PDA

View Full Version : Coanda & NOT Bernoulli



p-11.cAce
09-27-2006, 01:45 PM
http://www.aa.washington.edu/faculty/eberhardt/lift.htm
In the ever crazy world of hangar flying there are a few discussions that always amount to the "holy wars" of aerodynamics and flight - downwind turns, area of reversed command, spiral stability, etc. but none ever comes close to the passion behind how a wing works. For anyone who has never thought through the standard "Bernoulli" description of lift generation or for those who know that is really not what is going on check out this site!

NekoReaperman
09-27-2006, 05:18 PM
Bernoulli spoke about pressure and velocity, not just how it relates to hydrodynamics

NonWonderDog
09-27-2006, 06:01 PM
No, no, no, no, no, no, no, no.

It's not Coanda effect. Coanda effect does not generate lift. Wings can still generate lift when the flow is completely separated. If lift were due to the Coanda effect alone, stalled wings would generate no lift!

The air behind the wing will still have an overall downward movement when there's separated flow, but it will be turbulent, it will look nothing like those pictures, and the movement won't be due to Coanda effect at all. Newton always applies, but the second law is an effectively impossible way to analyze lift.

And, ya know? If the flow is incompressible and irrotational, the pressure distribution generated with Bernoulli's equation gives the the exact result for lift. Finding the velocity field is rather complex, but Bernoulli's equations are not just some fiction made up for schoolchildren. If it's compressible, you can still use the Euler equations to find the pressure distribution on the wing, and lift can still be found by integrating the pressure over the wing. For turbulent flow, you use the Navier-Stokes equations in a CFD analysis to find the pressure distribution.

The air is deflected, and the deflection satisfies Newton's second law, but the air deflection isn't really the reason for lift. It's more useful to say that the air is deflected *because* of the speed and pressure changes created above and below the wing -- especially for turbulent flow.

JTFM
09-27-2006, 06:04 PM
Clearly, the theory of equal transit times is a load of bunk.

However, as a CFI, I had to show this to my students in order for them to pass the written exam--even some oral exams--but then I always and immediately went outside the syllabus and explained the Newtonian side of things, including Coanda, viscosity, boundary layer, etc.

NonWonderDog
09-27-2006, 06:13 PM
I don't know what this "theory" of equal transit times is, or who is pushing it. It has nothing to do with aerodynamics, and results wholly from a misunderstanding of what is going on. The air won't be suddently decellerated in order to rejoin the free stream, and there's no reason whatsoever that it should be. Likewise, the air behind the wing will not be deflected to the horizontal in subsonic flow.

That does NOT mean that lift is not due to the pressure distribution, or that the Bernoulli equations do not describe lift!


Note, however -- and this is where it gets weird -- that in supersonic flight it is entirely possible to get flow after the wing to be parallel to the flow in front of the wing. With a diamond-shaped airfoil, the air would actually be deflected upwards. This will happen while the wing still generates positive lift! The only way to analyze this is by computing the shocks and integrating the pressure distributions. This happens because the shocks change the density of the air above and below the wing independently of the pressure, and this will never happen in subsonic flow. If you really cared to compute the momentum flow above and below the wing, you'd probably find that the second law (as F = d/dt(mv)) is still satisfied, but due to changes in mass rather than velocity. This isn't the normal case, but it shows that lift cannot be computed from the tail deflection angle.

raaaid
09-28-2006, 02:03 AM
wings generate vortices, vortices generate antigravity and free energy according to schauberger therefore lift and fuel saving

G.J.deRue
09-28-2006, 01:02 PM
Originally posted by raaaid:
wings generate vortices, vortices generate antigravity and free energy according to schauberger therefore lift and fuel saving ?

Akronnick
09-28-2006, 11:05 PM
Arrgghh, Lotsa big words in this thread, we're naught but 'umple err.. Pilots!

Whirlin_merlin
09-29-2006, 03:21 AM
A little experiment to try.
Hold a peice of paper along one edge by finger and thumb at each end.
Hold it up so that if it were stiff enough the paper would be parallel to the ground, as it's not stiff enough it should flop down away from you.
Now blow over the TOP of the sheet of paper (often best if your lower lip rests on the leading edge). Abit of fine tunning and the paper should lift and striaghten out as you blow.
Sorry if the description is hard to follow but give it ago it has alot to do with how a wing works but I'll let you decide which explination it supports.

Sergio_101
09-29-2006, 03:37 AM
Originally posted by p-11.cAce:
http://www.aa.washington.edu/faculty/eberhardt/lift.htm
In the ever crazy world of hangar flying there are a few discussions that always amount to the "holy wars" of aerodynamics and flight - downwind turns, area of reversed command, spiral stability, etc. but none ever comes close to the passion behind how a wing works. For anyone who has never thought through the standard "Bernoulli" description of lift generation or for those who know that is really not what is going on check out this site!

