Does a Flat-Plate wing really FLY?

[Texas Buzzard]

I started out building and flying Free Flights and Tow-Line gliders for contests. Much attention was paid to the airfoil of the wing. Then some 15 years ago NASA and the Univ. of Illinois built low speed wind tunnels. Before that only high speed tunnels were built.

We have learned a lot more about low-speed airfoils in the new tunnels.

When I built my first foam kit for electric power the FLAT PLATE WING I didn't trust it to fly "right". But, it flew. Sure the low wing loading helped a lot I think. Then I looked at ther flat-plate wing as a lifting body. But my smart (alack ) engineer son has changed my opinion.

He reported to me that the flat-plate wing, at speeds les than 20 mph will compete very well with "normal" airfoils. The slightly curved flat-plates the Wright Brothers flew were good at low speeds too.

The wind tunnel told us that a flat-plate airfoil at speeds less than 20 mph do act like a normal wing.....but they have a very sharp STALL. Have you noticed how your 3-D stalls just before it touches down? Can't help that either....sharp stalls come with the territory.

[BMatthews]

A flat plate is as valid an airfoil as any other shape. Like many of the usual options it has its strong and weak points that suggest when it's good to use it and when not.

The sharp stall is why the models can pretty much stop in the air on demand. A valuable trait for this style of flying.

[JetPlaneFlyer]

I think that the difficulty some people have with understanding why flat plates fly is due to the way they were taught about lift at school or college. The 'standard' explanation used to be (and perhaps still is in some places)that lift was produced due to the air molecules having to travel further over curved upper surface of the wing than the flatter lower surface. The theory went that in order to 'keep up' with the air below the wing, the air above, having further to go, must travel faster (the equal time of transit theory) : http://www.grc.nasa.gov/WWW/K-12/airplane/wrong1.html

Bernoulli theorem says that faster moving fluid will be at a lower pressure than slower moving fluid, so the pressure above the wing would therefore be less and lift would be produced.

The equal time of transit theory completely ignores the fact that symmetrical airfoils and flat plates fly just fine and that aircraft with cambered airfoil can often fly inverted :rolleyes:

[vintage1]

Its not that Bernouilli doesn't apply it's that the airflow over a flat plate is not as simple to calculate or to picture.

[MarkusN]

The "travel distance" explanation of lift is seriously flawed from the beginning. What outer power forces the same molecules to recombine at the trailing edge that before were separated at the leading edge? None at all.

Who was it that said: "For every complex technical problem there is a solution that is simple, elegant, and wrong."?

And, yes, that explanation is still taught. In theory courses for the soaring PPL, even.

[Brandano]

an F104 would fly essentially on a flat plate wing at mach 2. The real issue is the tradeoff between lift and drag, that's where the sharp leading and trailing edges of the 104 make a difference. And naturally the reynold numbers weight in heavily.

[NX-687]

with my flat plate plane at a 30 Degree AOA approx the wing goes up
Does this mean anything for flight , :confused:
I thought that explained why they fly completly , well for me it does any way

[nmasters]

W had a conversation like this a few months ago: http://www.rcgroups.com/forums/showthread.php?p=7994852#post7994852

[billsmithjr]

Bernouilli only accounts for about 30 percent of the total lift. The rest is produced by the bottom of the wing.
As long as the surface is not stalled it will produce greater amounts of lift/drag as the aoa is increased.

[NX-687]

If I round the LE and TE , even make them slightly pointy , this will only have good effects , no bad effects , yes :confused:

I assume so,,,,,, 6mm plate airfoil :)

Most of my flight is at high alpha , rounded LE wont hurt the high alpha will it , :)

[MarkusN]

Pointy makes for a sharp stall. But then, a flat plate wing is always kind of pointy-nosed.

[MarkusN]

Bernouilli only accounts for about 30 percent of the total lift. The rest is produced by the bottom of the wing.
As long as the surface is not stalled it will produce greater amounts of lift/drag as the aoa is increased.
It's kind of moot to differenciate between "Bernoulli" and "Newton" as is sometimes done (meaning that Bernoulli's law explains lift on the upper, Newton's lift on the lower side.)

Both are responsible for the whole thing. Force acting on the wing is the resultant of pressure distribution around the wing's outer surface. This can be calculated by Bernoulli's law (both the negative and the positive pressures).

The wing accelerates a packet of air downwards. This explains lift in terms of Newton's law. but usually it's simpler to calculate that downwash from lift than vice versa.

[vintage1]

Yes, Newton and Bernoulli are the same thing in the end.

Its a question of which is an easier starting point for calculations.

Bernouilli is easier in smooth laminar flow in incompressible liquids..turbulent flow means its a bad one, and even newton doesn't fare well unless you take every vortex into account. Hard work.

[Aio_1]

Navier Stokes works okay....... but takes a while!

[MarkusN]

Navier Stokes works okay....... but takes a while!
And is basically Bernoulli/Newton + effects of viscosity expressend in three dimensions.

[Aio_1]

And is basically Bernoulli/Newton + effects of viscosity expressend in three dimensions.
I think that's a little simplistic or perhaps a little over complicated depending on how you look at it. The Navier Stokes equations are pretty much the mechanics of a flow from first principles making no assumptions. All of the other flow models are simplified to some degree, usually very significantly. But then that's essential if you want to come out with a solution.

[Texas Buzzard]

Good info here. It made me go find a kite book I obtained long ago.

It had the usual diagram found in Physics texts for kites and sails. This was a vector diagram showing how "lift" is developed. It's good enough for me.

My first home brew electric was a conventional layout. I could not resist the desire to bend the flat plate wing about 35% back from the L.E. to make a slightly curved airfoil ( like the Wright Bros.) It flew O.K. Now I just go with the flow and use flat-plate wings. They work. But just before touchdown they stall and the plane goes, "Plop".

[Aio_1]

an F104 would fly essentially on a flat plate wing at mach 2. The real issue is the tradeoff between lift and drag, that's where the sharp leading and trailing edges of the 104 make a difference. And naturally the reynold numbers weight in heavily.
Transonic/Supersonic flow is a whole different kettle of fish. The viscous effects that rule our world go out the window and the requirements of compressible flow take over. For supersonic flow the ideal is the thinnest wing available and the shape of the aerofoil to a some extent is a result of both the need to fly sub-sonically at times and also structural aspects. The wing must be strong enough to carry the flight loads and it might be desired that it's thick enough to fill it with fuel, landing gear, weapons, etc (this is part of the reason why deltas have been popular. A thin aerofoil can have useful volume if the chord is long enough)

[Aio_1]

.....The equal time of transit theory completely ignores the fact that symmetrical airfoils and flat plates fly just fine and that aircraft with cambered airfoil can often fly inverted :rolleyes:
Symetrical aerofoils (including flat plates) do not have symetrical airflows unless the angle of attack is zero. With zero angle of attack they produce no lift.
If you look at the flow around a symetrical aerofoil with an angle of attack you'll see that the streamlines close to the surface over the upper surface originate on the underside of the LE and are indeed longer than those over the lower surface. An asymetrical flow is necessary to develop lift.;)

Aidan

[JetPlaneFlyer]

Aiden,
Did I imply that such airfoils had symmetrical airflow:confused:

[BarrelRoll]

Actually, the transit TIME over the upper surface of a wing has to be SHORTER (not equal) than that under the bottom surface. In equal transit time case the circulation would be zero and therefore we have no lift at all (at least for ideal fluid model).

[Aio_1]

Aiden,
Did I imply that such airfoils had symmetrical airflow:confused:
I thought so but perhaps I misinterpreted what you said. My understanding of the comment that I quoted was that you were refering to symetrical aerofoils and flat plates as special cases. I was simply pointing out that when they're generating lift they're asymetrical just like any other aerofoil.

[Neil Stainton]

Actually, the transit TIME over the upper surface of a wing has to be SHORTER (not equal) than that under the bottom surface. In equal transit time case the circulation would be zero and therefore we have no lift at all (at least for ideal fluid model).

Only initially, when circulation starts. In steady flight lift is constant, so circulation is constant, so time would be the same for upper and lower surfaces.

Neil.

[BarrelRoll]

Only initially, when circulation starts. In steady flight lift is constant, so circulation is constant, so time would be the same for upper and lower surfaces.

Neil.

If the time is the same then the circulation is zero. I think it would be impossible to create a downwash in incompressible fluid if there is no extra air moving over the upper surface:
http://www.av8n.com/how/htm/airfoils.html

[Gordonq]

An asymetrical flow is necessary to develop lift.

