Debunking lift theories

[Islander2]

I don't often read Today's Pilot, but for those that have been thoroughly confused by the arguments aired on this and other forums, I would commend you to seek out a copy of the November 07 issue and study John Gibson's letter entitled "Newton, Bernoulli and Lift Theory".

Physically correct and historically accurate, it does the most succinct job I've seen of debunking the amateur attempts over the past ten years or so to re-write aeronautical science around Newton's laws of motion.

It's clearly one of a series of correspondence on the subject but, although I've not seen the other letters, I don't believe that matters ... it stands very adequately on its own. John Gibson has spent a lifetime in aerodynamics at Warton, and in my book that makes him rather better qualified than those guys at Fermi and the University of Washington that have done so much to promote this nonsense amongst the less technically literate.

[Shaggy Sheep Driver]

Can you give us the info here? We don't all read that mag.

SSD

[Islander2]

SSD,

Sorry, I feel it would be improper to reproduce the lengthy letter verbatim, and to paraphrase could distort its content and accuracy.

It's on the magazine sellers' shelves now, if you're interested you can always go for a browse!;)

[Fly-by-Wife]

Sorry, I feel it would be improper to reproduce the lengthy letter verbatim

If you haven't got a scanner, say so - otherwise that is the most pathetic load of mealy-mouthed bollocks I have heard for a while.

Unless of course you work on the circulation promotion side for T's P, in which case it is just pathetic.

FBW

[LH2]

If you haven't got a scanner, say so

:) :D :ok:

[EvilKitty]

Fly-by-Wife said:
bollocks

Why?

Why do you expect Islander2 to reproduce that letter here? Because you're too lazy arsed to go read it yourself? Islander2 says quite rightly it would be improper to reproduce it here - after all, John Gibson did write it to Todays Pilot and not to PPRuNe.

If you feel that strongly that it should appear here then you should contact John Gibson yourself and ask for his permission to reproduce it here. Otherwise just go to the newsagents and read it there.

[Islander2]

I've no doubt that Danny is concerned about the laws of copyright, even if others here would be happy to disregard them.

[Shaggy Sheep Driver]

I think Islander2's response is quite a correct one. If he doesn't think it can be simply paraphrased, it would be improper to reproduce the letter verbatim on here. I'll go and have a sneak read in WH Smiths!

SSD

[Final 3 Greens]

FLy by wife

What a stupid and intemperate posting.

[Andy_RR]

If you start from first principles, Newton, Bernoulli and circulation theory are all one and the same, excluding the effects of viscosity.

[IO540]

Nobody is going to, nor should they, take the time to go to a shop and spend £3.50 or whatever on a magazine, so they can reply to a posting on pprune.

As regards gas flow theories, Newton is right, but that approach is not exactly useful because nobody is going to run a simulation of every molecule moving around the wing or whatever. Approximations have to be made. That is where Bernoulli etc came in, with far simpler and much more usable theories, many years ago.

AIUI, one could derive Bernoulli from Newton. Not very useful but possible.

[LH2]

FLy by wife

What a stupid and intemperate posting.

Intemperate maybe, stupid not at all. I believe what he was alluding to is that if the OP was not himself prepared to summarise in his own words why the article in question is important, perhaps he should have refrained from making such loaded comments such as "amateur attempts ... to re-write aeronautical science", etc. I do think the tone of the original post is a bit unfortunate--maybe Islander2 would care to retouch it a bit so that it sounds a bit less antagonistic.

As for the bit about not having a scanner, that was a brilliant insult. Should see them more often :ok:

[Andy_RR]

As regards gas flow theories, Newton is right, but that approach is not exactly useful because nobody is going to run a simulation of every molecule moving around the wing or whatever.

That's what integral calculus is for, or CFD, or whatever other way you wish to analyse the problem. The point is, the fundamentals are the same.

Anyway, Newton didn't explain how aerofoils work, whereas Bernoulli (and Prandtl et al) did, however Bernoulli and Prandtl were building on Newton's theorys, not tearing them up and starting again.

A

[Fly-by-Wife]

I've no doubt that Danny is concerned about the laws of copyright, even if others here would be happy to disregard them.

Before anyone bleats on about copyright, they ought to establish the facts.

British copyright law has a set of exceptions to copyright known as fair dealing. Fair dealing is much more restricted than the American concept of fair use and only applies in tightly defined situations -

s29.—(1) Fair dealing with a literary, dramatic, musical work … for the purposes of research for a non-commercial purpose does not infringe any copyright in the work provided it is accompanied by a sufficient acknowledgement …

s30.—(1) Fair dealing with a work for the purpose of criticism or review, of that or another work or of a performance of a work does not infringe any copyright in the work provided that it is accompanied by a sufficient acknowledgement and provided that the work has been made available to the public.

Discussion on PPRuNe is non-commercial use, and can certainly be considered criticism or review (although perhaps not research, but I have left s29 in, as others might consider it applicable). Since the letter was published in a magazine, then it has been made available to the public. Simply acknowledging the author and the magazine is all that needs to be done.

When (if) I have the time and am near a newsagent that stocks Today's Pilot, I will buy the issue and scan the article for your benefit. Then perhaps we can have a discussion about lift theories (interesting) and not copyright (PITA).

FBW

[ProfChrisReed]

I almost didn't post, but then thought that PPRuNE might be pleased if posters avoided infringing copyright.

FBW has fallen into the trap of reading a statute as if it existed in a vacuum - s. 30 is more complex than that, as its meaning has been elaborated over the years by court decisions. In simple terms:

1. S.30 only applies for the purposes of criticism or review by the person copying. If FBW copies the letter to allow others to criticise or review it, s. 30 will provide no protection.

2. S.30 also requires fair dealing in the copying. Posting a long letter with the comment: "Good/Rubbish innit" would not be fair dealing. You only get the right to copy the parts you need to undertake your criticism or review.

If posters start copying large chunks of copyright works to the site, PPRuNE will begin to receive formal notices requiring these posts to be deleted, and will have to do so to avoid liability. In some countries (and these fora are available pretty much everywhere except Burma at the moment) PPRuNE might be liable even without receiving such a notice.

Better to seek permission, or to write your own post.

[Final 3 Greens]

Well said Prof Reed.

FBW, time you established the facts, methinks.

[Fly-by-Wife]

ProfChrisReed,

Thanks for clarifying those aspects of s30.

Perhaps you can answer a further question or two:

When a person writes a letter (as opposed to an article) to a periodical or newspaper for publication, who owns the copyright, the author or the publisher - or both?

If not the author, then would not (even) the entirety of the letter be considered to be only that part of the entire publication (the magazine) necessary to carry out the review?

And if it had been a very short letter, would we even be having this discussion? ;)

FBW

[Tone]

What a wonderful combination of Private Flying and Jetblast. We should have more of it.

[Final 3 Greens]

FBW

Keep digging the hole :E

[Henry Hallam]

"If you start from first principles, Newton, Bernoulli and circulation theory are all one and the same, excluding the effects of viscosity."

True, but without viscosity, you can't have any circulation and therefore lift is zero.

[IO540]

True, but without viscosity, you can't have any circulation and therefore lift is zero.

Really?

[ProfChrisReed]

When a person writes a letter (as opposed to an article) to a periodical or newspaper for publication, who owns the copyright, the author or the publisher - or both?

I hope no-one expected a simple answer - why do you think I can afford to fly?

The author owns the copyright, unless he/she has assigned it to someone else.

However, there is a second copyright in the typographical layout of the published version which belongs to the publisher.

If you re-type the letter, you infringe only the first.

If you scan it and post an image, you infringe both.

And if it had been a very short letter, would we even be having this discussion?

Yes, unless the letter was so trivial as not to amount to a work (and that's very trivial indeed).

All this can be complicated by concepts like the public domain and implied licences, which I will save for another day.

Leaving the legal details aside, I'd say a pretty good test is to ask yourself how you would feel if you'd written the letter. If you wouldn't mind it being reproduced here, then probably the author won't complain. Common sense, and common courtesy, is often a good guide.