Yup, you are correct.
"Bernoulli" was not wrong, and he lived hundreds
of years before men flew. http://forums.ubi.com/images/smilies/shady.gif

Yes, it's Coanda & NOT Bernoulli.
Wings are effectivly air pumps. http://forums.ubi.com/images/smilies/winky.gif
Air is pumped down to get an equal and oppisite reaction. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Need a graphic demonstration? http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif
Stand under a helicopter.
That's a rotary wing.
Note the rather violent downdraft. http://forums.ubi.com/groupee_common/emoticons/icon_biggrin.gif

A "propeller" is also nothing more than
a rotary wing. http://forums.ubi.com/images/smilies/winky.gif
Props stall, and develope shock waves just
like wings. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

The airfoil shape on a wing
is important to keep the fluid (air) following
the countour. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif
But the airfoil is only there to efficently
run the air over the wing's structure in
an efficent manner. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

The old highschool textbook fairy story of
the air traveling a longer distance over
the top of the wing thus making a vacum
and therefore lift is 100% fantasy. http://forums.ubi.com/groupee_common/emoticons/icon_biggrin.gif

Truth is that many wings have the longer
surface on the bottom. ("super critical airfoils"). http://forums.ubi.com/images/smilies/mockface.gif
US F-111 and C-5 use supercritical airfoils and manage
to fly reasonably well. http://forums.ubi.com/images/smilies/winky.gif

Coanda was correct, he described how a wing really worked. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif
Bernoulli was not wrong, just a hundred years
of bad textbooks and popular press mistakenly
used his description erroniously. http://forums.ubi.com/images/smilies/shady.gif

Sergio

Ratsack
09-29-2006, 05:00 AM
Originally posted by raaaid:
wings generate vortices, vortices generate antigravity and free energy according to schauberger therefore lift and fuel saving

http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

Good answer.


Ratsack

Sergio_101
09-29-2006, 02:44 PM
There is lift generated by the bottom of the wing.
In a full stall, if there is airspeed, there is
still lift.
But your lift at the point of stall is cut
to less than half.

One only has to point a air nozzle at a flat
piece of material and shoot air at it
to see this is true.
There is lift, more like a planning boat
generated by the bottom surface.
At supersonic speeds that is primarily what
keeps you flying!
The near total loss of lift generated by the
Coanda effect at supersonic speeds is why
the huge full moving tail surfaces of modern
fighter planes need to be so big.

Note the SR-71/A-12 "Blackbird".
the bottom of the fuselage is shaped like
a "V hull" boat. That's no accident.
Kelly Johnson knew well that there is a lot
of lift to be had in that area.

Note also the drooping wingtips of the XB-70.
Air pressure trapped between the engine nacells
and those wing tips generated
a lot of lift at max cruise speed.

Again, Bernoulli was not wrong, just a hundred years
of bad textbooks and popular press mistakenly
used his description erroniously. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Sergio

NonWonderDog
09-29-2006, 03:30 PM
Note, however -- and this is where it gets weird -- that in supersonic flight it is entirely possible to get flow after the wing to be parallel to the flow in front of the wing. With a diamond-shaped airfoil, the air would actually be deflected upwards. This will happen while the wing still generates positive lift! The only way to analyze this is by computing the shocks and integrating the pressure distributions. This happens because the shocks change the density of the air above and below the wing independently of the pressure, and this will never happen in subsonic flow. If you really cared to compute the momentum flow above and below the wing, you'd probably find that the second law (as F = d/dt(mv)) is still satisfied, but due to changes in mass rather than velocity. This isn't the normal case, but it shows that lift cannot be computed from the tail deflection angle.

This still stands, and I think I have a Matlab code with a little demo I can send. The angle the air leaves the wing doesn't matter. It is a result, not a cause. The ONLY methods by which nature can influence an airfoil are pressure and shear stress. The pressure distribution can be found from Bernoulli's equation at speeds below Mach 0.3, if the velocity field is known. Thus, it is not incorrect to say that lift results from the "Bernoulli Effect." It's simple, but entirely less wrong than anything to do with the Coanda effect.

I hate quoting Wikipedia for anything, but they've actually got a better qoute than I could come up with. My aerodynamics books see this Coanda effect nonsense as beneath contept, and thus refuse to address it at all.



Some people have attempted to explain how a wing generates lift, by invoking the Coandƒ effect. However, this theory does not produce quantifiable data, and so it is unable to predict such things as the thickness of the boundary layer. Professional aerodynamicists regard this theory as a fallacy. For example, the theory states that air €œsticks€ to the surface because of its viscosity. This implies that if the viscosity of the fluid changes, the amount of lift an airfoil produces should change in proportion. Experiments show that the amount of lift produced by a real wing is independent of viscosity over a wide range. The real Coandƒ effect requires turbulence, so it occurs only if the viscosity is sufficiently low. Furthermore, the air speeds up above a wing's upper surface. The theory assumes that the relative air-flow meets the wing at the same velocity as in free air and then follows the curve. This understates the pressure gradients by an order of magnitude.



The big problem with this debate, however, is the fact that no one ever provides a mathematical definition of Coanda effect. I don't even know if one exists. Flow attachment can be perfectly described by pressure, momentum, and everything else aerodynamics is based on without resort to some undefinable Coanda effect.

