FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1938
1938 - 1112.PDF
SUPPLEMENT TO FLIGHT 391 h APRIL 21, 1938 THE AIRCRAFT ENGINEER TESTS OF THE N.A.C.A. Ii-A MODEL HULL IN THE R.A.E. SEAPLANE TANK. By L. P. Coombes, B.Sc... and F. G. Brown. R. & M. No. 1784. (12 pages and 4 diagrams.) September 23, 1936. Price 2S. net. An exchange of models was arranged with the N.A.C.A. Tank at Langley Field in order that the results obtained in the two tanks might be compared. As the American model was large for the R.A.E. Tank, the tests also served as a check on the results of the tank calibration made previously This report describes the tests of the American model, the N.A.C.A. 11-A hull. Measurements of drag, moment and depth of immersion were made at four angles and several different loads. The range covered was considered to be sufficiently wide to make a reliable comparison possible. The difference between the R.A.E. and N.A.C.A. results, based on mean values, was generally very much less than 5 per cent, for drag and 7J per cent, for moment measurements. The agreement is considered satisfactory, and can be accounted for by experimental errors and the like. No evidence of wall interference or depth effect was found, and the conclusions of the tank calibration tests were thus supported. COMPARISON OF RESULTS OF TESTS OF THE SINGAPORE II.C MODEL HULL IN FIVE TANKS. By J. P. Gott, Ph.D. R. & M. No. 1785. (16 pages and 18 diagrams.) November 26, 1936. Price 2s. 6d. net. Models of the Singapore II.c hull have been tested in five tanks, and, althoughthe tests were not always intended to be comparative, the results have been collected and examined 'to find what measure of agreement exists among the results oidifferent tanks. The five series of tests are as follows :— A 1/16th scale model was tested in three tanks :—1. The Yarrow tank at the National Physical Laboratory. 2. The tank of the designers. Messrs. Short Eros., at Rochester. 3. The Canadian tank at Ottawa.4. Models of 1/18&, l/12th, 1/eth scale were tested in the R.A.E. Tank. These tests were for the purpose of tank calibration and covered only a very limited range of loads and attitudes.5. A l/12th scale model has recently been tested in the N.A.C.A. tank at Langley Field. This model is the l/12th scale model used in the R.A.E.tests in series 4. Comparison of the drag measurements in various tanks shows general agreement,but the detail differences are considerable. These differences are probably due to differences in technique, to experimental error and to the use of several models.Comparison of the pitching moments measured in three tanks shows systematic differences of the order of 10 per cent, to 20 per cent. These differences have notbeen explained. Recent experience suggests that any future comparisons should be made on models 6 or 7 feet long. A REVIEW OF THEORETICAL INVESTIGATIONS OF THE AERODYNAMICAL* FORCES ON A WING IN NON-UNIFORM MOTION. By H. M. Lyon, M.A., A.F.R.Ae.S. ~R. & M No. 1786. (33 pages and 8 diagrams.) April 12, 1937. Price 4s. 6d. An investigation of published work on the application of aerodynamical theoryto problems of non-uniform motion was undertaken as a preliminary step in an attempt to develop a method for tbe calculation of the aerodynamical derivativesof flutter theory for a finite wing. A review of the papers consulted in the course of this investigation is presented in this report. The theory of the steady motion of thin aerofoils is briefly explained in theintroductory paragraphs, and special attention is paid to methods of investigation which are applicable to cases of non-uniform motion. Two distinct methods ofcalculating the effect of the wake on the lift of a wing of infinite span in non-uniform motion are described, aud the development of each method is traced through thework of various investigators. In conclusion the application of the results to the cafeujssion of the aerodynamical derivatives of flutter theory is discussed. ESTIMATION OF INCREASE IN LIFI DUE TO SLIPSTREAM. By R. Smelt, B.A., and H Davies, M.Sc. K. & M. No 1788. (20 pages and 16 diagrams.) February 3, 1937. Price 3s. net. This report gives a method of calculating the increase in the lift coefficient ofa wing due to the mounting of airscrews ahead of the leading edge. Distribution of velocity in the straight and inclined slipstreams behind an airscrew in theabsence of a wing is first obtained , the effect of this velocity distribution on wing lift is then found by considering two extreme conditions, viz., slipstream width-»- wing chord (1) very large and (2) very small. The formulse thus obtained are checked by comparison with experimental results on several combinations of wingand airscrews, with and without engine nacelles and fuselage. Comparison with experiment indicates that with any conventional arrangementof airscrews and wing it is possible, by means of the formula; derived, to give an estimate of the lift coefficient during take-off and climb, provided that flaps arenot fitted to the wing. TESTS OF RIVETS