FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1950
1950 - 2270.PDF
FLIGHT, 28 December 1950 631 CLEAR-AIR GUSTS ... It was quite clear that many more data were needed to presenta reliable picture to the aircraft designer and to the airline operator. The Comet episode underlined this need. In considering meteorological aspects, Dr. Hislop observedthat, if there was a large number of fully reported cases of clear turbulence, it might be possible to deduce the principalcause or causes by statistical methods alone. The other method of tackling the problem was to examine the immediate physicalcauses of atmospheric turbulence, and then attempt to obtain experimental coniirmation by actual observation in turbulentregions (with a cross-check in non-turbulent regions). These physical causes then had to be related to the generating meteoro-logical conditions, so providing a basis for forecasting. Since turbulence involved the dissipation of energy, energyhad to be supplied at a rate sufficient at least to counterbalance the dissipation There were three main sources of turbulentenergy: thermal instability, wave motion, and wind gradients. The possibility that thermal instability was a general causecould be disregarded, as in the great majority of cases a most noticeable feature was that the temperature lapse rate wasmarkedly stable. Atmospheric waves were known to exist in association with mountains, but the wavelength of the resultingdisturbance would merely result in a change in height of the aircraft, and not in the encountering of turbulent motion. Coldfronts had also been reported to give rise to atmospheric waves. There was a possibility that under certain conditions atmo-spheric waves might "break"—as a sea-wave does—and thus degenerate into turbulent motion, but there appeared to beno madiematical case covering such a breakdown. The third source of energy was believed to be the mostprobable cause of dear-air turbulence. If, in the presence of a wind gradient in the vertical, no turbulence was present, thenthe purely viscous forces in the atmosphere were small. In practice, some turbulence would exist, and the eddies woulddiffuse in the vertical and horizontal, tending to break down the wind gradient. The eddies could be regarded, in fact, asa means of dissipating energy. The original theoretical approach linking turbulence with vertical wind gradient wasdeveloped by Richardson, who stated that, if the rate of supply of energy exceeded the rate of dissipation, existing turbulencewould tend to increase. Wind/Temperature Gradients Realizing the importance of the local wind gradient in thevertical, the B.E.A. unit gave a great deal of thought to its accurate determination. Direct methods of attack, such as byuse of a pitot trailed two or three hundred feet below the aircraft, or by laying vertical smoke trails and observing therelative shift, were considered in detail, but, whilst ultimately practical, had to be rejected as they required developmentbeyond the resources of the unit. An indirect approach was then proposed by Mr. D. M. Daviesof B.E.A., who suggested that the relation between the vertical wind gradient and the horizontal temperature gradient, asexpressed in the familiar thermal wind equation, might be tried. Hence, if observations of temperature taken at frequentintervals over a distance of at least 50 miles showed a marked temperature gradient, the presence of a marked vertical windgradient would be deduced. Dr. Hislop then referred to some flight researches made onthis basis, in which some confirmation was given that, in the turbulent regions, it had been shown that the observed hori-zontal gradient of temperature implied the existence of a high vertical wind gradient, and also that the latter, when coupledwith the observed temperature lapse rate, satisfied the Richard- son criterion for turbulence within reasonable limits of experi-mental accuracy. In one case, a radio-sonde ascent was available, correspondingclosely in time and position to the turbulent area. The radio- sonde measurements showed a vertical wind gradient of only4-5 knots/thousand feet. This, however,, was not necessarily inconsistent, since the ladio-sonde at 35,OOOft would give amean wind over a height range of 3,000-4,000ft. However, it did illustrate the inadequacy of radio-sonde data for an analysisof this sort. A meteorological condition which, by definition, was boundto contain regions of high wind shear, was a jet-stream. This was generally taken as a narrow belt of high wind (100 knotsor greater) embedded in an airstream of markedly lower velocity. The jet-stream was associated witfi high horizontal temperaturegradients—frontal conditions in the upper air—which led to high wind gradients and. hence, to high winds when thegradients extended over an appreciable height range. It mighi be expected that clear-air turbulence would be asso-ciated with jet-streams, and this was sometimes the case. Seven turbulence had twice been found in a jet-stream; on the otho- hand, on more than 20 other