The idea of this pair of lectures is to give some idea of the problems which face an official organisation when it is given the task and the responsibility of making recommendations to allow a helicopter to operate in icing conditions. The principles of the problems are not new— they will be recognised by all quality assurance people, from those who have to vet the design of the Severn Bridge to those who have to certify the safety of a new dog food. We are engaged on another chapter of the everlasting question: how do we define the environment; how do we find the test condition; and, how do we interpolate or extrapolate from what we have tested to what will be found by the operator?

A comprehensive survey of advanced materials for civil aircraft structures from the research point of view would of course include new aluminium alloys, titanium alloys, steels, composites, transparency materials and adhesives. Such a survey is, however, out of the question within the scope of this part of the paper and attention will therefore be concentrated on fibre-reinforced composite materials, partly because as a primary structural material they are now widely recognised to offer much the greatest prospect for really substantial improvements in structural efficiency (weight saving of the order of 30%, for example Fig. 1.1 shows comparative specific properties of various structural materials) and for lower costs, and partly because to realise these improvements a wide and in many ways novel range of research problems throughout the whole spectrum of material development and characterisation, design, fabrication and testing needs to be solved. It will nevertheless be possible, even with this limitation, to do no more than to give a brief sketch of the present position in some of the main research areas, primarily from the UK point of view.

Many investigators have studied the rolling-up of the vortex wake behind a wing of finite span. The earliest step in this study was the work of Kaden who found an analytical solution for the rolling-up with time of a semi-infinite, straight, two-dimensional vortex sheet. This solution must represent the situation very close to the edges of a finite-span vortex sheet of zero thickness, in two or three dimensions, during the initial stage of the rolling-up process. An important result following from Kaden’s work is that, from the very onset of rolling-up, due to the infinite velocity at the sheet edge, a spiral of near-axi-symmetric form, with an infinite number of turns, is established at the edge. This is a consequence of the assumption of zero thickness for the sheet.

At present it is still very difficult to predict for a given incidence, sweep-angle, leading-edge shape and Reynolds Number, whether the flow around a wing will remain attached or separate either from the upper surface as a bubble or from the leading edge as a vortex sheet which rolls up above the wing. Various attempts have been made to remove this uncertainty and it was suggested by Küchemann that it should be possible to design wings which combine coiled vortex sheets, originating from the inner part of the leading edge, with attached flow over the leading edge of the outer part of the wing. In this way some of the properties of slender wings could be combined with some properties of swept wings. His original model consisted of a swept-wing planform modified by inboard leading-edge wing strakes, the purpose of the latter being to provide the inboard vortex sheets. A similar idea has now been used in the wing design of the YF-16 lightweight fighter.

Advances in computational methods are producing a need for accurate and detailed experimental results for test cases. At the same time economic trends force many of these experiments to be made in small wind tunnels. In some cases the small scale of the model does not provide enough space for the instrumentation and so the model must be distorted from the ideal shape. This paper is concerned with one aspect of this problem, that is, with the need to make wings thicker than required, either for strength or to provide volume for pressure tubes and other measuring devices. In particular the paper deals with the conditions under which the pressures measured on the flat surfaces of wings with bevelled or triangular cross-sections as shown in Figs. 1(b) and 1(c) are a good representation of the pressures which would be measured on the corresponding flat surface of thin wings (Fig. 1(a)). This problem has been touched on in a number of papers reporting experimental results but it does not appear to have been tackled systematically. The discussion of this paper is based on these earlier tests and on some new test results.