[Nathan Parker]

True, but without viscosity, you can't have any circulation and therefore lift is zero.

Really?

Yep. Circulation is initiated when the airflow separates trying to flow around the sharp training edge. The resulting vortex produces an equal and opposite vortex around the wing, the so-called "bound vortex" that produces lift.

[bookworm]

True, but without viscosity, you can't have any circulation and therefore lift is zero.

True in theory but a little misleading perhaps. It only needs a trace of viscosity to set up the flow patterns. If viscosity increases beyond that trace, nothing substantially changes (until you get to viscosities much denser than air).

It's a little like standing a pin on its point and saying "the acceleration due to gravity determines which way it falls". It's true, but it doesn't really matter if g is 10 m/s2 or 10 m/yr2, the pin still falls the same way. It's the phenomenon, not the value, that is important.

[IO540]

Surely one can get "lift" through simply redirecting some material that is flowing along. The "material" could be air, or a load of bricks. I don't think bricks will have a whole lot of viscosity.

Bookworm's post is intriguing and I am sure he is right (he is more or less always right). But the viscosity required must be negligible.

[mm_flynn]

IIRC the amount of 'lift' one gets from 'deflecting' the 'air bricks' that are directly intersected by the wing's forward movement is negligible. It is because of viscosity that this initial movement of a relatively small airmass generates the circulation of quite a large airmass which results in an extensive downward deflection of air (and us flying :) ) .
I seem to remember, in something like a TB-20, all of the air within about 15 feet above the wing is involved in this deflection. It is also why minor interruptions in the flow along the top surface (ie. a bit of rough frost) can so catastrophically kill the lift the wing generates.

[tigerbatics]

Islander2, is this what you disagree with?

http://www.aviation-history.com/theory/lift.htm

If so, why?

[bjornhall]

Regarding the link tigerbatics posted, in my view:

They are throwing out the baby with the bath water when they say Bernoulli is wrong. What they are right in is that you can't use the "equal time" argument to determine the velocity field around the wing. But if you can determine the velocity field by some other means, then you can determine the pressure field almost trivially (and quite correctly!) using Bernoulli. The problem is that it is very tricky to obtain the velocity field, and hand waving approaches like the "equal time" concept fail miserably. But that is hardly Bernoulli's fault!
People criticizing descriptions like those in the link, usually have gotten it all backwards: They believe the "Newtonian" approach amounts to claiming that the air is deflected down by "bouncing off the bottom of the wing", as if the air behaved like particles. That would be nonsense, and it is not what the authors here claim.
In blaming the deflection of the air (and thus the generation of lift) not on Bernoulli but on the Coanda effect, they are replacing one "easily understood" but totally wrong description with another equally false one. The Coanda effect deals with jets, not with bulk fluids. The Coanda effect comes into play if you use some rare aviation gizmos like blown flaps, but it has nothing to do with why most aircraft fly.I really like how these things are described in "See How It Flies", available online at http://www.av8n.com/how/ (see, e.g., http://www.av8n.com/how/htm/spins.html#sec-coanda-fallacy for why the Coanda effect is not the explanation for lift). The author, John S. Denker, is both a physics professor and a flight instructor, making him uniquely qualified to describe things in a way that is both correct and useful! :ok:

[Nathan Parker]

for why the Coanda effect is not the explanation for lift

Denker doesn't explain very well why the Coanda Effect isn't involved. I followed up on the reference he gave and the paper doesn't really address the issue. I agree with his conclusion, but mainly due to the fact that out of 40 odd aerodynamics books that I have, none of them mentions the Coanda Effect.

[BackPacker]

Now where is G-EMMA when you need her?

(At least I know for sure she's not doing barrel rolls in a C172...)

[Islander2]

Islander2, is this what you disagree with?

http://www.aviation-history.com/theory/lift.htm

If so, why?In a nutshell, because from their attempt: a) to demolish various trivial myths used in the explanation of lift (trivial in the sense that they never featured in any serious explanations anyway) and b) to attribute superior status to a description of flight based on the application of Newton's laws (the proper application of which, aside from the erroneous sine-squared law, have never been said to be invalid by the serious texts), the authors end up: c) dismissing out of hand the entire velocity- and pressure-distribution basis of lift explanation that is (and has been pretty much since the Wright Bros) used right across the scientific community and the aircraft industry both to understand and to calculate lift, and in the process d) introduce a long string of misconceptions every bit as invalid as the myths they set out to demolish!

Their paper was submitted in 2001 to the American Journal of Physics, but not accepted for publication. I fancy the peer review would make for interesting reading. Here's one damning critique by the av8n on-line aeronautics book author:

http://www.av8n.com/fly/lift.htm

Anderson and Eberhardt subsequently produced an expanded version of their erroneous ideas in a book entitled "Understanding Flight". Serious students of aerodynamics may prefer to re-title this: "Misunderstanding Flight"!

However, for an excellent and authoritative overview of the Bernoulli vs Newton debate, I commend to you the John Gibson letter (and no, I'm not going to reproduce it here!) :ok:

[Troy McClure]

Good to see av8n.com given a bit of an airing again and nice to see the site's still working. Prof Denker seems to know his stuff and I've recommended his website to many over the years.

[tigerbatics]

Many thanks for the explanation and links Islander2.

Very interesting in an academic way and well worth thinking about; but for safe flying it seems a pilot can get by with one of several approximations to the truth provided he/she does not really try to understand it!

[Saab Dastard]

FBW,

I have followed this thread in its "interesting" directions, and I thought it would be a useful idea to contact Today's Pilot to actually ask their permission to reproduce the letter.

I received permission to do so from Dave Unwin, Editor of Today's Pilot,
this afternoon, so if anyone has access to the letter (and a scanner) ;), then please go ahead and put up a copy.

I won't post the email correspondence here (to protect identities), but I have sent BRL a copy to prove the bona-fides!

I hope that someone will oblige, and the discussion can then become more "uplifting". Sorry, just came to me :O

SD

[Islander2]

As a trainee pilot it is just as useful to me to think of lift as 'downwash'Welcome back, G-EMMA.

I'd agree that, for pilots, a deep understanding of the cause of lift is unnecessary; what we need to concentrate on is the effects of lift.

The downwash explanation is fine for students of flight. As you're probably aware, though, serious aerodynamics texts tend to put it into the category of effect rather than cause.

It certainly produces some interesting tangles for those attempting to build a lift theory around downwash when they come to explain two-dimensional lift. A wing with infinite length (which can readily be simulated in a wind tunnel by having the wings protrude through the tunnel walls) generates strong lift with zero net downwash. In other words, a molecule of air that is initially displaced upwards as the wing approaches (upwash) is then displaced downwards (downwash) back to its original position.

This 'inconvenient fact' leads Anderson and Eberhardt to offer the ludicrous proposition that there is a net downwash in two-dimensional lift with the wing diverting "an infinite amount of air at zero velocity"!

Even Denker struggles with this one, leading him to assert that: "both the incoming upwash and the outgoing downwash make positive contributions to lift". Whilst that statement is true enough, it's hardly supportive of downwash as a lifting theory based on Newton's third law!

[Nathan Parker]

Part of the reason I enjoyed studying it was that I realised there is some 'magic' going on we can't explain yet.

I don't buy that at all. Much of the simple mathematics gives very accurate results and CFD takes over from that. There isn't much mystery left here.

[Contacttower]

There isn't much mystery left here.

I don't think that's completely true, scientific theories come and go, and centainly I live in hope that a much better one will come along than the ones we have at the moment to explain lift.

Some may remember the "Downwind turn" thread, which while is not directly related to lift theory demonstrated that there is still sufficient inconstistency in some parts of theory of flight to enable a lot of myths to be peddled.

[BackPacker]

I bought TP on recommendation from the folks here just yesterday, to read said letter. I was disappointed to see that:
a. It seems to be a rebuttal to a letter responding to an original article that John Gibson wrote earlier, probably in August. I do not have the original article, nor the response letter.
b. It only talks about the theory names, without going into detail on what each theory claims and why that claim is unfounded
c. I did not understand half of it. :rolleyes:
(Fortunately there are a few other articles in this TP worth reading...)

