines, bombs, etc.) off of the bottom of their wings if the bottom of the wing is so important for flight
I guess that since the top of the wing generates the fast flowing air which then generates the low pressure that generates the lift, by slowing the air on the bottom of the wing with bombs and engines is only going to increase this pressure differential and hence increase the lift (at the expense of drag).
The definitely don't seem to hang anything off the top of the wing.
> Ask yourself why planes can hang tons of massive crap (engines, bombs, etc.) off of the bottom of their wings if the bottom of the wing is so important for flight
I guess that since the top of the wing generates the fast flowing air which then generates the low pressure that generates the lift, by slowing the air on the bottom of the wing with bombs and engines is only going to increase this pressure differential and hence increase the lift (at the expense of drag).
The definitely don't seem to hang anything off the top of the wing.
Structurally, it's easier for something to hang (well in places that have gravity as a constant) than it is to prop up. You could mount something above the wings but than you would need additional hardpoints to keep it from falling over and smashing into the wing. If it's hanging you only need one hardpoint.
The reason I didn't believe the curved-upper-surface explanation as a kid was that meant that the volume of low-pressure air would have to be similar to the volume of a blimp for it to be useful for lifting the plane, even if it were as efficient as hydrogen; the wings would have to be many times the size of the fuselage.
Geek Cruise lecture "Why Airplanes Fly - A Modern Myth" (audio only, 3 parts):
http://podcast.geekcruises.com/index.php?search_string=airpl...
http://podcast.geekcruises.com/index.php?search_string=airpl...
I say they fly because the shape of the wing makes them fly. If you built an airplane with a wing shape that didn't fly, it wouldn't sell. Therefore, it's really a case of natural selection.
A few planes have been built with the Coanda effect in mind (http://en.wikipedia.org/wiki/Boeing_YC-14), but in general it isn't very important.