This informal history of the ARC outlines the background to its establishment in 1909, the subsequent growth and widening of its research activities for the advancement of aircraft design and construction, particularly around the two great wars, and finally its gradual decline until brought to a close in 1980. The story is told from a personal viewpoint and touches on the many personalities involved.

Simple first order surface singularity methods based on:

• (i) Smith and Hess uniform source panels plus a uniform vorticity around aerofoil profile,

• (ii) piecewise linear vorticity around aerofoil profile, with different assumption for Kutta condition,

have been applied to attached flows and separated flows past an aerofoil/spoiler configuration, assuming an inviscid model.

For separated flows, piecewise linear vorticity methods give reasonable results as long as small panel elements are taken in the region of the separations at the spoiler tip and aerofoil trailing edge. The Smith and Hess method gives results which do not agree too closely with the vorticity methods. There is doubt concerning uniqueness.

Results have been compared using two different wake models; in one, the total head inside the wake is taken to be uniform, in the second, the static pressures along the separation streamlines are taken to be uniform. There appears to be a difference of about 5% in CL. It is not known why.

A surface singularity method has been formulated to predict two-dimensional spoiler characteristics at low speeds. Vorticity singularities are placed on the aerofoil surface, on the spoiler surface, on the upper separation streamline from the spoiler tip and on the lower separation streamline from the aerofoil trailing edge. The separation region is closed downstream by two discrete vortices. The flow inside the separation region is assumed to have uniform total head.

The downstream extent of the separated wake is an empirical input. The flows both external and internal to the separated regions are solved.

Theoretical results have been obtained for a range of spoiler-aerofoil configurations which compare reasonably with experimental results.

The model is deficient in that it predicts a higher compression ahead of the spoiler than obtained in practice. Furthermore, there is a minimum spoiler angle below which a solution is not possible; it is thought that this feature is related to the physical observation that at small spoiler angles, the separated flow from the spoiler reattaches on the aerofoil upper surface ahead of the trailing edge.

While the bending analysis of cylindrical containers with wall thickness varying linearly has attracted much attention, it seems the general variation in wall thickness has not been considered. This is because of the difficulties which have been encountered due to the complex nature of the differential equations involved. In this paper, the container’s wall thickness is chosen to be of general variation and the differential equation is perturbed to give rise to a sequence of differential equations. It is shown that this sequence can be easily solved when the form of variation in wall thickness is specified. A complete solution is obtained in a particular case when the wall thickness varies linearly and the tank is subjected to hydrostatic pressure. For this particular case the numerical results are compared with the ones available in the literature which seem to be in good agreement.

The usefulness of zero pressure gradient, flat plate closure relations in providing approximate solutions for the boundary layer momentum and energy integral equations is examined. Expressions are obtained for skin friction, surface heat transfer rate and local Reynolds analogy factor under general compressible flow conditions. For laminar flows the predictions are compared with well known similarity solutions, with some exact solutions for non-similar flows and with experimental heat transfer data for high speed flow over a blunt cone. Consideration is also given to situations in which the surface temperature is a function of position. For turbulent flow situations comparisons are made with experimental data obtained from two-dimensional and axi-symmetric tests. Conditions vary from low Mach number incompressible flows through to high Mach number compressible flows with highly cooled walls. Some comparisons are also made with other prediction techniques.