Nevertheless, since Saab Dastard has obtained permission and nobody else seems to have gotten around putting it up, here goes:

NEWTON, BERNOULLI AND LIFT THEORY

Rob Davies (Letters, September issue) is upset by what he calls my opinions on Newton and lift theory. They are rather more than that. I prefer to call them historical facts plus knowledge gained in 55 years in employment and consultation at Warton, mostly within the aerodynamics department. It has been claimed recently that aerodynamicists have always been wrong (if you believe that, you can believe anything), because they have used the "equal transit time" theory. That claim is nonsense. The theory, rightly described as incorrect in that NASA website, is a fairy tale that is sometimes taught by pilots to other pilots. It was never part of lift theory or presented in serious textbooks, and NASA does not say it was.

I did not say that Newton's laws are invalid or nonsense. He was first to prove theoretically that fluid forces on a body are proportional to the fluid density, the body dimensions and the velocity squared. He formulated the theoretical structure of frictional shear stress in viscous fluids. His second law plays a part within the general three-dimensional fluid flow motion equations of Euler and Navier-Stokes of the 18th and 19th centuries, but these were insoluble until high power digital computing arrived long after two-dimensional aerofoil lift theory was well established and proven. He also proposed a fluid force model assuming a stream of individual particles colliding with an object in motion through it, giving up the component of their momentum normal to the surface and then moving off tangentially to it. When applied later to flat plates at small angles of attack, this led to the notorious sine squared law with very low lift and high drag, making flight impossible if true. Cayley knew about two French experiments proving the law was false. Nobody took much notice and he turned for a while to balloons and airships. Later events also disproved it, up to Langley's 1890 finding of 20 times more lift on a flat plate at five degrees AoA than predicted by the law. Up to the time of the Wrights' first flights, the few people believing in powered flight were widely dismissed as cranks. The fact that Newton had expressed uncertainty about this model was ignored, or more likely forgotten. Lanchester conceived the circulation theory of lift in 1893 from the principles of fluid vortex and streamline behaviour derived after Newton's time. Kutta and Joukowski established its mathematical structure soon after, from which the classic low speed aerofoil shape of a moderate thickness, well-rounded nose and sharp trailing edge was first obtained. Its essence is that aerofoils generate a circulating flow which is combined with the free stream velocity, increasing the local flow velocity above and decreasing it below the aerofoil. The integration around the aerofoil of these velocity variations gives the circulation strength. The lift per unit span can be found from the basic equation "lift equals circulation times velocity times density", without the direct use of either Bernoulli or Newton, as can be proved by measurement. In two-dimensional flow (no tip effects) there is no net downwards air movement. At negative AoA, negative lift is provided by the reversal of the circulation direction.

As noted in the NASA website, applying the Bernoulli equation to the local velocities obtains the pressure distribution around the aerofoil, and hence the total lift. 'Bernoulli' is not a lift theory, nor did Bernoulli write it. It was derived by Euler from a subset of his fluid motion equations, and gives the relationship between velocity and pressure along streamlines. Together they show that lift is generated largely or entirely by reduced pressure above the aerofoil, first proved by measurement in Eiffel's 1911 wind tunnel. Though it is essentially a low speed, or more specifically a low Mach number theory, it is still used today in greatly developed forms. It cannot be said that the current enthusiasts for the sole application of Newton's laws to explain lift have succeeded in rubbishing circulation theory, because they appear not to know it exists. Their ideas are assertions that have utterly failed to provide any quantitative measures of lift, or to explain the physical basis of lift generation, or to suggest how an aerofil might be desgined. It is these ideas that I referred to as nonsense.

John Gibson

(For the record, this letter appeared on page 56 of Today's Pilot, November 2007, and is copied here with permission obtained by Saab Dastard. All typos are my fault.)

[Islander2]

none of the theories we currently have explain lift in a way ..... that stands up to much scrutiny, each falls apart in its own way given the right set of circumstances.Sorry, G-EMMA, but I really have to take you to task over that!

In my experience, the theories that have existed for a century or more explain lift very well, stand up to intense scrutiny, and for incompressible flow (say <Mach0.5) quantify lift with a very close match to experimental observation.

In all the instrumented wind tunnel test results I've seen (and I spent a number of years in aircraft design with Hawker's in my formative engineering days), the measured lift force equates extremely well with the integrated pressure differential around the wing, and the pressure at any position equates equally well via Bernoulli's prediction to the streamline flow velocity at that position.

[Nathan Parker]

I don't think that's completely true, scientific theories come and go

That's what the US Creationists say, as if to dismiss the theories they don't like! I'd be surprised if you could find a theory well-supported by a century of experimental evidence that was tossed in the dustbin.

One example often cited is the replacement of Newtonian physics by Einsteinian, but that's not really true. We're still taught Newtonian physics in college. Einstein's equations reduce to Newton's when velocities are low.

Some may remember the "Downwind turn" thread, which while is not directly related to lift theory demonstrated that there is still sufficient inconstistency in some parts of theory of flight to enable a lot of myths to be peddled.If you think there is some inconsistency in the science of flight due to the "Downwind turn" debate, you're very mistaken. Those who advocate a problem don't understand Galilean physics. Why should we create a new physics when people don't understand the physics of the 17th century?

[Nathan Parker]

I hold a cigarette still there will be a portion of perfectly laminar flow as the smoke leaves the cigarette. An inch of so above the flow becomes turbulent, why does that happen? Got an equation for it, will CFD tell my why it happens?

The existence of turbulence is not a prerequisite for the existence of lift.

Still, your whole criticism is off-target. You're claiming that the difficulty in calculating lift in any particular situation is the result of a conceptual problem, when it's really a matter of practicality.

CFD doesn't give an explanation of liftNot to you, perhaps, but if it can calculate it, then it establishes that the factors that contribute to lift are well-known.

Apply simple mathematical equations to the wing of a model aeroplane. Work out mathematically how much lift it will produce (use Bernoulli if you really must).

What right have you to ask for "simple" equations? Reality isn't always simple, particularly when you're trying to predict the mass behavior of large numbers of independent actors.

You seem to think something can't be easily explained to you means that it isn't understood.

[Henry Hallam]

I think some of you guys should really take an aerodynamics course or read a textbook (Anderson is good) before saying that nobody understands how a wing works. If it could all be explained in a forum post then we wouldn't have extensive college courses on it, but that doesn't make it any less understood or understandable.

[bookworm]

What I'm saying is that it isn't the whole picture. If we had a really good explanation for lift, a real mathematical model that worked all the time, we could use it to design aerofoils? I could say, right I want it to do this, and then design the shape. As it is that doesn't work does it? We have some good bits of theory that let us work out some things such as predicting quantative lift at low Mach numbers for what I would call for want of a better term 'classic aerofoils'... but it isn't all of it.

Can you give any example of a physical model that works all the time without "some fudge factors thrown in"? I can't. The nature of physics is that we start with a simple model that crudely approximates reality and we add complexities to it so that the approximations become better and better.

[bjornhall]

Can you give any example of a physical model that works all the time without "some fudge factors thrown in"? I can't. The nature of physics is that we start with a simple model that crudely approximates reality and we add complexities to it so that the approximations become better and better.

Of course there are such models! In beam and ray optics, just to give one example, models are usually good enough that you can design from a model and get a predictable and useful result on the first attempt. Electronics usually as well. And those were just two...

Thing is though, when you have a model that is developed and understood at such a level, you tend to overlook it in discussions about "the unsolved problems of physics and engineering"... It's too simple.

If aerodynamics was understood at such a level (and I doubt it can ever reach that level, due to the nonlinearities of the Navier-Stokes equation), we could design a useable wing from first principles without using wind tunnels or even CFD, build it, fit it on a plane and it would work as designed on the first attempt.