Viper2005_
09-29-2006, 04:06 PM
The Coanda vs Bernoulli debate is rather like the blondes vs brunettes debate. Fundamentally if you close your eyes they both do the same thing. Any argument or debate over which is "better" is simply splitting hairs. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

However, since I'm in a pedantic mood, Navier-Stokes for teh win!!!!!111

NonWonderDog
09-29-2006, 04:29 PM
You know, though? It's been a long time since I've paid attention to the horrible high-school definitions of lift. I have to concede that the Coanda effect description IS better than anything you hear in high school.

Lets start at the beginning. High speed does not cause low pressure. No matter what the speed of a jet of air in the free stream, it will be at ambient pressure. The wing does not make the air go faster around some curve, and thus create a pressure differential. Blowing across the top of a piece of paper does not show lift, but simply turbulence at the opposite edge of the paper.

Rather, low pressure creates high speed if density is to remain constant. The speed above and below the wing, and thus the angle of the air when it leaves the wing, is a secondary effect. In supersonic flows, the tail-turning angle behind the wing will be POSITIVE. UPWARDS. It's not the reason for lift. Relative velocity is, however, the easiest way to compute lift if you have a potential flow. This is further generalized into circulation around the airfoil, making things even easier.

Just to say that flow does not separate because of "Coanda effect" is worse than useless. Flow will be perfectly attached to a cylinder, for example, only if it is inviscid... and if it is inviscid there will be no boundary layer. If Coanda effect is due to viscosity....

As the Reynold's number decreases from infinity, however, you'll get a boundary layer, you'll get separation, and you'll get turbulence. Does the Coanda effect somehow relate to Reynold's number? All I've ever heard as a definition of the Coanda effect is "A jet of fluid will stay attached to a body."


The only thing I can call Coanda effect in good faith is a jet placed on top of a wing to energize the boundary layer. And I don't say that lightly, becuase I still haven't seen an equation for Coanda effect.

Sergio_101
09-29-2006, 05:11 PM
Ok, just because of this hot button argument I
went downstaris and turned on the air compressor.
I got a piece of magic aluminum, bent it in
a simple curve.
Then I nailed it to a stick.
Then I used a single nail as a fulcrum and
blew air across the top of the sheet of aluminum.
Like magic the simulated piece of paper provided
substantial lift. Enough to make it hard to hold down!

Test #2, blowing the air across the bottom curve.
same result. Not having a gauge to test the resulting
lift pressure I will venture a guess that the lift
seemed the same. I do not have the facilities to
blow a large enough amount of air to test lift
with both sides accurately. But the stick disrupted
the air flow badly. (nailed to the top for the bottom
test, and to the bottom for the top test).

So, sorry, the idea that air blown across a piece of
paper is not providing lift is BUSTED. http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif
But I knew that before I tried it. http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

Next I tried a test I saw online.
I made my own "super critical" airfiol.
Sorry to say the lack of a simple wind tunnel
probibly made the test useless.
But strangely the crude model seemed to provide
lift directly from the angle of incidence reguardless
of being curved surface up or down!
Another myth busted by me and my airhose! http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

Coanda is the reason why wings work, Bernoulli is a
description of the side effect of what Coanda described.

Break out the text books, this is starting to be fun! http://forums.ubi.com/images/smilies/winky.gif

You guys have me going, I may build a simple wind tunnel. http://forums.ubi.com/images/smilies/53.gif
A leaf blower should do nicely to provide the air flow.

Oh, yes, I know this field very well, but I will not
spill the beans as to why or how.
I prefer to watch you guys post variations of the old Bernoulli myth.

Yesm the curving of the air flowing over the wing created a pull
on the top surface.
And there is as a result a thin low pressure area at the surface.
But the difference in velocity and distance traveled
between top and bottom has NOTHING to do with lift!

Sergio

NonWonderDog
09-30-2006, 09:13 AM
If you blow across a piece of paper, most of the "lift" is just surface drag pulling it straight. If you weren't holding on, it would just fly away from you and fall to the ground.

And don't get me wrong. I'm not saying that the Coanda effect doesn't exist. I'm saying that it does not in any way explain why an airfoil produces lift. I still want to see some kind of mathematical definition of Coanda effect, too. Aerodynamics just doesn't have these kinds of mystery effects with no definition.

If you blow an air jet (at ambient pressure, I'm not at all saying that fast air is at low pressure due to being fast) across a bent piece of metal, sure. The air, due to viscosity, would try to pull other air along with it. That would create a low-pressure region between the jet and the surface. The jet would be pushed towards the surface due to atmospheric pressure. The pressure differential is the only thing that is moving the airfoil, and I guess it's "created" by Coanda effect in this case.

That doesn't happen in a wing. The wing creates high-speed flow over the top surface all by itself, without the help of a compressor. Coanda effect provides NO mechanism by which to explain this, but the huge speed differential is symptomatic of the REAL lift created by the wing.

And no, two air molecules, when separated at the leading edge of the wing, don't meet up again at the trailing edge. I'm not saying that. Rather, the air on top gets there first. I think this is important to understand, as it debunks both the high school "Bernoulli effect" nonsense and the pseudo-intellectual "Coanda effect" nonsense.

