No one can explain WHY planes fly...

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

PagoBay

Well-Known Member
Joined
Nov 16, 2013
Messages
71
Location
US Territory of Guam
How Wings Produce Lift - Professor Holger Babinsky - University of Cambridge

Very clear explanation and debunking of popular explanations including the Bernoulli model.

(Upper and Lower surface - Arriving same time theory debunked)

Slides from the Lecture for those who like details:

Actual 55 minute lecture

Published Articles:
https://www.aerosociety.com/media/4841/babinsky-s-demonstration-the-theory-of-flight-and-its-historical.pdf
and here:
http://www3.eng.cam.ac.uk/outreach/Project-resources/Wind-turbine/howwingswork.pdf
 
Last edited:

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
12,826
Location
Port Townsend WA
That last video would have been useful if the speaker had poured the glass of water on the camera operator. He made a flawless video of the back of his head.
 

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
2,934
Location
Grand Junction, Colorado
See Post #20 video. I agree we need a better video in English. Search the internet.
Search YouTube for "starting vortex". I didn't find anything like Prandtl's original film but some college students try to approximate it with stuff in their dorm kitchens. There are also some CFD symulations. None with helpful voice or text. Here's a brief verbal explanation of the starting vortex.
 

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
2,934
Location
Grand Junction, Colorado
It's a little easier to see what's going on in this CFD clip than the original aluminum powder on water film.

No. Stall is a viscosity effect. Bernoulli's theorem doesn't have anything to say about viscosity, just speed and pressure outside the boundary layer. Not sure about Prandtl but I don't think he quite got to predicting when and where transition and separation would occur. Both of those are really hard problems. The NavierStokes equations can do it but that's super computer territory (or a couple of days to solve for a simple case on a high end desktop). Panel codes use a shortcut to estimate the thickness of the BL and where it will transition to turbulent but separation is still not well modeled.


You seem to be confounding turbulence with stall. Stall is separated flow, a turbulent boundary layer resists separation from the surface better than a laminar BL but it also creates a lot more friction drag. When a boundary layer leaves the surface it separates the potential flow of the atmosphere from a small slug of air between itself and the wing. This isolated slug of air is turbulent but on average almost stationary relative to the wing so creates little friction but it's also low pressure so it creates pressure drag (low pressure on aft facing surfaces sucks in the wrong direction).

Here's a video of Prandtl's original experiment showing circulation. The little vortex that gets shed from the trailing edge is what actually starts the circulation by pulling on the upper surface flow with the low pressure at its core thus accelerating the upper surface flow. After that it's a self sustaining feedback cycle. A new starting vortex will be shed every time the plane pitches up.

 

PagoBay

Well-Known Member
Joined
Nov 16, 2013
Messages
71
Location
US Territory of Guam
That last video would have been useful if the speaker had poured the glass of water on the camera operator. He made a flawless video of the back of his head.
Yep, there are many comments about that camera operator. But the audio is clear, so that helps. Plus the slides were added later. I am wondering if the Bernoulli explanation for airfoil lift will someday bite the dust. Kinda think that will take a few more decades?
 

mcrae0104

Armchair Mafia Conspirator
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
3,147
Location
BDU, BJC

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
2,934
Location
Grand Junction, Colorado
Vorcity creates pressure gradients and flow variations across the wind tunnel test section. Wings producing lift introduce vorticity into the flow field. The vortex theory of lift describes this flow field and the creation of lift.

See Post #20 video. I agree we need a better video in English. Search the internet.
Part two of this Vorticity lecture uses clips from Prandtl's film to illustrate the part of the lecture about lift and goes into more detail. Sorry I was only thinking about the starting vortex before and only searched for that term. It's only a subset of Vorticity so searching for that yealded narrow results.
 

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
12,826
Location
Port Townsend WA
If you impose a few rules on the cars it may make a bit more sense.

No penetration, the cars all have to stay on the road.
No voids, the cars have to stay on the road and the passengers in the cars can't get too far apart ie they have to maintain approximately the same average spacing at all times.
The speed of individual passengers in the cars is constant regardless of the bulk flow.

Now think of each car as a slug of fluid and the molecules as the passengers. As the car approaches the bottleneck it must stretch out and get thinner in order to maintain the its volume. The passengers are still getting bounced around randomly at the same speed but now their movement is more restricted across the flow but less restricted in the direction of flow and since the no penetration rule doesn't apply to them (they must be in a car but not necessarily the same car) some of them are being thrown from the cars (small scale turbulence).

