FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1951
1951 - 0336.PDF
214 FLIGHT, 23 February 1951 FLIGHT RESEARCH ... at the R.A.E., for example, the number of insects picked up perflight never exceeded one per 10ft span, and in most morning flights no insect contamination occurred at all. Rainy conditionsand high winds also seem to tend to reduce the insect population. In warmer weather the number of insects increases and on theafternoon of a warm summer day there have been occasions when so many insects have been picked up that the effective transitionis brought near the leading edge over most of the wing surface. Even in mid-summer, however, the number of insects collectedin an early morning take-off is usually small. Insect contamination also tends to become a more serious problem in warmer climates.During some systematic observations made in Louisiana in the U.S.A., for example, one insect was picked up per square foot ofwing frontal area for every minute of a flight at 500ft. Unfortunately, the mere possibility of a premature transitionbeing caused by insects on some occasions is sufficient to prevent full use being made of laminar-flow wings in achieving long range.It should not, however, discourage efforts to achieve extensive areas of laminar flow as a means of attaining higher speeds underfavourable conditions. The area of wing which may be affected by insect contaminationis not very large, extending perhaps from about 0.20c on the lower surface to about 0.07c on the upper surface. The possibilityof protecting this surface during taxi-ing and take-off therefore arises, since the fly population decreases rapidly with increasingaltitude. During the R.A.E. experiments on the King Cobra, previouslymentioned, some tests were made with a paper cover stretched over the front part of the test section and extending to about 0.3con both surfaces. The paper was fastened to the wing round its edges and was ripped off after take-off by pulling a string passingfrom the cockpit round the paper at the leading edge. This method proved quite successful, but its application to the complete wingof a large aircraft obviously would be more difficult. An attempt has also been made to coat the wing surface witha volatile solid material such as camphor, which would sublime away in flight, taking the insects with it. This method has notso far been successful. Boundary-layer Suction The lecturer next turned to boundary-layer suction. Althougha large amount of theoretical and wind tunnel work has been done on boundary-layer suction [he continued] flight research on thissubject has been limited so far, mainly owing to the complication and expense of the preparations needed for such a test. A fewinteresting experiments have been made in flight. In order to investigate the practical problems involved inapplying the now well-known principle of the Griffiths suction wing to an actual aircraft, the outer wings of a Meteor 3 wereremoved and replaced by special sections of the form illustrated diagrammatically in Fig. 6. These sections had a maximum thick-ness : chord ratio of 16 per cent and were designed to have a favourable pressure gradient almost to 0.75c, where there wasa suction slot, nominally 0.04m deep, formed between the aileron and the wings as shown. The suction was provided by a modifica-tion of the blower normally used to cool the rear bearings on the Derwent 1 engines. This should theoretically have given valuesof suction coefficient CQ at the slots estimated to be necessary to prevent a separation or rapid thickening of the boundary layerbehind the slot, provided the boundary layer remained laminar to somewhere near the slot. In the form in which the aircraft was first tested at the R.A.E. the suction had no effect on the drag. This was found to bepartly due to the fact that the transition on both upper and lower surfaces of the wing, as observed by Gray's chlorine method, wasoccurring at about 0.06c, where there was a ridge almost o.oiin high in places caused by a butt joint in the skin. The surfacewaviness was also of rather a poor standard. With this forward transition the boundary-layer thickness atthe slot was such that it would have required twice the available suction to produce the desired effect. The surface, therefore, was painted and rubbed smooth,removing the ridge at the skin joint to negligible proportions, and reducing the waviness. The transition was then found to be about0.70c on the lower surface and 0.55c on the upper surface (at CL=O.2). The transition was thus slightly aft of the maximumsuction on both surfaces so that further improvement of the surface would have been unlikely to have had much effect. Even with this improved surface condition the suction stillseemed to have only a small proportion of the expected effect. This was eventually found to be due to the fact that the local CQactually being achieved at the test section was only about half the required value. This loss was found to be partly due to leaks inthe duct system and partly to non-uniformity in the spanwise distribution of flow in the slots, caused by variations in slot widthand by the internal design of the ducts. One of the wings was removed from the aircraft and tested inthe No. 2 iijft tunnel at the R.A.E. These tests confirmed that when the value of CQ at the slots was increased to that theoreticallyrequired (about 0.0006) the expected gain due to the suction was realized, the section profile-drag coefficient being reduced toabout 0.002. DUCT O2 O4 OS O»6 'CHORD POSITION OF PRESSURE HOLES (STBO.WINC ONLY) INDICATED T VO Fig. 5. Wedges of turbulent flow caused Dy insects on the leading-edge. Fig. 6. Section through Griffith suction wing (at test section) as fitted to a Meteor 3 outboard of the engine nacelles. The aileron effectiveness, although substantially lower than ona standard Meteor 3, was considered to be adequate, even with suction off, in spite of the disturbed flow over the aileron surfacesin this condition. There was some aileron oscillation, the severity depending on the upfloat angle. With zero upfloat angle theoscillation was negligible. The longitudinal and directional handling characteristics of the aircraft were the same as those ofa standard Meteor. These experiments, although not entirely conclusive, served todemonstrate the following points in relation to possible applica- tions of the Griffith wing principle:— (1) It is still necessary with this design to achieve a very high standard of surface finish if the transition is to be got near enough to the slot to enable the full advantages of the design to be realized. (2) The construction of the suction slot and the ducting must besuch as to guarantee a reasonably uniform flow distribution in the slot. (3) Lateral control characteristics can be made acceptable even with suction off, at any rate for the moderate wing thickness of 16 per cent used in these experiments. Concluding the first section of his pap'er, Mr. Davies referredto the measurement of overall drag in flight which fortunately is one of the few problems which has become easier since the adventof jet aircraft. It was fairly easy to measure the thrust of a turbojet in flight, using a pitot in the jet-pipe and, since the technique couldbe extended to apply in dives as well as in level flight, measure- ments of drag could be obtained at speeds well above the maximumlevel speed of the aircraft. Perhaps the most important objective of measurements of this kind were to provide a check on windtunnel tests. The lecturer said: "On the whole these comparisons were veryencouraging in confirming the reliability of predictions based on wind tunnel tests. Considerable discrepancies might be expected,however, when separation effects are present in the drag. Such effects usually contribute to the increase in drag of an aircraft withincreasing incidence, i.e., to the so-called induced drag factor K. Unfortunately the measurement of K in flight is not easy, becauseof the difficulty of separating "out the effects of varying Reynolds number and lift coefficients. "Lift coefficient also has an important effect on the drag rise athigh Mach number. It is not really possible at these high Mach numbers to represent the effect as a variation in K, since the totaldrag is no longer even approximately a linear function of CL-"
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
