FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1939
1939 - 0098.PDF
44 FLIGHT. JANUARY 12, 1939 ,° PARTIAL MODEL UNCORRECTED •''PARTIAL MODEL CORRECTED MODEL TWO-ENGINED AEROPLANE CALCULATED INIIUCED DRAG 100 200 300 400 LIFT REDUCTION DUE TO NACELLE (LB AT 100 FT./SEC.) Determining nacelle drag with partial models. The model is in the form of one nacelle on a length of wing of aspect ratio 4. It was explained that case (b) had unusually bad leaks and was included in the table to show that the calculation was likely to give results of the right order rather than as being representative of modern practice. Model tests also showed how large leak drags could be avoided by isolating high- pressure leaks. If leaks were made to face in a backward direction they would not cause a high interna] pressure. Increases in form drag arose from the necessity to depart, for practical reasons, from the pure aerodynamic form, and items such as windscreens, gun turrets, engines, tyres too large to be entirely housed inside when retracted, in creased the drag. The wind tunnel could discover what was being paid for these practical necessities, and could sometimes suggest compromises. For this type of work large-scale partial models were most successful. It was advisable to fix the transition point of fuselages and engine nacelles forward. Nacelle Drag For testing the drag of parts which did not alter the lift of the rest of the model, tests with partial models were straightforward. But this was not so if, for instance, the addition of an engine nacelle altered the lift coefficient. If a nacelle was placed far below a wing, it would cause a reduction in local circulation, and this would cause an induced drag. Measurements had shown that the differ ence in drag between a nacelle placed at the vertical level of the leading edge and one dropped well below it was mainly due to induced drag. It was important to realise this because the circulation could probably be restored by curling down the trailing edge of the wing. Partial- model tests could be used to locate any local disturbances of circulation of this type. Miss Bradfield obviously did not like the expression "interference," and thought it dated back to the dark ages of "bits and pieces." She did not believe that an aeroplane which could be taken apart mentally was a really good aeroplane. It should be a single aerodynamic unit and must be conceived as a whole. She thought it better to use the single conception of form drag, and to use the skin friction drag as a basis for comparison, rather than using the drag of each part considered separately. Surface roughness could not be treated in the ioft. tunnel assumed for the purpose of the lecture. Its re sultant drag could be measured in free flight by the "comb" method, and supplemented by tests in the com pressed-air tunnel. Assisting the designer in the matter of drag was rela tively simple because at least they did know, and were agreed, that it was low drag they wanted. Longitudinal stability was a different story, and there was no unanimity. The pilot probably wanted an aeroplane which was '' light on the stick," stable when the stick was released (and preferably slightly more stable as the stall was approached), which did not change its trim or the force on the stick when the engines were throttled back or the flaps pulled down, and which had ample elevator control for landing. All these conditions must be met over a wide range of e.g. Slipstream tests introduced difficulties with electric motors, and hinge moments of elevators and the behaviour of trimming tabs could not be made satisfactorily at the Reynolds Numbers of the complete model. These considerations left only measurements of pitching moment for the ioft. tunnel. Pitching moment tests with stick fixed were quick and straightforward and were among the most valuable that could be made in that class of tunnel. Pitching moment tests were still necessary because it was not easy to estimate downwash angle at the tailplane owing to the effects of body and, sometimes, split flaps. And it was not possible to estimate at all accurately the interference on the tail from engine nacelles, for instance. On the subject of cooling drag the authors stated that that part of it which was due to the cooling flow over the cylinder barrels or radiator matrix was beyond the control of the wind tunneller, and was easily calculable from the full-scale cooling characteristics. From total head and velocity measurements at the entry and exit of the cooling duct could be deduced what was the actual drag inside the engine or radiator passage. Model tests were con cerned with the reduction of the loss found by this com parison. Much of the loss was due to poor entry design. The choice of entry and exit positions and areas, and the design of suitable entry fairings fell within the scope of model tests. Cooling and Lift With engines developing ever-increasing power in the same small space, it had become a matter of touch and go whether the cooling flow during slow, full-power climb was adequate. If the engine characteristics were accurately known, the model tests would indicate the margin of safety of the cooling system. Finally, the cooling system might reduce the maximum lift of the wing. This was particu larly likely to occur in the case of the cooling air leaving the gill cowl at low velocity and directed over the upper surface of the wing. The scale effect was probably large, and small-tunnel tests should be taken as a warning that the matter needed investigation at a higher Reynolds Number. Miss Bradfield confessed a disinclination to tackle such things as Frise ailerons because of their sensitivity to small changes in dimensions, or to leaks into the gap between aileron nose and wing. Her advice to designers was: " If you have a good Frise aileron, stick to it." Some tests were in progress at the N.P.L. in which every care was taken to see that there were no uncontrolled gaps or leaks, and already the results looked much more systematic than ordinary test results. Measurement of maximum lift was a matter for the compressed-air tunnels, but there were many ways in which a designer could use his ioft. tunnel to adapt the informa tion to his own use. Examples were possible effect on front slat setting when adding a Fowler flap behind it, and of bringing slotted flaps close to the wing-body fillets, or the effect of high-lift devices on stability. One class of work likely to take up much of the time of a designer's wind tunnel was the measurement of local pressure distribution for stressing purposes. Nowadays, Model of twin-engined monoplane with engines placed low.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