Aidan

I'm going to add my personal beliefs which are always subject to revision so that hopefully they will eventually coincide with reality:

All that's necessary to develop lift for level flight is to accelerate a mass of air downward that is equal to the weight of the aircraft, at an average acceleration of 32 feet per second per second (to exactly counteract the force of gravity). How you do this is immaterial to the object of providing lift if you wish to ignore drag and stability. Any airfoil can be inverted and still provide this lift but you will probably have to increase the speed and A.O.A. to overcome drag. The usually nicely curved upper surface is mainly useful for providing low drag while providing for internal wing structural components, liquids, mechanics, etc. No aircraft hangs in the air on it's thin upper skin and rivets/fasteners. If anybody in the aeronautic departments at Boeing or Airbus acutally believes this then I should refuse to fly commercially again. There is some Bernoulli effect obviously but I would guess it's responsible for much less than the 30% of lift mentioned somewhere earlier, especially when you take into consideration the force of the compression of air at the upper leading edge prior to it's acceleration rounding the top of the wing. An easy test of lift theory should be obtainable by building a simple but valid wind tunnel with an airfoil section with about 50 evenly spaced pressure test ports and doing a simple pressure analysis at these points and combining the forces measured for various wind speeds and A.O.A. into a resultant vector. The constant teaching of Bernoulli's Principal for aeronautical lift should be put to rest.

Key points: the undersurface of a wing is vastly superior to the top of the wing for lift. Air weighs 1300 g per cubic meter (1.198 oz per cubic foot) at sea level. Displace a mass that is equal to the weight of the aircraft downwards at 32 feet/sec/sec and you can achieve enough lift for level flight.

OK, I may have stepped on some beliefs and missed some physics, but further discussion is encouraged.

[Aio_1]

I'm going to add my personal beliefs which are always subject to revision so that hopefully they will eventually coincide with reality:

All that's necessary to develop lift for level flight is to accelerate a mass of air downward that is equal to the weight of the aircraft, at an average acceleration of 32 feet per second per second (to exactly counteract the force of gravity). How you do this is immaterial to the object of providing lift if you wish to ignore drag and stability. Any airfoil can be inverted and still provide this lift but you will probably have to increase the speed and A.O.A. to overcome drag. The usually nicely curved upper surface is mainly useful for providing low drag while providing for internal wing structural components, liquids, mechanics, etc. No aircraft hangs in the air on it's thin upper skin and rivets/fasteners. If anybody in the aeronautic departments at Boeing or Airbus acutally believes this then I should refuse to fly commercially again. There is some Bernoulli effect obviously but I would guess it's responsible for much less than the 30% of lift mentioned somewhere earlier, especially when you take into consideration the force of the compression of air at the upper leading edge prior to it's acceleration rounding the top of the wing. An easy test of lift theory should be obtainable by building a simple but valid wind tunnel with an airfoil section with about 50 evenly spaced pressure test ports and doing a simple pressure analysis at these points and combining the forces measured for various wind speeds and A.O.A. into a resultant vector. The constant teaching of Bernoulli's Principal for aeronautical lift should be put to rest.

Key points: the undersurface of a wing is vastly superior to the top of the wing for lift. Air weighs 1300 g per cubic meter (1.198 oz per cubic foot) at sea level. Displace a mass that is equal to the weight of the aircraft downwards at 32 feet/sec/sec and you can achieve enough lift for level flight.

OK, I may have stepped on some beliefs and missed some physics, but further discussion is encouraged.
I'd like to point out that "belief" doesn't really come into a great deal. The area is well understood at the level we're discussing. Also it's ridiculous to say that "the undersurface of a wing is vastly superior to the top of the wing for lift." This statement is completely meaningless. Neither surface in and of itself does anything. The resultant force due to the pressure distribution over the entire wing surface is the total lift. You can divide this into a simplified but still useful 2D model and look at the aerofoil alone but you can't usefully take the upper and lower surfaces and look at them in isolation. They have to exist as a pair and neither is "superior". The lift and drag are a product of both surfaces combined and you can't usefully discretise them.
The test you suggest for measuring pressure distribution over an aerofoil is, as you are probably aware, one of the standard methods used to analyse fluid flow over a shape. There's not really any room for debating the outcome; it's well known.

You quoted my statement in your post as follows:
"An asymetrical flow is necessary to develop lift."
I'm not sure from your post if you intended to agree or disagree with that statement. Bear in mind that I said asymetrical flow NOT asymetrical aerofoil and this does support your proposition that air must be accelerated downwards to provide lift.

I think a big problem in these discussions is that people argue theories as though they were in opposition when in fact they are alternative models of the same behaviour. Equally valid and not mutually exclusive. I'm not aware of a conflict between any of the accepted fluid dynamic principles at this level.

Aidan

[biber]

Very well put!

(instead of rating threads it should be possible to rate single posts, I'd be happy to give this one a full house)

biber

[Gordonq]

Excerpted from Wikipedia

(In opposition to the below article, I would still like to add my personal belief that the underside of an airfoil contributes the major component of lift and it does this by deflecting, no, FORCING, a mass of air downwards. The article discounts Bernoulli's Principle - YEAH - but states the upper surface is still responsible for doing the major deflecting of air downwards. The article states in support of its argument that the engines of airliners are mounted on the underside to keep the upper wing surface more effective but I believe it's because the position greatly lowers the center of gravity thus increasing the stability factors of flight. It amazes me that at this late date there is still any trace of confusion left about how an aircraft flies. I wonder if it's a marketing conspiracy to have people believe that something rock solid like good ol' Bernoulli's Principle is keeping aircraft safely aloft and keep them confidently buying tickets. Or perhaps it's just good old educational inertia.)

------

Excerpted from www.wikipedia.org

Bernoulli's principle...

A common misconception about wings
While lift generated by an airfoil is often attributed to Bernoulli's Principle, it cannot be used to explain this. The popular and faulty explanation is generally written as this:
A wing is flat on the underside and curved on the top (is cambered). Thus air that moves over the top of the wing must travel a longer path to get to the end of the wing and therefore gets a higher velocity. According to Bernoulli's Principle, the faster moving air has lower pressure and the difference therefore results in lift.
This explanation is flawed in a critical aspect: it assumes that two parcels of air separated at the leading edge of the wing, with one traveling over and one traveling below, must meet again at the trailing edge of the wing. This does not happen. In fact the Kutta-Joukowski Theorem shows that if this was to happen, the wing would generate no lift.
The actual mechanism generating lift on an airfoil is Newton's Third Law of Motion.[8] An airfoil is always flown at an angle of attack against the air flow. As the wing deflects air downwards the opposing reaction force on the wing pushes it upwards. Note that nearly all of the lift arises from airflow over the top of the wing being deflected downwards, due to the Coanda Effect; the deflection due to the underside of the wing makes only a small contribution. (This is one reason why wing-mounted jet engines are suspended below the wing, rather than being placed on top of it - the disruptioin to flow over the bottom surface of the wing has much smaller effect of lift than mounting the engines above the wings.) Contrary to Bernoulli's Principle this explains why an aircraft with a thrust-to-weight ratio less than 1.0 can fly on a level flight path while being upside down (instead of being pulled dramatically towards the ground); why slim wings, such as those of the F-104 Starfighter or those of a paper plane, generate lift despite the camber being nonexistent; and why some lifting body aircraft can fly despite being very bulbous on the underside.
The airflow can be observed by solving the Navier-Stokes equations for the appropriate flow regime (turbulent or laminar) and the pressure along the wings edges directly calculated. As pressure is simply a measurement of force per unit area, integration of the pressure along the wings surface (both top and bottom) provides an overall force, which is the amount of lift provided by the wing.
Though Bernoulli's Principle cannot be used to explain the lifting mechanism, it can still be used to accurately analyze the airflow around an airfoil. If you know either the air pressure or the air velocity over a wing you can use Bernoulli's equation to calculate the value of the other property. The equation is very often used this way. This high frequency of use has been cited as the reason the misconception has arisen.[8]

[vintage1]

I think the left side of the body is more important than the right. I mean if we only had left hand sides, well all our problems would be solved!

Look in nature, see how many things that only have one side there are! Do THEY suffer from depression, start wars, or have pointless arguments about which side of a 3 dimensional object is the more important? No. They sit there like unicorns, secure in the knowledge of their own impossibility.

Oh dear,. It's the difference in airflow between the two surfaces that makes all the difference. So to speak. The velocity differential creates te pressure diifferential that creates the lift. Bernoulli DOES work, if you allow for the fact that air is not the perfect liquid, and therefore moving a flat plate through it does not create symmetrical flows lines above and below the plate. You get flow separation and vortex generation above it, and the majority of the air has to jump over those and take that longer path. All an airfoil section is, is filling in that turbulent area with a bit of wing shape, to reduce the drag. The lift will be substantially unaltered.

As to why jet airliners have the engines underneath? well sorry to disappoint you. Its for one very simple and boring reason. They are far easier to service there. Service costs indirectly dominate the running costs of aircraft: to maximise return on capital they must be in the air as much as possible..this is more relevant than fuel efficiency. You could get far better fuel efficiency with slower turboprops, but you wouldn't get as many passenger miles per year out of them. So jets flying subsonic with engines slung underneath is the best compromise on long haul routes.

The pods are slung as well..weight can be saved by having the mountings essentially able to withstand tensile stresses, not compressive. I diont think any commercial aircraft is stressed for (much) negative G at all.