The distruption of the airflow creates low pressure above the wing -- if the air didn't flow around the airfoil there would be a vacuum. This low pressure "sucks" air from in front (the stagnation line is always below the chord line at any positive angle of attack) over the top at high speed. As long as this force is greater than the momentum of the airstream , it will stay attached. (It's more complicated than that, really. Lots to do with viscoscity, boundary layer effects, etc.) As long as it stays attached, the pressure will stay low, and there will be lots of lift. If flow separates, the air will become turbulent and increase the local pressure. This all happens continously, not in discreet steps.

That's all the very simple explanation, but it's more accurate than just saying "Bernoulli Effect!" or "Coanda Effect!"


I'd forgotten just how wrong the "high-school" Bernoulli effect description was; I think I was in 5th grade when I last heard it in a classroom. I'm not defending that description. I will never say that high speed air is inherently at a lower pressure, or that any such speed differential is created by "the top is longer than the bottom," or that the camber of an airfoil is the reason for lift. (Take note: the Coanda effect description depends on camber alone. It is wrong. Flat wings create lift, too, and not just from air hitting the bottom.)

Bernoulli's Equation DOES still predict lift at Mach numbers below about 0.3, regardless of any false causality you may think I'm defending.

Sergio_101
09-30-2006, 10:41 AM
The USAF Museum still has the old Bernoulli fairy tale
on it's web site. http://forums.ubi.com/images/smilies/88.gif

In truth on most modern airfiols the air seperated
at the leading edge get back together at pretty
close to the same time and location at the trailing edge.
But they have been deflected down.
Equal and oppisite reaction, lift! http://forums.ubi.com/images/smilies/winky.gif
No getting around it.
Coanda was correct in 1910. Still is.

An interesting aside is the symetrical airfoils
on areobatic planes. They provide lift by changes
in AOA. In a level plane they provide only drag.... http://forums.ubi.com/groupee_common/emoticons/icon_cool.gif

Same situation for "supercritical" airfoils.
They should drive you down! http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif

But with changes in AOA, lift is defeloped! http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Again Coanda & NOT Bernoulli.

But poor Bernoulli can never be discounted.
300 years before men flew he got it right.

Sergio

NonWonderDog
09-30-2006, 11:48 AM
All right, I haven't been clear. If I haven't said this already, I should have:

If air were completely inviscid (Re = Inf), wings would still create lift. There would be no boundary layer, and there would be no drag (d'Alembert's paradox), but there would still be almost exactly the same amount of lift as in real air. A large portion of subsonic aerodynamics is based upon this fact.

If air were completely inviscid, the Coanda effect would not exist. Coanda effect is due to a jet of fluid influencing the fluid around it through viscosity and creating a pressure differential near a surface.

Clearly, Coanda effect cannot be responsible for lift.

I can't be any more lucid than that.

Sergio_101
09-30-2006, 01:32 PM
Coanda has been dead for 40 years, but the effect he described is real.

Try reading about the Coanda effect, there is plenty of information out there.

Try this link.
http://jef.raskincenter.org/published/coanda_effect.html

Jeff gives a great laymans description of how Coanda works.

Sergio

NonWonderDog
09-30-2006, 09:20 PM
It seems I was wrongly attributing the Venturi effect to Coanda. My description was of the Venturi effect, which would also be evident in the cases I described.

Nonetheless, Coanda effect, or simply "boundary layer attachment," depends entirely on viscous forces. It is due to viscosity. The friction in the fluid keeps it from leaving the surface through some effect I have yet to find a mathematical description of in any book -- at least under the title of "Coanda effect."

Lift is emphatically not due to viscosity. If there is no viscosity, there is no boundary layer, and there is thus no Coanda effect. But there is still lift. None of the equations for lift depend on viscosity, as the Reynolds numbers are in the millions. In common language, that means viscous effects on an aircraft's lift are less than 1/1000000 as large as the inertial effects.

Lift is not due to viscosity! Coanda effect is due to viscosity! Lift is not and can not be due to Coanda effect!

It's that simple!

Sergio_101
09-30-2006, 09:38 PM
Be as emphatic as you like. http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif
Coanda got it right. http://forums.ubi.com/groupee_common/emoticons/icon_cool.gif
I am correct. http://forums.ubi.com/images/smilies/winky.gif
You are grasping at straws. http://forums.ubi.com/images/smilies/784.gif

Sergio

NonWonderDog
09-30-2006, 09:45 PM
If you really want a simple explanation of why the air curves around the top of an airfoil, just state that "nature abhors a vacuum." If that's not good enough, you can reason out that phrase from the second law of thermodynamics. (Mentioning this is a mistake on my part, as very few people have an understanding of entropy. I'm not going to explain entropy here unless pressed VERY hard, as it seems that I am unable even to explain that lift is not due to viscous forces. The proper explanation of entropy is rather involved. Suffice it to say that entropy is anything but a measure of "randomness.")

If you stuck an inclined plane into the free stream, and if the air did not follow the top surface, there would be a vacuum. It should be obvious to you that there won't be a vacuum. Therefore, air moves in to fill the void. There is still a low-pressure region created above the airfoil, and this is what creates most of the lift. That the mass of air moved downwards by the wing fits in with Newton's second law is only to be expected; there is no free energy. However, only pressure can lift the wing. Air displacement is all well and good, but it can't act on the wing without a force.

