FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1952
1952 - 0485.PDF
22 February 1952 POWER STEERING for AIRCRAFT A Summary of the R.Ae.S. Lecture by K. G. Hancock and P. Person THERE are certain major components of the modern aircraft which are so apt to be taken for granted that their design and refinements are largely overlooked; and when a specialist in one or other of these units reveals the details of his subject in a lecture there are usually many among his audience who obtain a proper appreciation of his task for the first time. These thoughts are particularly applicable to the paper Power Steering for Aircraft, prepared by K. G. Hancock, B.Sc, A.F.R.Ae.S., and P. Person, M.A., A.F.R.Ae.S. for delivery to the Royal Aeronautical Society last night, February 21st. This was one of the R.Ae.S. "main" lectures held periodically in provincial centres; in this case Hull was the venue. In their introduction, the authors emphasized that in providing power steering for the typical tricycle-undercarriage aircraft of today the prob lem was to provide a satisfactory performance for the minimum installa tion weight. As always, the matter was a compromise between what the pilot would like and what the designer could give, and cost as well as weight penalties must be kept down. The manoeuvres which the pilot would require to make with the aid of his steering were as follow: (i) turns of small radii at very low forward speeds to facilitate manoeuvrability on the apron and close parking; (li turns of larger radii at higher forward speeds on the runway; (iii) maintenance of directional control during the early part of a yawed take off and during the latter part of a yawed landing; and (iv) for a multi engined aeroplane, the correction of the out-of-balance thrust effects of the engines in a yawed take-off in which one engine was inoperative. Such an occasion arose when an aeroplane was required to return to its base for repairs to a defective engine. From the aircraft designer's point of view the theoretical aspects of the steering problem were as follow: (i) estimation of the net steering torque and steering rate; (ii) methods adopted to suppress shimmy and the amount of damping torque involved; (iii) estimation of torque neces sary to centre the undercarriage before retraction; (iv) estimation of the gross steering torque which followed immediately from the foregoing calculations; and (v) evaluation of the characteristics of the actuating mechanism. Hypothetical Aeroplane For the purpose of description the authors then considered a hypo thetical aeroplane of 120,000 lb all-up weight, with a twin-tandem-bogie main undercarriage and twin-wheel nose undercarriage. The nosewheel caster axis was vertical, the mechanical trail 5.4m, the total trail 7.9m, the wheel spacing 25m, and the weight of wheel assembly castering parts 185 lb. As a preliminary to the estimation of the net steering torque and rate of turn, the authors briefly discussed cornering and longitudinal-slip characteristics of tyres. When a tyred wheel which was rolling forwards was rotated about its vertical axis, the elastic properties of the tyre were such that it was possible for the wheel to track forwards while it rotated in a plane inclined to the line of motion at an angle called the "cornering angle." The restoring forces which arose from ground-friction effects were found to resolve into a side force acting perpendicularly to the plane of the wheel. This force varied with the vertical wheel load, the tyre-to- ground coefficient of friction and the elastic tyre properties. The lec turers showed curves for fully effective tyre-to-ground sliding coeffi cients of friction of 0.4 and 0.7, which were the approximate limiting conditions for wet and dry concrete-runway friction. The local peak value, occurring at approximately 20 deg cornering angle, was the static friction coefficient corresponding to the fully effective sliding coefficient. The cornering force was generally offset from the centre of the tyre ground-contact area by a distance termed "pneumatic trail." This trail, usually considered positive when behind the centre of contact area, was found to vary chiefly with the cornering angle and the length of contact path of the tyre. For a castering wheel with a vertical caster axis, the distance between the caster axis and wheel axis was termed "mechanical trail." The algebraic sum of the mechanical and pneumatic trails was termed "total trail." After discussing longitudinal slip of two wheels on a common axle the lecturers went on to consider the path of the aeroplane in a turn, to define steering rate—the rate of turn of the nosewheel about its caster axis—and to discuss net steering torque. Forms of shimmy and their suppression then came under discussion. Nosewheel shimmy was divided into two basic forms—"large-angle" and "small-angle" shimmy. The former was a shimmy in which the tyre reached its sideways limiting friction and slid sideways for a portion of each swing. It usually had a frequency in the region of 8 to 10 cycles per second and was extremely violent. ...... r Small-angle shimmy was an oscillation within the adhesion range of the tvre and usually had a frequency greater than the large-angle variety It was known to be influenced by the following factors: (1) geometry and relative masses of the undercarriage and wheel assembly; (11) lateral and torsional stiffnesses of the undercarriage and its mounting; (111) backlash of the castering parts; and (iv) tyre-stiffness characteristics. Small-angle shimmy must be adequately suppressed since, apart from obvious detrimental effects on the motion of the nosewheel, it was likely to induce fatigue loads and invited resonant vibrations elsewhere in the Suppression of shimmy could be achieved by one or a combination