Finally, by placing the engines outside the wings, which are largely where the fuel tanks are, the chances of a thrown turbine blade puncturing a tank is far far less. And an on fire engine does not immediately threaten the whole aircraft.

Aerodynamics on airliners is merely something they need in order to not burn too much fuel. Up till recently that hasn't been a hugely overriding concern.

[nmasters]

(In opposition to the below article, I would still like to add my personal belief
{yawn} :rolleyes:

[nmasters]

All an airfoil section is, is filling in that turbulent area with a bit of wing shape, to reduce the drag. The lift will be substantially unaltered.

Yep, the skin is just a streamlined fairing wrapped around a mean line.

[biber]

In opposition to the below article, I would still like to add my personal belief that the underside of an airfoil contributes the major component of lift and it does this by deflecting, no, FORCING, a mass of air downwards.Succing air down is nothing less effective than pushing it downwards,
as long as you don't approach (even locally) zero pressure,
or the speed of sound maybe. The article discounts Bernoulli's Principle - YEAH - but states the upper surface is still responsible for doing the major deflecting of air downwards.You can see it different ways.
The lift is basically the netto pressure difference between top and bottom surface.
Even if the pressure of a currently nonlifting wing is equal on top and bottom,
the pressure on both surfaces is lower, than the free stream static pressure.
Thats due to the displacement every body has in a fluid.
The displacement leads to an acceleration of the streaming fluid in the body's vicinity.
That locally increased velocity of flow gives a lower pressure according to Bernoulli.
If you now let the wing generate lift, the top surface average absolute pressure will decrease even further.
The pressure on the bottom will increase and may be close to static pressure, maybe a bit less, maybe a bit more.
But in any case, the bottom surface average pressure will be closer to static pressure, that the top surface's one.
That might lead to the conclusion, the top surface is the more lifting one.
But you could also say, the top is just less getting sucked down than the top is getting sucked up. The article states in support of its argument that the engines of airliners are mounted on the underside to keep the upper wing surface more effective but I believe it's because the position greatly lowers the center of gravity thus increasing the stability factors of flight. I don't think either one of the above is right.
The stability wouldn't be considerably effected.
Nor would it effect liftability of the top surface.
See the Attas e.g.: http://www.bs.dlr.de/wt/fb/bs/pictures/ATTAS.GIF
It has been done at least.
Maybe it's for less noise in the cabin or to avoid interfering with top surface
local sonic shock waves, that might show up sometimes on cruise.
Or maybe it's for better access for maintenance rather.

Edit:
Vintage, you da man!
I was not only to slow, but #29 said it better and more clear than I ever could.

biber

[pmackenzie]

There is a neat little on line flow simulation program here (http://adamone.rchomepage.com/foil_sim.htm)

You can change camber, thickness, AoA,etc and see the results.

The unmistakable conclusion of properly applying fluid dynamics to the problem of subsonic incompressible fluid flow around a wing section is that the majority of the force on a wing is a result of suction on the upper surface.

(Flat plate comes close to 50/50, but still there is a bit more area between the curve for the top surface pressure and the mean line than for the bottom)

The "momentum" or downwash model is similar to the momentum model for analyzing propulsion systems.
It can tell you about what is going on for the sytem as a whole, but can't help in any useful way in the design of actual sections.

For that you need fluid dynamics to resolve the flow and then Bernoulli to turn the flows into forces.

Pat MacKenzie

[BMatthews]

Pat, that is a copy of the parent at the NASA site...

http://www.grc.nasa.gov/WWW/K-12/airplane/foil2.html

And yes, while it's limited in what it can do and the airfoil options it is a very handy simulator that can help us understand a wide variety of flight issues. I also like to use it to determine the rough lift coefficients that some of my models are flying at so I can more intelligiently pick and choose airfoil options.

[RCAV8R13]

Bernouilli only accounts for about 30 percent of the total lift. The rest is produced by the bottom of the wing.
As long as the surface is not stalled it will produce greater amounts of lift/drag as the aoa is increased.
That's the old "bunch of bullets" theory. It's bunk. Lift comes from down wash, off the top of the wing, not the bottom. If the lift came from the bottom of the wing then Da Vinci's spiral helicopter would have worked, but it didn't
RCA

[biber]

RCA, the downwash comes from the entire wing, not only one surface.

biber

[pmackenzie]

Downwash is an effect, not a cause.

Pat MacKenzie

[robin andrew]

Hi, this a true story. Some years ago I ran a bungee competition and along came a chap with an 'ORANGE BOX' glider. His son assembled it and it went up the line OK and received a time of 2 1/2 mins. Upon retreival it was found to have the main wing put on backwards so the next flight was going to do better!! We spent some of the afternoon trying to beat the 2 1/2 mins of the first flight but didn't do it. So the TE was sharpe and hitting the air firstly ; so much for airfoils, Cheers Robin Andrew, Uk

[biber]

Razor sharp LEs are the way to go for real low Reynolds numbers.

biber

[Neil Stainton]

Razor sharp LEs are the way to go for real low Reynolds numbers

Really? Why is that?

TIA,

Neil.

[biber]

Hm, what does TIA mean?

At very low Re you can benefit of a sharp LE because it will destabilise the laminar flow.
And that is something you want to happen there, because at low Re you otherwise might suffer from severe laminar separation.
That would be by far the bigger drag penalty, than the turbulent BL is, at that point.
Try that on balsa or depron gliders.

biber

[nmasters]

A sharp LE will have clean separation at any Re which will roll up into a leading edge vortex with turbulent reattachment downstream AKA a separation bubble. Above some critical Re a bubble is just extra drag so you want to avoid it as much as possible. Below the critical Re an airfoil with a blunt LE will produce an unstable bubble or just stall early whereas the one with the sharp LE will hold its bubble in place and avoid a stall up to a higher AoA.

--Norm

[biber]

That's sound!

I guess TIA means thanks in advance?

biber

[Neil Stainton]

Thanks Norm and Biber, that is very clear. I thought a sharp LE would promote full separation, not create a safe separation bubble. That explains why Shockfliers fly so well. In the past I spent some time adding 1/2 round leading edges to Shockfliers to 'improve' them!

Yes TIA is short for thanks in advance.

Regards,

Neil.

[macboffin]

Succing air down is nothing less effective than pushing it downwards,
as long as you don't approach (even locally) zero pressure,
or the speed of sound maybe. You can see it different ways.
The lift is basically the netto pressure difference between top and bottom surface.
Even if the pressure of a currently nonlifting wing is equal on top and bottom,
the pressure on both surfaces is lower, than the free stream static pressure.
Thats due to the displacement every body has in a fluid.
The displacement leads to an acceleration of the streaming fluid in the body's vicinity.
That locally increased velocity of flow gives a lower pressure according to Bernoulli.
If you now let the wing generate lift, the top surface average absolute pressure will decrease even further.
The pressure on the bottom will increase and may be close to static pressure, maybe a bit less, maybe a bit more.
But in any case, the bottom surface average pressure will be closer to static pressure, that the top surface's one.
That might lead to the conclusion, the top surface is the more lifting one.
But you could also say, the top is just less getting sucked down than the top is getting sucked up. I don't think either one of the above is right.
The stability wouldn't be considerably effected.
Nor would it effect liftability of the top surface.
See the Attas e.g.: http://www.bs.dlr.de/wt/fb/bs/pictures/ATTAS.GIF
It has been done at least.
Maybe it's for less noise in the cabin or to avoid interfering with top surface
local sonic shock waves, that might show up sometimes on cruise.
Or maybe it's for better access for maintenance rather.

Edit:
Vintage, you da man!
I was not only to slow, but #29 said it better and more clear than I ever could.

biber Access for maintainnce and service ; plus when they catch fire, fusible connectors let go and the engine drops clear off the wing. This has saved quite a few airplanes!

[robin andrew]

Thanks for the explanation that shows how the thin flat sheet wings on a chuck glider can work. I used to be able to get 1.1/2 mins every time on an own design but when I rounded the LE and used a slightly thicker wing, with an airfoil section, it didn't do as well,cheers Robin Andrew,UK

[Texas Buzzard]

In POST 25 by gordong I read "Air weighs 1300 g per cubic meter (1.198 oz per cubic foot) at sea level."

I simply think gordong overlooked his units. It happens, but it should be pointed out that grams is NOT a unit of Weight.

Yes a gram is NOT a unit of weight, it is a unit of MASS.

Weight would be mass x acceleration . Basically F = MA.... right?

So on the earth we multiply the mass of an object by the acceleration of gravity ( 9.8 meters per second squared or in the cgs we say 980 centimeters per second squared.

In the example by gordong - he said air weighs 1300 g/meter 3.

That much air would have a weight of 1.3 kg x 9.8 meter/second/second

That would be 12.74 Newtons - - Not 1300 g :)

Weight is a Unit of Force while GRAMS is Unit of MASS

[Gordonq]

In POST 25 by gordong I read "Air weighs 1300 g per cubic meter (1.198 oz per cubic foot) at sea level."

