Theories of Flight.
Thread Starter
Join Date: Apr 2002
Location: somewhere else
Posts: 16
Likes: 0
Received 0 Likes
on
0 Posts
Theories of Flight.
When I was learning to fly, which was quite a while ago, I was told Bernoullis theory explained lift. However, for some time now it has been apparent that his theory (and it's still a theory) is wrong. It was developed by studying incompressible fluids in pipes. As explained at http://wright.grc.nasa.gov/WWW/K-12/...ne/right2.html Newtons second Law of Motion can be used to explain lift.
The next page (arrow on bottom of page) explains why Bernouillis Theorem is wrong. Is anyone still being taught this theory of flight, and if so, why? Or is this heresy?
Amazing that, a hundred years on, we're still not sure, 100%, why wings do fly.
The next page (arrow on bottom of page) explains why Bernouillis Theorem is wrong. Is anyone still being taught this theory of flight, and if so, why? Or is this heresy?
Amazing that, a hundred years on, we're still not sure, 100%, why wings do fly.
I have an aeronautics degree, in the course of which I learned 3 or 4 different explanations of how an aeroplane flies. They were at-least to some extent contradictory, but also all worked in certain circumstances.
Bernoulli is a vital part of an aeronautical engineer's toolbox, but is only one of various principles used. The important question is "what do I want to know". That might be how much lift, how much drag, what will the effects of alpha / beta / flap / slat / bank angle / etc. be. When presented by such a question, you pick the theory that answers the question best.
For pilots, Bernoulli's theorem has worked very well for years, because they don't need an ability to accurately predict aircraft behaviour - they need a reasonably feeling for why the aircraft flies and what will stop it. It's not heresy, it's an adequate working model.
In the meantime, bear in mind what a scientific or engineering theory is. It is the best model, to predict what is happening, for a given situation. TRUTH is a concept that scientists and engineers don't work in, they work in good and bad models, and in probabilities. TRUTH we leave to priests who deal in unprovable certainties, in science and engineering certainties are usually unprovable. If you meet a scientist or Engineer who offers you a clear certainty of anything, distrust them, they are a poor member of their profession.
G
Bernoulli is a vital part of an aeronautical engineer's toolbox, but is only one of various principles used. The important question is "what do I want to know". That might be how much lift, how much drag, what will the effects of alpha / beta / flap / slat / bank angle / etc. be. When presented by such a question, you pick the theory that answers the question best.
For pilots, Bernoulli's theorem has worked very well for years, because they don't need an ability to accurately predict aircraft behaviour - they need a reasonably feeling for why the aircraft flies and what will stop it. It's not heresy, it's an adequate working model.
In the meantime, bear in mind what a scientific or engineering theory is. It is the best model, to predict what is happening, for a given situation. TRUTH is a concept that scientists and engineers don't work in, they work in good and bad models, and in probabilities. TRUTH we leave to priests who deal in unprovable certainties, in science and engineering certainties are usually unprovable. If you meet a scientist or Engineer who offers you a clear certainty of anything, distrust them, they are a poor member of their profession.
G
The next page (arrow on bottom of page) explains why Bernouillis Theorem is wrong.
Bernoulli's theorem simply relates the pressure in an airflow to its velocity. It's correct for low-speed inviscid flow, which is a model that has been producing very consistent and accurate predictions of aerodynamic quantities for 90 years.
The air flowing over the top of the lifting wing is moving faster than that flowing below it, and the pressure is therefore lower. The fallacy is in the explanation for why the air over the top is moving faster.
You may not be sure why wings fly, but aerodynamicists and engineers have been for most of the last century!
Thread Starter
Join Date: Apr 2002
Location: somewhere else
Posts: 16
Likes: 0
Received 0 Likes
on
0 Posts
It's not that I am unsure of why wings fly: I've read quite a bit on the subject. The whole thing is very well explained at http://wright.grc.nasa.gov/WWW/K-12/...e/bernnew.html
What I was trying to point out is that very few pilots have heard anything beyond Bernoullis' Theorem. In three years sim training, I have only come across two pilots who'd ever heard of any other explanation.
I was merely making the point that this theorem appears to be used to make people learning to fly think that they understand the principles of aerodynamics. I was certainly never told of any Newtonian explanation when I wrote my ATPL exams. Surely this should be part of the curriculum. Or is it taught on a "need to know" basis? As long as pilots believe it will fly, they don't need to really understand why?
