Flying...
Eric Rannaud
I recently listened to an iterview of David Anderson on NPR
(http://www.npr.org/features/feature.php?wfId=3875411). He gave
interesting insights about the physics of flying. Interesting to
discover that Bernouilli is not really the guy to invoke when
explaining why a plane can fly... I've heard college physics
professors saying that. It's amazing.
Anyway, the main explanation given by Anderson (see his book with
Scott Eberhardt, "Understanding Flight") is based on the viscosity of
the air. I won't give any more details, I don't really know what I'm
talking about here.
He dismissed the classical explanations (the Bernouilli principle and
the so-called ground effect), revealing that the FAA still teachs them
as a requirement to obtain a pilot license!
But some surfing led me to this page:
http://www.av8n.com/how/
And it seems that John Denker does not really agree with Anderson and
Eberhardt, saying that the Coanda effect is not producing the lift on
a standard wing (see http://www.monmouth.com/~jsd/fly/lift.htm). But
his explanations are of the "Look, it's obvious" kind... I am not
really convinced.
Where could I find a clear and definite account (and possibly highly
technical) on flight principles and in particular on lift?
Thanks.
John T Lowry
"Eric Rannaud" <eric.r...@myrealbox.com> wrote in message
news:265ad44e.04090...@posting.google.com...
IMO Coanda and viscosity are minor add-on effects. You can explain lift
through either Bernoulli or more elaborate "panel" methods, but the
alternatives are somewhat like explaining the physics of a body sliding
down a ramp through either (1) force, or (2) energy considerations.
You might try Prandtl and Tietjens's Applied Hydro- and Aeromechanics or
Jack Moran's An Introduction to Theoretical and Computational
Aerodynamics. They're both Dover books.
John T. Lowry, PhD
Flight Physics
eric.r...@myrealbox.com
lift
> through either Bernoulli or more elaborate "panel" methods, but the
> alternatives are somewhat like explaining the physics of a body
sliding
> down a ramp through either (1) force, or (2) energy considerations.
Well, Anderson and Eberhardt reportedly made the following computation:
on a Cessna wing, using only Bernouilli and the "rule" that the
airflows above and below the wing must meet at the same time after the
wing, forcing the airflow above to go faster and lowering the pressure
there, a speed of 400mph would be necessary to take off with such a
plane... instead of 70-80mph in reality.
Note that they also say that the "rule" about the speed of the airflows
is flawed. Doing this computation they say: even if this rule was
correct, this wouldn't provide enough lift. Moreover, they say that
this wouldn't take into account the fact that a flat wing can provide
lift.
Do you have any comment on that?
Thank you for the references, though. They seem to be the kind of books
I was looking for.
BllFs6
>on a Cessna wing, using only Bernouilli and the "rule" that the
>airflows above and below the wing must meet at the same time after the
>wing, forcing the airflow above to go faster and lowering the pressure
>there, a speed of 400mph would be necessary to take off with such a
>plane... instead of 70-80mph in reality.
>
>Note that they also say that the "rule" about the speed of the airflows
>is flawed. Doing this computation they say: even if this rule was
>correct, this wouldn't provide enough lift. Moreover, they say that
>this wouldn't take into account the fact that a flat wing can provide
>lift.
>
>Do you have any comment on that?
>
>Thank you for the references, though. They seem to be the kind of books
>I was looking for.
>
How about doing the calculation that assumes that
one, the wing is flat plate
two, the air is made up of large non-viscous billard balls
three, a wing gets lift by moving through the billard ball air at speed and AND
at an angle....and the lift comes from the wing
"knocking" large numbers/masses of air downwards
And we can ignore what happens ABOVE the wing....
Ive never done that calc....but it wouldnt surprise me if it gave a good
number....
Because when push comes to shove (no matter the details of wing operation) a
plane gets if lift from "throwing" downwards large masses of air....
Fancy wing shapes are just to do that process in the most efficient way
possible give a certain set of mechanical and operational considerations.....
take care
Blll
eric.r...@myrealbox.com
>
> two, the air is made up of large non-viscous billard balls
>
> three, a wing gets lift by moving through the billard ball
> air at speed and
> AND at an angle....and the lift comes from the wing
> "knocking" large numbers/masses of air downwards
Well, I didn't do this calculation either, but I've read that if this
effect does exist most of the time, it was too small.
> Because when push comes to shove (no matter the details of wing
> operation) a plane gets if lift from "throwing" downwards large
masses of
> air....
Anderson and Eberhardt say that these large amounts of air thrown
downwards are a consequence of the Coanda effect.
Actually, this comment gives a clear account of the situation, weighing
the different points of view:
http://www.aa.washington.edu/faculty/eberhardt/Discover%20Magazine_files/featphysics.html
You can also find an article by Anderson and Eberhardt:
http://www.aa.washington.edu/faculty/eberhardt/lift.htm
Anyway, it seems that all the explanations usually given are partially
correct. But from what I have seen so far (arguably only in popular
accounts), no one really agrees on which is the most significant, or if
the sum of all these effects is enough to produce enough lift.
At least, the "principle of equal transit times" is bullshit.
I definitely need to read the previously mentioned technical books.