I simply think gordong overlooked his units. It happens, but it should be pointed out that grams is NOT a unit of Weight.

Yes a gram is NOT a unit of weight, it is a unit of MASS.

Weight would be mass x acceleration . Basically F = MA.... right?

So on the earth we multiply the mass of an object by the acceleration of gravity ( 9.8 meters per second squared or in the cgs we say 980 centimeters per second squared.

In the example by gordong - he said air weighs 1300 g/meter 3.

That much air would have a weight of 1.3 kg x 9.8 meter/second/second

That would be 12.74 Newtons - - Not 1300 g :)

Weight is a Unit of Force while GRAMS is Unit of MASS


From Wikipedia:

In everyday usage, mass is more commonly referred to as weight, but in physics and engineering, weight means the size of the gravitational pull on the object; that is, how heavy it is, measured in units of force. In everyday situations, the mass and weight of an object are directly proportional to each other, which usually makes it unproblematic to use the same word for both concepts. However, the distinction between mass and weight becomes important:

for measurements with a precision better than a few percent, due to slight differences in the strength of the Earth's gravitational field at different places
for places far from the surface of the Earth, such as in space or on other planets

Since we're pretty much earth-bound and every-day folks, I'll keep with the grams. When buying food outside of the United States folks aren't going to carp about getting charged for force instead of mass when they purchase their vegetables :-) If you have a cubic meter container full of air on one side of a balance beam and on the other side have the "same" container with a perfect vacuum, you would find you would need to add 1300 grams to balance the beam. Interestingly, a 1300g aircraft would have to apply the force to accelerate one cubic meter of air downwards at the acceleration rate of gravity to maintain level flight.

[Gordonq]

I think the left side of the body is more important than the right. I mean if we only had left hand sides, well all our problems would be solved!

Look in nature, see how many things that only have one side there are! Do THEY suffer from depression, start wars, or have pointless arguments about which side of a 3 dimensional object is the more important? No. They sit there like unicorns, secure in the knowledge of their own impossibility.

Oh dear,. It's the difference in airflow between the two surfaces that makes all the difference. So to speak. The velocity differential creates te pressure diifferential that creates the lift. Bernoulli DOES work, if you allow for the fact that air is not the perfect liquid, and therefore moving a flat plate through it does not create symmetrical flows lines above and below the plate. You get flow separation and vortex generation above it, and the majority of the air has to jump over those and take that longer path. All an airfoil section is, is filling in that turbulent area with a bit of wing shape, to reduce the drag. The lift will be substantially unaltered.

As to why jet airliners have the engines underneath? well sorry to disappoint you. Its for one very simple and boring reason. They are far easier to service there. Service costs indirectly dominate the running costs of aircraft: to maximise return on capital they must be in the air as much as possible..this is more relevant than fuel efficiency. You could get far better fuel efficiency with slower turboprops, but you wouldn't get as many passenger miles per year out of them. So jets flying subsonic with engines slung underneath is the best compromise on long haul routes.

The pods are slung as well..weight can be saved by having the mountings essentially able to withstand tensile stresses, not compressive. I diont think any commercial aircraft is stressed for (much) negative G at all.

Finally, by placing the engines outside the wings, which are largely where the fuel tanks are, the chances of a thrown turbine blade puncturing a tank is far far less. And an on fire engine does not immediately threaten the whole aircraft.

Aerodynamics on airliners is merely something they need in order to not burn too much fuel. Up till recently that hasn't been a hugely overriding concern.


LOL. To clarify what is meant...

Bernoulli's principal is often given as the sole or major factor in lift. What I'm implying in my fascination with the underside of an airfoil is that the difference in under-wing air pressure versus local surrounding static air pressure is considerably greater than the difference in over-wing pressure versus static pressure. Thus the underside contributes the majority of lift. Admittedly I say this without research or empirical evidence because this is non-professional venue and I'm lazy and have a right to my beliefs as long as they don't interfere with the flight of your aircraft :-)

Since working airfoils are not two dimensional entities drawn on paper, what are you going to do if your aircraft develops a small hole on the underside that allows the interior of your wing to become somewhat pressurized at the underside air pressure. With your aircraft relying in Bernoulli to suck it up, you've got the entire aircraft mass hanging on your top skin. Be sure to put lots of glue on the top skin, maybe even double plank it.

What explains the violent snap up manouver of full up elevator? Is this suction on demand?

How does your Bernoulli enabled aircraft fly inverted so beautifully?

I can perhaps design some airfoils that are Bernoulli-enhanced that have gentle wavy tops to cause that airflow to travel a bit farther and really have to speed up thereby creating incredible lift.

No, I'm glad that Bernoulli is not designing aircraft. (I retract my misconceptions regarding lift in post #60)

[vintage1]

Bernouilli works exactly IF you calculate the TRUE streamlines due to a wing. Hint: they don't follow the surface, due to turbulence and friction.

There is MORE turbulence on top of the wing.

The pressure drop ON TOP of the wing is MORE than the pressure rise under it.

As can be seen from where the vapour forms..

i.e. the wing top is probably MORE important when considered alone, BUT since the bottom of the wing is necessary to give the top the properties it has, its slightly silly to consider it in total isolation.

[biber]

that the difference in under-wing air pressure versus local surrounding static air pressure is considerably greater than the difference in over-wing pressure versus static pressure.Not correct.
Admittedly I say this without research or empirical evidence because this is non-professional venue and I'm lazy and have a right to my beliefs as long as they don't interfere with the flight of your aircraftDoesn't change the fact, that you're assumtions do not correspond with the real world.

biber

[nmasters]

I can perhaps design some airfoils that are Bernoulli-enhanced that have gentle wavy tops to cause that airflow to travel a bit farther and really have to speed up thereby creating incredible lift.

If you do you’ll be in good company. A fellow named A. Einstein did that and it didn’t work because he had just as bad a grasp of the problem as you do. Wings do not work like water skis. If they did there wouldn’t be any stall because upper surface attachment wouldn’t matter but it does.

[Aio_1]

....What I'm implying in my fascination with the underside of an airfoil is that the difference in under-wing air pressure versus local surrounding static air pressure is considerably greater than the difference in over-wing pressure versus static pressure. Thus the underside contributes the majority of lift. Admittedly I say this without research or empirical evidence because this is non-professional venue and I'm lazy and have a right to my beliefs as long as they don't interfere with the flight of your aircraft :-)...
Take a look at my Avatar.

It's a typical computational fluid dynamics image showing the airflow around a symetrical airfoil at a moderate angle of attack. I can't remember now if I was plotting velocity or pressure for that particular image but the features are similar for both. As you can see the pressure gradient on the upper surface is far greater than the lower surface in complete contradiction of your statement. This pressure distribution is confirmed if you hook up manometers to a line of tappings along the aerofoil. I've done the tests myself as have thousands of others. It's well understood and has been for most of a century. It's not a mystery and it's not fiction.

You can stick to your beliefs if you choose but I assume from the fact that you've posted them here that you either want to learn more or want to influence others to believe the same. The former is admirable the latter is a problem.

Aidan

[Brandano]

Incidentally, the old Porsche Carreras used to take off beautifully at high speed, even though the pressure under their bodywork was lower than ambient pressure. That is because the domed bodywork cause much lower pressure above the car, and lifted the wheels off the asphalt. They've since been fixed, by adding a large spoiler, sideskirts, larger wheels, a larger wheelbase, moving the engine forward of the rear axle, lowering the engine, lowering the car's riding height, adding an even bigger spoiler, adding a front spoiler integrated with he bodywork, adding ESP, Traction Control and ABS. NOW the car sticks to the ground, but all of this would have been unnecessary if they didn't have to keep making the car pretty much the same shape because people wanted to buy a car looking like that. That said, another couple of Porsche GT cars had some spectacular takeoffs during a Le Mans 24 because they were tailing the car in front of them so close that there wasn't enough downforce on the front of the car.. Here's one that was caught on tape (don't worry, the pilot is fine) http://uk.youtube.com/watch?v=8M3NQeDzedk
[edit] One Porsche GT, not a couple. And in 1999 Mark Webber took off twice in a Mercedes CLR, during the qualification and in the warm up lap. He left the honor to get the car airborne in the actual race to his teammate, Peter Dumbreck

[Gordonq]

Take a look at my Avatar.

It's a typical computational fluid dynamics image showing the airflow around a symetrical airfoil at a moderate angle of attack. I can't remember now if I was plotting velocity or pressure for that particular image but the features are similar for both. As you can see the pressure gradient on the upper surface is far greater than the lower surface in complete contradiction of your statement. This pressure distribution is confirmed if you hook up manometers to a line of tappings along the aerofoil. I've done the tests myself as have thousands of others. It's well understood and has been for most of a century. It's not a mystery and it's not fiction.

You can stick to your beliefs if you choose but I assume from the fact that you've posted them here that you either want to learn more or want to influence others to believe the same. The former is admirable the latter is a problem.

Aidan


Well then, I'm certainly fortunate that inquisitions have become unpopular. There is certainly room for discussion.

