FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1936
1936 - 2409.PDF
d FLIGHT. SEPTEMBER 3, 1936 More Full-scale Stalling (continued from top of previous page). stalled regions vary greatly, which means that the downwash angles must vary correspondingly. In R. and M. 1491 there are given the results of swept wing tests made in 1930, and published in 1933; these were on wings of uniform chord, and Figs. 17 to 20 of that work clearly show that there must have been some drastic change in the type of flow with varying amounts of sweepback and sweepf orward. Since the foregoing was written Mr. Lock's letter has ap peared, and perhaps I may be allowed to give attention to the criticisms contained in it: broadly, Mr. Lock seems to say, politely, that the transverse pressure gradient idea is all non sense ; well, that may be so, but let us study his arguments. One must be very careful indeed in applying the results of reasoning based on assumed conditions, as Mr. Lock says they have been assumed for the Prandtl theory, to a set of real and different conditions such as an aeroplane in air—even though the imaginary conditions may help us as a simplifying step. Unlike Professor Einstein, Dr. Prandtl has not found any slight errors in Newton's laws of motion, and Newton's simplicity- is always refreshing and is worth keeping in mind : until one is able to study the flow of. non-viscous air over the infinitely long wings of the Lancaster-Prandtl aeroplane in free flight one cannot use arguments based on its presumed behaviour and apply them with any certainty to real conditions. Now Mr. Lock says that with two dimensional flow in an ideal fluid, there will be no transverse flow (in line 15 he seems to use " transverse " in a different sense) with yaw ; it is equally true, of course, that there is no downwash, in spite of there being lift. The explanation is that the lift is obtained by action on an infinite depth of air, so that the velocity im parted to it is infinitely little and therefore the kinetic energy left behind is infinitely small; there is thus no induced drag. In the same way there can be a transverse force without any transverse flow. Real Conditions But in a real finite aeroplane which can only influence a limited quantity of air, there is a definite downwash created in the generation of the lift, and similarly, if there is a transverse force on a yawed wing, as there is, there simply must be a transverse flow. If we take an extreme case and set a yawed wing at 45 deg. incidence, it is obvious that there is a large side load and a considerable lateral bending of the flow ; a yawed wing, even below the stall, is only a mild case of this kind. (In parenthesis, it is perhaps of interest to note here, for the benefit of those who hate complexities, that the ad vantage of a big span for an aeroplane as a means of keeping down induced drag lies in the fact that lift depends on the momentum imparted to the air, while the energy left behind depends on the kinetic energy imparted; as the momentum is proportional to the downwash velocity and the kinetic energy is proportional to the square of that velocity, it ob viously pays to deal with a large quantity of air, by means of a big span, and give it a low velocity.) Mr. Lock wishes to resolve the flow over the yawed wing into planes normal to the wing and a component along the span—which, he says, will not upset the equilibrium of his component of ideal flow along the chord (if one understands him correctly). He then wishes to apply this method to an infinite wing in real air and says that there will be a " tendency for the boundary layer to be carried along by the transverse flow." But here it must be pointed out that his transverse component is purely imaginary and that the only real flow- is down-stream, and how this down-stream flow can carry the boundary layer in a direction other than down-stream is very difficult to see; for the fact remains, as shown by the photographic evidence, that the surface flow is often very- far from being down-stream. It would be a great help if Mr. Lock could explain what he thinks causes this change of direc tion and why it is impossible and fallacious in his opinion to consider it as due to the cross-wind pressure gradient as origin ally suggested ; he has offered us no explanation showing why the flow cannot be bent as indicated in Fig. 2 of the original article; does he dispute that there is a cross-wind pressure gradient? According to Newton the flow should continue straight unless acted on by a side force. It would also be help. ful if he could say whether he considers the viscosity sufficient to enable his transverse '' component'' to get a grip on the boundary layer and carry it along both before and after the