But we are not on that level with aerodynamics. I suspect that is what G-EMMA means; not that we have no clue how our planes really work...:rolleyes:

[Andy_RR]

yeah, but CFD is simply a numerical solution to the Navier-Stokes equation. Just because you can't find an analytical solution, doesn't make it a bad (or poorly understood) method.

A

[IO540]

In electronics, one can model resistors and capacitors 'exactly' but semiconductors are all approximations full of fudge factors. Digital logic can be modelled exactly for function of course but not exactly in timing.

Incidentally, one sentence in that letter caught my eye

In two-dimensional flow (no tip effects) there is no net downwards air movement.

How can this be true, while the weight of the aircraft is being supported? There has to be some downward acceleration of air mass.

[bookworm]

Of course there are such models! In beam and ray optics, just to give one example, models are usually good enough that you can design from a model and get a predictable and useful result on the first attempt.

As we do in aerodynamics. Beam and ray optics is fine until you get to distances of the order of the wavelength. Can you use it to explain diffraction aand interference? And try applying "beam and ray optics" to near-field situations! Incompressible flow is fine until you get velocities of the order of the speed of sound. But it's good enough to design aerofoils with.

Physics is about peeling the layers of the onion to get closer to reality. Aerodynamics is no different.

[Contacttower]

That's what the US Creationists say, as if to dismiss the theories they don't like!


Maybe, but what US Creationists say is not really relevant to this, science draws the distinction between laws and theories because some ideas are more proven and established than others. Ideas relating to lift tend to fall into the latter, and I think one is being naive if you say hand on heart we know exactly how a wing works and can predict with total accuracy what it will do throughout the speed and angle of attack range.


If you think there is some inconsistency in the science of flight due to the "Downwind turn" debate, you're very mistaken.


The poster who constantly advocated the "downwind turn" claimed to be an RAF helicopter pilot (and I have no reason to believe he was not) and was basing his ideas on an observation.....now I don't believe in them...and I doubt many do but I just thought it demonstrates that this is far from a closed debate.

[Henry Hallam]

"In two-dimensional flow (no tip effects) there is no net downwards air movement.

How can this be true, while the weight of the aircraft is being supported? There has to be some downward acceleration of air mass."

This is true for two-dimensional, *inviscid* flow. There is no circulation (flow turning) and therefore no lift. There is also no drag (d'Alambert's paradox). Of course all actual flows are viscous, but the inviscid equations are easier to solve, so it's often convenient to approximate them as inviscid over almost the entire volume except for point or sheet vortices that we add in to provide the circulation.

p.s. the Wikipedia article on D'Alembert's Paradox is full of crap.

[Islander2]

This is true for two-dimensional, *inviscid* flow.John Gibson's comment is not referring to inviscid flow, however, he is describing what one observes in a wind/smoke tunnel when simulating two-dimensional flow by taking the wings to the tunnel walls or beyond (no tip effects). There is no net downwash (i.e the downwash is balanced by the upwash), yet the measured lift force is greater than for the equivalent three-dimensional flow.

As I said in an earlier post, this is a highly inconvenient fact for those that wish to built a lift theory around Newton's third law. Of the various inventive arguments put forward by the proponents of modern-day lift theories to explain away this inconvenience, the prize has to go to Anderson and Eberhardt who state that there is a net downwash in two-dimensional flow, it's just that it's produced by the wing diverting "an infinite amount of air at zero velocity!"

[Genghis the Engineer]

Oh no, not again! The dreaded Newton .v. Bernoulli debate.

If anybody wants a really clear explanation of the various forms of aerodynamic theory read Glauert's book "aerofoil and airscrew theory", published in 1926, still in print, and far more complete than anything I've seen in PPL or ATPL theory books.

G

[IO540]

I do not believe that it is necessary to have a wingtip (i.e. that it is necessary to have a wing of a finite length) in order to get lift.

To me, it seems intuitively obvious that one cannot get lift unless one is accelerating some air downwards.

Is that wrong?

[Henry Hallam]

"To me, it seems intuitively obvious that one cannot get lift unless one is accelerating some air downwards.

Is that wrong?"

That's right, at least as far as I'm concerned and explicitly what Prof. Darmofal taught my class.

[bjornhall]

As we do in aerodynamics. Beam and ray optics is fine until you get to distances of the order of the wavelength. Can you use it to explain diffraction aand interference? And try applying "beam and ray optics" to near-field situations! Incompressible flow is fine until you get velocities of the order of the speed of sound. But it's good enough to design aerofoils with.

The crucial difference, in my view, that made me pick that example, is that beam and ray optics is self-contained. It is not complete, there are limits to its applicability, but within its range of applicability you can use that and only that to obtain a useful result.

Simple aerodynamics, such as inviscid flow approximations, as far as I am aware, is not thus self-contained. For instance, ignoring viscosity you have to pull things like the Kutta criterion out of your hat to explain what the circulation will be; details about the boundary layers such as thickness and where they go turbulent influence the results (especially at high angle of attack) but you can't readily calculate those details from first principles, etc.

I think that is the main point about these "pilots' explanations" of lift, such as the equal-time argument etc: There is no simple and correct way of explaining it. It's either simple, or correct, or most commonly neither. Whereas better understood and more easily understood (and thus far less fascinating! :) ) topics can be explained to a sufficient level within some parameter range.

I also take the view that just because one can calculate something does not in-itself mean that one understands it. The engineer-me is happy with just an equation and a numerical solution, but the physicist-me wants more... ;)

[BackPacker]

So... If one of my non-pilot friends asks me how this whole flying thing works, what do I tell them?

So far I've tried a combination of "equal distance over the top and bottom" (hence the shape of the wing) and "impact lift" (hence angle of attack), and that the truth is somewhere in between, but not completely understood from a theoretical point of view. And then added that a lot of empirical data has shown that aircraft, indeed, can fly. And, so far, they've bought it. But what if a really clever chap comes along?

(I did not spend the better part of an hour retyping Gibsons article just to hear the phrase "we really don't know" over and over again... :rolleyes:)

[BackPacker]

So inbetween the pax briefing about doors and seatbelts, if the question comes up, would this be a more or less correct reply?

"Originally there were two theories, commonly referred to as Bernoulli and Newton, although both scientists never put forth those theories themselves. Very briefly put, the Bernoulli camp uses the equal transit time theory (hence the shape of the wing) to explain a pressure differential between the top and bottom of the wing, while the Newton camp uses the impact theory (hence angle of attack). Both theories hold some merit of truth but on their own are not enough to explain lift, and elements of both theories have been proven to be wrong in any case. Euler took the work of Bernoulli and Newton and put forth a comprehensive set of equations that explain lift in a lot of cases, and Navier-Stokes expanded on that to include viscosity effects, which is needed to explain boundary layer behaviour, among other things. Euler and Navier-Stokes are now the essential formulas used in CFD, which is used to design modern wings. All this is essentially for laminar flow: turbulence is still far from understood from a theoretical point of view."

Let's hope their eyes start glazing over by the time I speak that second sentence so that I can start extending that stick below the fuselage to hold the plane up.:O

[soay]

Let's hope their eyes start glazing over by the time I speak that second sentence so that I can start extending that stick below the fuselage to hold the plane up.
Mine uses a sky hook. Vorsprung durch Technik!

[barit1]

BackPacker - you need to edit that tome so it's a single sentence with a minimum of punctuation :}

[BackPacker]

Nervous pax: "How does this whole flying thing work then?"
Me: "Well, you see, onions have layers, just like Ogers."

Anyway, "The wing, due to its shape and forward motion, displacing the air downwards, thus generating an aerodynamic counterforce more or less upwards (let's just mention lift at this stage and ignore drag) which keeps the plane in the air" works for me. And then add "all the slightly more complex explanations have all been proven wrong somehow and the theory that explains almost all of it correctly (except turbulence), Navier-Stokes, is so complex that it takes a mathematical genius to understand and apply it. And, by the way, the sickbags are located in this here pocket."

[bjornhall]

Should add to your nervous pax tho' ;) that Navier-Stokes explains turbulence just fine!