Please take a peek at this:

http://www.eskimo.com/~billb/wing/airfoil.html

[BMatthews]

A while back I had the "honor" of moderating over a group of aerodynamics types about the age old question of "what creates the lift". It got quite heated and rather exciting. I learned a lot watching the back and forth and supportive arguments. Keep in mind that these folks were in the industry designing and running wind tunnels and other direct application stuff. They were also quick to quote some pretty heavy texts. Take all that for what you will....

Out of all the broohaha I learned that if you study the upper and lower pressure distributions and integrate them all into one value that you can show that the lower upper surface pressure combined with the raised lower surface can account accurately for the lift of the wing. And it was agreed by all that the lower upper pressure was more noticable than the raised lower pressure. Sorry Gordonq.

The other camp countered by showing that the redirection of the air off the trailing edge compared to the direction out in front was able to show that the purely Newtonian acceleration of the air mass was able to account for the total lift required for flight.

So we've now got two ways to show the same amount of force to hold up the airplane. Funny thing is that one can't exist without the other any more than Vintage's single sided planarian can drink a latte. They go hand in hand and are locked together. In fact you can say that the pressure distribution and the newtonian redirection are all caused by the same action of driving our wing through the air. So really they ARE the same thing since they both describe the same work being done. After all, you may be able to demonstrate the lift value in two different ways but we're still only lifting one airplane so there can't be twice the lift, right?

Interestingly enough if you study the pressure distribution around the wing you'll see that the forces all seem to act to support the airfoil at around the 25% chord point just as we'd expect from the Cm values and the agreed upon "aerodynamic center" being at the 25% point. However if you look at the streamlines for a newtonian answer you'll notice how the lines all lift up and then turn over the hump and flow dowwards at the 25% point as well.

So we come back to what came first, the chicken or the egg. Does the pressure gradient cause the airflow to redirect itself at a point over the wing and we get lift from the newtonian acceleration of the air? Or does the wing lift due to the lower pressure above and the higher pressure below pushing it upwards and the airflow is just affected by coincidence? Or is it just one mutually locked combination that cannot be separated but can be analyzed in two ways that makes it appear to be separate?

[Gordonq]

A while back I had the "honor" of moderating over a group of aerodynamics types about the age old question of "what creates the lift". It got quite heated and rather exciting. I learned a lot watching the back and forth and supportive arguments. Keep in mind that these folks were in the industry designing and running wind tunnels and other direct application stuff. They were also quick to quote some pretty heavy texts. Take all that for what you will....

Out of all the broohaha I learned that if you study the upper and lower pressure distributions and integrate them all into one value that you can show that the lower upper surface pressure combined with the raised lower surface can account accurately for the lift of the wing. And it was agreed by all that the lower upper pressure was more noticable than the raised lower pressure. Sorry Gordonq.

The other camp countered by showing that the redirection of the air off the trailing edge compared to the direction out in front was able to show that the purely Newtonian acceleration of the air mass was able to account for the total lift required for flight.

So we've now got two ways to show the same amount of force to hold up the airplane. Funny thing is that one can't exist without the other any more than Vintage's single sided planarian can drink a latte. They go hand in hand and are locked together. In fact you can say that the pressure distribution and the newtonian redirection are all caused by the same action of driving our wing through the air. So really they ARE the same thing since they both describe the same work being done. After all, you may be able to demonstrate the lift value in two different ways but we're still only lifting one airplane so there can't be twice the lift, right?

Interestingly enough if you study the pressure distribution around the wing you'll see that the forces all seem to act to support the airfoil at around the 25% chord point just as we'd expect from the Cm values and the agreed upon "aerodynamic center" being at the 25% point. However if you look at the streamlines for a newtonian answer you'll notice how the lines all lift up and then turn over the hump and flow dowwards at the 25% point as well.

So we come back to what came first, the chicken or the egg. Does the pressure gradient cause the airflow to redirect itself at a point over the wing and we get lift from the newtonian acceleration of the air? Or does the wing lift due to the lower pressure above and the higher pressure below pushing it upwards and the airflow is just affected by coincidence? Or is it just one mutually locked combination that cannot be separated but can be analyzed in two ways that makes it appear to be separate?

Excellent post. The intuitive approach didn't work for me so I decided to go to the NASA Technical Reports Server. I saw some amazing things there and now have joined the negative pressure on the upper surface camp so I've done a little more than a 180 degree change since I was willing to intuitively give the upper surface 30% lift. Now I'm not willing to give the lower surface even that much. I'll post some pressure charts of a super critical airfoil and a Clark Y when I get a decent means (soon?).

[MarkusN]

Looking at theory of thin wing sections, upper and lower surface at design AoA contribute both 50%. It's just that this is superimposed by the (overall negative) pressure distribution of the section of finite thickness. This adds to the negative pressure on top, and subtracts from the positive at the bottom and thus makes the top look more important. Still, it's contribution is neutral overall. Things change at higher AoA, though.

[2phar]

Flat plate wings really DO fly.
So do some squirrels.
And not a one of these damn squirrels ever talked to Bernouilli.
Airplanes work by throwing air at the ground- period.
regards, chris

[Gordonq]

Excellent post. The intuitive approach didn't work for me so I decided to go to the NASA Technical Reports Server. I saw some amazing things there and now have joined the negative pressure on the upper surface camp so I've done a little more than a 180 degree change since I was willing to intuitively give the upper surface 30% lift. Now I'm not willing to give the lower surface even that much. I'll post some pressure charts of a super critical airfoil and a Clark Y when I get a decent means (soon?).


Nothing beats a picture and as I was shopping around for pictures I found a very rich educational website run by NASA. You can have a lot of fun here designing stuff! (be sure to have a recent version of JAVA). What amazes me about my ongoing discoveries of airfoils is that the peak lift force comes from the upper surface at just a few% of the chord in just about every practical airfoil and tapers off to about zero towards 100% chord. The bottom DOES contribute lift pressure but the top surface is what's best in redirecting air mass downward. In just about every practical airfoil the sum of lift pressures (positive and negative) is about 25% back of the leading edge due to the pressure taper. Also amazing to me is that the upper airstream is quite far downstream when it rejoins the lower airstream. This is perhaps why the top surface is so prevalent in achieveing negative relative pressure. I will admit that my attitudes have just completely changed and were probably based on extending my hand out the car window as a child. (Nowadays kids are belted into the center of the rear seat and maybe won't have to suffer from my particular misconception.) I also found that a flat plate (the main theme of this thread) has very similar pressure distributions as a standard airfoil (you can interactively manipulate airfoils at this site and see the pressure distributions).

Three wrong airfoil theories (yup, mine is there)
http://www.grc.nasa.gov/WWW/K-12/airplane/wrong1.html
http://www.grc.nasa.gov/WWW/K-12/airplane/wrong2.html
http://www.grc.nasa.gov/WWW/K-12/airplane/wrong3.html

Some text lifted from the site:

HOW IS LIFT GENERATED?

There are many explanations for the generation of lift found in encyclopedias, in basic physics textbooks, and on Web sites. Unfortunately, many of the explanations are misleading and incorrect. Theories on the generation of lift have become a source of great controversy and a topic for heated arguments. To help you understand lift and its origins, a series of pages will describe the various theories and how some of the popular theories fail.

Lift occurs when a moving flow of gas is turned by a solid object. The flow is turned in one direction, and the lift is generated in the opposite direction, according to Newton's Third Law of action and reaction. Because air is a gas and the molecules are free to move about, any solid surface can deflect a flow. For an aircraft wing, both the upper and lower surfaces contribute to the flow turning. Neglecting the upper surface's part in turning the flow leads to an incorrect theory of lift.


Here's some more of that rich content.

URL: http://www.grc.nasa.gov/WWW/K-12/airplane/index.html

FoilSim II: (80KB) FoilSim II computes the theoretical lift of a variety of airfoil shapes. The user can control the shape, size, and inclination of the airfoil and the atmospheric conditions in which the airfoil is flying. The program includes a stall model for the airfoil, a model of the Martian atmosphere, and the ability to specify a variety of fluids for lift comparisons. The program has graphical and numerical output, including an interactive probe which you can use to investigate the details of flow around an airfoil.

CurveBall: (39KB) Using the CurveBall applet, students learn more about aerodynamics by controlling the conditions of a big league baseball pitch. You can vary the speed of the pitch, the spin on the ball, the release point, and the location of the stadium which affects the atmospheric conditions and the amount of curve on the ball. The program will compute balls and strikes and tells you how far your pitch passes the center of the plate .

EngineSim: ( 455KB) EngineSim is a simulator that models the design and testing of jet engines. The program works in two modes: Design Mode or Tunnel Test Mode. In the Design Mode, you can change design variables including the flight conditions, the engine size, the inlet performance, the turbo machinery compressor and turbine performance, the combustors or burner performance, or the nozzle performance. For a turbofan engine design you can also vary the fan performance and the bypass ratio. When you have a design that you like, you can switch to the Tunnel Test Mode which simulates the testing of a jet engine on a test stand. You can then vary the test altitude, flight speed and throttle setting. Several existing engines are also modeled in EngineSim.