AND BACKWARD-LAPPED JOINTS IN THE COMPRESSED- AIR TUNNEL By D H. Williams, B.Sc., A.F.R.Ae.S., and A. F. Brown, B.Sc R. & M. No. 1789. (6 pages and 6 diagrams.) May 22, *937- Pnce is 3d net An aerofoil, of section N.A.C.A. 0012, was tested over a range of Reynolds numbernp to R — 8 X Iff5, with 7, 8 and 12 rows of rivets, and with i backward-lapped joints, on each surface. The effect of 8 rows of rivets was to increase CB min. of the smooth aerofoil byabout 0.0007 over the whole range, the 7 rows gave the same increase up to R = 2 x 106, but the effect was less at higher Reynolds numbers; the 12 rowsincreased C D min. considerably over the middle part of the range, but the effectdecreases at high values of R. The lapped joints increased C D miu. by about0.0012 over the whole range. The lapped joints have no effect on Ct max.; the 7 and 8 rows of rivets havelittle effect at high Reynolds numbers ; the 12 rows have little effect at R — 4 x 10*' but tue maximum lift falls rapidly for higher values of R. THE INFLUENCE OF END CONSTRAINT ON THE TORSIONAL STIFFNESS OF A RECTANGULAR SECTION TUBE. By J. C. K. Shipp, B.Sc. R. & M.No. 1790. 7 pages and 4 diagrams.) March 22, 1937. Price is. 3d net. In calculating the torsional stiffness of a metal or plywood covered wing it i=customary to employ Bathos formula in conjunction with an empirically deter- mined effective value of the skin modulus of rigidity. By using an effective modulusynstead of that appropriate to the skin material, some allowance is made for the fact that the actual wing conditions differ in a number of ways from those assumedin the derivation of Batho's formula. In practice, for example, tension fields may form in the skin material in contradistinction to the uniform shear conditionsassumed by Batho, or the wing root section may be subject to axial constraint instead of being free to warp. This report is concerned with the latter kind ofdeparture from the Batho conditions, and describes the results of calculations on the effect of end constraint on the torsional stiffness of a rectangular section tuberepresentative of a monocoque wing structure. For a metal tube with proportions representative of wing construction the increase of torsional stiffness due to end constraint, when measured at a section in the outer third of the tube length, is of the order of 10 per cent. For a wooden tube this increase rises to roughly 25 per cent. For short wide tubes, however, and for torques applied near the wing root, this increase in stiffness may be much greater. THE DEVELOPMENT OF A HIGH-SPEED INDUCED WIND TUNNEL O» RECTANGULAR CROSS-SECTION. By A Bailey, M Sc, A.M.Inst.C.E., and S. A. WOOD, M.SC. R. & M. No. 1791. (16 pages and 8 dia- grams.) February 23, 1937. Price 2s. 6d. net. Experiments have been made on the N.P.L. high-speed jet to find out whether asatisfactory design of high-speed tunnel, to run from the C.A.T. exhaust and to measure 3 ft. X 1 ft., can be found. A wind tunnel was constructed with a workingportion 0 in. x 3 in. in cross-section and 6 in. long, which was subsequently length- ened to 9 in. The design was based on experience witb previous circular tunnels,the high-pressure ejector nozzle being in the form of a slot, of plain convergent profile, surrounding the tunnel downstream of the working portion. The experiments made included modifications of the inlet flare and of the diver-gence of the working portion, and the determination of the effect of variations in the pressure on the ejector nozzle. These first tests were made without a modelin the tunnel. Experiments were also made, with model aerofoils of thickness 0.24 in., 0.125 in. and 0.07 in. respectively in position, and the effect of modifyingthe longitudinal profile of the working section, to compensate for the presence of the model, was investigated by the use of liners. Working speeds up to the local speed of sound can be obtained by the inducedprinciple in a rectangular wind tunnel having a suitable divergence, with no model in position. The longitudinal variation of speed can be controlled by varying thedivergence of one opposite pair of sides, and the lateral variation of speed is very small. The introduction of a model aerofoil reduces the maximum value of u/a attain-able, the reduction increasing as the projected area of the model increases. With a projected model area 1/25 that of the tunnel, the maximum approach speed is0.77a, and with a projected area 1/86 that of the tunnel, the speed is 0.85a. Tbe rise in air speed downstream of the model, which occurs with a parallel tunnel,can be obviated by suitable modification of the profile of the walls parallel to the aerofoil. The maximum air speed attainable can be slightly increased by lining down theworking section, but the increase is limited by the relative increase in the projected area of the model. In view of the likelihood that a different tunnel profile will be required for eachdifferent thickness of aerofoil and working speed, it is suggested that it would be convenient to use flexible walls