What I was trying to point out is that very few pilots have heard anything beyond Bernoullis' Theorem. In three years sim training, I have only come across two pilots who'd ever heard of any other explanation.
I was merely making the point that this theorem appears to be used to make people learning to fly think that they understand the principles of aerodynamics. I was certainly never told of any Newtonian explanation when I wrote my ATPL exams. Surely this should be part of the curriculum. Or is it taught on a "need to know" basis? As long as pilots believe it will fly, they don't need to really understand why?
Last edited by CaptainCargo; 12th May 2002 at 10:56.
Flatthatt, if you go and find the bookshop at any University with an aeronautical Engineering department, you'll find a selection of textbooks with titles along the lines of "aerodynamics for engineers", most of which are a bit excessively mathematical but will usually cover a fair selection of theories.
When I was a student the standard book was "Aerodynamics for Engineering Students", by E.L. Houghton and N.B. Carruthers. I have my 10 year old copy in front of me which lists the following in the contents page...
- Bernoulli (2d inviscid flow theory)
- 2D viscous flow theory
- Finite aerofoil theory (using the simplified horseshoe vortex model)
- Estimation of pressure distributions from boundary layer theory.
- Mach number effects and supersonic flow / thin wing aerodynamics.
(so that's five methods, I'd forgotten the BL method, which is usually only used for drag estimation rather than lift).
Interestingly I also recently acquired a 1930 reprint of a 1926 textbook by Glauert called "aerofoil and airscrew theory" which has Bernoulli's theory on page 10, basic aerofoil theory (the 2D viscous flow method) on page 117 and 3D viscous flow on page 125. He also lists the same two methods (momentum method and blade element method) as Houghton for how propellers work. What I really love about Glauert's book however is the separate chapters on monoplane and biplane aerofoil effects, the latter is sadly missing from most modern textbooks.
Hope that help, it's also fascinating that 3 out of the 5 methods I learned in the 80s and 90s were already published in the 20s.
G
N.B. Bookworm, Engineers and Scientists are most certainly not sure about why wings fly. We are pretty confident that we can predict whether they will and how well, but as to why - no. I asked this in my office (containing four qualified aeronautical Engineers from CEng to HNC) recently and nobody was prepared to say that they did.
When I was a student the standard book was "Aerodynamics for Engineering Students", by E.L. Houghton and N.B. Carruthers. I have my 10 year old copy in front of me which lists the following in the contents page...
- Bernoulli (2d inviscid flow theory)
- 2D viscous flow theory
- Finite aerofoil theory (using the simplified horseshoe vortex model)
- Estimation of pressure distributions from boundary layer theory.
- Mach number effects and supersonic flow / thin wing aerodynamics.
(so that's five methods, I'd forgotten the BL method, which is usually only used for drag estimation rather than lift).
Interestingly I also recently acquired a 1930 reprint of a 1926 textbook by Glauert called "aerofoil and airscrew theory" which has Bernoulli's theory on page 10, basic aerofoil theory (the 2D viscous flow method) on page 117 and 3D viscous flow on page 125. He also lists the same two methods (momentum method and blade element method) as Houghton for how propellers work. What I really love about Glauert's book however is the separate chapters on monoplane and biplane aerofoil effects, the latter is sadly missing from most modern textbooks.
Hope that help, it's also fascinating that 3 out of the 5 methods I learned in the 80s and 90s were already published in the 20s.
G
N.B. Bookworm, Engineers and Scientists are most certainly not sure about why wings fly. We are pretty confident that we can predict whether they will and how well, but as to why - no. I asked this in my office (containing four qualified aeronautical Engineers from CEng to HNC) recently and nobody was prepared to say that they did.
Last edited by Genghis the Engineer; 12th May 2002 at 11:29.
Join Date: Feb 2002
Location: oxford
Posts: 247
Likes: 0
Received 0 Likes
on
0 Posts
I have decided to remove my original post from this slot, beause my argument over-simplified the case. There is some pretty sophisticated engineering further downstream on this thread and my simple model does not stand up to that sort of level of analysis.
Oxford Blue
Oxford Blue
Last edited by oxford blue; 17th May 2002 at 14:43.