This effect is entirely independent of viscosity. Lift is entirely independent of viscosity. Coanda effect is due entirely to viscosity, and a quick search will find that every site that attempts to explain Coanda effect makes this clear.

If Coanda effect is due to viscosity, and lift is independent of viscosity, how the h*ll do you intend to tell me that lift is due to Coanda effect? Exactly how am I grasping at straws?

Jetbuff
10-01-2006, 12:44 AM
Interesting discussion guys. If I may though, this confuses me NonWonderDog:

That the mass of air moved downwards by the wing fits in with Newton's second law is only to be expected; there is no free energy. However, only pressure can lift the wing. Air displacement is all well and good, but it can't act on the wing without a force.
Why wouldn't Newton's second law account for at least a part of the lift force? The way I see it, the lower surface of the wing acts as a continuous impact surface for the oncoming air molecules. And, while the individual deflections may be small, on aggregate, they amount to a net force at 90 degrees to the wing's angle of incidence. This force can then be resolved into vectors both up through the wing (lift) and backwards along the chord. (drag) Am I wrong to hold this simplistic notion?

Also, regarding the low pressure area that forms above the wing, you say it pulls air up over the wing from in front of the leading edge. I find this surprising given the commonly experienced phenomena of the sipstream whereby the path of least resistance for air molecules to fill the 'gap' is from behind and slightly below the wing. So my question is, why this apparently unnatural behaviour? Or does the latter only occur when there is airflow separation?

NonWonderDog
10-01-2006, 10:22 AM
It's wrong to treat air as deflected molecules unless you're at VERY low density. If you're explaining the Space Shuttle reentry, for example, you have to treat air this way (I can't for the life of me remember what this is called).

For aircraft, however, the air is best treated as a continuum. There are mathematical descriptions like the Knudsen number, the continuum hypothesis, ect., but mostly it's just because interactions between the individual air molecules happen MUCH more often than interactions with the wing.

In that case, we can only speak of the forces on a wing in terms of pressure and viscous surface forces. If there is zero viscosity, it turns out that the air always follows the airfoil perfectly, with no turbulence. There is still a pressure differential, and the air above is still accellerated, and there is thus still lift. The interesting thing, however, is that once you sum the forces, there is absolutely no drag whatsoever. This is called d'Alembert's paradox, and you might gather from the name that it's rather hard to explain. (After looking into it, I've read online that the lift vanishes at zero viscosity in some way as well. To explain airfoils the air is apparently treated as both infinitely viscous and inviscid at the same time. I rarely trust what I read on the internet, but I am somewhat confused on this point now. It's still true that lift is mostly independent of the viscosity of the medium, however, and this has been shown to be true time and time again. I'm just relaying what I've learned in my aerodynamics classes, in any case.)

As it turns out, drag and turbulence are entirely due to the viscosity of air. Air forms a boundary layer with any surface it's in contact with. Within this layer, the velocity decreases from the velocity above it to nearly zero at the surface. This doesn't really have any impact on the lift unless at high angles of attack, where the boundary layer becomes separated and turbulent. As Coanda effect is just a way to desctibe the attachment of the boundary layer to an object, Coanda effect does not explain lift.

I suppose, on reflection, that you could say there would be almost no lift without Coanda effect in real air, as the wing would be perpetually stalled -- but only if you generalize "Coanda effect" to mean "boundary layer attachment." Boundary layer attachment enables lift, then, but it is horrendously poor at explaining it. It's like saying your car accellerates by throwing the road backwards, instead of analyzing the frictive forces between the tires and the road. In the same way as that simple explanation cannot explain burnouts and understeer, Coanda effect cannot explain stalls or the intricacies of supersonic wings. The simple fact is that Newton's second law depends on forces. The only forces in evidence are pressure and surface drag. I'm trying my best, therefore, to simplify circulation theory (which is all about the forces) into something readily understandible. I appear to be doing poorly.


The reason for upwash is a bit hard to explain on its own, actually. If you look at any streamline picture of an airfoil you will see the upwash, though. It's especially evident in completely flat wings -- the stagnation point is a bit below the leading edge of the wing. There are some good pictures on this page: http://www.windsofkansas.com/atoz2.html . It's about dragonfly wings, but it seems to give a pretty decent desciption of lift anyway.

The very basic reason is that it's actually a freer path for the air slightly below the leading edge of the wing to go up and over the wing than to go below it. If the air goes below the wing, it will be slightly slowed down against the free stream and will increase in pressure. The air is much more likely to go towards the low-pressure region, even if it means the streamlines become more complicated.

Sergio_101
10-01-2006, 12:43 PM
For every action there is an equal and oppisite reaction.
Newtonian physics works in everyday life.
And so it does on an airplane.
Air is pumped down by a wing, lift is therefore generated.
Air striking the bottom/facing surface will generate lift.
Air flowing over the top and deflected down
also generates lift.
The description of how the fluid (air) follows
the surfaces is commonly referred to as
the Coanda effect after Henri Coanda.

No amount of alchemy and hokus pokus will change
the facts.