Undergraduate Computer Programs: Here is a group of Java programs which were designed to help undergraduate engineering students learn about the basics of aerodynamics and propulsion. There are programs to solve for the flows across shock waves and expansion fans, the flow through crossed and reflected shocks, and the properties in regions of isentropic flow. There are also special versions of EngineSim and FoilSim for undergraduates.
RangeGames: (487KB) This program presents a variety of multiple choice math and physics problems involving aircraft performance. The student can choose from several different types of aircraft and must answer questions about the range, fuel usage, acceleration, velocity and location of the aircraft during take-off. RangeGames can record your answers for teacher evaluation, or you can just play for fun.

RocketModeler: This program lets you design and study the flight of a model rocket. You can vary the size of the rocket, the number of fins, and the materials used to construct the rocket. You can choose from a variety of available model rocket engines and test fly your rocket on the computer. The program computes the stability of your design and the flight trajectory. Output includes the maximum altitude which the rocket achieves. You can then compare the computed and actual performance of your model rocket.

KiteModeler: This program lets you design and study the flight of a kite. You can select from five different types of kites and then vary the length, width and types of materials used to construct the kite. You then trim the kite by setting the length of the bridle and tail and the position of the knot attaching the control line to the bridle. Finally, you test fly your kite on the computer by setting the wind speed and the length of control line. The program computes the aerodynamic forces, weight, and stability of your design and the shape of the control line as it sags under its own weight. Output includes the maximum altitude which the kite achieves. You can then compare the computed and actual performance of your kite design.

Atmosphere Applet: This program lets you study how pressure, temperature, and density change through the atmosphere. You can study the atmosphere of the Earth or of Mars. Since speed of sound depends on the atmospheric gas and the temperature, you can also output the local speed of sound and the Mach number for a selected aircraft velocity. You can either input a selected altitude, or change altitude using an aircraft slider.

GasLab Program: Here is a group of computer animations which were designed to help high school chemistry students learn about the basics of the gas laws and the equation of state. The state of a gas is determined by the pressure, temperature, mass, and volume of the gas. The program lets you fix two of these variables and observe the relation of the other two variables by changing the value of one of them.


Lastly, there's two photos showing the rejoining of the upper airflow with the lower airflow. URL for these: http://amasci.com/wing/airgif2.html

[MarkusN]

Flat plate wings really DO fly.
So do some squirrels.
And not a one of these damn squirrels ever talked to Bernouilli.
Airplanes work by throwing air at the ground- period.
Stop beating Bernoulli for the incredibly flawed run-length explanation of lift that is unfortunately named after him. The currently handed around theory of throwing air at the ground is just as wrong but even more difficult to beat to death.

People read their Anderson/Eberhardt and then say: "Yes! I knew it all along! It's Newton all the way!" But they ignore that the objective of that publication is to give pilots a grasp of power requirements of flight.

So A/E mostly talk about induced drag. (They also clearly call it that on the webpage touting their book.) And of course induced drag is associated with downwash, and that downwash is related to lift.

BUT SAYING THAT THAT DOWNWASH IS THE COUNTERFORCE TO LIFT IS HOGWASH!

On a high-aspect ratio wing most of the momentum change of the airflow over the wing is upwash in front to downwash behind, with the downwash quickly petering out in a potential flow that is perfectly described (not explained!) by Bernoulli's law.

The downwash that remains behind is actually a "sealing loss" created by the less than perfect wing planform. It's the part of flow that creates drag, not lift.

[Odysis]

OK, I've been watching this thread for a while, the heated discussion is interesting!

The downwash that remains behind is actually a "sealing loss" created by the less than perfect wing planform. It's the part of flow that creates drag, not lift.

This doesn't make any sense whatsoever. Drag is perpendicular to downwash, so how can one be created by the other? And downwash isn't created as a result of an imperfect wing, any three dimensional lifting surface will create downwash. How much? w~=L/(p.pi.b^2.V) That much. No drag term...

One thing I think a lot of people here aren't grasping. Physics / engineering / whatever science doesn't define our world, it describes it. In my thesis, I used three different theories of lift (momentum, pressure distribution and circulation). They don't contradict each other, one is not more correct than any other - they all describe the same phenomenon, just in different ways.

Does a flat plate fly? Yup - my foamies fly quite nicely!

[MarkusN]

This doesn't make any sense whatsoever. Drag is perpendicular to downwash, so how can one be created by the other? And downwash isn't created as a result of an imperfect wing, any three dimensional lifting surface will create downwash. How much? w~=L/(p.pi.b^2.V) That much. No drag term...

Then Ludwig Prandtl doesn't make sense whatsoever and we can stop buliding wings with high aspect ratio and lift distributions approximating an ellipse. Because wings with low aspect ratio create more downwash and therefore must create more lift.

Folks, read up on your basics.

I've stated it before and I will have to repeat it until I die:

Yes, the wing does chage momentum of air flowing over it to create lift: From +v (up) to -v (down). In a flow field that is perfectly described by potential theory (Bernoulli), which feeds the wing with an upwash and takes a downwash off it and flattens that out to no vertical movement. Apart from the imperfections left by finite aspect ratio, that is. Those leave a vortex field behind the wing that keeps moving until it dies down due to friction, taking energy away from the field.

The drag component, BTW, comes from the very downwash you think creates lift: That downwash tilts the flow over the wing. Tilting the lift vector by the same amount. Backwards. See a drag component now? (This is what Prandtl describes in his theory on induced drag.)

Sorry to sound so condescending, but I hate what happens right now. Hate, hate, HATE it. One terribly flawed theory replaced by an even more simplistic one. The world, and especially aerodynamics, aren't as simple as we'd like them to be.

If you don't believe me, maybe you believe NASA: http://www.grc.nasa.gov/WWW/K-12/airplane/foil2.html
(Even though even these guys intertwine induced drag and lift-generating downwash on one page, I leave it up to you to find it)

Or this guy:http://www.diam.unige.it/~irro/profilo_e.html

And if you read up on the guys who apparently started the whole mess: http://home.comcast.net/~clipper-108/lift.htm
you will notice that even they call the energy put into downwash 'induced drag'.


Oh and BTW: of course a flat plate creates lift. In a flow field that very well is described by Bernoulli's law.

[Odysis]

Downwash is required to create lift - the elliptical circulation distribution is there to create a constant spanwise downwash distribution. This is the most efficient because induced drag is not linear. Doubling the downwash doesn't double the induced drag.

In even a two dimensional flow field, it can be shown an aerofoil creates downwash. Unfortunately the mass term approaches infinity, so the downwash approaches zero. A result often misinterpreted as not being there. But any wing does create a finite downwash.

A great "real-world" example of downwash? Go stand under a helicopter.

And once again, I re-iterate. Downwash distributions don't create lift. Pressure distributions don't create lift. Vortex lines and sheets don't create lift. They are however tools we use to describe the interplay between all these things. Without downwash, there is no lift. Without lift, there is no downwash. Two sides of the same coin...

[MarkusN]

OK, apparently you know what you're talking about. That's not something one can expect on an internet forum. Also quite aparently (sorry, I ignored part of your first (?) post, I apologize for that) you have investigated these matters in great detail.

So what's your take on this?

I am very well aware that pressure distribution calculations, calculations on vorticity, and those looking at momentum balance are models describing the same phenomenon and arriving at very similar results.

I have racked my brain about this for hours, but I don't really have enough maths in my toolbox to come to a final conclusion.

A wing of infinite span in an inviscid fluid can be regarded as a perfect machine, right? Perpetuum mobile of the first order. So no net energy transfer is needed (or to be expected) from the wing to the fluid. But any net downwash in a two-dimensional field would have to be regarded as such.

Yet, to generate lift, we must have momentum change of the flow over the wing. No way around that. But the flow field is symmetrical (at least it is over the rotating cylinder; can't say for the transformed plane, that's where my rusty math toolbox comes in).
Still: for the wing of infinite span, isn't it so that the wing does not "throw air at the ground" as 2phar so aptly stated, but BACK at the ground? That there really is no net downwash, but rather a change of upwash to downwash, in a field that, due to its symmetry, takes the remaining momentum behind the wing out of the flow by remote actions of pressure. (Sorry for the clumsy language, English is not my first.)

Such a perfect wing is the energetical equivalent of a table carrying a weight. Or a piston with a perfect, frictionless seal holding a force against pressure behind it. (Yeah, both comparisons don't really fly (no pun intended)). Anyway, you get my drift: No energy being converted, no energy flow, just forces in equilibrum.

Coming from that model, to me the picture still makes sense that any net downwash is a result of imperfections of the real world, 3D system.

[Odysis]

Your analogy of a table carrying weight I think is very close to an aerofoil in an inviscid fluid. It doesn't actually do anything really. There is no drag, therefore the wing does no work. In this case, there is no downwash (or an infinitely small amount).