Jim
Airplanes, like everything else in the universe, follow the path of
least resistance. :)
Jim
Jan-Olov Newborg
John S. Denker use the mathematical "Potential/Circulation Theory" as
Physical Lift explanation!
But once asked what causes the Circulation, Dr. John S. Denker could
not answer.
Angle of Attack (AOA) of the wing causes the Lift of the Wing that
keeps the aircraft flying. The wing at an AOA causes "downwash" and
vertical acceleration of inert air masses and Newton´s 3 law explains
by "action and reaction" what happens.
There are many good websites today explaining the Lift process
physically corret!
Jan-Olov Newborg
der.Todesvogel
he thinks lift is the product on air being viscously "pulled"
(accelerated) down across the top of the wing. That is totally
impossible because air does not support tension forces (just as a
length of yarn does not support compression, or "push").
Air fills the space above a moving wing surface because air pressure
existing far above the wing maintains, to the extent possible, static
air pressure exerted on the top surface of the wing. That is, air
pressure surrounding the wing prevents a "vacuum" (a volume of space
that is airless) from forming above the wing.
Air mass/inertia prevents the instantaneous space-filling, with air,
above and behind the moving wing. If air had no inertia, wings would
produce no lift because air would jump in and out of the wing's way
instantaneously, and with no exerted force.
The True Simple Explanation of Flight: Wings experience a differential
between the upper and lower surface (total) air pressures due to air
inertia (air can't immediately get out of the way of the wing's bottom
surface, and the air can't immediately fill the space above the wing,
as the wing passes through the stationary air).
Todesvogel
eric.r...@myrealbox.com (Eric Rannaud) wrote in message news:<265ad44e.04090...@posting.google.com>...
Edward Green
Oh joy! Another thread about aerodynamic lift!
> I recently listened to an iterview of David Anderson on NPR
> (http://www.npr.org/features/feature.php?wfId=3875411). He gave
> interesting insights about the physics of flying. Interesting to
> discover that Bernouilli is not really the guy to invoke when
> explaining why a plane can fly... I've heard college physics
> professors saying that. It's amazing.
Don't worry. No matter what anybody says in a popular tidbit, it will
be wrong.
> Anyway, the main explanation given by Anderson (see his book with
> Scott Eberhardt, "Understanding Flight") is based on the viscosity of
> the air.
There is no one-line catch phrase which is going to explain much.
Saying "it's not really Bernoulli, it's the viscosity of the air" is
merely substituting one piece of non-knowledge for another.
> I won't give any more details, I don't really know what I'm
> talking about here.
You are elevated to the pantheon of the gods for your restraint. :-)
> He dismissed the classical explanations (the Bernouilli principle and
> the so-called ground effect),
This is no more the "classical explanation" than some often repeated
story about eohippus is the "classical model" of evolution. Nobody
who has thought seriously about it thinks that "the Bernoulli
principle" is more than a throw away oolie. (And the "ground effect",
while real, just names the increase in lift noted when the wing is
close to the ground).
> revealing that the FAA still teachs them
> as a requirement to obtain a pilot license!
Big deal. About as relevant to piloting skills as some of the silly
questions one has to answer to obtain a driver's license.
> But some surfing led me to this page:
> http://www.av8n.com/how/
> And it seems that John Denker does not really agree with Anderson and
> Eberhardt, saying that the Coanda effect is not producing the lift on
> a standard wing (see http://www.monmouth.com/~jsd/fly/lift.htm). But
> his explanations are of the "Look, it's obvious" kind... I am not
> really convinced.
>
> Where could I find a clear and definite account (and possibly highly
> technical) on flight principles and in particular on lift?
The plane exerts a force F on the air. By Newton's third law, the air
exerts a force -F on the plane. The component of -F directed in the
vertical direction when the plane is in approximately level flight is
called "lift". Through creative flow management, which in passing may
invoke concepts like "Bernoulli" or "Coanda", but in the end comes to
the Navier-Stokes equations, it is sometimes possible to create a
given F_vertical more efficiently than by batting the air down with a
slat.
Edward Green
> >Well, Anderson and Eberhardt reportedly made the following computation:
> >on a Cessna wing, using only Bernouilli and the "rule" that the
> >airflows above and below the wing must meet at the same time after the
> >wing, forcing the airflow above to go faster and lowering the pressure
> >there, a speed of 400mph would be necessary to take off with such a
> >plane... instead of 70-80mph in reality.
> >
> >Note that they also say that the "rule" about the speed of the airflows
> >is flawed. Doing this computation they say: even if this rule was
> >correct, this wouldn't provide enough lift. Moreover, they say that
> >this wouldn't take into account the fact that a flat wing can provide
> >lift.
> >
> >Do you have any comment on that?
> >
> >Thank you for the references, though. They seem to be the kind of books
> >I was looking for.
> >
>
> How about doing the calculation that assumes that
>
> one, the wing is flat plate
>
> two, the air is made up of large non-viscous billard balls
>
> three, a wing gets lift by moving through the billard ball air at speed and AND
> at an angle....and the lift comes from the wing
> "knocking" large numbers/masses of air downwards
>
> And we can ignore what happens ABOVE the wing....
>
> Ive never done that calc....but it wouldnt surprise me if it gave a good
> number....