of methods: (a) caster axis friction; (b) wheel coupling; and (c) hydraulic damping. Discussing these three methods, the authors stated that built-in caster-axis friction should be employed to deal with small-angle shimmy and hydraulic-orifice-type damping used to suppress large-angle shimmy. Next they estimated the gross steering torque for the hypo thetical aeroplane. This was evaluated in terms of- P, the nosewheel reaction, and c, the total trail which, for the maximum nose load, were 14,500 lb and 7.9m respectively. Shimmy damping torque provided by leg friction and the multi- plate friction damper was taken as cP/16, i.e., 7,160 lb/in. This caster- axis friction torque would have only a small effect on the freely castering properties of the leg. The centring device, incorporated to ensure that the undercarriage was centred before retraction had to provide 1,000 lb/in torque if the design case of one g sideways was assumed. In most cases, the centring torque had to be overcome by the steering torque when steering the aeroplane. The maximum net steering torque was fixed so as to limit the cornering angles of the nosewheels to a maximum of 8 deg on a surface with a sliding coefficient of friction equal to 0.7. From a cornering curve shown the corresponding side force coefficient was seen to be 0.6 and so the requisite torque was 0.6 cP, i.e. 68,000 lb/in. Summing these torques and allowing 7 per cent for friction losses in the steering jacks, the gross steering torque was 82,500 lb/in, i.e., 0.72 cP. In their discussion of gross steering torque the authors mentioned the calculation of the steering path of the aeroplane, and the case of landing with one tyre deflated. Steering Circuits On the subject of steering circuits, the main requirements were: (1) to enable the pilot to steer the aeroplane; (2) to provide hydraulic shimmy damping; (3) to centre the unit before retraction; and (4) to permit free castering. As usual, the designer was faced with the choice of using hydraulic, electric, or pneumatic power; the authors selected for their purpose a main hydraulic system of 3,000 lb/sq in. The system of controlling the hydraulic power could be one of two forms having the following characteristics: (i) rate of steering proportional to the amount of error; and (ii) rate of steering independent of the amount of error, but a func tion of the frequency with which the error changed. Sensitivity could be denned as the ratio of maximum angular error between the pilot's control and the steerable wheels, to the full steering range. The factors giving loss of sensitivity were: (i) backlash on either side of the selector; that is, in the manual or power sides of the follow-up; (ii) excessive valve overlap and travel to the fully open position; (iii) insufficient velocity ratio at the lost-motion device to operate the selector; (iv) backlash in the lost-motion device. Sensitivity could be improved by mechanical means, such as reducing backlash in cable-runs by adequate pre-tcnsioning, or by increasing the velocity ratio at the point in the follow-up where the selector was oper ated. It could also be improved in the selector by using a negative- overlap scheme. The authors illustrated two basic circuits used to obtain shimmy- damping and free-castering characteristics, and discussed methods by which the nose undercarriage could be centred before retraction. For example, this could be done by centring cams engaged by the shock- absorber at take-off, spring loaded centring cams, and so on; but if one of more steering jacks were included on the nose undercarriage, it was an advantage if these could be used for centring purposes. Finally, the lecturers referred to back-pressure thermal effects and torque relief. Some back pressure in a steering .circuit was desirable to remove sponginess due to dissolved air in the fluid. The recommended value of this pressure was between 200 and 500 lb/sq in. The steering circuit must be so designed that excessive pressures were not produced by thermal expansion of fluid. Also, if fluid was lost from the circuit during expansion, the jacks should not be allowed to cavitate on fluid contraction, otherwise efficient hydraulic shimmy-damping would not be achieved. Torque relief was desirable to prevent excessive loads being imposed during ground manoeuvring. Authors' Proposed Circuit The lecturers then proposed and illustrated a steering and shimmy- damper circuit for their hypothetical aeroplane. It employed a slide selector giving rates of steering proportional to the error signal ; and mechanical follow-up with toggles round the retraction pivot. Shimmy- suppression was by a certain amount of built-in caster-axis friction plus hydraulic damping. The circuit was maintained full of oil under all con ditions by a pressure-maintaining valve set at 200 lb sq in, which also produced sufficient back pressure to remove sponginess. Free castering was obtained by allowing the jack and tank ports in the selector to be connected in the neutral position. Centring on retraction was by separate pistons incorporated in the steering jacks, and connecting the steering circuit to the undercarriage up and down lines so that pressure was applied to both sides of the jacks during retraction. This also ensured that steering was inoperative during flight. An effective steering torque from o to ± 40 deg was achieved and the unit was free to caster through ±90 deg without manual disconnection. The pilot's control was by a small handwheel and, with this circuit, an error between the pilot's wheel position and steerable wheels of approximately ± 4 deg at the central position would be expected. For permission to publish their paper the authors gave acknowledge ments to Electro-Hydraulics, Ltd.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
