FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1912
1912 - 1238.PDF
DECEMBER 28, 1912. Edited by Wind Pressure on Aerocurve Surfaces. DesiRlNG to obtain some information with respect to the wind pressure on the windward and lee side of an aerocurve, and being unable to obtain any published information either in English or French" text books save the well-known classical experiments on plane surfaces at impossible angles of flight, the writer had recourse tome time ago b> the one and only method left, viz., that of personal experiment. SPRN IS DIMENSIONS Of flEROCURVEL CNTERINS TO TRAILING EDGE 4-25' Figl CHORD 4 A ^- ..-* Y TRAILING EDGE -1 LOWER THAN ENTERING ED6C WILING WIND . DIRECTION f TRAILING AND EDGES LEVEL. TRfllLINS WIND-^ DIRECTION CNTERIN6 ED6E 7 LOWER *" _s. THAN TRAILING EDGE TRAILING WIND DIRECTION S£5==ft ~ -A, B The results obtained are shown graphically in Fig. andC. The curves do not 5hoW the pressures in any especial unit, but regarding a vertical section at any point as representing some particular unit, the rest of the curve gives a correct proportionate representation in that same unit. The experiments were made with great care—three different kinds of manometers were used—an open tube connected by flexible tubing with :i thread of coloured alcohol in a U-shaped tube; in another case, a modified form of Pitot tube in conjunction with a more sensitive gauge, and in the third case a tube similar but somewhat larger than the first, over which a very thin membrane was stretched, connected with a drop of alcohol in a fine horizontal tube, the movements in the latter case being very small. Suitable optical means were provided for observing them on an enlarged scale. In every case the tube mouth or membrane was quite flush with the aerocurve surface. This adjustment had to be made with considerable care, as the smallest error gave rise to marked dis crepancies. No devices, sach as a cheese cloth, &c., were employed to render the wind other than turbulent in this sense, but the velocity was kept fairly constant. The respective pressures on the underside of the aerocurve at the points A, B, <_', O, D were com pared relative to one another, all the various combinations, A with B, A with C, B with C, &c, being tested. The same was done with respect to the respective pressures on the top of the plane. Then the top, or, as it is usually termed, negative pressure at A, was compared with the positive pressure at a portion of the aero curve, slightly on one side of A, but having the same position as A in so far as the leading and trailing edge are concerned. The same was done at B, C, O and D. Other combinations were also tried. The Fig. l A, B, C summarises all these results. The positive pressure gave no trouble at all, the results agreeing fairly closely—nor did the negative in so far as the rear half of the * Tha above experiments were made prior to the publication of Eiffel's experiments. E. JOHNSON, M.A. aerocurve was concerned—but with respect to that portion contained between the entering edge and middle considerable difficulty was experienced. The camber of the aerocurve was purposely made somewhat excessive. The graphs show quite clearly the more efficient part of the aerocurve so far as the lift is concerned. It will be noticed that even in the case of I C there still exists considerable lift, although the trailing wind is now horizontal.t The direction of the trailing wind was ascertained by means of smoke blown over the aerocurve, by lycopodium powder scattered on the same and then blown off, and by fine threads of unspun silk. In I B it will be noticed the trailing wind is slightly inclined upward compared with the tangent to the trailing edge. The following experiments were made with respect to what is known as the cyclic up-current. In Fig. 2, if A be the scale pan of a small and fairly sensitive balance with parallel motion so that B C must move vertically up and down, on placing a piece of light cardboard or paper on A, slightly larger than it and on balancing it carefully—and on blowing above or along the top of it (not on it) using a piece of cardboard C D as a screen (but not allowing it to touch A) so that all the " blow" shall pass along or above the cardboard it was found the scale pan at once rose, showing apparently diminished atmospheric pressure on the top of A. This is not, however, the correct explanation as can be shown as follows : In Fig. 3, let A B be a sectional or side-view of the cardboard A. The air streams by the plane A B along the lines of O P, but when passing A the front edge of the plane, owing to viscosity^ (the kinematic viscosity of air is 14 times that of water) drags some of the air from e along with it (and from c as well, but we are not concerned with that), and we have what is known as a cyclic up current at A, tending to lift the first part of the plane. If the plane A be a piece of paper, it will, if strongly blown across, finally lift up in front, be caught in the air current and blown away. Experiments with smoke fully confirmed this. If, instead of shielding the plane we have it inclined—if only at a very small angle to the horizon, this cyclic up current is still formed even with the air blowing along both sides. Obviously the same holds with respect to an aerocurve, with a dipping front edge in order that it may meet the relative upward trend of the wind with minimum of shock. Fine threads of unspun silk attached to the underneath surface of the aerocurves shown in Pig. 1 A, B, C clearly proved its existence. In some experi ments made with considerably less camber than the one illustrated, the negative pressure was proportionately less, but in every curve or plane experimented with it was always found to exist. CARDBOARD C TWs d°eS not on. PAPER mean that anything in the nature of a partial vacuum exists on the upper side of an aerocurve—as so many still appear to think —which per se assists the lift. The average pres sure of the air is some 2,116 lbs. per square foot, and the maximum lifting effect of an aero curve not more than 5 lbs. per square foot.§ Divide 2,116 by 5, and we have 423. The greatest vacuity that could possibly be produced would be less than one-quarter of one per cent, the" n^h?palth0f C°U™' **" * do*nwa«" acceleration of the air with respect to SSV^S^-" ^^ t0 °De is exeSdWL ** Wn"en &m have^il is *™. »w» cases where this SHIELD I23O
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