For an efficient wing, I believe this would ignore a significant
fraction of the lift. You are effectively looking at the pressure
increase below the wing, but ignoring the pressure decrease above it.
It is the mechanism of the pressure decrease above the wing, invoking
more or less subtle concepts of fluid flow, that gives the air of
mystery to the thing and the oolie-ish potential.
Among the many sources of confusion here, "lift" is used in two
different senses. First it is used merely to label a coordinate
direction: lift is the vertical component of air force in level flight
(1) , drag the horizontal -- or perhaps taken relative to the
direction of travel of the airfoil (1a). But then it is also used to
label the _mechanism_ of the vertical force (2) -- part of which, as
you note, is easy to understand, part of it a little more complicated.
> Because when push comes to shove (no matter the details of wing operation) a
> plane gets if lift from "throwing" downwards large masses of air....
That is correct. However, part of the "throwing downward" is
accomplished by pulling the chair out from under the air above the
wing, which comes crashing down ... just after the wing has departed.
:-)
> Fancy wing shapes are just to do that process in the most efficient way
> possible give a certain set of mechanical and operational considerations.....
Yes.
Edward Green
> > one, the wing is flat plate
> >
> > two, the air is made up of large non-viscous billard balls
> >
> > three, a wing gets lift by moving through the billard ball
> > air at speed and
> > AND at an angle....and the lift comes from the wing
> > "knocking" large numbers/masses of air downwards
>
> Well, I didn't do this calculation either, but I've read that if this
> effect does exist most of the time, it was too small.
Right. As I mentioned to Bill, that's half the story for an efficient
wing (possibly even for a slat).
> > Because when push comes to shove (no matter the details of wing
> > operation) a plane gets if lift from "throwing" downwards large
> masses of
> > air....
>
> Anderson and Eberhardt say that these large amounts of air thrown
> downwards are a consequence of the Coanda effect.
By which they simply substitute another buzz word for "Bernoulli".
"The Coanda effect" is not a fundamental law of physics, is it? It's
simply an observation of some regularity in certain fluid flows.
I believe the story is this: given that we are dragging a shape
through the air (our wing), the lift (component of force perpendicular
to the direction of dragging) undergoes a jump when the flow regime is
such that "Coanda effect" are words of power. The sudden loss of the
Coanda effect is called "stall".
It's like noticing an empirical effect in crowd behavior, labeling it
(the "Rannaud effect"), and then making this label an explanation for
whatever happens when the Rannaud effect is observed -- a syndrome is
confused with etiology. First we conflate "lift", a general term for
a force couple in a direction perpendicular to aerodynamic transit
with the detailed fluid mechanical account of such force, and then we
notice that in a particular flow regime there is a jump in this force
for a given speed compared to outside this flow regime, then we notice
that this jump is accompanied by a characteristic qualitative
variation in the flow ("it stays glued to the wing") which we label
("Coanda effect") and, in a final logical coups de grace (that is, the
death of logic) make this label an "explanation" of the force in
general, while it is merely a label of an accompanying effect in
certain flow regimes.
Need I say more?
> Actually, this comment gives a clear account of the situation, weighing
> the different points of view:
> http://www.aa.washington.edu/faculty/eberhardt/Discover%20Magazine_files/featphysics.html
>
> You can also find an article by Anderson and Eberhardt:
> http://www.aa.washington.edu/faculty/eberhardt/lift.htm
>
> Anyway, it seems that all the explanations usually given are partially
> correct. But from what I have seen so far (arguably only in popular
> accounts), no one really agrees on which is the most significant, or if
> the sum of all these effects is enough to produce enough lift.
That's nonsense.
What you can perhaps deduce from the lack of a clear popular picture
is that the experts have taken a "shut up and calculate" approach.
Popular nonsense rushes into this vacuum. In this sense popular
nonsense, conveying no information directly, serves as an indicator of
professional silence.
John Denker
> Hello,
>
> I recently listened to an iterview of David Anderson on NPR
> (http://www.npr.org/features/feature.php?wfId=3875411).
Yeah. That interview should have been called "Anderson's
Fairy Tales."
> He gave interesting insights about the physics of flying. Interesting
> to discover that Bernouilli is not really the guy to invoke when
> explaining why a plane can fly... I've heard college physics
> professors saying that. It's amazing.
It would be amazing, if true. It turns out that Prof. Bernoulli
was a pretty smart guy. Bernoulli's theorem is an actual
theorem. That is, it is a provable corollary of the laws of
motion, subject to certain mild provisos. It is used literally
every day, correctly, by aircraft designers if they want to
calculate pressure given the velocity. It is not the only way
of approaching the problem, but it is often convenient.
It should go without saying that Bernoulli's principle can be
misapplied. But that means nothing. Students misapply Newton's
laws every day, but that doesn't mean F=ma is invalid.
> Anyway, the main explanation given by Anderson (see his book with
> Scott Eberhardt, "Understanding Flight") is based on the viscosity of
> the air. I won't give any more details, I don't really know what I'm
> talking about here.
Neither does Anderson. Viscosity, while not entirely irrelevant,
is verrry far from being a proper basis for explaining how a
plane flies. The usual test of an alleged cause-and-effect
relationship is to change the cause and see what happens to the
effect. Well, if you change the viscosity by an order of magnitude
in either direction (other things like density being equal) the
lift will not change in proportion. You'd be doing well to
perceive any change at all.
