A laser doppler anemometer (LDA) was configured to provide simultaneous ID velocity readings from two separate points in the contracting flow regime upstream of 0·63 m diameter model propeller operating under high thrust conditions. Cross-correlations between the velocity signals were used to map the progress of turbulent eddy structures throughout the distortion, even within the highly oscillatory region close to the blades. These measurements were used to provide data on the strength and velocity of the eddies, as well as their average size and shape in two dimensions. The results show both elongation of the turbulent structures in the streamwise direction and distortion along their length, caused by the contracting flow and the high Reynolds stress. Linearised isotropic theory was used to predict the changes in the turbulent kinetic energies, and was found to be accurate for the longitudinal component only.

The non-equilibrium fluid mechanics and thermodynamics of two-phase vapour-droplet and gas-particle flow are considered. The formation of the droplets as well as their subsequent interaction with the vapour are discussed. Five topics have been given particular attention: (i) CFD application to unsteady condensation waves, (ii) CFD application to shock waves moving through a vapour-droplet mixture, (iii) a new theory of nucleation of water droplets in steam turbines based on Monte Carlo simulation (steam turbines are responsible for 80% of global electricity production and the presence of moisture significantly reduces the turbine efficiency costing £50m per annum in the UK alone), (iv) a unified theory for the interpretation of total pressure and total temperature in two-phase flows and, (v) a unified theory of particle transport in a turbulent flowfield.

Simple straight and curved beams of unidirectional glass-epoxy were tested in fatigue to investigate failure under overall interlaminar tension and shear combined with local stresses at cut plies. Similar behaviour was observed as under static loading, with overall and local stresses interacting to cause delamination. When normalised, all the results lay close to, or above, a line determined from previous tension tests on specimens with cut plies, suggesting that this could be used as the basis for through thickness fatigue strength calculations.

A moving mesh method for the computation of compressible viscous flow past deforming and moving aerofoils is described. It is based on a well established time-marching finite-volume scheme, which has been widely used for steady compressible flows. An implicit real-time discretisation is used and the equations are integrated via a dual-time scheme. This involves the introduction of derivatives of a fictitious pseudo time. The solution at each real-time step involves seeking solutions which are steady-state solutions in pseudo time. This approach decouples the stability and accuracy limitations of the scheme and permits large real-time steps to be chosen. Also well-proven convergence acceleration techniques developed for steady flows such as local-time stepping, residual averaging and multigrid may be used in the pseudo-time stepping scheme without compromising real-time accuracy. A sequence of body-conforming grids and corresponding grid speeds is required, where the inner and outer boundaries of the grid move independently. The required grids and speeds are found using a transfinite interpolation technique. Applications of the method to external compressible flows are shown, including results from a parallel version of the computer program.

Integrated modular avionics has been proposed as a means of reducing avionics development and operating costs, by standardisation of avionics hardware and non-application-specific software. With the introduction of integrated modular avionics, there is the possibility of dynamically reconfiguring available resources, to preserve the most critical functions, when failures occur. This paper examines the feasibility of dynamic reconfiguration within the system architectures proposed by ARINC 651 and assesses the potential benefits. The analysis shows that at least two of the architectures proposed by ARINC 651 are well suited to reconfiguration and that although there are certification problems that must be considered, these problems do not appear intractable. Significant benefits, in terms of reduced redundancy, improved availability and higher levels of safety can potentially be obtained. The paper also shows that reconfiguration is only required locally, within a cabinet, and that large benefits are still obtainable even with relatively small cabinet sizes. This reduces the complexity and cost of any reconfiguration scheme and increases flexibility so that any reconfiguration scheme developed can be easily adapted to differing aircraft requirements. The development of an autonomous reconfiguration scheme, in which individual modules determine their own function is particularly attractive, as it can offer reduced susceptibility to common mode failure, and provides fault tolerance within the reconfiguration process itself.

The precise birth date of the aerospace composites industry cannot readily be identified; perhaps one should really talk about its rebirth as the first aircraft relied on natural composites such as wood. McMullen gives 1946 as the date that work on cellulose based composites for aircraft use was abandoned in favour of much more stable inorganic reinforcement fibres. This change in the direction of approach was crucial to further developments and can be thought of as marking the start of the aerospace composites industry that can be seen today. Whatever the exact date the industry is now about 50 years old so this golden jubilee edition seems an appropriate place to look at the constraints on the use of composite materials and at recent work at Bristol University aimed at reducing these constraints.