I would like to thank the directors of British Aircraft Corporation (Operating) Ltd for permission to give this lecture, but I must emphasise that the opinions expressed are my own. I would also like to acknowledge the considerable contribution that has been made by every Value Engineer in BAC to the pioneering work of introducing Value Engineering into our Corporation, and thank all those, both in BAC and in other aerospace companies in this country and in the United States, with whom I have discussed Value Engineering, and whose ideas have contributed to this paper.

In most one-parameter methods of calculating laminar boundary layers it is assumed that the non-dimensional parameters H=δ1/δ2, I=δ2τW/μu1 and L = 2{I−λ(H+2)}, depend only upon the pressure gradient parameter λ=u1δ22/v. In this paper it is shown theoretically that a more accurate, two-parameter representation is

L=F0(λ)−μG0(λ)

I2=F1(λ)−μG1(λ),

where μ=λ2U1U1/(U1)2. Careful examination of the available range of exact solutions of the boundary layer equations has enabled the four functions F0, G0, F1, G1, to be tabulated, and the above functional forms agree with the exact solutions to a remarkable accuracy.

In view of the fact that a reasonable first approximation to L is usually , we write

,

and it is then shown that the momentum integral equation becomes

This equation is easily solved by iteration, setting g=0 in the first approximation, and convergence is extremely rapid.

The method is, in effect, a refinement of that due to Thwaites, which is universally accepted as one of the better of the existing calculation methods. Detailed calculations made by the present method indicate that the errors are only 5% of those given by the Thwaites method.

Measurements are presented of the amount of air which flows axially towards the surface of a disc rotating at various speeds when the nature of flow is turbulent.

The method used is similar to that of Ricou and Spalding for the case of the axisymmetrical turbulent jet. The area to one side of the disc is surrounded by a chamber, leaving a small gap between the chamber and the disc. Air is injected into this via a porous baffle until the pressure in the chamber is uniform and atmospheric, a condition which is taken to signify that the “entrainment appetite” of the disc is satisfied.

Existing theories predict the results obtained well at high disc Reynolds numbers, but give greater entrainment rates than experiment at low disc Reynolds numbers. More precise measurements are needed to substantiate these conclusions: in particular the use of a hot-wire anemometer seems essential.

The experimental methods in current use for the determination of the dynamic properties of an arbitrary finite-freedom system are based on restrictive assumptions which may often be of questionable validity. The most common assumption is that the damping matrix can be reduced to a diagonal form by a congruent transformation using the modal matrix of the undamped system. Further, the damping is normally assumed to be hysteretic, or light if viscous.

Bishop and Gladwell give an excellent account of the theoretical background of the present experimental techniques. These can be divided into three groups:

• 1. The method of Kennedy and Pancu which obtains experimentally the complex receptance of the system and extracts its dynamic properties from vectorial plots.

• 2. The method of Trail-Nash and Asher which is based on the experimental determination of the in-phase response of the system and requires computation to be incorporated with the experimental procedure.

The basic theory of the compressible non-swirling rotational flows through ducts with varying hub radii, associated with the concept of actuator discs, is described. The problem is simplified by considering a single-parameter in the radial velocity profile. Particular attention is given to the combined effects of the taper of inside walls and also of compressibility on the radial velocity and the axial velocity profile. The derived ordinary differential equation with non-homogeneous terms can be reduced to the well-known formula of classical actuator disc theories for the cylindrical passage. A numerical example is also presented.

In recent years, manufacturers of centrifugal blowers for ventilation systems have abandoned the traditional arrangement of radial vanes shown in Fig. 1(a) for the “backward-aerofoil” arrangement shown in Fig. 1(b), which is the same as a conventional centrifugal water pump. Unlike the older type, the aerofoil-type blower does not depend on “centrifugal force” to produce a pressure rise despite the misleading accounts given in some hydraulics textbooks (the circumferential component of the air velocity is zero at entry and small at exit) although it does depend on the increase in blade peripheral velocity from the entry to the exit (which explains the curious blade arrangement).

A simple expression is obtained for the lift induced upon a wing by the pressure field of the fin mounted upon the wing in supersonic flow. This expression holds for all fins with sharp edges provided that the wing trailing edge is normal to the free stream and that the Mach lines from the apex of the fin do not intersect the wing leading edge. If the fin has a blunt trailing edge the further restriction that the wing trailing edge is upstream of that of the fin is also required.

Results for the induced lift and pitching moment are presented for a delta fin with variations in size and position of the fin relative to the wing trailing edge.

Seide and Weingarten have, recently, shown conclusively that in isotropic cylinders, the critical stress for bending stability is essentially the same as that in the corresponding compressive stability problem although the buckle pattern is quite different. The deflection pattern obtained in ref. 1 as a result of elaborate computations presents certain interesting features. The dominant feature is that the buckle pattern is an asymmetric one, because any axisymmetric mode is incompatible with the bending problem.

In recent years a number of authors have analysed the bending of clamped skew plates under uniform pressure. Many of these solutions are very approximate. Favre and Dorman adopted the Ritz method and their expressions for displacement are restrictive. Based on the generalised method postulated by Lardy, Mirsky has presented a solution which involves considerable algebraic and arithmetic work. Komatsu's paper contains a solution using conformal mapping. Quinlan has given a powerful method to solve skew plate problems involving different boundary conditions and various types of loadings, but no numerical work has been done.