During the past few years, the introduction of the large jet air liner has introduced new noises into the sky which has caused the public to be more aware than ever of the increasing nuisance value of aircraft taking off, landing, or on the ground. Much of this is being blamed unfairly on the jet air liner but the problem in any case remains with us.

For example, recent measurements given before the Society of Automotive Engineers are very revealing; here Baranek plotted the maximum noise levels on the ground when aeroplanes took off with a 5° take-off path.

The title of this paper could cover a number of interesting problems ranging from the ultrasonic scream from compressors and its (alleged) effect in producing “jet sickness,” to the noise from supersonic propellers but, since it is “in particular Jet Noise” which is under discussion, this paper deals entirely with that subject. Some new facts are presented, mostly obtained at my own firm, with the active encouragement of the Ministry of Supply, and some of their implications are discussed.

The heading aerodynamic noise includes all cases where sound is generated by fluid flows. It includes aeolian harps, edge tones and related phenomena as well as vortex, boundary layer, wake and jet noise. In the strict sense it does not include propeller noise and the direct noise generated by; bodies in motion. However, in view of their importance as sources of aircraft noise some of their more important aspects will be discussed briefly and compared with the similar properties of pure aerodynamic noise.

In general it is audible aerodynamic noise which is of major interest at present, but it would be unwise to neglect entirely the ultrasonic aspects. These have received little attention as yet.

Before discussing the noise associated with supersonic flight, it is pertinent to restate the sources of noise at subsonic speeds.

As Lilley has pointed out, noise at subsonic speeds is due to turbulence, such as occurs in jets, wakes, boundary layers, and regions of separation, and to vortices and regions of shear flow. In all these types of flow viscosity plays a predominant part, and therefore at subsonic speeds the noise of an aircraft depends very much upon the viscosity of the fluid: in an inviscid fluid, an aircraft would make no noise, because, at subsonic speeds, the pressure changes induced by its motion would be too gradual to be audible.

In cases of noise nuisance, measurement of the noise has several uses. It helps in deciding the degree of noise reduction which is desirable; it may serve to identify the most important source or sources of the noise where this is in doubt; and it is of assistance in devising methods for reduction of the noise. For the first purpose a single figure to characterise the noise would be preferable, but for the others some form of noise analysis is needed.

To the sufferer from the noise the most important characteristic is the annoyance it causes, or its nuisance value. This may be assessed very differently by different individuals, and consequently it is not very amenable to measurement.

The problems of noise on the ground coming from aircraft on the ground, while still not simple to resolve, are rather less difficult than those associated with noise coming from aircraft in flight. The source of noise—the aeroplane—is no longer travelling rapidly through the air with three degrees of freedom but is either stationary on the ground or taxying at a comparatively slow speed across an airport. In the first case, the aeroplane is at rest, either for a short period when the pre-take-off check is being performed near the end of the runway, or for a much longer time while a post-servicing engine check is being made in a maintenance area. During each check, up to maximum engine power may be used, although for jet aircraft the pre-take-off check is relatively short and appears to be becoming part of the take-off itself. In the maintenance area prolonged running is likely and will consist of operation at a medium engine power with occasional bursts of maximum power. Of the three ground cases, this is the one on which attention has to be focused. Indeed, it is not unreasonable to confine an examination of the problems of ground-to-ground noise to this and to this alone.

In the design and production of structures it is commonly assumed that the raw material supplied in the form of bar, sheet, forgings and castings is homogeneous, isotropic and stress-free. Some failures in service and more in the laboratory have drawn attention to the possible dangers arising from inhomogeneity, directional properties and residual stresses. It is the purpose of the present paper to review the situation with regard to the last, which the production engineer, who has to cope with distortion on machining, at least cannot ignore.

The torsional stiffness of a thin-walled tube of given perimeter and thickness varies as the square of the enclosed area. The shape giving maximum torsional stiffness is therefore a circle, this being the curve with maximum area for given perimeter. We consider here the nature of the corresponding curve for maximum bending stiffness. The quantity to be made a maximum is now, not the area within the curve, but the second moment for the whole curve about the bending axis. This will give the greatest “moment of inertia” for a tube of uniform thickness. The theory is given in the Appendix and typical forms are shown in Fig. 3. A finite length of base has to be assumed; otherwise the curve reduces to a single straight line normal to the axis.

Propeller moment is the name given to a dynamic phenomenon which causes considerable trouble in the design of variable pitch propellers and helicopter rotors. The development of the fundamental equation has not previously been published, to the author's knowledge, and for this reason the steps leading up to it are given. The propeller moment on any section tends to place its major axis in the plane of rotation, and the fundamental equation is equation. Its use to determine the total moment at a blade root results in lengthy calculations, with the possibilities of delays and errors.

In solving for the moments at the supports of a continuous beam, recourse is often made to Professor Hardy Cross's method when the number of supports is comparatively large. This paper describes a graphical method of moment distribution and it is proposed first to work out a specific problem by Cross's method, so that the validity of each step of the graphical method can be checked.

We wish to propose that the word “receptance” should be used instead of “admittance” as applied to mechanical systems (see R. and M. 2,000, 1947). By this, we mean that if a generalised disturbing force Fse iωt is applied at the s th generalised co-ordinate q s of a linear oscillatory system, then the response at the r th generalised co-ordinate is given by

q s = αrs Fse iωt

where αrs is the “Cross-receptance” between q r and q s. If r = s, then αrs would be the “direct receptance” at q s.