The possibility of obtaining two-dimensional turbulent jet flow from a series of closely-spaced uniform holes in line has been investigated both theoretically and experimentally. The case studied was that of a jet discharging into still fluid of similar density at incompressible speeds. Such a quasi-two-dimensional jet is a particular example of a multiple-interfering jet group.

A large deflection theory is established for the bending and stretching of aeolotropic plates with elastic moduli that vary in the x, y, z directions. The theory is formulated in terms of two simultaneous non-linear differential equations, with variable coefficients, for the transverse deflection w and an Airy stress function F. Small deflection heterogeneous plate theory and some other sub-classes of plate problems are formulated by specialising the general non-linear plate equations. Effects of initial irregularities and of a variable elastic foundation are considered.

The concept of an idealised “unique” stage characteristic is employed to analyse and understand the performance of axial compressors. It is shown that there exists a “matching” line across the compressor characteristics at any point of which all the stages are operating at the same point of their “unique” stage characteristic, and that this matching line is readily obtainable, almost from inspection. Simple calculations lead to a derivation of both this “unique” characteristic and the effective area ratio of the compressor. The behaviour of the stages at other points than the matching line is readily understandable and presents a simpler picture of compressor performance than is often obtainable from actually measured stage characteristics.

This paper covers an investigation into a new way of assisting the pilot to hover an unstabilised jet-lift aircraft. In a normal control system the aircraft is controlled by a moment which varies linearly with stick position. In this paper control systems were considered in which the controlling moment was a linear function of the position and rate of change of position of the stick. The investigation was primarily limited to optimising such a system with freedom restricted to rolling motion and the associated lateral displacement. Effects of limited control power and engine failure were considered. With the optimum control the mean errors in position recorded when hovering were under half those achieved with a normal manual control and the results were almost as good as for an ideal rate-stabilised aircraft. Extending the investigation to both pitch and roll planes led to an even greater improvement, even when the effects of gyroscopic coupling between the two planes were taken into account. The precise engineering of a “mixed” control system of the type discussed needs much thought and it will require considerable development, but the results achieved indicate that it may prove an attractive alternative to an autostabiliser

Existing experimental data on the occurrence of attached flow around the leading edges of swept plane thin wings is summarised. It is shown that the change from separated to attached flow occurs while the Mach number normal to the leading edge is subsonic and that the Mach number required for the change increases with incidence. It is also shown that the Mach number and incidence conditions which hold on the boundaries dividing the two types of flow on wings with straight edges also apply locally at points on the leading edges of “Gothic” wings. The implications of this result on the flow over general curved-edge wings is briefly discussed.

The elastic stress concentration factors for the torsion of solid and hollow shouldered shafts have been determined by means of a pure resistance electrical analogue. Fillet radii ranged from 0.05 to 1.0 times the diameter of the smaller shaft, and the shoulder diameter increased from 1.0 to 8.10 times the diameter of the smaller shaft. A comparison is made with the results of other techniques. A study has also been made of the formation of a plastic region in the neighbourhood of the fillet.

The paper presents the results of an experimental study of the buckling of transversely stiffened shear webs when the stiffeners have a thickness less than that of the web-plate.

It is shown that the effectiveness of such stiffeners is greatly influenced by two factors, the ratio of the thickness of the attached leg to that of the web-plate and the position of the connecting rivet. An empirical design rule relating these two factors is presented.