FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1934
1934 - 0571.PDF
FLIGHT, JUNE 7, 1934 tioa of. a, region in which the " total head " falls below the constant value which it would have in installed flow, but in which there is no definite separation of the stream from the surface. During this stage, which precedes the true stall, the air stream may separate from the leading edge, but quickly rejoins the surface. But this is only of theoretical interest and merely gives warning that the true stall is approaching. It is in accord- ance with what happens in the next stage that the aero- foils fall into one of the three previously-mentioned groups. In the first case the definite separation of the stream from the upper surface occurs suddenly and completely from the leading edge, so that the boundary of the main stream passes well clear of the remainder of the aerofoil, and a permanent region of " dead air " of uniform low pressure was formed over the entire upper surface. This change coincides with the discontinuity in the force curves, and occurs only when there is a large difference between the values of the lift before and after the critical change. If the difference is not so large, the first complete separa- tion is followed by a remarkable and inexplicable pheno- menon. It appears that the boundary of the main stream may alternate between two fairly distinct forms. One is that in which the boundary passes well away from the trailing edge and is clearly defined towards the front. In the other form the boundary is also clearly defined towards the front, but remains within three or four inches (on full scale) of the surface, for about half the width of the aerofoil, and then becomes difficult to define. The alternations of the flow between these two forms are very Tapid when thought of in terms of measurement of forces on a small model, but they are surprisingly slow in terms •of the distance travelled by the aerofoil through the air. Either kind of flow can persist while the aerofoil travels through a distance equal to 20 or more times the length of its chord. These alternations are responsible for the violent fluctuations indicated by the sudden widening of the space between the lift and drag curves. Flow Separation Disregarding the shallow " bubble " of turbulence which may form behind the leading edge at relatively low inci- dences, the first clear sign of the separation of flow from the leading edge always coincides either with a complete •discontinuity in the force curves or with the onset of violent fluctuations. This, however, does not necessarily happen at the incidence of maximum lift ; with some aero- foils, particularly when they are thick and have the maxi- mum camber far back from the leading edge, the first de- finite separation of the flow occurs near the trailing edge. This region expands gradually with increasing incidence and lift falls gradually without serious fluctuations or dis- 'continuities. Eventually, however, the flow separates suddenly from the leading edge and the characteristic fluc- tuations appear, but at an incidence considerably greater than that of maximum lift. It appears that* as the incidence of a wing of moderate thickness increases, there are two points in the boundary layer where the conditions tend towards separation of the flow from the surface, one of which is near the leading edge and the other near the trailing edge. A kind of race •develops between these two conditions. If the " front •stall " wins, discontinuities or violent fluctuations mark the attainment of maximum lift, whereas, if " rear stall " wins, all such violent phenomena are postponed to inci- dences considerably higher than that of maximum lift. The occurrence of " rear stalling " modifies the main flow so as to reduce the danger of " front stalling," and hence when " rear stall " wins the race, even by a small margin, all violent fluctuations or discontinuities may be postponed until many degrees beyond the incidence of maximum lift. It does not take much imagination to see how this state of affairs may account for the peculiar sen- sitivity of some wings to small changes of shape and to other variables. A general study of the published records of experiments on wings suggests that while front separation is post- poned by increasing Reynolds number, rear separation is •encouraged. This will be found to explain most of the complicated effects of change of scale which have hitherto ibeen so puzzling. It appears that the flow may alternate through wide limits, even when the incidence is maintained strictly •constant. In manoeuvres of an aeroplane in flight, sudden changes of the forces on paxts of the wings may have verydisconcerting effects, and the point is that it is these sudden and local changes which are responsible both forthe principal remaining dangers of the stall and for the difficulties which we experience in predicting behaviourfrom calculations based on model experiments of conven- tional type. ft should therefore be our aim to eliminate these suddenchanges altogether or postpone them to incidences which cannot be reached in ordinary manoeuvres, and since wecannot hope to locate them by studying curves of lift and drag obtained in the conventional way, we must find othermeans of doing so. In the experiments on aeroplanes in flight at Cambridge,the flow is studied by watching, or photographing with a small cinematograph, the movements of little tufts of woolfastened to the fabric, of the wings, and to very light posts temporarily erected upon their upper surface. In unstalledflight these tufts remain practically still, pointing back- wards, but when the wing is completely stalled they areviolently agitated for a certain distance from the surface, but beyond this distance they stand out steadily in theair stream. The dividing line, of course, marks the boun- dary between the main stream and the dead air. There aremany variations in the behaviour of the tufts, but at present we are concentrating upon broad features only. The wings of the aeroplane used have a section forwhich the flow separates first from the leading edge and then fluctuates between the two forms previously men-tioned. The incidence at which these flow variations begin coincides exactly with the incidence at which the pilotbegins to notice those erratic plunging movements which everyone knows as characteristic of the stall. The Remaining Problems When measurements of lift and drag were obtained, it wasnoticed that though the lift curve changes direction rather abruptly at 16 deg., there was no indication of any suddenchange in the forces themselves. When, however, this aero- plane was stalled in a manoeuvre in which incidence changesprogressively, the change of flow over the unslotted parts of the wings, from the unstalled to the fully-stalled state,may occur very suddenly and permanently so that the fuli effects of a discontinuous force change may be felt.This phenomenon of sudden stalling was met for the first time when we were attempting to discover the cause of avery disconcerting sudden change of behaviour of the aero- plane in steep turns, begun at some 40 per cent, above thestalling speed in straight flight. It was found then that the sudden flow change occurred at exactly the sameinstant as the change of behaviour and that its etlects were greatly aggravated by sympathetic stall of the tail,caused no doubt by the change of the down wash from the wings. This sudden change is not nearly so dangerous as thedeadly spiral turn which may follow a stall on an unslotted aeroplane, but we now believe that it is responsible forthe greater part of the few remaining accidents which still occur from stalling on slotted aeroplanes when they aresteeply turned near the ground. Summarising, it seems that the ofder methods will needto be supplemented by new methods of experiment which will enable sudden local changes of force and flow to bedetected. On the model scale, this calk for quick-acting balances which record changes of lift and drag whilstincidence is progressively changing, and for some means of studying flow to supplement force measurements. Onthe full scale the examination of the flow< by some simple methods which can be easily applied in flight will benecessary for investigations. By such means iti should be possible in time to recognise the forms of wing which areliable to sudden flow changes in any of their parts and either cease to use them or use only those on which thetrouble is postponed to incidences greater than those employed in ordinary manoeuvres. Until we know when and how the stall occurs oneach part of the wings in flight we shall not reach any clear understanding of the reasons, why aeroplanes behaveas they do. At the conclusion, Lt. Col. J. T. C. Moore-Brabazonthanked the lecturer in a very amusing little speech, and the room moved on to refreshments and a conversazione. The awards were as follow: —First British Gold Medal toCapt. G. de Havilland ; Simms Gold Medal, to Sir Gilbert Walker ; the Wakefield Gold Medal to Sefior J. de laCierva ; and the Taylor Gold Medal to Mr. A. Plesman.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
