FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1939
1939 - 0101.PDF
JANUARY 12, 1939 FLIGHT. 45 so 30 20 FROM SYSTEMATIC TESTS FROM COMPLETE MODEL TESTS !; O NACELLES ON CHORD LINE X LOWERED $3% CHORD Q NACELLES ON CHORD LINE + NACELLES BELOW CHORD LINE f u £ . S •S*W m^-"1" (5) ww, m s**& fj) • 0 10 H) 30 40 SO CO 70 I » DISTANCE OF NACELLE NOSE FORWARD OF WIN6 LEADING EDOE(? CHORD) Change in slope of pitching-moment/lift-coefficient curve, due to nacelle. local dynamic pressures of the order of 300 or 400 lb./sq. ft. had to be met. Pressure holes could be put into the model already used for the other tests. If pressure. dis tribution was needed on a rudder or elevator horn, the largest possible model was usually the best. Half of a 14ft. span tailplane fixed to the wall of the tunnel would be a suitable kind of model. Hinge moments were some times more conveniently deduced from pressure plottings than from direct tests. An example was the variable gills of engine cowls. Smaller, low-powered tunnels could be of almost as great use as the 10ft. tunnel, provided more intelligence was put into using them and more tests were made. The most direct extension to the atmospheric tunnel test was the test at high Reynolds Number in the compressed- air tunnel, which was the chief means of studying maxi mum lift. It was quicker and easier to work in the atmo spheric tunnel, and the C.A.T. was the place to check the more important drag results rather than an alternative to the smaller tunnels. In America they had gone the whole way towards correct representation of at any rate the smaller types of aero plane by building the 60ft. by 30ft. N.A.C.A. tunnel. Here stalling problems could be dealt with at the full-scale Rey nolds Number. In this country we were, with the 24ft. Farnborough tunnel, in the same state as the man with a 5ft. tunnel instead of a 10ft. tunnel. With ingenuity it was possible to cover most of the same range. The main work for which the Farnborough 24ft. tunnel was built was the testing of engine nacelles. " Partial-model " tech nique was used, with real engines and airscrews. This tunnel gave the quantitative values of pressure drop across the engine, and the cooling flow which was assumed in the small-tunnel tests. The tunnel had also shown most convincingly on full-scale engine installations the large drag losses due to leaks. Of the new high-speed tunnel the lecturer said that it was only possible to drive a 10ft. high-speed tunnel with a reasonable horse power by reducing the air pressure, and it was intended to study high speed at a low Reynolds Number. It would be necessary to check the validity of this course of action. Finally, the paper referred to free-flight tunnels, pointing out that in spite of the low Reynolds Number the vertical spinning tunnel had made it possible to predict and to cure bad spinning tendencies. The American free-flight tunnel had the air jet inclined. It was in its youth, and they would have to wait and see what the American National Advisory Committee workers could do with it. THE DISCUSSION MR. FEDDEN commented that the authors had given a most impartial review of the subject of wind tunnel tests on models. Whilst they had emphasised the value of such tests, they had frankly admitted their limitations. He called upon Mr. Wimperis to open the discussion. MR. H. E. WTMPERIS expressed regret at the absence of Miss Bradfield. There were certain things he had wished to say in her presence, but he would say them in her absence. Miss Bradfield had been engaged on aeronautical research for 20 years, and while he was Director of Scientific Research at the Air Ministry he had been impressed by her skill, energy and thoroughness. Her accuracy had come to be accepted by the aircraft industry. The interpretation of model results was becoming more diffi cult owing to the high speeds reached by modern aircraft, and there was a terrific lack of knowledge of what was happening, why and how it happened, and whether it always happened in the same way. He was puzzled by the statement in the paper that the extra induced drag due to a nacelle could be cured by just curling down the trailing edge. MR. ELLIS explained that badly placed nacelles on a twin- engined monoplane would split the wing into three of very low aspect ratio, and there would be six sets of wing tip vortices, all producing induced drag. The local reduction of lift could be avoided by deflecting the trailing edge. MR. J. D. NORTH pointed out that induced interference drag was measured in 1916 or 1917, but was not recognised as such at the time. He quoted the case of a report on tests on an R. E. model with a nacelle on top of the wing. The tests contained all the evidence of the existence of induced drag. Without disputing what the authors had said, he did not quite understand why it was not possible to measure MQ, the pitching moment, in the small tunnels. The forced oscil lation method was giving results which were in reasonable agreement with deductions from flight observations. He em phasised the ex lent to which misleading measurements were made of interference prior to the recognition of the effect on interference of moving the transition point. A matter which caused him some anxiety was that the stability of flow around a body appeared to change with Rey nolds Number. He thought a great deal of simplification from the structural point of view could be effected if we could use geometrical forms which were satisfactory at high Reynolds Number, but were not necessarily found to be good at low Reynolds Number. As an instance, lie quoted an early R. and M. Report on tests of the drag of a number of struts. The curves indicated that, as the Reynolds Number increased, the strut shape did not matter very much, provided it was smooth. MR. ELLIS replied that in order to measure MQ one must have a model which was free to oscillate not only about a transverse axis, but vertically and fore-and-aft as well. He thought that when the drag of a fairly poor form improved at high Reynolds Number this was due to breakaway effects being reduced rather than to a reduction of true form drag. The form drag would still be high at high Reynolds Number, but the poor form might in addition have a turbulent wake due to breakaway at lower Reynolds Number. Overtime at Farnborough? MR. R. K. PIERSON agreed that model tests could give the right answers if properly carried out, but the models must be truly representative of the full scale. He suggested that ex periments in connection with hinge moments should be carried out in the 24ft. tunnel at the R.A.E., and said he would have thought it desirable to employ a night shift in order that the tunnel might be employed to its full capacity. MR. A. E. RUSSELL asked why the tunnel did not show in creased wing drag with increased wing thickness. He also mentioned that there was a kink in the pitching moment curve which indicated that the tail was subject to buffeting and asked whether the authors had had experience of that in their tunnel work. Mr. Ellis said tunnel tests had indicated in crease in drag with increase in wing thickness. The most consistent results had been obtained at the D.V.L. on large- scale models and in a tunnel comparable with our 24ft. tunnel. He agreed that there was a connection between the kink in the pitching moment curve and the fact that the tail was going through a wake from the nacelle. MR. H. KNOWLER suggested that the paper was mainly an exposition of how to "cook " wind tunnel results. He agreed that model results were, in the main, rather pessimistic com pared with full scale. One aspect which occurred.to him was that frequently models showed directional instability which was not found to be repeated in the full-sized machine, and he asked for information. He believed there had been trouble with directional stability in some large American aeroplanes, although the Americans spent money lavishly on wind-tunnel tests. Some had developed from one fin and rudder to two and even to three. MR. ELLIS, replying to the accusation of "cooking" results, said that much depended on the definition of "cooking." If one could place a model in a wind tunnel and obtain a com plete answer to one's problems without having to make cor rections, aircraft manufacturers could probably dispense with the whole of their design staffs, except for one man in each works to manipulate a slide rule! With reference to weathercock stability, he had not had
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