> He dismissed the classical explanations (the Bernouilli principle and
> the so-called ground effect), revealing that the FAA still teachs
> them as a requirement to obtain a pilot license!
The Bernoulli effect is 100% real. Ground effect is 100% real,
and even more relevant to pilots. Pilots *should* be taught
this stuff. (There are a few things the FAA gets wrong, but
these two items are not on the list.)
> But some surfing led me to this page: http://www.av8n.com/how/ And it
> seems that John Denker does not really agree with Anderson and
> Eberhardt, saying that the Coanda effect is not producing the lift on
> a standard wing (see http://www.monmouth.com/~jsd/fly/lift.htm). But
> his explanations are of the "Look, it's obvious" kind... I am not
> really convinced.
We have an 1800-word document:
http://www.av8n.com/fly/lift.htm
and a 12,000-word document:
http://www.av8n.com/how/htm/airfoils.html
It is going a bit far to summarize them in three words ("Look,
it's obvious"). If you read more carefully, you find a bit
more meat on the bones.
> Where could I find a clear and definite account (and possibly highly
> technical) on flight principles and in particular on lift?
How highly technical do you want? Calculus? Conformal
transformations? CFD?
You said you found
http://www.av8n.com/how/
Presumably that means you found the annotated bibliography
http://www.av8n.com/how/htm/bibl.html
Are the references cited there not good enough?
> Anderson and Eberhardt reportedly made the following computation: on
> a Cessna wing, using only Bernouilli and the "rule" that the airflows
> above and below the wing must meet at the same time after the
> wing,....
The Bernoulli part isn't the problem. The equal-time "rule" is
the problem. It's completely bogus, as discussed in detail at
http://www.av8n.com/how/htm/airfoils.html#fig-delay
> Anderson and Eberhardt say that these large amounts of air thrown
> downwards are a consequence of the Coanda effect.
That's yet another of their fairy tales.
Using the Coanda effect to explain the operation of a normal
wing makes about as much sense as using bowling to explain
walking. To be sure, bowling and walking use some of the same
muscle groups, and both at some level depend on Newton’s laws,
but you can walk just fine even if you haven't got a bowling
ball. For details, see
http://www.av8n.com/how/htm/spins.html#sec-coanda
> (arguably only in popular accounts), no one really agrees on which is
> the most significant, or if the sum of all these effects is enough
> to produce enough lift.
In the serious literature, there is 100% consensus on a
useful and highly-accurate description of how lift is
produced. The essential ideas have been around, unchanged,
for many decades. For instance, the beautiful result now
known as the Kutta-Zhukovsky theorem was first published in
1902.
You can easily use the Kutta-Zhukovsky theorem to prove that
what Anderson said about ground effect is completely false.
For details, see
http://www.av8n.com/how/htm/airfoils.html#sec-k-z
Jim <lose...@workfromhome.com> wrote:
>> Airplanes, like everything else in the universe, follow the path of
>> least resistance. :)
No, they don't. Consider a glider following a 20:1 glide
slope. I retrim it to follow a 30:1 slope. Then I trim
it to follow a 10:1 slope. Which of these is the path of
least resistance?
Edward Green <spamsp...@netzero.com> wrote:
>>> There is no one-line catch phrase which is going to explain much.
>>> Saying "it's not really Bernoulli, it's the viscosity of the air" is
>>> merely substituting one piece of non-knowledge for another.
Yes. Exactly so.
>>> I believe the story is this: given that we are dragging a shape
>>> through the air (our wing), the lift (component of force
>>> perpendicular to the direction of dragging) undergoes a jump when
>>> the flow regime is such that "Coanda effect" are words of power.
>>> The sudden loss of the Coanda effect is called "stall".
That's not the correct story. There is such a thing as the
Coanda effect, under certain very special circumstances, but
these circumstances are never met for a normal wing in
normal flight. Stalling also is not related to Coanda (and
is usually not sudden).
>>> (And the "ground effect", while real, just names the increase in
>>> lift noted when the wing is close to the ground).
That's not right either. Ground effect has more to do with
a decrease in drag than an increase in lift. For details, see
http://www.av8n.com/how/htm/airfoils.html#sec-circulation-vortices
especially
http://www.av8n.com/how/htm/airfoils.html#sec-soft-field-why
>>> About as relevant to piloting skills as some of the silly
>>> questions one has to answer to obtain a driver's license.
A proper understanding of ground effect is actually quite
relevant for things like takeoff and landing, especially
soft-field takeoff.
der Todesvogel <spamjun...@home.com> wrote:
> he [Anderson] thinks lift is the product on air being viscously
> "pulled" (accelerated) down across the top of the wing. That is
> totally impossible
We agree what he said is impossible. There are several reasons
why it's impossible.
> because air does not support tension forces (just
> as a length of yarn does not support compression, or "push").
That's not the best argument. That depends on points of view,
and depends on taste. If you measure _absolute_ pressure,
there is only push. If you measure _gauge_ pressure, there is
both push and pull. De gustibus non disputandum.
