It is a privilege to pay tribute through this Cierva Memorial Lecture to a great pioneer and an honour to do so in this, the Society's Centenary Year. In 1918, when 24 years of age, Cierva designed and built a three-engined aeroplane which stalled during a turn and, after spinning, crashed. He recognised the limitations of the conventional aircraft and turned to the rotating wing.

Westland's association with rotary wings began with the five-seater 600 hp Cierva C.20, built in 1934. Their first series production rotorcraft, the S.51, to the design of Igor Sikorsky, came some 14 years later. With the passage of further time, bringing some successes and some frustrations, a sound foundation has indeed been laid for future progress.

In the past few years, man powered flight has become a reality. The remarkable achievements in England by the Hatfield Man Powered Flight Group prove that man may take off and fly with his own muscle power without external aid.

Such a possibility seemed very remote to the early pioneers of man powered flight in the late thirties in Germany and Italy. Even after the Second World War, some of the designers of these pre-war aircraft still believed that the take-off of a man powered aircraft would never be possible without some outside assistance.

In a supersonic transport aircraft, the propelling nozzle is required to operate at high efficiency over a wide range of conditions — typically a five to ten-fold variation of exhaust pressure ratio. This paper offers some discussion around the problems of geometry which such requirements entail, and explains the reasons for which a nozzle of centre-body form with parallel outer shroud appears an attractive solution. In support of the claims for this arrangement of nozzle, test data are presented from both single and twin models under representative conditions of external flow. (The tests were made at the National Gas Turbine Establishment between 1960 and 1965.)

In this paper we examine the hypothesis that aeroplane accidents are grouped, ie that individual accidents are not randomly spaced in time. While it has not been possible to find any explicit statements of the non-randomness hypothesis, it seems to be a widely-held belief. In its least precise form, the belief is that as soon as there is one aeroplane accident, there will be “several more within a relatively short period of time”. Clearly one does not expect equal or even near-equal spacing between aircraft accidents; the interesting question is what amount of bunching leads to the rejection of randomness as an adequate description of the underlying distribution. And if non-randomness is encountered in practice, how is it explained or interpreted?

The stability of a compressed plate is examined, using the energy method, with transverse deflections defined by a function giving the exact shape, but never specified. By this means it is possible to make observations about the form of the plate buckling coefficient. This form is examined tor two cases of loading, and it is demonstrated that a close approximation to the complete curve linking the buckling coefficient with plate aspect ratio can be deduced. The method is applicable to single plates or thin walled members consisting of an assembly of plates, and enables families of curves connecting buckling coefficients with aspect ratios to be summarised in a single relationship.

This paper describes a preliminary investigation of the torsional stress concentration factors for circular shafts containing grooves determined using an analogue computer. The range covered by this analysis is for grooves which increase in depth from a minimum of 0·05 of the diameter of the shaft, with radii which vary from 0·5 to 0·05 of the diameter at the minimum section. Results are presented in the form of a data sheet, and a comparison is made with the approximate expression of Neuber. An investigation is also made of the modifications to the stress concentration factor if the flank is not perpendicular to the centre-line of the shaft.

A simplified analysis indicates that the increase in profile drag of an aerofoil due to an isolated roughness element is, in general, different from the drag of the element measured on a flat plate with the same free-stream velocity. This “magnification” effect is caused chiefly by the effect of the pressure gradients on the boundary layer downstream of the roughness element.

The degree of magnification is not closely approximated by the ratio of local to free-stream dynamic pressure and, in many typical cases, the contribution to the drag due to roughness elements may be seriously under-estimated in this way.

Measurements of the effect of the initial boundary-layer thickness on the subsequent development of a turbulent boundary-layer in an adverse pressure gradient support the theoretical conclusions.

Experimental results are presented of the measurement of skin friction in fully developed turbulent flow in concentric annuli using Preston tubes situated on the inner and outer annular surfaces. Both Preston's calibration and Patel's calibration were used to evaluate the results. It was found that the latter gave excellent results. Several radius ratios were investigated with a reasonable range of the annulus Reynolds number. The good agreement was not affected by radius ratio or smallness of core tube within the range of these parameters investigated here.

A need exists for empirical criteria by which the behaviour of hypersonic boundary layers flowing in adverse pressure gradients can be predicted, thereby placing the design of high Mach number intake/diffuser systems on a rational basis.

The main results of the inviscid flow performance analysis of a two-dimensional “isentropic ramp” intake diffusing to supersonic Mach numbers over the free stream Mach number range 6 to 10 are presented. Limitations on the degree of diffusion, and hence geometry, have been deduced in terms of the combustion length requirements of an idealised hydrogen/air combustion process with com-bustor inlet conditions given by the intake. The analysis can be considered as a first order approximation defining the ranges of continuous pressure gradients and oblique shock wave strengths relevant to the experimental investigation of boundary layer flow in the mixed adverse pressure gradients associated with hypersonic intakes.

It is well known that in the case of turbulent flow over smooth boundaries, the velocity distribution in the neigh- i bourhood of the wall is given by the law of the wall, c written as

where u is the turbulent mean velocity at a normal distance of y from the boundary, u* is the shear velocity equal to being the boundary shear stress and ρ the mass density of the fluid and v is the coefficient of kinematic viscosity. In the viscous sublayer, eqn. (1) becomes

and in the turbulent flow region above the sublayer and the transition region, for yu*/v> 30, eqn. (1) becomes

where A and B are coefficients.

Recently Alexander suggested modifications to the inviscid theory of Stanton Jones for improving the agreement between theory and experiment on the cushion pressure of hovercraft. However, these simple inviscid theories can hardly be expected to represent even approximately the behaviour of peripheral jets in hovercraft and a more plausible theory taking mixing into account can easily be developed from some work of Bourque and Newman on the re-attachment of plane jets to adjacent surfaces. This theory, which is somewhat similar to a theory of Chaplin, can be used to predict the behaviour of peripheral jet models both in and out of ground effect but it is the intention here simply to compare the predictions of the mixing theory in ground effect with the simple inviscid theories which seem so commonly used in cushion craft work. Also, the advantages and potential of the mixing theory will be discussed briefly.

In a recent technical note Rajaratnam has pointed out the possibility of using the pressure difference measured between two static pressure holes of different diameter as a means of determining skin friction. The device exploits the fact that for any hole there is an error in measurement of static pressure which depends both on the hole diameter and on local skin friction, as Shaw has demonstrated. Consequently the difference in pressure measured between two adjacent holes of different diameter will be a function of the skin friction. In this note the device will be called a “static hole pair”. The larger hole will be referred to as the “test” hole, and the smaller hole as the “reference” hole.