Note the effect of "spoilers" on lift.
Just like in my stick and aluminum sheet experiment
placing a disruption in the airflow over or under
a wing will totally disrupt airflow and cease
the lifting of a wing.


Sergio

NonWonderDog
10-02-2006, 09:27 AM
Originally posted by Sergio_101:
The description of how the fluid (air) follows
the surfaces is commonly referred to as
the Coanda effect after Henri Coanda.


This is not precisely true. Coanda effect describes how a thin jet of fluid follows a surface due to a centripetal force exerted by viscosity. As a packet of fluid (in a Laplacian frame) hits a curved surface, the fluid nearest the surface is slowed to a very low speed by its viscosity. The fluid next to that is slowed down slightly less, the fluid next to that fluid is slowed a lesser amount, etc. The net result is a roation resulting in a centripetal force on the packet of fluid. If this force is great enough, the fluid will stay attached.

It's a boundary layer effect, which applies to thin jets. It doesn't even try to explain how the air outside the boundary layer follows an airfoil, say.... and it doesn't need to. That motion is emphatically NOT due to Coanda effect, but simply the second law of thermodynamics and continuity. If the air didn't follow, there would be a discontinuous pressure: a sudden vacuum.


If you had an invisid fluid, the air would still follow the airfoil without Coanda effect. Where it gets confusing is d'Alembert's paradox -- an inviscid fluid applies no force as it supports no vorticity. The air should still follow an airfoil, but it would do so in such a manner that the pressures above and below, and in front of and behind, are equal.

Zhukowsky (I think it was Zhukowsky) got around that by introducing a bound vortex, centered on the airfoil. If we accept that airfoils create circulation in the airflow and cleverly choose this circulation, we get exactly the amount of lift generated from Bernoulli's equation, and the computed streamlines are exactly what we see in real attached flow.

There's still no drag, so Prandtl (I'm sure it was Prandtl) introduced the concept of a boundary layer, in which the flow is not inviscid. The skin friction within the boundary layer and the turbulence created by a separated boundary layer is what creates drag, while it is the circulation that creates lift.


That's the engineering way to go about it. The formulas are excessively complicated once you actually need to find the necessary parameters, so it's not taught until you're halfway through a bachelor's degree. After learning this, though, it's just not possible to reconcile the actual lifting force with that predicted by Coanda effect alone. Coanda effect, as applied to the wing, would be barely evident. Most of the air influenced by the wing is at a Reynold's number of several million. It's effectively inviscid.

Trying to ignore that just by saying "flow turning = Coanda effect" is wrong.

In any case, it's MUCH more instructive to identify the forces actually acting on the wing. The forces are pressure and surface friction. The correct desctiption of lift must explain those forces, rather than feebily explaining the mass flow and assuming the forces must be there.

p-11.cAce
10-02-2006, 10:19 AM
NonWonderDog meet Weltner, Klaus and Ingelman-Sundberg, Martin (Department of Physics, University Frankfurt, Postfach 11 1932, 60054 Frankfurt, Germany
*Stockholm, Sweden)

http://user.uni-frankfurt.de/~weltner/Flight/PHYSIC4.htm (http://user.uni-frankfurt.de/%7Eweltner/Flight/PHYSIC4.htm)

The conventional explanation of aerodynamical lift based on Bernoulli€s law and velocity differences mixes up cause and effect. The faster flow at the upper side of the wing is the consequence of low pressure and not its cause.

The generation of lift by an airfoil can be explained correctly and simply taking the downward acceleration of air into consideration. This approach allows to derive the dependency of lift from angle of attack, flow velocity and the air€s density in a streightforward and coherent way.

NonWonderDog
10-02-2006, 11:34 AM
I'm in no way defending the horrible grade-school idea that pressure differences are due to differences in path length.

To be honest, I don't really care what the exact causality is as long as I have a way to compute the lift and moment on an arbitrary airfoil. Circulation theory gives me that. Just blaming the whole mess on Coanda does not. In order for an explanation to be useful to me, it has to have not only explanative power, but predictive power.

Again, just saying lift is due to the Coanda effect is like saying your car's accelleration is due to throwing the road backwards. It's somewhat correct, but ultimately useless.

In contrast, saying lift is due to different path lengths is like saying your car accellerates because it's painted red. I'm not defending that idea.

Jetbuff
10-02-2006, 02:24 PM
Originally posted by NonWonderDog:
For aircraft, however, the air is best treated as a continuum. There are mathematical descriptions like the Knudsen number, the continuum hypothesis, ect., but mostly it's just because interactions between the individual air molecules happen MUCH more often than interactions with the wing.
I don't get it. Why would the fact that air particles interact more often with each other nullify the sum total of impacts with the under surface of the wing? Why would it matter? Every time an air particle strikes the undersurface of the wing it will impart a tiny force a component of which will be up in the direction of lift. Why do we care whether the particle collides with other air particles before and after the event?

p-11.cAce
10-02-2006, 02:48 PM
I'm confused by that as well - isn't the lifting force of a planing boat created by the impact force of the water? I would think that water, being much denser, would have particles that intereact more fequently than air...yet they support the weight of a planing boat.