> The True Simple Explanation of Flight: Wings experience a differential
> between the upper and lower surface (total) air pressures due to air
> inertia (air can't immediately get out of the way of the wing's
> bottom surface, and the air can't immediately fill the space above
> the wing, as the wing passes through the stationary air).
Einstein said any theory should be as simple as possible, but
no simpler. Alas that "True Simple Explanation" is too simple.
Pilots need an explanation that consistently explains all the
stuff they have to deal with. The get-out-of-the way theory
utterly fails to explain the operation of stall warning
indicators ... which are among the first things student pilots
ask about. For details, see
http://www.av8n.com/how/htm/airfoils.html#sec-stall-wrng
der.Todesvogel
information on the relationship between maneuvering speed, G loading,
and vehicle mass. Tv
Edward Green
field, wrote:
> Edward Green <spamsp...@netzero.com> wrote:
>
> >>> There is no one-line catch phrase which is going to explain much.
> >>> Saying "it's not really Bernoulli, it's the viscosity of the air" is
> >>> merely substituting one piece of non-knowledge for another.
>
> Yes. Exactly so.
Thanks! I'm glad I troubled to read down far enough in your combined
reply to see I merited a few lines of personal rejoinder.
> >>> I believe the story is this: given that we are dragging a shape
> >>> through the air (our wing), the lift (component of force
> >>> perpendicular to the direction of dragging) undergoes a jump when
> >>> the flow regime is such that "Coanda effect" are words of power.
> >>> The sudden loss of the Coanda effect is called "stall".
>
> That's not the correct story. There is such a thing as the
> Coanda effect, under certain very special circumstances, but
> these circumstances are never met for a normal wing in
> normal flight. Stalling also is not related to Coanda (and
> is usually not sudden).
Sorry to spread misapprehension then. I thought, in general terms,
that the "Coanda effect" amounted to the emprical observation that
airflows sometimes seem to stay glued to convex surfaces, the way a
water jet might?
My understanding of stall was that, in the pre-stall regime, laminar
flow prevails on both wing surfaces and, post stall, laminar flow has
come unglued from the upper wing surface, leaving an intervening
staganant or turbulent region, accompanied by a loss of lift. This
suggests that pre-stall we have a species of Coanda effect --
streamlines in one case parallelling the surface closely, in the other
breaking away. Perhaps the taxonomic distinction is on "thin jet",
which is lacking here?
You say:
"The name Coanda effect is generally applied to any situation where a
thin, high-speed jet of fluid meets a solid surface and follows the
surface around a curve."
> >>> (And the "ground effect", while real, just names the increase in
> >>> lift noted when the wing is close to the ground).
>
> That's not right either. Ground effect has more to do with
> a decrease in drag than an increase in lift. For details, see
> http://www.av8n.com/how/htm/airfoils.html#sec-circulation-vortices
> especially
> http://www.av8n.com/how/htm/airfoils.html#sec-soft-field-why
Hmm... It is presumptuous of me to question your account, but this is
Usenet, where presumption is king, so here goes. In the section above
the soft-field procedure, you say:
"Here are some more benefits of understanding circulation and
vortices: it explains induced drag, and explains why gliders have long
skinny wings. Induced drag is commonly said to be the ?cost? of
producing lift. But there is no law of physics that requires a
definite cost. If you could take a very large amount of air and pull
it downward very gently, you could support your weight at very little
cost".
I find it hard to unreservedly accept this.
The observation that we can buy a given amount of dynamic downward
momentum flow (lift from downward movement of air) at a cost which
goes down with the velocity of the downward moving air (because of the
dependence of KE on v^2 vs. momentum on v), is a common one. But it
doesn't follow that because this cost is variable it is therefore
negligible (true or not as it may be, it certainly doesn't follow).
You go on to say:
"The cost you absolutely must pay is the cost of making that trailing
vortex. For every mile that the airplane flies, each wingtip makes
another mile of vortex. The circulatory motion in that vortex involves
nontrivial amounts of kinetic energy, and that?s why you have induced
drag."
Ok. You adduce another source of drag, but as to why the work done on
the air to set it into wingtip vortices dominates that done on the air
to given it gross downward momentum, this is left to pure assertion
(at least in the limited context of this paragraph)?
(Am I correct in my impression that you are asserting some necessary
relation between circulation around the long axis of the wing, and the
circulatory in the trailing wingtip vorticies, are am I reading this
in wrongly? Since the circulation are about different axes, the
connection would not be obvious.)
Hmm... In fact, I could see how the kinetic energy wrapped up in the
wingtip vorticies might be the same kinetic energy imparted to the air
in the service of the prior sense of induced drag -- that accruing
directly from moving the air downward. The air moved down by the
passing wing is only moved down over the wingspan. So we have a column
of downward moving air directly adjacent to still air. Angular
momentum about an axis in the flow interface and perpendicular to the
flows (i.e., parallel to one extending aft from the wingtip) is
already implied in this situation. Add a soupcon of viscosity, and
this angular momentum manifests itself in actual vortices.
It appears that the kinetic energy wrapped up in the trailing wingtip
vorticies may be derived from the kinetic energy injected into the
downward moving air after all, the previous explanation of "induced
drag" -- it's not one or the other, but different manifestation of the
same effect.
