FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1939
1939 - 0097.PDF
JANUARY 12, 1939 FLIGHT. THE USE of WIND TUNNELS Miss F. B. Bradfield and Mr. D. L. Ellis, in R.Ae.S. Paper, Review Value of Model Data to Aircraft Designers WE have lost our early optimism that we could measure everything, but we have also lost our later pessimism that we could measure nothing correctly." This sentence from the paper by Miss F. B. Bradheld and Mr. D. L. Ellis, read to the Koyal Aeronautical Society last Thursday by Mr. Ehis, sums up the present general attitude towards results obtained with scale models in the ordinary wind tunnel. The title of the paper was '' The Use of Model Data in Aeroplane Design," but actually the treatment was such that a more appropriate title might have been "The Uses of Different Types of Tunnel." Mr. Roy Fedden, president of the R.Ae.S., was in the chair, and in introducing the authors he mentioned that Miss Bradfield was, unfortunately, prevented by illness from being present. Discussing the authors' qualifications he said that Miss Bradfield, after a most distinguished academic career, had started her professional work in 1918; she had been engaged on aeronautical research work ever since, and her many reports were familiar to those con cerned with aeronautical science. Miss Bradfield had been almost entirely engaged on wind-tunnel work throughout the whole period. Mr. Ellis, after having taken an engineering degree at Glasgow, had had some practical engineering experience there, and later had been engaged in the British Thomson- Houston Company's works at Rugby. After that he had spent some years as a technical officer at the National Physical Laboratory, and later had devoted some time at Farnborough to aeronautical research work. At the moment he was in charge of the Vickers tunnel at Weybridge, and he was also secretary of the Society's local branch at Weybridge. Thus, the experience of both authors had been most extensive. Simple Tests The purpose of model tests is to foretell by means of simple, controllable tests that can be made both quickly and cheaply how flight behaviour will change with specific variations in design." This definition was emphasised in the paper because Miss Bradfield considered that the essence of good ad hoc model work lies in the question of simple, controlled testing. If this concept of the scope of model tests were accepted it carried the implication that the most useful tests which the research establishments could carry out for designers were those in which systematic variation was made, on an aeroplane reasonably like his, of some of the most important variables, such as radiator position, elevator balance etc. The simple type of test, which could be done quickly, was not expected to tell the whole story, but chiefly is useful in foretelling the effect of variations from a known type. The first part of the Bradfield-Ellis joint paper dealt with questions of the partial nature of model tests and 'vith their utility if the limitations were appreciated and the results correctly applied. The second outlined how the utility of model tests could be enlarged by the use of special apparatus. In measuring the drag of a complete model in order to tell the aircraft designer the speed of his new aeroplane it was usual to supply the "wind tunneller " with a fairly smooth, solid wooden model to a scale of perhaps one- eighth. The tunnel speed was low, so that the model had to be tested at rather less than half of the maximum speed. The turbulence, Reynolds Number and surface r oughness all differed from full scale, so that the transition point, or point at which the flow changed from laminar Partial model for testing ducted nose radiator. to turbulent, was different on model and full scale. These and other difficulties meant that, in the 10ft. tunnel taken as typical of those available to most wind tunnellers, the overall drag could not be estimated accurately. There were, however, means available for getting around these problems (or for "cooking" the results, as one speaker termed it during the discussion). Induced drag was a function of lift and was calculable. In any case it was low at high speeds. Turbulent skin friction of the sur faces could be estimated at appropriate values of Reynolds Number. In general, the wing, fuselage and tail surfaces were estimated separately. This might give a value 1.2 times the skin friction of the equivalent flat plate surface area. The actual value of the profile drag as deduced from flight tests might be from 2 to 2.5 times the flat plate skin friction. The object of wind-tunnel tests on drag was to track down and show how to reduce this gap between 1.25 and 2.25 times skin friction. The extra drag might arise from engine cooling, from leaks, from high-form drag of certain parts, from interferences, or from surface roughness. Finally, it. might arise in the form of extra induced drag when the lift distribution was bad. Experience had shown that drag due to leaks could be quite large. A value of 2-6 lb. for an engine nacelle at 100ft./sec. represented good modern practice. Incident ally, when the air was deliberately wanted for cooling and the passages were arranged to give the desired flow with a minimum of losses, the term ducts was used, the expression leak being reserved for unwanted flow. The rate at which air would flow through a hole, and consequently the drag created, depended upon the external pressure at the par ticular point, on the direction in which the hole faced, and on the internal pressure. In other words, on the pressure- difference. The most difficult part of estimating the drag of a leak system lay in the determination of the internal pressure. Typical values of internal pressure were: +0.55 V2 for a radial engine nacelle ahead of the fireproof bulkhead ; +0.08 for a liquid-cooled engine; and from o to —0.2 for the general cabin space. Typical estimates of the difference in drag when the leaks were sealed are given in the following table: — TABLE I. DRAG, LB. FULL SCALE AT 100 FT. SEC. Type of Model. (a) Liquid-cooled engine nacelle with nose radiator, mounted on wing ... (b) Air-cooled engine nacelle (c) Air-cooled engine with duct cooling Drag. Calculated. Measured. 2i 12 51 24 101 4 Quantity, cu. ft sec. at at 300 m.p.h 53 229 lOfi
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