N.B. Bookworm, Engineers and Scientists are most certainly not sure about why wings fly. We are pretty confident that we can predict whether they will and how well, but as to why - no. I asked this in my office (containing four qualified aeronautical Engineers from CEng to HNC) recently and nobody was prepared to say that they did.
That's tautologous. By that standard no one can ever say "why" anything in nature is the way it is, we can only explain it in a quantitative and predictive way in terms of other models.
If you see an apple fall to the ground, can you say "why" it fell? You can certainly predict whether it will and how fast, but "why"? That would seem to require the mind of God.
Low-speed aerodynamics is as well understood as any other area of applied physics, and to maintain that science has been baffled for 100 years by the nature of lift is simply daft!
Join Date: Mar 2001
Location: Florida, USA
Posts: 632
Likes: 0
Received 0 Likes
on
0 Posts
Don't mean to hijack the thread - but whilst engineers are talking about information given to student pilots...I asked recently about the "engineering" view on "How many Forces" act on an aircraft in flight.
The purists seem to be starting to say Three Forces, not four...as taught to pilots.
Only answers I got were from pilots - who all said "four" - anybody disagree?
The purists seem to be starting to say Three Forces, not four...as taught to pilots.
Only answers I got were from pilots - who all said "four" - anybody disagree?
CaptainCargo
I apologise if I misunderstood the motivation for your post. I did not interpret it from your original message. If your point is that this is an aspect of science that is both difficult and generally poorly taught, I wholeheartedly agree.
Before you start the campaign to strike out mention of Bernoulli's theorem in any elementary texts though, I would caution you to consider carefully the logic of the arguments presented.
The super little Java simulator in one of your links demonstrates very nicely that the flow pattern around an aerofoil is predictable. Once you know the flow pattern, you can calculate the lift. I'd be very surprised if the FoilSim applet does much integration of momentum fluxes in a Newtonian way -- more likely it uses Bernoulli's theorem to calculate the pressure at each surface knowing the velocity (you'll see they calculate the pressure coefficient).
The key part of the "why" that is so difficult to explain simply is "Why is the flow pattern the way it is?" The physicist's explanation is that it is calculable from the governing equations and the boundary conditions -- just do the sums.
But that's very unsatisfying for the layman, who doesn't want to be told, for example, that gravity obeys a relatively simple set of field equations if you'd just solve them for the earth and the moon, but rather prefers to be given Keppler's laws that describe the motion of the planets and satellites. So we're left with a need to give a simple explanation of why the flow pattern around a wing looks the way it looks.
The Equal Transit Time" assertion (pleeeeeaaaaase don't call it "Bernoulli's theorem") is one little untruth designed to do just that. It gets the right result for the wrong reason.
So knowing that it's untrue, you understandably reach out for an alternative. How about "air gets deflected downwards"? That seems to be the gist of the "turning flow" argument that you want to prefer. But does that explain why so much airflow changes direction around the wing? Before you say "yes, it's obvious", take a look at the Skipping Stone theory, put your hand on your heart and tell me that this is not the picture at the back of your mind when you say "of course the air is deflected downwards". But as an explanation, it's just as wrong as the first one.
I have to admit defeat here. I cannot come up with a good intuitive non-circular explanation of why the air moves faster over the top of a lifting aerofoil than over the bottom, or why a huge tranche of air is deflected in its vicinity, at least not one that stands up to detailed inspection. The best I can do is along the lines of asserting that the flow must be parallel to the trailing edge, and the only allowable flow that fits the symmetry of the situation is something that circulates around the aerofoil. But that's more of a qualitative explanation of the way the maths works than an explnation in its own right. Even so, the inability to provide a simplistic explanation is not, IMHO, the same as saying that we don't understand something. For anyone who wants to, the theory is there, and it's been perfectly well tested and proven over a hundred years of practial flight.
I apologise if I misunderstood the motivation for your post. I did not interpret it from your original message. If your point is that this is an aspect of science that is both difficult and generally poorly taught, I wholeheartedly agree.
Before you start the campaign to strike out mention of Bernoulli's theorem in any elementary texts though, I would caution you to consider carefully the logic of the arguments presented.
The super little Java simulator in one of your links demonstrates very nicely that the flow pattern around an aerofoil is predictable. Once you know the flow pattern, you can calculate the lift. I'd be very surprised if the FoilSim applet does much integration of momentum fluxes in a Newtonian way -- more likely it uses Bernoulli's theorem to calculate the pressure at each surface knowing the velocity (you'll see they calculate the pressure coefficient).
