An adaptive method for the computation of steady, viscous compressible flows over a wide range of Reynolds and Mach numbers is presented. The method is based on a well-proven scheme for the solution of the compressible Reynolds-averaged Navier-Stokes equations and employs the flexible AMR-grid structure for the adaptation of discretisation to the computed flow. The basic numerical scheme and the grid structure are reviewed and the sensors employed for the detection of the main flow phenomena are presented. Special attention is given to several flows computed with the adaptive method. The numerical results are partially compared to experimental data and generally a good agreement is observed. It is also recognised that, with the present adaptive method, flow features can be computed with high resolution using greatly reduced computing power. At this time, the method is limited to the simulation of two-dimensional flows, with the extension to three space dimensions currently under development.

A simplified trailing vortex concept has been used to model unsteady aerodynamic effects onto the mathematical model for an augmented unstable aircraft. An output error method with, and without, an equation decoupling technique in the frequency domain has been used for the estimation of parameters. The two approaches of estimation are compared for the cases where unsteady effects are included or excluded from the estimation model. It is shown that the incorporation of the equation decoupling technique makes the estimation algorithm robust with respect to initial guess values of the parameters. Finally, the inclusion of unsteady aerodynamics is shown to reduce the variation in the parameter estimates due to different control input forms.

Self-preserving turbulent flows in the farfield of binary gas jets in an arbitrary stream are investigated. The governing equations solved are valid for jet flows only and are not applicable to wake flows where a linearised set of boundary layer equations are more appropriate. Gaussian error-functions are assumed for the mean excess velocity and mass fraction distributions. Thus formulated, the eddy diffusivities for momentum and mass are evaluated by solving the governing equations and are shown to vary across and along the jet. A condition is imposed on the eddy momentum diffusivity so that it correctly approaches the limiting behaviour for the case of an incompressible free round jet. This condition gives rise to an auxiliary equation for the determination of the decay of centreline properties and jet growth. The new and more general auxiliary equation is shown to model fluid entrainment and to reduce correctly to previous equations derived for the special cases of free gas jets and incompressible heated jets in a co-flowing stream. Thus deduced, the new auxiliary equation is solved together with other conservation equations to yield a set of growth rate and decay laws. The derived growth rate and decay laws correctly approach the various laws derived earlier for different special cases and correlate well with experimental measurements reported in the literature. The present solutions also represent solutions for compressible jets in a general stream where the pressure field is constant. Finally, this approach of analysing the self-preserving mean jet flow is also applicable to the study of wake flows where the linearised boundary-layer equations are solved.

Shells, stability under various types of loadings and their combinations have not yet been completely investigated. This is due to incomplete knowledge of shell philosophy. The aim of this study is to provide a complete and integrated method for treating shell problems on a pure statical basis. This study introduces shell orthotropic behaviour, discusses hoop stress resultant actions, and applies the shell equilibrium equations to a uniaxially compressed cylindrical shell with finite length and fixed ends. Theoretical results are also compared with previously published experimental work. The behaviour of shells in elastic and plasto-elastic range is discussed for a wide range of shell geometries (dimensions), material properties, methods of manufacture, material imperfections, and methods of load application. From this study we conclude mainly that shells behave orthotropically. Their flexural rigidities are related to their radii of curvature and the resultant buckling mode. The buckling strength of a shell depends on shell geometry, material properties, buckling mode and shell coefficient H. This shell coefficient is affected by the method of manufacture, existing imperfections, degree of plasticity, and the method of load application. H can be evaluated by experiment.

A pressure instrument has been designed for the measurement of local skin friction in turbulent boundary layers on smooth and non-smooth walls. The instrument consists of three pitot tubes placed in the logarithmic mean velocity region of a turbulent boundary layer.