> >>> About as relevant to piloting skills as some of the silly
> >>> questions one has to answer to obtain a driver's license.
>
> A proper understanding of ground effect is actually quite
> relevant for things like takeoff and landing, especially
> soft-field takeoff.
I was reacting to the mention of "Bernouilli principle" rather than
"ground effect", and I think I imagined that the question included the
discredited "equal times" argument. But no matter.
Touching on the ground effect, I had a rather different take on it.
You write:
"When the aircraft is in ground effect, it ?sees its reflection? in
the ground. If you are flying 10 feet above the ground, the effect is
the same as having a mirror-image aircraft flying 10 feet below the
ground. Its wingtip vortices spin in the opposite direction and
largely cancel your wingtip vortices --- greatly reducing induced
drag."
Or, the air thrown downward by the wing is partially confined between
the wing and the ground, increasing pressure and reducing the velocity
of subsequent air thrown down to maintain this high pressure pocket,
reducing induced drag. This might be another way of saying the same
thing. (I always thought of ground effect as a species of the
hovercraft effect -- a very leaky but still not totally inefficient
hovercraft in the case of a wing).
By the way, I noticed your very proper insistance that, in
discrediting some simple old textbook explanation of lift we do not
throw away the Bernoulli with the bathwater: of course this remains a
real effect, however often it is misapplied or misunderstood. (The
derivation of the relation however makes no reference to potential
energy, which idea I think you mention somewhere in your very nice
site (showing us that false-oolies are everywhere :-)).
Very Respectfully,
Ed Green
P.S.
> Einstein said any theory should be as simple as possible, but
> no simpler.
If you have a reference for this quote, I'd like to see it, but I
believe it's apocryphal! Of course it's an excellent precept no
matter who said it.
John Denker
>
> I thought, in general terms,
> that the "Coanda effect" amounted to the emprical observation that
> airflows sometimes seem to stay glued to convex surfaces, the way a
> water jet might?
Hmmmm. This is a semi-quagmire. There are very few people
who could pass for experts on the Coanda effect ... and even
they don't 1000% agree on how to define it. But there is
full agreement on what it is *not*. It's safe to say that
the basic notion that you can have non-separated flow
predated Mr. Coanda by at least 100 years. Also, a highly
accurate understanding of how lift is produced predated
Coanda by many decades.
Coanda filed a number of patents, most of which involved
curvature-enhanced turbulence (mostly for the purpose of
mixing things), and AFAIK all of which involved a very
high-speed jet. The physics is quite tricky, and depends
on having a velocity that *increases* as you get farther
from the curved surface ... a condition that is impossible
for anything resembling a normal wing in normal flight.
This peculiar velocity distribution causes an instability,
the same way as a fluid heated from *below* undergoes a
Rayleigh instability. It is this instability that I
associate with the term Coanda effect. Among people who
know what they're talking about, this is the most common
meaning of the term, and the other meanings are at least
similar.
If all that seems arcane ... don't blame me. I didn't
bring it up. Yes, it is arcane. You can spend a nice
lifetime designing airplanes without ever needing to
know anything about the Coanda effect.
To repeat, if all you mean is attached flow, call it
attached flow and leave Coanda out of it.
> My understanding of stall was that, in the pre-stall regime, laminar
> flow prevails on both wing surfaces and, post stall, laminar flow has
> come unglued from the upper wing surface, leaving an intervening
> staganant or turbulent region, accompanied by a loss of lift.
There are several misconceptions all mixed together there.
Laminar is not the same as attached or vice versa. Turbulent
is not the same as detached or vice versa. For details, see
http://www.av8n.com/how/htm/spins.html#sec-boundary-layer
Also separation is not quite synonymous with stalling, just
as water is not quite synonymous with drowning. See
http://www.av8n.com/how/htm/spins.html#sec-bl-control
> This
> suggests that pre-stall we have a species of Coanda effect --
> streamlines in one case parallelling the surface closely, in the other
> breaking away.
You're working waaay too hard.
1) If you leave Coanda out of it, and just call it flow
pattern and another flow pattern, it's less pretentious
and much more nearly correct.
2) It's not even a different "species". A wing one degree
above stalling angle-of-attack produces almost exactly
the same amount of lift as it would one degree below
stalling angle-of-attack. The flow pattern is mostly the
same and the principles involved are the same.
> Perhaps the taxonomic distinction is on "thin jet",
> which is lacking here?
In my book, the essential property that defines the Coanda
effect is the Rayleigh-like instability associated with
a 'backwards' velocity-versus-radius profile.
> The observation that we can buy a given amount of dynamic downward
> momentum flow (lift from downward movement of air) at a cost which
> goes down with the velocity of the downward moving air (because of the
> dependence of KE on v^2 vs. momentum on v), is a common one. But it
> doesn't follow that because this cost is variable it is therefore
> negligible (true or not as it may be, it certainly doesn't follow).
I didn't say it was negligible. I just said that there's
no nontrivial lower bound; the lower bound is zero. How
closely you can actually approach zero depends on all sorts
of engineering details. Among other things, I point out
it's hard to make a wing that is long, thin, and strong.