The key part of the "why" that is so difficult to explain simply is "Why is the flow pattern the way it is?" The physicist's explanation is that it is calculable from the governing equations and the boundary conditions -- just do the sums.
But that's very unsatisfying for the layman, who doesn't want to be told, for example, that gravity obeys a relatively simple set of field equations if you'd just solve them for the earth and the moon, but rather prefers to be given Keppler's laws that describe the motion of the planets and satellites. So we're left with a need to give a simple explanation of why the flow pattern around a wing looks the way it looks.
The Equal Transit Time" assertion (pleeeeeaaaaase don't call it "Bernoulli's theorem") is one little untruth designed to do just that. It gets the right result for the wrong reason.
So knowing that it's untrue, you understandably reach out for an alternative. How about "air gets deflected downwards"? That seems to be the gist of the "turning flow" argument that you want to prefer. But does that explain why so much airflow changes direction around the wing? Before you say "yes, it's obvious", take a look at the Skipping Stone theory, put your hand on your heart and tell me that this is not the picture at the back of your mind when you say "of course the air is deflected downwards". But as an explanation, it's just as wrong as the first one.
I have to admit defeat here. I cannot come up with a good intuitive non-circular explanation of why the air moves faster over the top of a lifting aerofoil than over the bottom, or why a huge tranche of air is deflected in its vicinity, at least not one that stands up to detailed inspection. The best I can do is along the lines of asserting that the flow must be parallel to the trailing edge, and the only allowable flow that fits the symmetry of the situation is something that circulates around the aerofoil. But that's more of a qualitative explanation of the way the maths works than an explnation in its own right. Even so, the inability to provide a simplistic explanation is not, IMHO, the same as saying that we don't understand something. For anyone who wants to, the theory is there, and it's been perfectly well tested and proven over a hundred years of practial flight.
Thread Starter
Join Date: Apr 2002
Location: somewhere else
Posts: 16
Likes: 0
Received 0 Likes
on
0 Posts
The skipping stone theory is obviously flawed, as it does not take into account the flow over the upper surface. Neither does the equal transit time theory work, as it does not explain
symettrical aerofoils.
I'm not saying ditch Bernouillis'Theorem, as it does work, and can be used to predict the lift generated by an aerofoil. I just find it a bit strange that the Newtonian explanation has not found its way into the piliots curriculum, at least it hadn't when I last saw a copy of the ATPL syllabus. Bernouillis' Theorem is taught as the only explanation of lift, when it is, in fact, exactly as its name suggests, a theory.
As I said in my first post, and genghis concurred,no-one is really sure what the correct explanation for lift is, they just have models that (usually)work. I think pilots should be taught this.
symettrical aerofoils.
I'm not saying ditch Bernouillis'Theorem, as it does work, and can be used to predict the lift generated by an aerofoil. I just find it a bit strange that the Newtonian explanation has not found its way into the piliots curriculum, at least it hadn't when I last saw a copy of the ATPL syllabus. Bernouillis' Theorem is taught as the only explanation of lift, when it is, in fact, exactly as its name suggests, a theory.
As I said in my first post, and genghis concurred,no-one is really sure what the correct explanation for lift is, they just have models that (usually)work. I think pilots should be taught this.
Last edited by CaptainCargo; 12th May 2002 at 18:11.
I'm not saying ditch Bernouillis'Theorem
To say that it's "just a theory" is all very well, but as Genghis conjectures, everything in science is "just a theory". Next time you want to find the length of the hypotenuse of a right-angled triangle, be careful when applying Pythagoras's Theorem -- it is, after all, "just a theory".
Ultimately, all the mechanics we do is derived from Newton's Laws of motion -- even Bernoulli's theorem. The Newtonian approach that you cite is fine as far as it goes, but I have two issues with it:
1) It's just another way of calculating the lift once you know the flow field. It's not a very direct one.
Imagine I placed a watepaper bin in a corridor (in a very litter conscious office
) and wanted to know the weight of the contents at the end of the day. Compare the following methods:a) Weigh every person entering the corridor during the day as they enter. Weigh every person leaving the corridor during the day as they leave. Subtract the totals and deduce the weight of the contents added to the bin during the day.
b) Weigh the contents.