An instrument was built and tested in the boundary layer produced on both smooth and rough walls. The rough walls consisted of two sandpaper abrasives of sizes #40 and #24. The experiments were performed over a range of momentum thickness Reynolds numbers, Rθ, from about 5 × 103 to 18 × 103, and in a small favourable pressure gradient given by –0·0005 < (υ/ρU3τ)(dp/dx) <–0·0001. The non-dimensional equivalent sand roughness, ks+ ≡ ksUτ/υ varied from zero to approximately 85, thus spanning all three roughness regimes from smooth to fully rough. Skin friction coefficients calculated using the three tube instrument were compared to those obtained using a skin friction balance. For the smooth wall measurements, skin friction relations and Preston tubes were also used as a basis of comparison. The results were found to agree fairly well for both the smooth and non-smooth walls, and was independent of the instrument's position within the logarithmic region. However, due to measurement difficulties the scatter and uncertainty were larger for the three tube instrument than the other techniques.

Numerical smoothing of the results indicated that the agreement is expected to improve with better measurement of the pressure differences between the pitot tubes, or direct measurement of the velocity differences.

Experiments to study the behaviour of a compressible free jet issuing from an axisymmetric convergent nozzle are reported. Detailed measurements of the mean flow field for subsonic, sonic and underexpanded sonic jets was made within the pressure ratio range of 1.13 to 4. The results show that the spread and decay rates of the jet vary significantly with pressure ratio. The results further indicate the radial velocity/total pressure collapse within the two-dimensional shear layer data. The underexpanded jet propagates to a greater distance downstream than the low speed jets; this leads to enhancement of the far field mixing.

This paper presents the results of an investigation of the flow around rectangular and chamfered cavities at high subsonic and low supersonic speeds. Pressures measured on the faces of the cavities are integrated to find the pressure drag of the cavities. The types of cavity tested range from simple sawcuts to cavities so long that the two ends can be regarded as independent and the results for these are compared with the sum of the drags of isolated forward and rear facing steps.

Although the Reynolds numbers of the tests are similar to those in flight conditions the maximum depth of the cavities tested is only 6 mm so that the pressure resolution on the vertical faces of the cavities is limited. In spite of this it is estimated that the maximum error in the drag of any particular cavity is less than the skin friction drag on a smooth surface equal in area to half the plan area of the cavity.

In this paper some experimental results related to a calibration model (the Standard Dynamics Model, SDM) are presented.

The wind tunnel test activity is based on static force measurements and flow visualisations. The behaviour of the fuselage on its own is analysed and a correlation with some available prevision theories for the onset of asymmetries on ogival forebodies is proposed.

A complete set of static data for the SDM is introduced, evaluating the main aerodynamic interactions between the various elements of the configuration.

A series of flow visualisations was performed to verify with qualitative results the validity of the conclusions found from the analysis of the static coefficients.

After deriving the Stream Force Theorem, it is shown how this simplifies many problems of propulsion theory and makes the mathematics direct and concise. Applications discussed include forces upon ducting systems, intake entry losses, spill drag, mixing losses, errors in measuring intake losses by pitot traverse, and use of measured intake lip suction force in assessing entry losses and spill drag.

Interpreting test data or a mathematical function is often aided by plotting results in the form of contours. Many routines are available to accomplish this, but if a qualitative survey of the data is sufficient then a simple method can be employed using the basic colour screen graphics facilities found on personal computers. The principles of the method are described in this note.

To form a set of contours a base area of the function plane must be defined and the method uses the screen as this base area.

The permanent urgency of aircraft design during the author’s career at Hawker Aircraft, a.k.a. British Aerospace Kingston, inhibited publication of his advances of theory. So he collected those on propulsion and presented them to the Royal Aeronautical Society in a lecture “Some Unpublished Contributions to Propulsion Theory” on 11 March 1992. The meeting proposed that the material should be published in full, and F. W. Armstrong helpfully suggested that the best way for future reference would be to divide the material into Technical Notes on related topics, each capable of standing alone. These have been reworked to stand separately, as follows:

• 1) Fuel consumption formulae and the general range equation

• 2) The Stream Force Theorem and applications to propulsion

• 3) Non-isentropic duct flow and the general wake traverse

• 4) Propulsion streamtubes in supersonic flow and the first supercritical intake cowl

• 5) Rotary jet thrust augmenters

• (a) fallacy of Foa’s theory and correction

• (b) the jet fan augmenter.

