FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1946
1946 - 2244.PDF
Blind- Approach Problems Second Part of R.Ae.S. Lecture by Mr. H. C. Pritchard, B.A. : Some Suggestions : The Discussion (Continued from page 474) IN the first half of his lecture, which was summarized in lastweek's issue, Mr. H. C. Pritchard described present-dayblind-approach systems and also one of the latest types ofautomatic pilot. He then continued to describe some pre- liminary experimental work on automatic approach' using theSCS.51 system in conjunction with a conventional automatic pilot. In this, the runway localizer signal was introduced tothe automatic pilot to cause a proportional rate of turn. Such a system, however, produced an unstable approach. Thiscould be cured by introducing means of using the difference between the known runway direction and the aircraft headingas given by the directional gyro. A typical approach with automatic control is shown belowin diagrammatic plan form. In this the aircraft turned rapidly on to the centre-line until, at the point "A," theengines were throttled The drift angle was about 4 deg. The northward departuie of the aircraft after starting the glidewas due to change of rudder trim when the engines of the Lancaster were throttled back, but the deviations were ironedout before the runway was reached. Consideration of a number of experiments showed that highgearing in an automatic pilot was required to give precise stabilization, and that large variations of cross-wind mightprove to be the limiting factor with the particular system used in the experiments.Mr. Pritchard thought that the use of the aircraft heading as a measure, of approach to the beam might be supersededby one derived from the beam itself. Given some measure of the range from the localizer, the distance from the runwaycentre-line could be derived from the signal, and differentia- tion in relation to the time factor would provide a rate ofapproach. With such a system the errors arising from varia- tions of cross-wind should be minimized. The problems of automatic vertical guidance seemed, hethought, to be somewhat easier, largely because the aero- dynamic damping in pitch was usually better than that in yaw. Mr. Pritchard made it clear that much experimental work 5000 FLIGHT NOVEMBER 7TH, 1946 4OOO 3OOO 2OOO WND-1OOOFT. -TCoSoTS ^O'M,2KNOTS MILES FROM LOCALISER An example of an automatic approach among those carriedout in this country. The directional over-corrections are explained above. The A.12 automatic pilot as installed in an R.A.F. Liberator.This has been successfully modified in America for auto- matic approach work. with different aircraft types and with variations of flap settingwould be required before any final type of automatic control could be established.So far the lecturer had dealt with methods of fixing the air- craft's* position during the approach, but the urgency was fargreater during the final landing stages. He observed that con- ditions of zero visibility were somewhat exceptional even in thiscountry, but that since in the South of England the incidence of a 600ft ceiling was four times as frequent as that of a 150ftceiling, the possibility of merely reducing the safe operating ceiling would not go a long way towards eliminating cancella-tions. In American operations bad terminal visibility accounted for 70 per cent of the cancelled flights. Accurate Position-fixing Any final choice of a position-fixing system would un-doubtedly be influenced by aircraft control requirements. The conception of a fixed glide path meeting the runway at aparticular point was not attractive for blind landings. Another suggestion was that the aircraft should be controlledon the fixed glide path down to a particular height, after which vertical control would be taken over by gyroscopic means soas to flatten out the path. Such a system would be distinctly critical and was not inherently reliable; it had, however, beenused in pilotless aircraft work. Another idea was that the height of the aircraft above therunway level should be measured continuously and smoothly, and that primary control should depend on this measurement.The most promising means of obtaining this result was by means of some form of radio altimeter. Those so far used wereaccurate only to within 10ft, but a substantial improvement could no doubt be obtained with developments in centimetrevalve technique. There was one difficulty—that the ground on the approach was often far from level and it would b:necessary to apply corrections. In German experiments it hac been found that the height of the aircraft varied exponentiallywith the time of touch-down, and the FU.G.IOI radio alti-meter was arranged so that no current flowed when the resultant equation was satisfied. The pilot, consequently, merely main-tained the needle of the instrument against a fixed mark, and this technique theoretically provided an accurate glide pathto the ground. The use of centi metric wavelengths made it possible to pro-duce very narrow beams and the accuracy obtained might go a long way towards removing the reflection difficulties. Since a3-centimetre radiation could be directed within an* accuracy of plus or minus 0.5 deg and a 1 deg obstruction angle wasnormally specified for runways, there should be immunity from reflection troubles. If the localizer radiation could be keptabove this minimum angle and restricted in the horizontal plane, with both transmitting aerials in one place, an attractiveapproach scheme would be presented. Precise short-range navigation would, however, be required, since the coverage
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