on the sides of the tunnel parallel to the axis of theaerofoil. This would also enable the working speed to be controlled by means of a variable throat, downstream of the working section, where the speed would alwaysbe that of sound. Experience indicates that such an arrangement would result in more stable conditions than if the speed were controlled by varying the pressureon the ejector. NOTE ON THE DRAG OF A GAUNTLET AS MEASURED IN THE 24 FT. i TUNNEL. By W. G. Tennings, B.Sc R. & M. No. 1792. (8 pages and 3 diagrams.) March 23, 1936. Price is. 6d. The Gauntlet was, at the time of its introduction, representative of the bestmodern practice Service type aircraft design, and its measured performance exceeded expectations. Drag tests have been made on the actual aeroplane in the R.A.E.24 ft. tunnel to investigate whether the high performance could be attributed to some particular feature in the design, e.g., low body drag. The aeroplane wasonly available for a short time and the tests were not, therefore, of a very com- prehensive nature. The opportunity was, however taken to include in the testssufficient measurements from which, with the assistance of the available engine data, the performance of the aeroplane could be estimated and compared with themeasured flight performance. This provided an interesting check on how far the analysis of tunnel tests on a complete aeroplane with wings extending beyond thejet boundaries could be usefully employed in predicting performance. The drag of the Gauntlet body, including cowled engine, tail unit and under-carriage (less wheels), was 53.0 lb. at 100 ft./sec. With the engine removed and replaced by a faired nose the drag was 40.3 lb. The corresponding drag coefficientsbased on wetted area, i.e., total surface area, are 0.0126 and 0.0000, and the corres- ponding ratios of drag coefficient to equivalent fiat plate skin friction are 4.8 and 3.7. A drag estimate (at zero lift) of the complete aeroplane was made from the tunneltests where the wings were projecting beyond the jet boundaries; the estimated drag was 111.8 1b. at 100 ft./sec, and the wetted area drag coefficient ratio 3.1.All these figures show that the drag is appreciably greater than for modern clean monoplane design. The top speed of the aeroplane estimated from tbe tunnel drag tests and theengine test bed power data was 234 m.p.h. at 15,000 ft., compared witb a measured speed of 230 m.p.h. ft appears that if the actual engine power developed at altitude is known thetop speed performance of an aeroplane can be estimated with a fair degree 0! accuracy from drag tests in the 24 ft. tunnel even when the wings project outsidethe jet boundary; this conclusion must be treated with reserve until further evidence from similar experiments is collected. FORCED OSCILLATIONS OF AEROPLANES, WITH SPECIAL REFERENCE TO VON SCHLIPPE'S METHOD OF PREDICTING CRITICAL SPEEDS FOR FLUTTEK.By R. A Frirer, B.A., D.Sc, and W."P. Jones, B.A. R. & M. No. 1795. (26 pages and 2J diagrams.) October 21, 1936. Price 3s. Od. net The subject of the forced oscillations of aeroplane structures has assumed import-ance in recent years in relation to the prediction of critical speeds for flutter. The present paper deals principally with a method proposed by von Schlippc, in whichthe aeroplane is given forced oscillations in flight and the critical speed for flutter is estimated from measurements of the maximum forced amplitude correspondingto various flight speeds. The mam purpose of the investigation is to examine whether the method would givt reliable estimates of the critical speed if themeasurements were restricted to flight speeds well removed from the actual critical dj Fart 1 describes the theoretical behaviour of a model wing which is subjected to forced oscillations at various airspeeds below the critical speed. Part 2 sum- marises the results of some wind tunnel experiments. Tbe Appendix deals with the general theory of the forced oscillations of dissipative systems which obey linear laws. The conclusions of the report are as follows :— (i) When the damping forces are large, the forcing frequencies corresponding to maximum forced amplitude depend on the particular co-ordinate chosen for observation of the motiou. The term resonance is, in general, inapplicable to such systems. fii) A dynamical system with two degrees of freedom can show at most three maxima of forced amplitude. More generally, with a system having m degree^ of freedom the greatest number of maxima is 2m — 1 (see (v) of the Appendix) (iii) In the case of an aeioplane wing the complete "Schlippe diagram snowir. the maxima of forced amplitude plotted against airspeed comprises several curvt- one or more of which may end or begin abruptly at particular speeds. (iv) The critical speed for flutter cannot in general be estimated satisfactory by von Schlippe's method, unless some measurements of maximum forced aioplit*'' " corresponding to airspeeds elose to the critical speed are included.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