> "The cost you absolutely must pay is the cost of making that trailing
> vortex. For every mile that the airplane flies, each wingtip makes
> another mile of vortex. The circulatory motion in that vortex involves
> nontrivial amounts of kinetic energy, and that?s why you have induced
> drag."
>
> Ok. You adduce another source of drag, but as to why the work done on
> the air to set it into wingtip vortices dominates that done on the air
> to given it gross downward momentum, this is left to pure assertion
> (at least in the limited context of this paragraph)?
Is there a question there?
Note that the *entire* motion of the air can be attributed
to a set of vortex lines. Vortex lines form a basis for
describing virtually any meaningful solutions to the
equation of fluid motion. So in some sense you can't speak
of the vortex-associated motion as being "additional" to any
other motion. For a whole lot more detail on this, see
http://www.av8n.com/fly/vortex.htm
> (Am I correct in my impression that you are asserting some necessary
> relation between circulation around the long axis of the wing, and the
> circulatory in the trailing wingtip vorticies, are am I reading this
> in wrongly? Since the circulation are about different axes, the
> connection would not be obvious.)
Yes, there is a "relation". An exact relation. There is a
law that says vortex lines cannot end. They're like magnetic
field lines. This is not merely a law of physics ... it's a
mathematical identity. The vortex lines that are bound to
the wing spill off near the wingtips and become the trailing
vortex. This is mentioned at various places including
http://www.av8n.com/fly/vortex.htm#sec-endless
> Hmm... In fact, I could see how the kinetic energy wrapped up in the
> wingtip vorticies might be the same kinetic energy imparted to the air
> in the service of the prior sense of induced drag -- that accruing
> directly from moving the air downward.
Yes, it's all the same air. It's all the same energy.
The vortex lines are a basis for describing the entire
motion.
> The air moved down by the
> passing wing is only moved down over the wingspan. So we have a column
> of downward moving air directly adjacent to still air. Angular
> momentum about an axis in the flow interface and perpendicular to the
> flows (i.e., parallel to one extending aft from the wingtip) is
> already implied in this situation. Add a soupcon of viscosity, and
> this angular momentum manifests itself in actual vortices.
Yes. To say the same thing, the trailing vortex is right
at the boundary (you might say it _is_ the boundary) between
the air that was hauled down by the wing and the air that
wasn't.
> "When the aircraft is in ground effect, it ?sees its reflection? in
> the ground. If you are flying 10 feet above the ground, the effect is
> the same as having a mirror-image aircraft flying 10 feet below the
> ground. Its wingtip vortices spin in the opposite direction and
> largely cancel your wingtip vortices --- greatly reducing induced
> drag."
>
> Or, the air thrown downward by the wing is partially confined between
> the wing and the ground, increasing pressure and reducing the velocity
> of subsequent air thrown down to maintain this high pressure pocket,
> reducing induced drag. This might be another way of saying the same
> thing.
The way I read that, the notion of increased pressure sounds
more like changing the lift than changing the drag.
Sure, there are ground-related lift effects, but they are
usually not very noticeable (except in some exceptionally
badly-designed airplanes). The dominant effect, which is
guaranteed to happen in all airplanes, is the reduction
in drag. So I prefer to focus on that. And that is most
simply explained in terms of vortex-cancellation.
Economic Girly Man
John Denker wrote:
<major snippage>
>
>> "The cost you absolutely must pay is the cost of making that trailing
>> vortex. For every mile that the airplane flies, each wingtip makes
>> another mile of vortex. The circulatory motion in that vortex involves
>> nontrivial amounts of kinetic energy, and that?s why you have induced
>> drag."
No, not correct. The trailing vortex system contains ZERO energy (relative to
the undisturbed air).
Remember that Bernoulli's equation is simply a statement of Conservation of
Energy. The increase in Kinetic Energy due to the "circulatory motion" of the
vortex is exactly equal to the decrease in Potential Energy, or pressure
(p0=p1+.5rhoV^2)
EGM
John Denker
> The trailing vortex system contains ZERO energy
> (relative to the undisturbed air).
That's a new one. Call that assertion [1].
> Remember that Bernoulli's equation is simply a statement of Conservation
> of Energy. The increase in Kinetic Energy due to the "circulatory
> motion" of the vortex is exactly equal to the decrease in Potential
> Energy, or pressure (p0=p1+.5rhoV^2)
That's not right.
The law of conservation of energy doesn't say that each thing
you point to has the same energy as every other thing you point
to. Are we supposed to believe a baseball has the same energy
before and after it gets hit by the bat?
If you wish to consider Bernoulli's theorem as a consequence
of conservation of energy, you need to first prove that the air
parcel is not exchanging energy with its surroundings before
you apply the theorem. You must not *assume* constant energy
(to make the theorem applicable) and then turn around and
apply the theorem to "prove" the energy is constant.
BTW there are other ways of deriving Bernoulli's theorem,
without assuming in advance constancy of energy. See e.g.
Feynman volume II chapter 40, especially equation 40.13.
However that derivation is premised on other requirements --
notably steady flow -- that do not obtain in this situation.
For every mile of flight, each wingtip makes a mile of *new*
trailing vortex, so the situation is never truly steady.
In any case, the argument quoted above is invalid.