Sergio_101
10-02-2006, 03:47 PM
Originally posted by p-11.cAce:
I'm confused by that as well - isn't the lifting force of a planing boat created by the impact force of the water? I would think that water, being much denser, would have particles that intereact more fequently than air...yet they support the weight of a planing boat.

p-11.Ace, you are on the right track!
There is more than one source of lift.
Compression lift, the planing boat thing,
and coanda.
Air pressure on the bottom of the wing at high enough AOA
also provides lift. At supersonic speeds it provides
most of the lift!
But at subsonic speeds Henri Coanda got it right.
For every action there is an equal and oppisitte reaction.
Yes, on some wing designs Bernoulli comes into
the picture.
Coanda and Bernoulli are linked.


Sergio

NonWonderDog
10-02-2006, 06:00 PM
The idea is there's high pressure at the bottom, yes, as well as the low pressure above.

The extension of that idea is that you cannot treat this pressure as a bunch of particles hitting the bottom surface. Newton did, and his result basically said wings were impossible.

You have to say, instead, that the fluid increases in pressure due to the decrease in velocity. The air is squeezed between the bottom of the wing and the streamlines below it, slowing down. You have to find the velocity change from circulation theory. From Bernoulli's equation (which is really just another way to write conservation of energy), we know that the pressure increases when the air slows down. The pressure exerts a force on the wing.

It seems convoluted, but it's the only correct way to actually solve for the pressure. And once you've done that, it's actually one step less to find the total lift from circulation theory. All without invoking Coanda.

How would you find the lift distribution from "air deflection and Coanda effect," pray tell? It seems to me a completely useless theory unless it can actually produce results.

p-11.cAce
10-02-2006, 06:17 PM
Ok - I like the "circulation theory"...never been exposed to it before this discussion. Is the circulation "around" the chord of the wing? Does the circulation change at the stall? Is it the reversal of flow on the bottom of the wing which activates the stall vane on most light aircraft?

NonWonderDog
10-03-2006, 09:55 AM
After doing some digging as to what Coanda effect really is, it seems that I must soften my position. Coanda effect is not irreconcileable with established theories of lift, and actually helps explain vortex sheet theory. Careful, though, this doesn't mean what you might think it does.

Firstly, the idea that wings produce lift for the same reason that water from the faucet follows the back of a spoon is almost completely wrong. In the same way as the path-length "theory" is a bastardization of Bernoulli's law, what I'm going to call the "spoon lift theory" is a bastardization of Coanda effect. It is absolutely wrong to say that Coanda effect makes the air above the wing go down, and the air below just kind of hits the wing and pushes it up. Absolutely wrong. This is what people almost always mean when they say Coanda effect is responsible for lift, and I see now that it's the reason for my prejudices against any Coanda effect explanation.

Lets start at the basics. An inviscid potential flow will follow any object perfectly. It will also exert no force on the object. Circulation theory gets around that by hypothesizing a bound vortex centered on an airfoil -- the vortex speeds up the air above it and slows the air below it, creating a pressure differential. The mathematical answers, after a huge mess of very difficult calculations, end up being absolutely correct for attached flows... except that there's no drag. Prandtl's way to account for drag is to theorize a boundary layer, in which viscosity is important, and to say that the boundary layer is responsible for drag. There's also no explanation of what causes the circulation, because it really doesn't matter to the calculations.

Circulation theory, on its own, is exceedingly difficult to calculate for complicated airfoils. A more powerful method is vortex sheet theory. Here, instead of one vortex centered on the airfoil, we treat the entire surface of the airfoil as a sheet of infinitely many vortices. This seems to be even harder to explain, but it works.

Here's where Coanda comes in. Coanda effect describes how the viscosity of a fluid, when in contact with a surface, imparts a rotational force on the fluid. As this happens along the entire surface of the airfoil, top and bottom, Coanda effect is the vortex sheet. Thus, it is seemingly correct to say that Coanda effect is the root cause of circulation around the wing -- as long as you know what you're talking about.

Note, however, that it is absolutely NOT correct to say that Coanda effect is the reason the flow follows the wing. Any potential flow, and we're still talking about potential flows, will follow the wing at any any angle of attack, with or without circulation. It is only correct to say that Coanda effect is the reason for the vortex sheet, and the reason for circulation. The circulation slows the air below the wing and speeds up the air above the wing. These accellerations cause the pressure differentials that create lift.


As you can see, that's a complicated explanation. It's much easier just to say the pressure differential causes lift, because that's what happens.

Some non-engineers didn't know why the pressure differential causes lift, so they just said it was because the top surface was a longer path. This is wrong. Other non-engineers, unsatisied with this answer, looked deeper into the theory and found Coanda effect. After holding spoons in the faucet, they decided that Coanda effect says the air will go down when it touches the wing, and that pressure doesn't matter. This is even more wrong. Please don't believe either of these explanations.

NonWonderDog
10-03-2006, 10:01 AM
Oh, and stall warning horns are very simple devices. The easiest to understand is just a paddle-switch on the leading edge of the wing, below the chord line.

At low angles of attack, the stagnation point will be above the switch and the airflow will push it down. At high angles of attack, the stagnation point will be below the switch, pushing it up. It's triggered so that the horn blows when the switch is pushed up.