The problem is not finding an equation that contains turbulence in its solutions; Navier-Stokes already does that. The problem is solving the equation...

[homeguard]

At the end of this very interesting debate it has never been in doubt that we would end up back at the beginning, and so we have with the scientists amongst us concluding that there is no simple way to brief a student PILOT on how their wings/flight control surfaces work. And those that it would appear most knowledgeable on the matter lost for words.
I said at the beginning because, there is nothing wrong to state; " that the thickness of an aerofoil is designed such that the airspeed over the upper cambered surface increases with an associated reduction in pressure, thus the pressure above the aerofoil becomes less than that beneath the wing - where it is not necessary for a change of speed to take place. The force thus generated is called lift and is said to be perpendicular to the surface of the aerofoil. The lift force generated, in order to maintain height, must equal weight (G). The lift force generated will vary with changes in air density, airspeed and Aof A.
I have never heard a pilots briefing done by a knowledgeable instructor different from the above and it needs no further explanation - well, perhaps in the bar.

[foxtrot-oscar]

Backpacker,

You could always say to your passenger to stick their hand out of a car window and tilt to to experience different angles of attack. If they do this I’m sure that (like me ) they will FEEL the air below their hand pushing the said hand upwards (rather than feeling a vacuum above their hand sucking it up)

But I still byt the big stick theory

[bookworm]

The crucial difference, in my view, that made me pick that example, is that beam and ray optics is self-contained. It is not complete, there are limits to its applicability, but within its range of applicability you can use that and only that to obtain a useful result.

Simple aerodynamics, such as inviscid flow approximations, as far as I am aware, is not thus self-contained. For instance, ignoring viscosity you have to pull things like the Kutta criterion out of your hat to explain what the circulation will be; details about the boundary layers such as thickness and where they go turbulent influence the results (especially at high angle of attack) but you can't readily calculate those details from first principles, etc.

I think you're being unfair in your comparisons.

When your ray hits a glass surface, you apply Snell's Law to work out by how much the ray refracts. Isn't that "pulled out of your hat" in the same way (unless you happen to remember that the light is a wave, of course, but then you're moving beyond that self-containment)?

According to Newton's Laws I can balance a pin on its point on my desk. Experience says that there's enough vibration around that the unstable equilibrium will never be maintained and the pin will topple. So do Newton's Laws lack this "self-containment" because my pin-balancing experiment fails?

It doesn't take a detailed quantitative understanding of viscosity to conclude that the Kutta condition works, just an empirical understanding of fluids clinging to surfaces. I think singling out aerodynamics as being in some way "incompletely understood" is inappropriate. It is, undoubtedly, harder than maany areas of physics to explain convincingly to the layman in a way that is physically accurate.

[bookworm]

The force thus generated is called lift and is said to be perpendicular to the surface of the aerofoil.

Not sure I like that bit. Lift is by definition the component of the total aerodynamic force that is perpendicular to the freestream airflow (and drag is the parallel component). If you explain it in that way, you're storing up a lot of room for misconceptions when the student starts thinking it through to the next level ("so what happens to lift and drag when I pitch the aerofoil up?").

[bookworm]

To me, it seems intuitively obvious that one cannot get lift unless one is accelerating some air downwards.

Not "wrong" but potentially misleading.

Consider a long wind tunnel with a roof and a floor, as well as sides, with an aerofoil section running across it. The air goes in parallel to the tunnel centreline upstream and comes out parallel downstream -- if it doesn't come out parallel, please extend your wind tunnel by another mile and repeat the thought experiment. ;) The aerofoil will experience lift, but there is no "net" acceleration of the air downwards. There is local acceleration, but all you end up with is an aerofoil being lifted (for the sake of argument) upwards, and an equal and opposite total force on the walls of the tunnel.

You might think it an unrealistic situation but it's related to ground effect, and to the realisation that we don't actually fly in an infinite volume of air -- at some point below the aircraft the earth has got to do some pushing up.

The relationship between pressure forces on a notional surface in a fluid and the momentum flux passing into and out of that surface are connected by a fluid mechanics equation usually called the momentum theorem. It's through that equation that the various forces and "accelerations" of the air can be related. How they add up in detail depends on where you draw the notional surface.

[bjornhall]

bookworm:

1) Of course, all scientific models are "pulled out of one's hat" at one level or another. That in itself does not make it less self-consistent. It is when you find you have to start adding bits and pieces here and there to solve specific problems that you know you are in trouble; e.g., the Kutta criterion.

2) Newton's laws do not say what you are claiming. Finding a stationary solution does not imply that you will come across such a solution in real life. For that you need to make a stability analysis, which will readily prove that the solution of a pin balancing on its point, while stationary, is unstable. Therefore, with a correct application of Newton's laws, you will conclude that the pin falls.

3) Surely you will agree that the degree of understanding of a physical field should be related to how much the experts understand, and not what one might explain to the layman. If we had a full understanding of aerodynamics, we would have no need for wind tunnels and we would not have a problem with turbulence modelling. But we do.

homeguard: That's my preferred explanation too! :ok:

[bookworm]

It is when you find you have to start adding bits and pieces here and there to solve specific problems that you know you are in trouble; e.g., the Kutta criterion.

That's a pejorative description of something that physicists do with their models all the time: add conditions that tie the model to observed behaviour when prediction of the behaviour is beyond the scope of the model. Isn't refractive index just such a "fudge factor" in beam and ray optics?

Surely you will agree that the degree of understanding of a physical field should be related to how much the experts understand, and not what one might explain to the layman.

Certainly.

If we had a full understanding of aerodynamics, we would have no need for wind tunnels and we would not have a problem with turbulence modelling. But we do.

Do you think Zeiss don't prototype their microscopes just because beam and ray optics is "fully understood"?

[bjornhall]

bookworm, I'm rather losing the plot of your argument there... Physicists don't develop models the way you claim that we do, refractive index is certainly not a fudge factor but a material parameter, prototyping is mostly about production matters and not basic design principles... Maybe we'd better give this a rest?

[bookworm]

Physicists don't develop models the way you claim that we do,

We'll have to agree to differ on that one.

refractive index is certainly not a fudge factor but a material parameter,

Ah the "material parameter"! It doesn't get much more empirical than that, does it?

prototyping is mostly about production matters and not basic design principles...

And that's why aeronautical engineers use wind tunnels -- production matters and not basic design principles. The fundamental design of aircraft is done using CFD these days, not balsa wood and fans.

Singling out aerodynamics as mysterious or undiscovered just fuels the dumb conspiracy theories that Gibson was condemning.

[bookworm]

One really needs to look to the motivations behind aerodynamic theories before dismissing anything as 'dumb'.
Prandtl was inspired by the success of Lilianthal and the Wrights....

Whoa! Hold on. Perhaps my mistake for not reading Gibson's original article to see exactly what he condemns, but Prandtl's work is not at issue here. The "conspiracy theories" I'm thinking of are those that persist that aerodynamics has been "wrong" all these years, that engineers don't know how wings "work" and that a naive application of Newton's laws rides in to the rescue like the seventh cavalry.

[bookworm]

Fancy explaining it to a non-lay person in a way that is physically accurate? I must have been asleep in that lecture.

What do mean by "it" in that context? I used the article to mean aerodynamics as a whole, and I think you'll agree that might take a while... ;)

Circulation isn't really what is happening - it isn't an attempt at an explanation for lift, it's just another tool used in design to calculate values that we wouldn't be able to otherwise.

Eh? Not really happening?? Of course it's really happening. How do you explain how the calculated values are correct if it's "not really happening"?

[bookworm]

I understand how to calculate all sort of useful things that would allow me to design a practical working wing, but I still don't get how it works.

So what do you "understand" the workings of? What allows you to say "yes, I get that"? I'd wager it's your familiarity that makes you comfortable with assumptions and models in some cases, but not in others (which is the point I was trying to make with bjornhall and the beam-and-ray optics). Subjectively that's fine, but I don't think it's appropriate to generalise your subjective comfort level to an objective assessment of the state of the art.