Using the potential flow field around a cylinder, a Joukowski (sp?) conformal mapping transform, and finding the circulation given by the Runja-Kutta (sp?) condition will yield the same result - no drag, but plenty of lift. The Runja-Kutta part is what gives us a particular lift for a given AoA, but still doesn't add any "real" effects, ie viscosity or finite span. This does give rise to an assymetric flow field. Because the aerofoil is assymetric, so must be the flow field.

So far, I agree with you entirely. In none of these flows is there a net downwash. Even in the transformed, asssymetric flow field, the upwash is equal to the downwash.

Things change a lot when you no longer have an infinite amount of air to play with. I guess if you call a finite wing an imperect wing, I'd agree. I think we were both looking at this from two different points of view, and using slighty different terms for the same thing.

Why is it a first year question "how does a wing create lift?" get harder to answer the more study and research you do?!?

[MarkusN]

Heh! My world is in equilibrum again.

The trailing edge condition is named after Kutta and Joukowski, by the way. Runge-Kutta is a method for numerical integration.

Things change a lot when you no longer have an infinite amount of air to play with. I guess if you call a finite wing an imperect wing, I'd agree. I think we were both looking at this from two different points of view, and using slighty different terms for the same thing.
Agree completely. My angle on things: I am often in the situation of having to explain polar data and their application to real world problems to users. Having a lift model that inherently intertwines two-dimensional and three dimensional flow does not help then.

Why is it a first year question "how does a wing create lift?" get harder to answer the more study and research you do?!?My thoughts exactly.

[Odysis]

The trailing edge condition is named after Kutta and Joukowski, by the way. Runge-Kutta is a method for numerical integration.
It's been a few years :rolleyes: All this is unfortunetly just of the top of my head - thanks for the clear-up!


Having a lift model that inherently intertwines two-dimensional and three dimensional flow does not help then.
Which model is that? And polars can be made up for wings just like aerofoils. That helps with the "real-world" part.

Odysis

[MarkusN]

Which model is that? And polars can be made up for wings just like aerofoils. That helps with the "real-world" part.
Well, a lift model that requires net downwash to be present. Whose effect I will later have to explain has to be added to the calculation as induced drag.

It's happened on the German Wikipedia. They went to great lengths explaining lift according to the "Plane throws Air at the ground" model. Calculating power required for this.

And then adding the drag components to the mix: Viscous drag, induced drag, wave drag. Adding induced drag twice in the process :(


Of course the final objective is to have a wing or even plane polar. But data has to be taken from polars for infinite span. So I'd better be able to explain the meaning / interpretation of those.

[2phar]

The calculus was invented to try to quantify fluid flow.
its a deep subject.
the fact that a parcel travels farther along the top surface than the bottom is strictly a function of the foils thickness- nothin else.
chris

[Odysis]

the fact that a parcel travels farther along the top surface than the bottom is strictly a function of the foils thickness- nothin else.
So flat-plate theory is wrong? I.e. Cl=2.pi.alpha?
Camber and angle of attack both change the flow. If it didn't, flaps wouldn't work, and a wing would change lift when the nose pitches up.

[vintage1]

The calculus was invented to try to quantify fluid flow.



No it was invented to solve many problems, Newton was particularly interested in mechanics. Others were interested in calculating areas of mathematically described,but odd shaped, objects.

http://en.wikipedia.org/wiki/Calculus

The calculus as applied to fluid flow is a fiendishly nasty three dimensional one known as tensor calculus.

Its so nasty there isn't a decent wiki explanation. I have to admit getting rather bored with calculus at that point, and deciding that it wasn't relevant to what I wanted to do, and more or less ignoring it completely.


its a deep subject.
the fact that a parcel travels farther along the top surface than the bottom is strictly a function of the foils thickness- nothin else.
chris

Don't understand that..its to do with lots of things, like the behaviour of viscous compressible fluids under pressure variations.

[markschaffin]

The flow around a rotating cylinder is not symmetric. With your rotating cylinder model (basic potential flow) the stagnation points (think leading edge and trailing edge (kutta)) move towards each other. It is this circulation that causes lift. When the stagnation points meet, the maximum lift has been achieved. This works out to be a lift coefficient of 4*pi, if I am remembering correctly.

You don't get lift from a wing/airfoil without circulation. You don't get circulation without viscosity. Without viscosity you don't get the Kutta condition at the trailing edge. When using potential flow methods, you have to force the Kutta condition to be met, i.e. you shed a wake at the trailing edge. This wake essentially represents the flow separation off the airfoil at the trailing edge. The same thing happens at the wing tip. One has to be very careful when one talks about inviscid flows. Most inviscid flow calculations have some aspect of "viscosity" in them, such as the forced Kutta condition in potential flow modelling, or artificial (numerical) viscosity in the more modern computational methods.

As for a flat plate wing/airfoil, the flow is probably separated and at high angles of attack the flow is probably dominated by a beautifully complex flow dominated by vortices being generated at the sharp leading edges. Again, circulation.

[MarkusN]

The flow around a rotating cylinder is not symmetric.

Symmetric about the vertical axis I meant. Should have been more precise about that.

I completely agree with everything you say, but I stand by my point: It is NOT necessary, to explain lift, to have net mass being "thrown at the earth". It does happen in all real world flow patterns, but it always shows up in the final balance as a LOSS to what we want to achieve.

[vintage1]

Indeed. It takes no power to maintain height, as any balloon knows.

Air 'thrown at the ground' represents a net increase in energy of the air mass, and therefore a net loss of power from the aircraft.


I remember also reading that without viscosity there would be no drag, but no lift either, or something.

I also remember very early on in maths and physics, realising how actually you could cancel terms in complex equations and get results like 'no drag = no lift'..that scientific models as such are only models..and you pick the one that makes the maths easy..

[markschaffin]

I am an aerospace engineer who spends a good part of his time doing wing design. If it is any comfort, this stuff is debated among the folks that are supposed to know all this stuff. Does it really matter what happens to the wake far downstream? What really matters is the effect on the wing. We can not possibly model completely the flow field about a real wing, and the minute you go to an infinite wing (2-D/airfoil), you have to start making assumptions. You can't even really test an airfoil in a wind tunnel as there will ALWAYS be three dimensional flow, and not only at the walls but down in the boundary layer and wake.

Also, beware of posting sites that seem above question when making these arguements. I know some well educated folks who violently disagree with sites like these. I myself usually try to stay out of these arguements.

Mark

[MarkusN]

I remember also reading that without viscosity there would be no drag, but no lift either, or something.
Yup. It's what Mark states as the Kutta condition: Without viscosity the flow would zip around the trailing edge and leave the profile at some point upstream, resulting in no circulation and no lift.

that scientific models as such are only models..and you pick the one that makes the maths easy..
Again, yes. But 'circulation corresponds to lift' is a well proven model. And the Newtonians can find their flow of momentum in the circulation as well, even without a downwash wake. Just as the Bernoullians can find their pressure differences. :D

[Texas Buzzard]

Then Ludwig Prandtl doesn't make sense whatsoever and we can stop buliding wings with high aspect ratio and lift distributions approximating an ellipse. Because wings with low aspect ratio create more downwash and therefore must create more lift.

Folks, read up on your basics.

I've stated it before and I will have to repeat it until I die:

Yes, the wing does chage momentum of air flowing over it to create lift: From +v (up) to -v (down). In a flow field that is perfectly described by potential theory (Bernoulli), which feeds the wing with an upwash and takes a downwash off it and flattens that out to no vertical movement. Apart from the imperfections left by finite aspect ratio, that is. Those leave a vortex field behind the wing that keeps moving until it dies down due to friction, taking energy away from the field.

The drag component, BTW, comes from the very downwash you think creates lift: That downwash tilts the flow over the wing. Tilting the lift vector by the same amount. Backwards. See a drag component now? (This is what Prandtl describes in his theory on induced drag.)

Sorry to sound so condescending, but I hate what happens right now. Hate, hate, HATE it. One terribly flawed theory replaced by an even more simplistic one. The world, and especially aerodynamics, aren't as simple as we'd like them to be.

If you don't believe me, maybe you believe NASA: http://www.grc.nasa.gov/WWW/K-12/airplane/foil2.html
(Even though even these guys intertwine induced drag and lift-generating downwash on one page, I leave it up to you to find it)

Or this guy:http://www.diam.unige.it/~irro/profilo_e.html

And if you read up on the guys who apparently started the whole mess: http://home.comcast.net/~clipper-108/lift.htm
you will notice that even they call the energy put into downwash 'induced drag'.


Oh and BTW: of course a flat plate creates lift. In a flow field that very well is described by Bernoulli's law.
>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<

BRAVO!

[JetPlaneFlyer]

Indeed. It takes no power to maintain height, as any balloon knows.

It takes no power to maintain height 'providing' your aircraft is as light as the air that surrounds it (like a balloon) :rolleyes:
For aircraft that are heavier than air to maintain height then lift must be generated and with lift inevitably comes drag (as you stated)... and to overcome drag needs thrust and to generate thrust needs power.