Now replace some of the passengers with smoke particles in a wind tunnel. When the smoke is pulsed the smoke line shows the length of a section of the stream tube (a car) and the spacing between lines shows the width. As the tube gets stretched it gets narrower and the smoke lines get closer together. Since the length of a given volume of the tube is longer in the bottleneck, and the individual partials must maintain their speed, the drift in the bulk direction is greater than the drift in the cross flow direction.

There... the Walt Disney version of flow acceleration around a curved surface at low Mach numbers.
After watching the Lippisch Secret of flight videos, I think I see what you are getting at here.
The question is why does the air increase in speed when going around the curve? Lippisch did not address this question.
I think the answer is that air molecules can indeed stretch out by making the space greater between them and they speed up. It's because each molecule is sort of connected to all the nearby molecules as if by springs*. The springs can stretch or shrink some but don't break free unless a strong force breaks them and that makes a stall or eddy.
So the air from far above is connected by springs and is dragging these molecules all the way down near the wing surface and making them speed up.
Does that make sense?
If cars on the freeway were all on a controller autopilot, the computer controlling all of them could speed them up as they near a bottleneck and fit more cars in between them and then slow them back down on the other side.

* this idea of springs connecting air molecules was from my old Richard Feynman lecture on cassette tapes, I think. Might try to dig them out.
edit, Here: https://www.feynmanlectures.caltech.edu/I_01.htmlc
 
Last edited:

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
2,934
Location
Grand Junction, Colorado
After watching the Lippisch Secret of flight videos, I think I see what you are getting at here.
The question is why does the air increase in speed when going around the curve? Lippisch did not address this question.
I think the answer is that air molecules can indeed stretch out by making the space greater between them and they speed up. It's because each molecule is sort of connected to all the nearby molecules as if by springs*. The springs can stretch or shrink some but don't break free unless a strong force breaks them and that makes a stall or eddy.
So the air from far above is connected by springs and is dragging these molecules all the way down near the wing surface and making them speed up.
Does that make sense?
If cars on the freeway were all on a controller autopilot, the computer controlling all of them could speed them up as they near a bottleneck and fit more cars in between them and then slow them back down on the other side.

* this idea of springs connecting air molecules was from my old Richard Feynman lecture on cassette tapes, I think. Might try to dig them out.
edit, I think is it https://www.feynmanlectures.caltech.edu/I_01.html
No. There is no significant attractive force between neutral molecules. The springs that Feynman was talking about are the strong nuclear force which only affects the nucleus, not the bulk flow. The only interaction between molecules of a neutral gas is collisional. To contenu the traffic analogy imagine that the inside car hits a pothole (the adverse pressure gradient) and slams into the car next to it and the debris from that wreck takes out Cars in several adjacent lanes. The no leaving the cars ruel still applys to the passengers so now they're running around in the wreckage with very little average speed relative to the wall while the cars in lanes far from the wall that aren't completely blocked by debris can Swerve around it and contenu relatively unimpeded.
 
Last edited:

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,730
I'm sick of people dissing Bernoulli. If something needs to be debunked, how about incorrect understandings of Bernoulli?
 

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
2,934
Location
Grand Junction, Colorado
I'm sick of people dissing Bernoulli. If something needs to be debunked, how about incorrect understandings of Bernoulli?
Nothing wrong with Bernoulli's theorem except that it can't be used in a cause and effect argument because it's reversable. It also says nothing about viscosity which is a property of all fluids at normal temperatures. Euler's formulas do an OK job of describing the flow around an airfoil except that they have the flow separate at a point forward of the trailing edge (it doesn't satisfy the Kutta-Joukowski condition) and no lift is produced. Adding an arbitrary circulation term (Gama) to Euler to push the separation point to the trailing edge solves the separation before the TE problem and lift is equal to Gama. Vorticity is what stretches out the flow lines and in real fluids viscosity is what creates Vorticity.
 
Last edited:

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
2,934
Location
Grand Junction, Colorado
=only (neglectable) gravitational forces !
When two molecules bounce off eachother at 750 mph their velocity is much higher than required to ensure that gravity will never pull them back together. The mass of the wing isn't even enough to pull them back. The mass of the Earth is the only gravitational influence of importance here. Earth's gravity is what creates pressure.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,730
Norm:
Of course you're right. Expecting Bernoulli's to completely describe the behavior of a viscous fluid probably counts as an incorrect understandings. One ignores the prof's numerous qualifiers at one's peril. I seem to recall that there were usually at least five, but the ones I remember are inviscid, irrotational, and only on Sundays.

It's been a long time since I've gone through the math.
 
2
Group Builder
Top