occasions the search aircraft flewacross the course of, or near, a jet-stream without incident On one occasion, the aircraft flew out from base in smooth air,and returned on a reciprocal course about 30 minutes later to find severe turbulence, despite crossing the jet-stream boundaryboth on the outward and return trips. A possible explanation was that, even in jet-streams, conditions became critical only inregions of limited extent. The lecturer thought it unlikely that the effect of high groundwould extend to high altitude except through some form of wave motion. During the B.E.A. gust research flights, twocases of turbulence had occurred near the Alps. The first was associated with a marked low-pressure region, and the secondwith a high horizontal temperature variation. The only other case where apparent wave motion had made itself evident wasat 35,000ft over Cranfield, where there was no definite evidence relating it to high ground. It could thus be inferred that highground was not an important source of turbulence except, per- haps, in a minority of cases. An arrangement whercoy details had been sent to B.E.A.of any dear-air turbulence encountered either by B.O.A.C. air- craft or by the de Havilland Aircraft Company's aircraft, hadbeen in existence for some little time. So far, upwards of 30 cases had been reported, but these did not constitute a greatdeal of evidence, mainly because they were bare reports of what had been encountered without any attempt to gatherspecialized information. B.O.A.C.'s results showed a marked grouping at a position about 55 deg north 40 deg west, i.e.,some hundreds of miles past of Newfoundland. Investigation into the meteorological conditions prevailing at the time of theincidents was still in progress. Aircraft-design Implications In the concluding part of his paper, Dr. Hislop observed thatit was probable that eddies in the atmosphere might range in size from about lft to several hundreds of feet until they mergedinto the general large-scale movements of the atmosphere. Within this very wide band were those eddies or gusts whichgave rise to " bumps " felt by aircraft, and such eddies probably had a linear dimension lying roughly in the range 50ft to 500ft.Eddies smaller than 50ft would be felt by a fast aircraft, net so much as a succession of bumps but more- as an irregularvibration. On encountering eddies much larger than 500ft, the aircraft attitude would tend to adjust itself to the change indirection of airflow, that is to say, it would " ride" eddies of this size. These remarks applied, of course, to present sizesof aircraft, and the data given in the paper were generally applicable to aircraft of orthodox design at present envisaged. Because they were met with little or no warning, clear-airgusts would tend to be associated with the normal cruising speed of the aircraft, irrespective of the magnitude of the gusts.On the other hand, the basic gust envelope of the British Civil Airworthiness Requirements associated the largest gust velocitvwith a speed less than the design cruising speed. Dr. Hislop suggested that the implications of this difference on the designrequirements for high-flying aircraft should be examined. There was the possibility that clear-air turbulence might beof a less irreguiai nature than that normally met, e.g., in cloud. The records showed here and there that for a period of afew seconds the accelerations might occur at a sensibly con- stant frequency which miglit be as high as three per second ita speed of 350 m.p.h. Whilst this frequency was well below die wing fundamental vibration for aircraft such as theMosquito, it nevertherless coincided with the natural frequency for an aircraft with a span of 130ft. The possibility of a seriesof eddies giving rise to accelerations at a frequency coinciding with the natural frequency of a large wing could not be over-looked. In such circumstances, the usf of the normal alleviation factor which was based on the concept of a single up or downgust became misleading as the lag of the wing in responding to a given series of up and down impulses had to be taken intoiccount. On some occasions, the effect of a given gust would be aggravated by the wing motion, and sometimes it would beeduced; but, emphasized the lecturer, in this light the orthodox ' flat-topped" gust conception used for design purposes waslearly unrealistic Dr. Hislop ended his lecture by listing precepts which coulde offered to operators and> especially, to aircrews: — (a) Avoid flying in areas where jet-streams or high thermalgradients in the horizontal are known ro exist. fb) Avoid flying within 2,000ft of the tropopause.c) If heavy turbulence is encountered, assume that one is entering a )et-stream and (i) climb or descend untilturbulence is reduced and/or (ii) fly at right angles to the local wind direction. 1) Should it appear that wing oscillations are being excitedby the turbulence, alter Speed as much as possible, pre ferably in the increasing s
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