[vintage1]

It takes no power to maintain height 'providing' your aircraft is as light as the air that surrounds it (like a balloon) :rolleyes:
For aircraft that are heavier than air to maintain height then lift must be generated and with lift inevitably comes drag (as you stated)... and to overcome drag needs thrust and to generate thrust needs power.


Well. yes, but we were talking perfect worlds..

Oh thrust doesn't take power to generate at all. Ask my backside on the chair..

Only when something moves as a result..

[JetPlaneFlyer]

Well. yes, but we were talking perfect worlds..



Even in a 'perfect word' to generate lift must produce drag... is that what the previous discussion in this thread concluded? ie 'no drag = no lift' (quote Vintage1)....
And to generate thrust (in a moving aircraft) does indeed need power.

Or have I misread something :confused:

[vintage1]

Even in a 'perfect word' to generate lift must produce drag... is that what the previous discussion in this thread concluded? ie 'no drag = no lift' (quote Vintage1)....
And to generate thrust (in a moving aircraft) does indeed need power.

Or have I misread something :confused:

A few missing smileys.. :D :D :D

Its just the general lack of understanding that 'staying up in the air' doesn't actually take power by virtue of being there, and nor does generating thrust.

Given a low enough gear ratio, an IPS motor can lift an elephant..albeit rather slowly.



And the total confusion between 'power' and 'thrust' that leads to so much incomplete understanding.

I mean most of this thread has been about how aircraft DON'T fly, not how they do.

I think we should try to be pedantic and exact, if we want to help.
So thank you for rhe bit I have reddened above..

[JetPlaneFlyer]

Of course in an aircraft that was not moving it would take no power to generate thrust... However in an aircraft that was not moving there would be no drag so you would need no thrust anyway... There would also be no lift and no flight... so the whole issue is splitting hairs as far as I can see.

Back to the original issue... Is it not true that to maintain level flight in a heavier than air craft does require input of some power to overcome induced drag which is the inevitable 'byproduct' of lift?... even in a 'perfect' frictionless world?

Or is this not the case?

Steve

[Odysis]

Steve,
The "perfect" vs "real-world" discussion that was held earlier was two dimensional vs three dimensional.
An aerofoil does not induce downwash; a wing does. In an inviscid flow field, there is no profile drag; in a viscous one there is.
How simple do you want to make your math? In the simplest case (but still useable) we use a two-dimensional, inviscid flow, with a circulation added artificially to meet the Kutta condition. In the most complex case, you solve the Navier Stokes equations.

So in a "perfect" world, defined in this case by a 2d inviscid flow, there is no drag. But we don't have inviscid fluids, and we live in at least three dimensions, so this "perfect" case is an imaginary simplification. Engineers do this all the time, as sometimes Cl=2.pi.alpha is close enough. Sometimes it's not.

As we got into earlier, confusing 2d and 3d makes things very different.

[nmasters]

But we don't have inviscid fluids
Well... not normally, on this planet :p

[Odysis]

Well... not normally, on this planet :p
Yeah, but my poor little parkies become really hard to see once they leave orbit :D

[JetPlaneFlyer]

Steve,
The "perfect" vs "real-world" discussion that was held earlier was two dimensional vs three dimensional.

I take your point, it all depends on how 'perfect' your world is. I was thinking about a 'perfect' 3D world where wings are not infinite.

As I read earlier; even in the perfect 2D world there is downwash, but it's infinitely small (because the air mass is infinitely large)... but infinitely small still exists...

[vintage1]

I have a sneaking suspicion that without viscosity (and hence drag) wings wouldn't work at all.

[MarkusN]

Oh thrust doesn't take power to generate at all. Ask my backside on the chair..

Only when something moves as a result..
Very good remarks. Also, you can make a bench out of the plank that was nailed to my forehead.

I think I am a convert. What I could not wrap my mind around was that the very downwash that creates induced drag is the mass carrying the required counter-momentum for lift.

This coming from the observation that in the two-dimensional problem there is no downwash (not even infinitesimally small downwash, ask the rotating cylinder. Also, because the mass of air observed in the two dimensional problem is in similar relation to the lift as in real-world problems, there would be required that finite downwash was present, if this indeed were the mechanism at work). So if lift is possible without downwash in two dimensions, the same effect must be present in three, right?

Wrong. The two dimensional problem is a static one. The wing is upheld by pressure alone, and that pressure has equally large static conterparts. In the same way a piston with a load is held against a force in a cylinder. No mass acceleration there, no Newton's second law, only the third one. The pressure is kept in the system by the wing extending into infinity. Or by the sidewalls of the wind tunnel. Just like vintage's backside on the chair.

This changes when you move into three dimensions. The problem of enclosing that pressure switches from a static to a dynamic one. Now you must accelerate masses to create those forces enclosing the pressure system around the wing. And the character of the problem changes from piston in cylinder analogy to rocket on a jet of fire...

Will have to do some calculations to verify, but, as stated above, I think I'm a convert. (But Anderson / Eberhardt still have part of it backwards. They'd better leave the part of the field that is explained by the bound vortex alone, i.e. upwash and the two-dimensional part of downwash.)

[poulsbobill]

I am not sure whether this question applies here...but here goes. Is the above discussion the reason why a plane with an airfoil here ( http://i35.photobucket.com/albums/d191/H8SUVS/DSC04478.jpg) flies slower than a comparable plane with a flat plate? Same wing loading/cubic loading. I dont get why the plane i have with the airfoil flies so much slower then the same plane (really close) with a flat plate?

Thanks,

Bill

[pmackenzie]

More camber means higher maximum lift coefficient, so you can fly slower.

You can play around with this sort of thing here (http://adamone.rchomepage.com/foil_sim.htm)

Pat MacKenzie

[Odysis]

Bill,
The speed that an aircraft can fly at is determined by the lift it can create. Lift must equal weight, so:
weight = Cl * 0.5 * density * V^2 * Area
So, as the Velocity term decreases, one (or more) of the other terms must increase. 0.5 is constant, as is (for this discussion) density, so that leaves Cl and Area. By increasing area (less wing loading), we can slow down, or by increasing Cl.

At the slowest, we're only interested in Cl (max). For a flat plate, this occurs much earlier than for an aerofoiled wing. This is why for the same area, an aerofoiled wing can produce a higher Cl, hence fly slower.

[JetPlaneFlyer]

I dont get why the plane i have with the airfoil flies so much slower then the same plane (really close) with a flat plate?


Define slower... slower maximum speed or slower stall speed?... or both?

[poulsbobill]

Define slower... slower maximum speed or slower stall speed?... or both?
I mean slower speed before stall. ie indoor flight

pmackenzie,

Thank you That is cool

odysis

thank you. Will not increasing the AOA do the same thing?

Bill

[pmackenzie]

I mean slower speed before stall. ie indoor flight

pmackenzie,

Thank you That is cool

odysis

thank you. Will not increasing the AOA do the same thing?

Bill

If you go to the airfoil simulation link you can change it from "ideal flow" to "stall model"
This will better show what happens as AoA is increased.
If you did not find it you can click on the "SHAPE" buttion and turn the thickness down to zero and also play with the camber.

Pat Mackenzie

[BMatthews]

The Foilsim deal stalls at around 10 degrees which is pretty typical for larger size reynolds numbers. That is one part that it doesn't model very well as even little 3 inch wings still show the stall at 10 degrees. For "us" the stall angle is more like 5 to 6 for smaller wings and perhaps 6 to7 for larger ones and maybe 8 for the big giant scale stuff.

[JetPlaneFlyer]

Will not increasing the AOA do the same thing?

Bill
Up to a point increasing AoA does increase lift but increase it too far and you stall.
A 'proper' cambered airfoil will produce more lift for the same AoA and also be capable of operating at a higher AoA before stalling... That's why you observe that the 'proper' airfoil equiped model is capable of slower flight than the flat plate one.

[Texas Buzzard]

Up to a point increasing AoA does increase lift but increase it too far and you stall.
A 'proper' cambered airfoil will produce more lift for the same AoA and also be capable of operating at a higher AoA before stalling... That's why you observe that the 'proper' airfoil equiped model is capable of slower flight than the flat plate one.

......................

You are correct. That is why the last two scratch builts have had a Kline-Fogelman section.

[lukibob]

Which one allows less deflection at zero angle of attack from a force perpendicular to forward movement through air?

I speculate that it could depend on the velocity of the forward movement.

For low airspeed, less than 20mph, flat plate might be better? Greater than 20, symmetric airfoil?

Something to ponder. Please comment with experience and theory.

Matt

[JetPlaneFlyer]

Which one allows less deflection at zero angle of attack from a force perpendicular to forward movement through air?


:confused: Matt, I 'm not really sure what you are asking here? At zero angle of attack neither a flat plate nor a symmetrical airfoil will produce any lift at all so would provide zero resistance from any force perpendicular to the direction of travel... and BTW a flat plate IS a symmetrical airfoil