Note that the wing isn't necessarily stalled when the horn blows, and in fact the horn usually blows just before you're going to stall. It's really just a warning that you've crossed some angle of attack threashold.

Jetbuff
10-03-2006, 12:12 PM
NonWonderDog, I am aware that simple striking of the lower surface of the wing does not account for all lift. But earlier you appeared to be discounting it totatlly.

As for the pressure differential, I have much simpler (and probably wrong) visualization of how it develops: Wing deflects air; the shape of the wing basically scoops air downwards from it would normally be; this results in air being slightly compressed below the wing and slightly decompressed above it; voila pressure differential!

Simple, clean and easy enough for my non-physics mind to comprehend but probably wrong as I'm sure you'll tell me. http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

NonWonderDog
10-03-2006, 05:07 PM
I didn't mean to, if I did. For a bluff cambered airfoil it's true that the top creates most of the lift when measured at geometric angle of attack... but the camber and curvature kind of mess things up. A bluff airfoil will create low pressure above and below it at zero angle of attack as measured from the chord line, but will only change pressure above it if you orient it so that the bottom is flat -- and this is sometimes defined as zero angle of attack. This really needs pictures.

If you have an uncambered, thin airfoil, things get a bit easier. There will be equal lift created by the top and bottom surfaces. Most airfoils are rather thick and cambered, so it's tempting to say that the top creates most of the lift in real wings. It all really depends on how you measure the angles, though.


It's not completely wrong to say wings keep you aloft by pushing the air down. My argument is that this is (1) about three steps away from Coanda effect, (2) a very poor way of thinking if you actually want to know how much lift will be generated, and (3) not necessarily true at supersonic speeds. Oh, and it's actually a bit misleading when the wing is in ground effect... but explaining finite wings online is probably beyond me.

Sergio_101
10-03-2006, 05:18 PM
Originally posted by NonWonderDog:
I didn't mean to, if I did. For a bluff cambered airfoil it's true that the top creates most of the lift when measured at geometric angle of attack... but the camber and curvature kind of mess things up. A bluff airfoil will create low pressure above and below it at zero angle of attack as measured from the chord line, but will only change pressure above it if you orient it so that the bottom is flat -- and this is sometimes defined as zero angle of attack. This really needs pictures.

If you have an uncambered, thin airfoil, things get a bit easier. There will be equal lift created by the top and bottom surfaces. Most airfoils are rather thick and cambered, so it's tempting to say that the top creates most of the lift in real wings. It all really depends on how you measure the angles, though.


It's not completely wrong to say wings keep you aloft by pushing the air down. My argument is that this is (1) about three steps away from Coanda effect, (2) a very poor way of thinking if you actually want to know how much lift will be generated, and (3) not necessarily true at supersonic speeds.


At supersonic speeds the lift becomes primarily
compression lift. The Coanda effect fails
and the air flow no longer can follow the top
surface.
There is still lower pressure at the top, and
higher at the bottom of course.
But the air flowing over the top is no longer
channeled down.
Many things change at the trans sonic regime and
at supersonic speeds.

Point 3 is correct.


Sergio

gregpeters
10-04-2006, 09:39 AM
Stick with it NonWonderDog - you are on the side of the Angels (and their wings dont need Coanda, either!!)



It's not completely wrong to say wings keep you aloft by pushing the air down.


It IS wrong to say wings "work" by pushing the air down. But, you could say the main thing they rely on is PULLing the air down - onto their upper surface, through the agency of the partial vacuum created there. Of course, there are two aspects to this "pulling" (as Newton would remind us). The partial vacuum is both pulling the airflow down, and pulling the wing up.

As NonWonderDog points out, the aerodynamics experts dont even mention Coanda. Even Wikipedia knows this, saying that when Coanda effect is used to explain lift: "Professional aerodynamicists regard this theory as a fallacy."

One of these professionals (R von Mises: "Theory of Flight") gives the following:
for a standard aerofoil (NACA 4412 at 8% AOA), the negative pressure generated by airflow on top of the wing is around 2.5 to 3 times the positive pressure generated below, when both are averaged over the whole wing chord.

Hence the critical component of lift here comes from the top of the wing - not the undersurface (for low speed flight, at least). So, wings are mainly pulling the air down, rather than pushing it.

In his 600pp book, von Mises never mentions the name Coanda at all.

Regards to O Maddox & team, for this fantastic sim.

Sergio_101
10-04-2006, 05:47 PM
[QUOTE]Originally posted by gregpeters:
Stick with it NonWonderDog - you are on the side of the Angels (and their wings dont need Coanda, either!!)


Wings do indeed pump/displace air down!
Anything besides that required some weird
force to provide lift.

A bug flying in your airplanes cargo space
ads the same weight to your plane as if it
was sitting on the floor. http://forums.ubi.com/images/smilies/winky.gif

For every action there is an equal and oppisite reaction! http://forums.ubi.com/images/smilies/35.gif

Stand under a rotary wing "helicopter" and tell
me that air is not beimg pumped down. http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

Stand behind a rotary wing "propeller" and tell
me that air is not beimg pumped backwards!
http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

It's all about Coanda.

Sergio