In principle, both are possible and correct. Applying Newton's Second Law (using the Momentum Theorem) to find the lift of a wing is a bit like method a). It's good to know it works, and perhaps it adds some insight to the understanding of the problem. But if you know the speed of flow at each point over the surface of the wing, why not use it directly to find the pressure?
2) It often leads to some incorrect assertions. For example, it's common to hear that the "lift of the wing is equal to the rate at which air momentum is tranfered downwards" or worse that "an aerodynamic force can only be generated by changing the momentum of air". Neither give a very complete picture of the physics -- for example in ground effect you have to add in the increased pressure on the ground as well as any momentum changes. Even in free flow, you have to be very careful about the way you add up the momentum to make it work. I note, BTW, that the Lift from Turning Flow page avoids any of these horrors.
I'm sorry to rant. I'm just a bit fed up with seeing a load of "Shock horror -- Established Aerodynamics Proved Wrong" headlines from people who don't understand the physics. I appreciate that yours was a genuine attempt to highlight some important shortcomings in the way pilots are taught aerodynamics. I hope you've enjoyed the debate.
Join Date: Feb 2002
Location: Anywhere but Ireland...
Posts: 112
Likes: 0
Received 0 Likes
on
0 Posts
Ah, Genghis was right...
Guys,
Check out this link... extremely detailed but easy to understand.
http://www.monmouth.com/~jsd/how/htm/airfoils.html
Nial
Check out this link... extremely detailed but easy to understand.
http://www.monmouth.com/~jsd/how/htm/airfoils.html
Nial
Do a Hover - it avoids G
Join Date: Oct 1999
Location: Chichester West Sussex UK
Age: 91
Posts: 2,206
Likes: 0
Received 0 Likes
on
0 Posts
Oxford
Sorry if it seems to cut across your ideas, but Mr Bernoulli actually has no trouble with generating lift regardless of the wing cross section or any airframe related angle of incidence. It may help you to visualise why if you consider the task of carrying a sheet of 8x4 chipboard (or MDF if you are richer than some) from the B&Q door to your car in a strong wind. Edge on it is not very aerodynamic and easy to carry. At an angle of attack (AOA) of say 20 deg you might just find it leaves your sweaty grasp.
The reason that Mr Bernoulli’s explanation applies to this flat uncambered surface is thanks to something called the stagnation point (SP). The SP is that point (looking at a cross section of any wing) above which all air goes over the top surface and below which all air goes underneath.
If you are still bothering to read this, consider holding your board at a 90 deg AOA to the wind (no lift but plenty of drag) the SP is right in the middle of the board (this would be much easier with a pencil and paper) if you now rotate the board slowly back towards 0 deg AOA the SP moves slowly towards the leading edge (LE) but only gets there at 0 deg. At say 10 deg the SP is still a little way back from the LE (and of course on the undersurface of your board) and so the air that goes over the top actually has a longer journey than the air that goes underneath because it starts its journey from a little way back on the bottom surface.
Regards
Sorry if it seems to cut across your ideas, but Mr Bernoulli actually has no trouble with generating lift regardless of the wing cross section or any airframe related angle of incidence. It may help you to visualise why if you consider the task of carrying a sheet of 8x4 chipboard (or MDF if you are richer than some) from the B&Q door to your car in a strong wind. Edge on it is not very aerodynamic and easy to carry. At an angle of attack (AOA) of say 20 deg you might just find it leaves your sweaty grasp.
The reason that Mr Bernoulli’s explanation applies to this flat uncambered surface is thanks to something called the stagnation point (SP). The SP is that point (looking at a cross section of any wing) above which all air goes over the top surface and below which all air goes underneath.
If you are still bothering to read this, consider holding your board at a 90 deg AOA to the wind (no lift but plenty of drag) the SP is right in the middle of the board (this would be much easier with a pencil and paper) if you now rotate the board slowly back towards 0 deg AOA the SP moves slowly towards the leading edge (LE) but only gets there at 0 deg. At say 10 deg the SP is still a little way back from the LE (and of course on the undersurface of your board) and so the air that goes over the top actually has a longer journey than the air that goes underneath because it starts its journey from a little way back on the bottom surface.
Regards
Guest
Posts: n/a
I'm actually a student right now at a major university working on my private. They are definitely teaching Bernouillis principle here... the venturi effect if you will. To me anyway, it doesn't completely explain flight.