Are you being serious?

Perfectly.

[bookworm]

can you give me some insight as to your understanding of circulation and why you would believe it is 'what is really happening'

Yes, I will, but answer my question first please.

So what do you "understand" the workings of? What allows you to say "yes, I get that"?

Explain an aspect of physics of similar complexity in a way that allow you to say that you "understand" it.

[benhurr]

It is all down to PFM - pure magic.

[bookworm]

Your not stalling are you Bookworm? I understand the lift theories that currently exist, because I understand the mathematics and the motivation. I understand that they don't give an explanation for lift as they were not intended to.

Nope, I'm not stalling. I'm trying to find out what you mean by "understand" and "explanation" because you're being very selective about their use. If you don't offer a reference case, then whatever I say you're going to claim that it doesn't really "explain" it.

Perhaps the debate on the reality of circulation is separate and simpler? If I were to measure the velocity at closely spaced points on a closed loop around a lifting wing, for the sake of simplicity well away from either root or tip, and use that to construct a line integral of the measured values around the loop, I believe I would come out with a non-zero value, which would, furthermore, be roughly proportional to the lift. Do you dispute that? If not, how can you say that circulation is "not really happening"?

[bjornhall]

We'll have to agree to differ on that one.

And that's not the only one...

Ah the "material parameter"! It doesn't get much more empirical than that, does it?

Surely you will agree that there is a difference between an empirical model, and a parameter whose numerical value is obtained empirically?

Singling out aerodynamics as mysterious or undiscovered just fuels the dumb conspiracy theories that Gibson was condemning.

"Singling out" only to the extent that since this is a board about aviation, going too deply into physical fields other than fluid dynamics might be a tad boring to the unfortunate readers of this debate... Aerodynamics is certainly not the only field we have insufficient understanding about ("undiscovered" is your word, I think it is too strong to describe aerodynamics). Nonlinear optics is a good example, but let's not go into that...;)

I'd wager it's your familiarity that makes you comfortable with assumptions and models in some cases, but not in others (which is the point I was trying to make with bjornhall and the beam-and-ray optics). Subjectively that's fine, but I don't think it's appropriate to generalise your subjective comfort level to an objective assessment of the state of the art.

Speaking for myself, but I'm sure it is true for G-EMMA as well, I'm quite familiar with theories and models of both types; those where we have a very clear and complete understanding of what is going on, and those for which we only have crude and incomplete "calculation tools" rather than a deep understanding. I spent five years working to bring examples of the latter to the level of the former... Of course, when chosing to contrast aerodynamics with ray optics and optics of beams, I picked fields I am thoroughly familiar with. So it is hopefully a little more than "subjective comfort level"...

[bookworm]

Do you dispute that? If not, how can you say that circulation is "not really happening"?

You didn't answer the questions.

Thus calculating a quantative theoretical value for the lift that might be a good approximation to the empirical result...

No, not "calculating a quantative theoretical value". Measuring a physical quantity. Do you believe that magnetic flux "doesn't exist" because I can't measure it directly but rather have to measure magnetic field strength and aggregate it up in a similar way? Is magnetic flux a "theoretical value"?

To say that there is circulation around a lifting wing is just a quantification of the easily observed phenomenon that the air travels faster over the upper than the lower surface (if the wing is lifting upwards). What Joukowski did (and if you believe Anderson then it was Joukowski, not Prandtl) was to relate that to lift per unit span in an inviscid fluid model. You can call the relationship "inexact" or "theoretical" if you like, but that's a far cry from denying the existence of circulation.

[bookworm]

Surely you will agree that there is a difference between an empirical model, and a parameter whose numerical value is obtained empirically?

I can obtain the direction of flow at the trailing edge of the aerofoil with a cotton thread, which is probably rather simpler than even measuring a refractive index. Both are measurable quantities whose values can be fed in to useful physical models and are not themselves predictable by those models. If you look to one to prove that the model is not "self-contained", I think you'd be hard pressed to treat the other differently.

[Islander2]

G-EMMA wrote:
Clearly circulation isn't what happens at all when a wing produces lift, it is just a technique for calculating theoretically the quantative lift that might be expected, Prantdl made it up, to get design results.Having missed the last few pages of this thread, I'm reluctant to re-enter the fray at this stage! However, I would just like to say that I believe that statement to be historically incorrect, G-EMMA.

It is Lanchester who is generally credited with an explanation of lift based on circulation and he developed his ideas in the early 1890s, fifteen or more years before Prandtl developed his own wing theory. From a 1930s aerodynamic theory text by Giacomelli and Pistolesi: "With regard to Lanchester's contribution to aerodynamics, there are two great ideas conceived by him: the idea of circulation as the cause of lift, and the idea of tip vortices as the cause of drag known today as induced drag."

Importantly, Lanchester's circulation ideas (as eventually published in his 1907 Aerodynamics) were essentially non quantitative and very much not put forward as a design tool. It was mathematicians Kutta and Joukowski that independently produced the quantitative development of circulation theory that is still used today to calculate the lift generated by wings at airspeeds <Mach0.5.

If you have not read it already, G-EMMA (edited to say I now see you have!), I would urge you to take a look at John D Anderson's seminal A History of Aerodynamics and Its Impact on Flying Machines. As an eminent academic aerodynamicist, historian and one time Special Assistant for Aerodynamics at the Smithsonian, Anderson's book is IMHO the most authoritative and readable account ever written of the story of aerodynamics development. Who knows, it may cause you to question your belief that: "the science of aerodynamics remains in many ways in its infancy, both mysterious and undiscovered", which is a view I struggle to share.

[bjornhall]

I can obtain the direction of flow at the trailing edge of the aerofoil with a cotton thread, which is probably rather simpler than even measuring a refractive index. Both are measurable quantities whose values can be fed in to useful physical models and are not themselves predictable by those models. If you look to one to prove that the model is not "self-contained", I think you'd be hard pressed to treat the other differently.´

Now you're just playing with words... But you're no nearer to the point...

Look: In ray optics, you have a model, that is not merely empirical, in which you have a parameter: refractive index. You know that parameter has to be there, you know why and what it means, you just don't know its value. So you measure that parameter.

In non-viscid fluid dynamics, you have no notion of such a thing as 'viscosity', and you have no clue at all that there is 'something' that makes the zero-circulation solution unphysical, or what the air will do at the trailing edge (or even that there is something especially significant about the trailing edge). So you add a whole new concept to your model; you no longer have non-viscid fluid dynamics, but you now have that + the Kutta criterion.

In the first case you only measure a parameter, and one that you know what it means and why it is there; in the second, you have to 'measure' an entire concept without knowing why you even have to do that.

And it gets worse: If you go outside the most basic theory, you can even calculate refractive index from first principle. But in aerodynamics, especially at high alpha, you have trouble even finding the right fudge factors...

[bookworm]

However I believe Prandtl published in 1905 his circulation theory of lift in the same year that he gave us lifting line theory, or am I'm getting confused somewhere???

Well I can't find a mention of it in Anderson but it does get a mention here (http://www.scientistsandfriends.com/aerodynamics.html) but there's no citation.

So I may have to surrender on the history G-EMMA but I'm standing firm on the physics. ;)

[Islander2]

However I believe Prandtl published in 1905 his circulation theory of lift in the same year that he gave us lifting line theory, or am I'm getting confused somewhere???G-EMMA, I don't know. I think Prandtl's best known for his work on boundary layer theory, some early high-speed aerodynamics thinking and also, as you say, his lifting line theory. He could well have made a contribution to the circulation ideas, but the other names we've mentioned are generally taken to be the prime movers in this area.

[bookworm]

You know that parameter has to be there, you know why and what it means, you just don't know its value.
...
In the first case you only measure a parameter, and one that you know what it means and why it is there; in the second, you have to 'measure' an entire concept without knowing why you even have to do that.

No you know that the parameter has to be there, because you're familiar with waves, and the implications for the direction of travel of a ray, which are outside the self-contained scope of your ray optics. Refractive index is such an intuitive concept to you that you take it for granted. But give a child a prism and a laser (I know, you shouldn't let children play with lasers ;)) and they will be utterly stunned that the beam bends: it's magical.