============
This leaves us with the question of whether assertion [1]
might be true for other reasons. After all, if somebody
gives you an invalid argument in support of the proposition
that 2+2=4, the proposition might be correct anyway, for
other reasons.
In this case, however, it is a trivial exercise to prove
that assertion [1] is *not* true.
Start with a single long straight vortex line all by itself
in otherwise-still air. Prove that its total energy (KE + PE)
is in fact zero. That is, write down the 1/r velocity pattern
you expect a vortex to have, then calculate the pressure
required to be consistent with that pattern. This calculation
must be done using balance-of-forces, *not* using Bernoulli,
because we cannot use Bernoulli until *after* we have proved
the parcels have unchanging energy ... which we have not yet
proved. Also we cannot use Feynman equation 40.13, because
we have no way of knowing _a priori_ that different
streamlines have the same total energy. Thirdly we cannot
apply Feynman equation 40.14 because that only applies to
irrotational flow. So forget about Bernoulli and calculate
the pressure from the streamline curvature.
Next, (*not* using Bernoulli) calculate the KE density in
the obvious way. Finally observe that the pressure term
and the KE term are indeed equal-and-opposite. So yes, the
parcels have all the same energy.
That's the hard part. The math takes maybe half a sheet of
paper.
Using that as a foundation, you can immediately see that the
constant-energy claim cannot be valid for (say) the starboard
trailing vortex of a real airplane, since it sits in the
flow-field induced by the port trailing vortex.
Consider two points near the starboard vortex. For simplicity
take them to be along the line that runs spanwise toward the
other vortex. We have two contributions to the local velocity
field, namely one from the starboard vortex (call it v) and
one from the port vortex (call it c). The v contribution will
have lots of curvature, but if our two points are reasonably
near each other the curvature of the c contribution will be
negligible in comparison ... so we take c to be more-or-less
constant, i.e. independent of position in this neighborhood.
Therefore it contributes nothing to the pressure profile.
After all, pressure depends on streamline curvature. So
pressure(v+c) = pressure(v).
Mutter something about Galileo's principle of relativity
if you want ... shifting the velocity by a constant doesn't
change forces or pressures.
However ... there is no way that KE(v+c) will be equal to
KE(v). KE is a nonlinear function of velocity.
Summary: we have proved that although a single isolated
vortex-line involves no change in energy, a pair of vortex
lines does involve a change in energy.
When I fly an airplane, it creates a pair of trailing
vortices. It imparts energy to the air as it flies along.
A slightly different way of looking at this, with somewhat
more detail, and leading to the same conclusion, can be found
in the reference I cited in my previous notes:
http://www.av8n.com/fly/vortex.htm
Jan-Olov Newborg
> Eric Rannaud wrote:
> > Hello,
> >
> > I recently listened to an iterview of David Anderson on NPR
> > (http://www.npr.org/features/feature.php?wfId=3875411).
>
> Yeah. That interview should have been called "Anderson's
> Fairy Tales."
>
> > He gave interesting insights about the physics of flying. Interesting
> > to discover that Bernouilli is not really the guy to invoke when
> > explaining why a plane can fly... I've heard college physics
> > professors saying that. It's amazing.
>
> It would be amazing, if true. It turns out that Prof. Bernoulli
> was a pretty smart guy.
Professor em of Aerodynamics at Maryland University, John D. Anderson,
writes about the family Bernoulli and their understanding of pressure
(in History of Aerodynamics):
"None of family Bernoulli, father John or son Daniel, set up any
equation or theorem", they did not even understod "pressure property
of point"!
There is no "Bernoulli Effect" in real physics world. It´s always
mistaken for "Venturi effect"!
"Venturi effect" is due to "The Coanda effect" in the divergent part
of the pipe.
The telltale "High velocity causes low pressure" should mean that "the
speed of your car causes the power of the engine"!
That´s not the fact.
Mathematically one can set up reverse equations and reverse reactions,
but not in real life.
Forget every thought that a wing is a spilt venturi pipe, because
that´s totally false, despite all mathematical equations.
Can you explain LaPlace effect in Lift, John?
> The Bernoulli part isn't the problem. The equal-time "rule" is
> the problem. It's completely bogus, as discussed in detail at
> http://www.av8n.com/how/htm/airfoils.html#fig-delay
It´s not only "equal time rule" that is total bogus. Wing as split
venturi is 100% the same bogus!
>
> In the serious literature, there is 100% consensus on a
> useful and highly-accurate description of how lift is
> produced. The essential ideas have been around, unchanged,
> for many decades. For instance, the beautiful result now
> known as the Kutta-Zhukovsky theorem was first published in
> 1902.
>
> You can easily use the Kutta-Zhukovsky theorem to prove that
> what Anderson said about ground effect is completely false.
> For details, see
> http://www.av8n.com/how/htm/airfoils.html#sec-k-z
>
> Jim <lose...@workfromhome.com> wrote:
>
> >> Airplanes, like everything else in the universe, follow the path of
> >> least resistance. :)
>
The Russians flew their 570 tonnes "Kaspian Sea Monster" already in
1962 and shocked the US military trhen.
It flew all due to the aircusion below the Wing. It is all good
explianed on their webpage"
Jan-Olov Newborg