stall and more especially to carry along the '' dead air'' as well after the stall, without any assistance from pressure differ ences ; the idea seems to be widely held that it is the inability of the main stream to carry the boundary layer away with it that leads directly to the stall. It is interesting to note that Mr. Lock at least grants that with a yawed wing the dead air is carried along somehow and accelerates the stall at the down-stream end while delaying it at the forward tip, and that sweeping the T.E. forward similarly accelerates the stall at the root. What is even more interesting is his implication that '' the ordinary Prandtl theory of a monoplane wing" only holds good if the T.E. is straight; that seems to call for a reply from Gottingen. Finally, Mr. Lock, after referring to the imagined bending near the L.E., as a "further example of the fundamental fallacy" of the theory advanced goes on to say that a yawed L.E. will have " little effect, because the boundary layer there is thin;" he does not seem to feel sure that it will have no effect. It was never suggested that it would be likely to have much effect, and was regarded as unimportant. It still seems to me that it is the influence of the cross-wind pressure differences on the flow near the surface just before and during the development of the stall that determines very materially the manner in which the stall takes place. And turning again to a consideration of a swept-forward wing at a small positive incidence as being a mild case as compared with the same wing at 45 degrees incidence, it seems clear that there must be quite a considerable mass of air bent inward by each half of the wing, and not just the boundary layer alone; again the pressure differences seem the only reasonable explanation, as these extend to a considerable distance from the wing. For want of a comma, Mr. Lock almost gives the impression that I am connected with the N.P.L., and it should perhaps be made clear that I have never had that pleasure, although I have always found the N.P.L. staff very helpful and en couraging when I have come in contact with them, and I wish to thank Mr. Lock for joining in this discussion. Turning finally to Mr. Needham's second letter, it does not seem sound to consider arbitrary '' components '' of flow and their inter-actior. as though unaccompanied by the remaining components; it is like saying that because two cars meeting head-on at 2 m.p.h. come to rest in the road the same will happen with two cars moving at 20 m.p.h. and converging with a relative velocity of 40 m.p.h. Mr. Needham's explana tion regarding a difference in the strength of the tip vortices in a sideslip is welcomed, for his previous wording did not make his meaning clear; but whether it is the reduced vortex that causes the earlier stall or whether the reduction of lift on that side caused bv an accumulation of dead air causes the reduced vortex, is perhaps, debatable. The suggestion of model tests with fixed aspect ratio and span, to find the effect of sweeping the edges separately, remains impracticable, for the taper is being varied all the time and will confuse the results. I do not intend, at present, to carry this work any farther single-handed unless some new point is raised and requires investigation. It is hoped that enough has been done to arpuse interest in these questions, and that some small contri bution may perhaps have been added to our knowledge of stalling, and therefore, to safety. Forthcoming Events August 29-Sept. 6. British Gliding Association. Annual Competition Gt. Hucklow, Derbyshire. August 30. Gordon Bennett Balloon Race, Poland. September 4-5. Magyar Athletic Club Motor Section : Inter national Flying Meeting. September 5. Bristol and Wessex Aeroplane Club and Bristol Branch, R.Ae.S. : Garden Partv September 5-6. Aero Club of Hungary: Week-end Aerien. September 12. London Aeroplane Club: Hatfield Garden Party. September 12. Rochester Empire Air Week Display. September 13. Aero Club de France : Coupe Deutsch Race. September 19. Cardiff Aeroplane Club: London-Cardiff Race. 'All R.Ae.S. lectures take place at the Royal Society of Arts, 18, John St., September 19. Three Counties Air Race, Hereford. September 20. London Air Park Club : Display. September 29. London-Johannesburg Race. October 8. R.Ae.S. Lecture': "Tapered Wings," by Dr. G. V. Lachmann. October 22. R.Ae.S. Lecture*: "Economics of Airline Opera tion," by Mr. L. T. H. Greig. October 25. Aero Club de France: Saigon Race. November 12. R.Ae.S. Lecture*: "Skin Friction," by Dr. F. W. Manchester. November 13-29. International Aero Exhibition Paris London, W.C.2, at 6.30 p.m. Admission is by ticket obtainable through a member.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