By contrast isn't the Kutta condition pretty obvious to a child with a sheet of paper and a hair-dryer?

I'm not playing with words -- I'm accusing you of subjective bias, a bias that any physicist is susceptible to. We all have different views on what makes an "explanation" based on our education and experience. The semiconductor physicists try to explain to me about transistors by talking about where the "holes" go. I tell ya, I'm not buyin' it 'cos I know there's no such things as "holes", but still my PC manages to launch Windows... ;)

[bjornhall]

Sorry, that's just not how it works... This has nothing to do with whether a concept is intuitive or not. It has to do with what is contained in the theory itself, and what has to be added from 'the outside'.

I think you're highlighting the problem though, when you say "we all have our views on what makes an explanation" ... But for those of us who do physics for a living, that point of view does not work. What constitutes a self-consistent theory, and what makes a theory 'complete' and 'sufficient', is something we learn in training, are supposed to apply when creating the models, and will have a new one ripped for us by our peers if we don't observe. We're supposed to know that stuff. And while there's a degree of subjectiveness in that judgement, it's not sufficient to allow aerodynamics to pass as a sufficiently developed and understood theory...;)

But that does not imply that our current understanding is not sufficient to be quite useful to engineers, design working aircraft, and tell us a great deal about how lift relates to other observables. It is certainly useful in its current form; of that we can hopefully agree!

[Islander2]

But for those of us who do physics for a livingbjornhall, before this gets too 'them and us', you may care to ask bookworm what he does for a living ;)

[bookworm]

I found this on Prandtl, it doesn't really say much regarding circulation but it got the old cogs turning regarding the relationship between Kutta, Lanchester and the master himself.

Enough to convince you that circulation is very real? ;)

[bookworm]

Bookworm, this isn't the first thread on PpruNe to run to several pages where you have argued with aeronautical engineers on the subject of whether or not they 'understand' how lift is produced.

I would never presume to tell you what you or any particular group understand. I merely assert that the understanding is there for the taking. You need to decide if a particular explanation or model convinces you or not. So far you've declined my request to give an example of a phenomenon of similar complexity that you feel you "understand".

Edit: Having read back through I've muddied the waters with my use of the word 'circulation' instinctively I've used it refering to Circulation Theory - hence running rough shod over Lanchester's work and picking up on circulation at the point where it began to give useful quantitive results.

So does that put you in a position to answer my questions on whether you believe that circulation is "really happening" now? ;)

[bookworm]

Unfortunately Bookworm is it your viewpoint that stiffles students of aerodynamics. You think it is dumb to question our current knowledge. I realise it is full of gaping holes.

On the contrary. I think students of aerodynamics should get up to speed on "our current knowledge" before accusing it of being "full of gaping holes".

[bookworm]

Pictures? This one comes to mind. ;)
http://www.unc.edu/courses/jomc050/idog.jpg

[bjornhall]

On the contrary. I think students of aerodynamics should get up to speed on "our current knowledge" before accusing it of being "full of gaping holes".

"Get up to speed"...:rolleyes:

Just who are you trying to convince here? Obviously not someone knowledgeable; you've tried that and failed... Obviously not those you're arguing with; we've heard and considered what you have to say and concluded you have no idea what you're on about...

Can just picture it now: Having followed all this conversation, all the audience will go "Hmmmm ... Well he's been trying to discuss aerodynamics with aerodynamicists, physics with physicists, engineering with engineers (here and in previous threads), they all say he's wrong, they are all able to explain how and why he is wrong, none of them agrees with him... I guess he's right then!"

No, wait, I can't picture that at all... :E

[bookworm]

Obviously not those you're arguing with; we've heard and considered what you have to say and concluded you have no idea what you're on about...

I'll get back in m' kennel then. :{

[IO540]

Not wishing to argue with anybody here, but taking bookworm's point about taking a line integral around the wing and getting a nonzero value:

The air is not really flowing around the wing, is it? The airflow over the top is front to back, and the airflow underneath is also front to back.

Mathematically, the nonzero integral means there is some net "circulation" but IMHO the physical reality is not really "circulation" because that would imply e.g. the upper flow to be front to back and the lower flow to be back to front again :)

It's a bit like in structural analysis, you get a negative value for compression in a particular member. It means the member is in tension, not compression.

[bookworm]

Mathematically, the nonzero integral means there is some net "circulation" but IMHO the physical reality is not really "circulation" because that would imply e.g. the upper flow to be front to back and the lower flow to be back to front again.

Circulation is a well-defined term in aerodynamics and means that net circulation. Lanchester's contribution was to recognise that the flow was a superposition of the freestream flow and a purely circulatory component - you're used to superposition in linear models in electronics, I presume. That flow pattern is a reasonably realistic model of the real airflow around a wing beyond the boundary layer, and predicts lift in the way that Kutta and Joukowski suggested more than 100 years ago. It's not a perfect model, because of viscosity and compressibility, but it gives a very sound basis for the understanding of lift.

Only if you interpreted the circulation in circulation theory to mean your back-to-front flow would you conclude that circulation theory was a contived engineering model and that circulation wasn't what was happening at all. To suggest that Lanchester, Kutta, Joukowski and Prandtl didn't understand what was going on around a lifting wing because they chose to give a precise technical meaning to the word circulation would be particularly daft. ;)

(Oh sorry I seem to have wandered out of the kennel again... Now where was that bone? ;))

[homeguard]

I've stuck with the thread in a rather cynical certainty that I might actually learn something that would turn on its head all my twenty years of explaining lift somewhat incorrectly.
Now having stayed up all night, ears pricked and eyes wide open, everyone appears to be taking their bat and ball home. Ah! well, its the way it goes. I'll just carry on explaining it all wrong - well why not, it seems to work and I aint heard anything better. Last one out turn out the lights and drop the latch please.
Good night, sleepy heads. :confused:

[Another St Ivian]

I am curious to know how you can regard our understanding of aerodynamics as being in its infancy when we're seeing CFD models with upwards of 97% correlation to the 'real world' test data?

Would it not also be fairer to say that we understand well the way in which lift is generated, but that our approaches to precisely quantising such values are at best inelegant, and if being brutal, somewhat bodged?

ASI

[bookworm]

Would it not also be fairer to say that we understand well the way in which lift is generated, but that our approaches to precisely quantising such values are at best inelegant, and if being brutal, somewhat bodged?

Yes. What a delightfully succinct description of the state of the art!

[bjornhall]

Would it not also be fairer to say that we understand well the way in which lift is generated, but that our approaches to precisely quantising such values are at best inelegant, and if being brutal, somewhat bodged?

I would say it is exactly the other way around! We can calculate things pretty well in many cases, and the methods used for 'mere' calculation are quite often rather bodged, in any field, which is not a problem when all you're after is numbers... It is the detailed understanding and analytical descriptions that are lacking.