I was an AP physics student in High School so I know there are other explanations, but I don't have nearly enough authority or knowledge to conest it.
I was an AP physics student in High School so I know there are other explanations, but I don't have nearly enough authority or knowledge to conest it.
Join Date: Mar 2002
Location: KEGE
Posts: 154
Likes: 0
Received 0 Likes
on
0 Posts
Wow this thread has really taken off, 2 pages already and it was only posted yesterday. Wonder if it’s due to Bernoulli, Kutta, Joukowski, or simply the Magnus effect. Coanda was discussed earlier.
That’s all, nothing useful
Oh, one more thing as G the E said, engineers, and scientists deal with probabilities and not certainties, or more precisely they do not deal with 100% probabilities. These theories you all are discussing on the thread were borne from consensus. Bunch of researchers sat around in a conference and said, yeah that sounds reasonable; the math and research supports the model, we’ll adopt it. Except it is not quite so clean and tidy; politics and personal egos always get involved. Clearly certainties can be part of engineering and the sciences, but only when it is defined as a probability as in p>.0001 or some such nonsense. A measure of conviction is required for prudent action, whether you call that a certainty or a p value. Anyway carry on, it's most entertaining.
That’s all, nothing useful
Oh, one more thing as G the E said, engineers, and scientists deal with probabilities and not certainties, or more precisely they do not deal with 100% probabilities. These theories you all are discussing on the thread were borne from consensus. Bunch of researchers sat around in a conference and said, yeah that sounds reasonable; the math and research supports the model, we’ll adopt it. Except it is not quite so clean and tidy; politics and personal egos always get involved. Clearly certainties can be part of engineering and the sciences, but only when it is defined as a probability as in p>.0001 or some such nonsense. A measure of conviction is required for prudent action, whether you call that a certainty or a p value. Anyway carry on, it's most entertaining.
Last edited by '%MAC'; 16th May 2002 at 04:27.
Thread Starter
Join Date: Apr 2002
Location: somewhere else
Posts: 16
Likes: 0
Received 0 Likes
on
0 Posts
I don't pretend to be an engineer, the reason I started this thread was to see if anyone could explain why only one theory of flight is currently taught to pilots. That, and to see what the techies on the site had to say about it. My apologies for being deliberately contentious when I made my first post, but it did elicit the response I was hoping. It's been most interesting. More interesting, perhaps, would be to run a poll, preferably on a different forum (less techies), asking what percentage of pilots have ever heard of more than one theory of flight.
The first I heard of a Newtonian theory was when I went on a performance refresher a few years ago. At the end of the session, the chap teaching us said that he had a few hours before his flight, and was anyone interested in seeing a different explanation of lift to that we were taught when learning to fly. Only two of us stayed (out of about twenty), which perhaps proves that the majority of pilots are not particularly interested in why their aircraft fly. Thanks for your replies.
The first I heard of a Newtonian theory was when I went on a performance refresher a few years ago. At the end of the session, the chap teaching us said that he had a few hours before his flight, and was anyone interested in seeing a different explanation of lift to that we were taught when learning to fly. Only two of us stayed (out of about twenty), which perhaps proves that the majority of pilots are not particularly interested in why their aircraft fly. Thanks for your replies.
"The INTRODUCER"
Over the years I've asked umpteen pilots, engineers, lecturerers, etc, etc the question - in nom-mathematical terms how does a wing work? I've never had the same answer twice but I've hugely enjoyed the discussion - and this latest one.
Line pilots inevitably give the old chestnut of longer distance/same time and are amazed if you suggest that it might not be entirely, how can I put this, true. Professors tend to start speaking and then suddenly stop as they realise the tricky ground ahead.
I was recently greatly impressed by the answer from a young, working engineer with a major airframer. A full and frank admission of the difficulties and a sketch of the prime competing theories. So impressed indeed that we are in the process of hiring him (for other reasons too I hasten to add).
Line pilots inevitably give the old chestnut of longer distance/same time and are amazed if you suggest that it might not be entirely, how can I put this, true. Professors tend to start speaking and then suddenly stop as they realise the tricky ground ahead.
I was recently greatly impressed by the answer from a young, working engineer with a major airframer. A full and frank admission of the difficulties and a sketch of the prime competing theories. So impressed indeed that we are in the process of hiring him (for other reasons too I hasten to add).