Experiments have been conducted to study the impingement fiowfield, over a double wedge deflector, produced by the impingement of an overexpanded supersonic jet. The jet has an exit Mach number of 3.1 and an expansion ratio of 0.81. The distance between the nozzle exit and deflector apex was varied from 2 to 5 times the nozzle exit diameter. The experiment consisted of schlieren flow visualisations and measurements of pressure distribution. Load coefficient has been obtained by integrating the measured pressure distributions. Boundary layer thickness and skin friction coefficient on the deflector surface have been calculated by solving compressible turbulent boundary layer equations, using moment integral and moment-of-momentum integral equations in conjunction with the measured surface pressure distributions.

The preliminary design phase of an aircraft requires quick calculation methods to assess its performance. This note develops an analytical method to obtain the optimum range and associated flight parameters for a piston-prop aircraft. The aircraft is flying so that the altitude and airspeed are kept constant.

Such a flight is called “constant altitude-constant airspeed” flight or simply “constant h-V flight”. Transport aircraft generally utilise this type of flight profile. Such a profile has been analysed by Bert for turbojet aircraft, and Hale has discussed it for both turbojet and piston-prop aircraft. This note further improves the analysis of constant h-V flight by optimising it independently and without using the known values from any other flight profile.

After showing the extension of one-dimensional duct-flow equations to flows in which total pressure is lost from one station to another, it is shown how a pitot static traverse at a general station can be used to measure the drag developed upstream of that station. This was the first section of Ref. 1.

Low speed wind tunnel tests have been made to investigate the flow around a leading edge vortex flap at the maximum LID condition. Tests were also made to measure the performance of a vortex plate. The force measurements and flow visualisation tests were conducted on a 0·53 m span 60° delta wing model. Results indicate that the lift to drag ratio is a maximum for any given flap deflection angle that the flow comes smoothly onto the deflected vortex flap without forming a large leading edge separation vortex on the flap surface. The benefit of the vortex plate is seen in the drag results which are smaller than those for the datum wing. This benefit is due to leading edge suction acting on the forward facing region between the delta wing and the vortex plate.

The proper treatment of the pre-entry and post-exit streamtubes of supersonic aircraft and their “corresponding” windtunnel models was considered in some depth for supersonic fighter designs, and investigated in wave drag research, but as a result of the cancellations of the major supersonic fighter programmes the results were not published. This note sets out the salient conclusions.

The development of the first supercritical air intake cowl is also outlined.

The theory behind Foa’s rotary jet-thrust augmentor dates back to 1955, but there is a fundamental fallacy in it, which is proved by this note. When corrected, the augmentation value equals that of the static-jet augmentor.

The principles and theory of the jet-fan thrust augmentor are outlined. Thrust augmentation ratios of these will be around three times those of static-jet augmentors.

A major structural problem in the air transport industry is the very small profit margin i.e. 2-3% over a long period, e.g. 1970-1990. But the industry has to make heavy investment (12-15% of yearly turnover) on a regular basis in market growth and fleet renewal.

As airline cash flow is often insufficient they have to raise equity (limited returns in view of the low profit margins) and inevitably suffer from chronic debt. This is all the more expensive as an airline is often short of capital and made worse when interest rates are high (as is currently the case in France).

The current financial situation was caused by an unprecedented overcapacity crisis, in a time of economic stagnation. This has shown no sign of improvement since the Gulf conflict, and is all the more crucial today because it comes at a time when the air transport industry in general, and Air France in particular, are taking delivery of a large number of new aircraft.

This paper examines a specific case, Air France. This will enable some overall conclusions to be drawn, thereby making an effective contribution to restoring this sector to a sound economic balance in the medium term.

The problem does not just concern airlines. The aircraft financing sector is beginning to feel the effects of the crisis. They might do well to draw the necessary conclusions from this study. Also, one must not forget the aircraft manufacturers, who without doubt still have to face the bulk of the crisis ahead.