[jamestkirk]

If lift is so important, why do people put so much emphasis on learning about drag..What a waste of time.:eek:

[Another St Ivian]

G-EMMA,
I'm not really interested in joining in the bunfight you've all made for yourselves here, so the first point of your post is somewhat moot.
I really want to know the answer to whether Aeronautical Engineers have historically 'understood' lift generation for the past 100 years before moving on.
Surely it still stands that the case is we engineers have known how lift is generated by empirical means, i.e. through observation. We've been sticking aerofoils and wing sections in tunnels and measuring the forces, pressure fields and streams for (in technological terms) aeons. It's our mathematical approaches to describing these observed phenomena that come up short in various ways.
But as you asked, the question is not a case of whether CFD models work, the question is whether we 'understand' the phenomenon of lift generation
I don't think there ever will be a time when a leading thermo/aero/fluid/math genius will sit down and write a paper on lift, finishing with quod erat demonstrandum.
I feel it is fair to accept that there may never be a perfect, elegant, 'smart' singular mathematical solution to describing a natural phenomena with potentially more input parameters than we're even capable of imagining.
My point is that if we're generating these 'crude' CFD models which are approaching an infinitesimal deviation on real test data, then by definition we have reached an understanding. It might take weeks of raw number crunching of Monsieur Navier and Mr Stokes' fine equations, but ultimately the results we're seeing these days are quantising the natural phenomena, and thus it follows logically that we have reaching an understanding.
My viewpoint is not an affront to the 'state of the art' of current engineering practise and what can be achieved. It is more a rather childish excitement regarding what the future may hold. I can see the gaps still [...] Look at what has been achieved in the past 100 years, where will we be in another 100?
Don't get me wrong, every day I am in Uni I learn more and more about how we really know less and less in our field, but that being said, I can also see that we're producing results and are rapidly approaching the day where we don't really need our 'Einstein' moment.
It was a viewpoint instilled in me by lecturers, without it, it is rather hard to progress through a study of aerodynamics and deal with all the contradictions.
Don't worry, it's a viewpoint that is still shared by many of the staff, we students are all aware how fallible each theory is in isolation, but we're also aware that we're not stabbing at some mythical force which envelops a wing every time we wish to fly :E

[Another St Ivian]

You make some very good and succinct points, G-EMMA (Gemma, Emma?). To be honest I don't feel as though it's in my capacity to sufficiently debate either side of this one; Your points are entirely valid, but I also feel that there is a case to be made in purely logical terms for CFD and other associated numerical analyses being viewed as an extension to an understanding of lift generation.
Anyway, more pressingly, this all came to a head from a desire to find a theory of lift which could be related to PPL's. Personally, when I take friends flying, I prefer to relate the theory espoused by our military brethren (http://www.pprune.org/forums/showpost.php?p=3285718&postcount=81).

When I advance the throttle and the engine begins to roar, lift pixies are attracted to the noise and flock to the wings. They stick on and up we soar;

http://img516.imageshack.us/img516/5167/liftpixiesun4.jpg

Of course we can't pull the nose up too high, or manoeuvre too aggressively, lest they fall off.

And on landing, a touch of the spoilers to give them a gentle nudge off and let us return to the ground;
http://i21.photobucket.com/albums/b270/cumpas/Untitled-4-1.jpg

[SkyHawk-N]

I recently purchased a STOL kit for my Cessna 172 ....
http://www.csharpprogrammer.com/pictures/hortonstol.jpg

[llanfairpg]

John Gibson has spent a lifetime in aerodynamics at Warton, and in my book that makes him rather better qualified than those guys at Fermi

Harold Shipman spent a lifetime in medicine too !:ok:

Shouldnt this all be in INSTRUCTORS AND EXAMINERS, imagine someone thinking about learning to fly seeing all the above, a trifle off putting perhaps!

[Spinflight]

Why do aircraft fly?

Because fluid dynamics are just too damned weird.

What else do you need to know?

[Another St Ivian]

I can't really see me setting up a line integral and measuring the circulation on me wing on my next circuit session, can you?

Probably more straightforward than playing with that bloody whizz-wheel which has been imposed on students for far longer than seems necessary!

[Another St Ivian]

MDR all the way!

[moggiee]

I've stuck with the thread in a rather cynical certainty that I might actually learn something that would turn on its head all my twenty years of explaining lift somewhat incorrectly.
Now having stayed up all night, ears pricked and eyes wide open, everyone appears to be taking their bat and ball home. Ah! well, its the way it goes. I'll just carry on explaining it all wrong - well why not, it seems to work and I aint heard anything better. Last one out turn out the lights and drop the latch please.
Good night, sleepy heads. :confused:
I'm with you - does it REALLY matter how lift is produced as long as we know how to use it?

It doesn't matter if it's Bernoulli, Newton or Basil Brush - more AoA = more lift up until the point when more AoA = less lift. As long as the pilot knows how to recognise, avoid and recover from the dodgy bits of the flight envelope the detailed aerodynamic theory doesn't matter.

This thread has turned into a p1ssing contest - "I can get it higher up the wall than you!".

[Major Major]

Now that sounds useful.

I've found the old degrees=crosswind/nm per minute really handy of late.

So the other day I was given a diversion, cross wind was 6Kts, at 90Kts (1.5nm/min) I had an easy 9 degree correction angle, which did the trick - flew 279 instead of 270.

Though it took me more than 2 seconds, so your prepped sheet is probably better.

Sorry about the thread creep (though on this thread...).

[bookworm]

Some of the commonly given theories are wrong though,

Now you're going to tell me that the lift pixies aren't real either, aren't you? You're a hard one to convince, you know. ;)

I think every explanation of lift to a student should start with an empirical description: this is what happens with a flat plate, this is what happens when you change the AoA, when you introduce camber, when you introduce thickness, when you extend flap... That's a lot more use to the student than trying to explain why the lift is created. The rest can be saved for rainy days on PPrune...

[IO540]

I've just reversed engineered the whizzwheel and now have a spread sheet that will generate the sine and cosine waves for drift and GS for any given predicted wind so I can workout the heading to fly for diversions in 2 seconds in the air.

I don't think one needs to use the circular slide rule when airborne. One is not allowed to use electronic calculators in the ground exams or in planning the flight and there is thus an implicit assumption that you will have no option but to use the slide rule, but I don't think you actually have to use one.

What you have to watch is this: the CAA exams have tricks to catch you on common errors and if you get the answer 1 degree out you will fail that question. The multiple choices are rigged to catch this. The fact that not even a £20k autopilot can hold a heading within 1 deg is irrelevant.

The trig for wind calcs is trivial. There are also countless E6B programs around for PDAs. I have one myself, called Flightcalc or something like that.

[bookworm]

I'm troubled by the Lift Pixie theory...

I do wonder how they solve the turtle paradox (http://en.wikipedia.org/wiki/Turtle_theory) aspects, as well as how they manage to stay comfortable while sitting on our vortex generators.

Of course it's a simplification of Shafer's Lift Demons (http://www.messybeast.com/dragonqueen/liftdemon.htm) which is unquestionably the truth as far as lift goes. S'pose you're going to say that that proves that lift has only been properly understood since 1994 then... ;)

[tigerbatics]

I've gone through this thread and am confused. What is all this stuff about lift theories, Newton etc ?

None of this has anything to do with how and why an aeroplane flies. An aeroplane files because the paperwork is in order and it has a valid C of A.

It must be so because the CAA told me that without one my aeroplane could not fly.

[EvilKitty]

G-EMMA wrote:

All will become clear when I write my paper on the extendable stick theorem.I'll be very interested to read that.

Personally though, I believe that this approaches the problem from totally the wrong direction, and thereby propose the "string theory of flight". In this model, aircraft hang from high strength high tensile monofilament spidy-silk (which is why you can't see it). pitch, role and yaw movements are all based around the movement of the string at the attachment loops on the aircraft, and/or movement of the airframe within the loop. Turbulence is also easily explained by the interference of the string in contact with an external force or body that transmits as vibrations through to the airframe.

:\:}:8

[Major Major]

You're all not seeing the bigger picture.

It's a monofilament above and a stick below. When the two are in balance you're straight and level.

Trimming shortens or lengthens the monofilament, which we call Lift, but you always pivot on the stick, which is called Gravity.

The stick is obviously connected to the aeroplane at the Centre of Gravity.

In my scientific paper soon to be published, I will draw on this theory to illustrate how nanopixies circulate from the wings to the Gravity and Lift with little parcels of air to keep the whole thing working - they run very fast indeed and have little boots that stick to aeroplanes

Computational Pixie Dynamics anyone?

[Another St Ivian]

I would, but I still can't figure out how to write papers in LaTeX...! :uhoh:

[EvilKitty]

Don't they use paint on latex for their bodysuits?

[Spinflight]

Just out of interest....

If you were using a CFD package to model the laminar flow over a wing how many individual cells (modelled pockets of air effectively) would be needed in order to provide a reasonable calculation?

I'm assuming you can alter the degree of precision by increasing or decreasing the number, would this be correct?