Fifty years ago, the A.V. Roe Team took the Lancaster four engined bomber (Fig. 1) from concept to first flight in a period of about nine months. Under the driving pressures of War and with an outstanding approach to the challenges of the project, a number of key concepts emerged, which are still considered modern and innovative today:

1. Teamworking and leadership

2. Design for ease of production

3. Simultaneous engineering

4. Materials and parts supply logistics

This approach considered the aircraft engineering process as a whole and kept the ultimate aims of the project uppermost.

The project achieved remarkable manufacturing performance targets; assembly of 49 aircraft per week at peak production, combined with a three and a half day assembly lead time.

After the War, the need to manufacture aircraft in large volumes decreased and at the same time rapid technical advancements dramatically increased aircraft complexity. The nature of the industry changed and the balanced application of design engineering and manufacturing logistics that had made Lancaster manufacture so successful, became of secondary importance. During the 1970s and 1980s improvements in manufacturing within the aerospace industry were fragmented and now, at the beginning of 1990s, when the UK aerospace industry is compared with other sophisticated, high-technology industries, most business performance indicators are firmly at the bottom of the league (Fig. 2).

From the first, let me nail my colours firmly to the mast by quoting Beethoven’s eventual dedication of the “Eroica” Symphony. For this particular Lanchester Lecture we are “to celebrate the memory of a great man”. My meaning here, I hasten to add, has none of that bitter intention of the Republican Beethoven who, so the story goes, having ripped out his earlier dedication to Bonaparte then substituted those words to commemorate the man once revered but who had recently proclaimed himself Emperor of the French. My intention is that we celebrate the memory of a man whose thinking remains as alive, as vital, as the day he set pen to paper. Each day of our lives we see his vision made manifest reality in the vapour trails of the Big Jets high overhead. And to borrow such a dedication seems appropriate since, not only was Lanchester a great admirer of Beethoven’s music, but also Lanchester’s largeness of spirit, his grandness of vision, his capacity for unremitting sheer hard labour put him in the heroic mould.

A review of recent developments in helicopter noise reduction is presented. The paper includes a discussion of the present state of understanding of helicopter noise radiation mechanisms and a review of actual noise reductions achieved by the manufacturers over the past decade. An analysis of progress achieved in meeting certification limits is given, together with an evaluation of the impact on design. Future trends in helicopter noise reduction based on current noise research programmes are discussed and suggestions advanced for possible additional areas of study.

When I look back over the list of those who have given each of the previous 36 Roy Chadwick lectures since their inception in 1956, I observe a number of interesting features. The list includes some of the most illustrious names that have contributed to the pioneering and maturing phases of the design and development of gas-turbine powered flying machines. I was fortunate to have worked for and been in contact with — not to say under the lash of — at least two of them: Sir Stanley Hooker and John Coplin.

Heat transfer and pressure measurements were made on a model of the United States Space Shuttle Orbiter in the University of Queensland's T4 shock tunnel at three angles of attack, with stagnation enthalpies which varied by a factor of 11, from 2·1 MJ/kg to 23 MJ/kg and normal shock Reynolds numbers which varied by a factor of 38, from 2·1 x l04 to 8·l x 105. Leeward pressure results were obtained for comparison with flight data and equilibrium calculations, but the majority of the experiments were conducted to investigate the heat transfer distributions around the shuttle model. Both the windward and leeward heat transfer results exhibit the onset of transition to turbulent flow. The leeward results are compared with flight data as well as with conventional wind tunnel data, and are used to establish trends associated with variation in the major flow and geometry parameters. It was found that although high enthalpy effects could be important, Reynolds number effects played a dominant role in determining the flow.

Willy Messerschmitt’s most notable impact on aviation history occurred during the wartime period 50 years ago. While there is a general belief that the business approach under the extraordinary circumstances of this period is in no way relevant to today’s business environment, this paper will, nevertheless, attempt to illustrate that some elements of Messerschmitt’s approach may still be worthy of consideration.

Messerschmitt’s approach had not been that of “big business” but rather comparable to that of an artist-craftsman. Since the approach of an artist is best reflected by reviewing his “oeuvre”, this paper will be proceeding accordingly. Figure 1 shows Messerschmitt in 1935.

Longitudinal vortices have tremendous practical utility for flow control (control by vortices) and in some cases have exhibited tremendous potential for causing harm if uncontrolled (i.e. control of vortices is required). Vortex control has thus far been carried out via multitudinous approaches in an empirical fashion, aided by the essentially inviscid nature of much of longitudinal vortex behaviour. Further refinement and several applications of vortex flow control require knowledge regarding the detailed flow physics of longitudinal vortices such as transition, transitional flow regimes, turbulence structure and modelling, and interaction with shock waves, other vortices and surfaces. This paper summarises vortex control applications and extant techniques for the control of longitudinal vortices produced by bodies, leading edges, tips, and intersections.

It was while lecturing at Cranfield in the late 1970s and early 80s on the aerodynamic design of aircraft wings that the ideas discussed in this paper began to develop. My work on wing tip sails had led me to look in some detail at the flight of birds and their ability to vary wing camber. I realised that this was an attractive way to overcome the need to compromise between the design requirements associated with the different regimes of flight necessary with fixed geometry.

To be economically attractive, civil aircraft have to be capable of operating over a wide range of flight conditions. For example, an executive jet may be required to fly fast and low at light weight over a short stage length, such as London to Paris, on one day, and fly heavily loaded a much longer stage length the next, when it is desirable for the aircraft to fly high and at an economical speed.

Juan de la Cierva was born in Mosei, Spain nearly 100 years ago on 21 September 1895. At the age of 15 he constructed with two friends, a glider of his own design and contined to show great interest in aviation and aircraft design. He gained an engineering degree and in 1918 built his first tri-motor aeroplane in Spain. His concept was to provide a flying machine having a parachute capability for improved safety in the event of an engine failure and he worked in the direction of the helicopter concept, but concluded that true helicopter flight was not technically feasible with the state of the art at that time. He concentrated therefore on short take-off and landing combining both the fixed wing and the helicopters physical characteristics and called his invention the autogyro. The first successful flight with his autogyro was on January 9th 1923 at Getafe Aerodrome, Madrid. By 1925 development aircraft had achieved reliability and he began demonstrations in France, England and the United States before starting to produce the autogyro in quantity in both England and the United States. It was Juan de la Cierva's main rotor blade design and creation of the articulated rotor system that played such an important role in the first successful helicopter which flew in 1936 and in all subsequent helicopters.

The experimental results of an oblique Blade-Vortex Interaction (BVI) study are presented. The quality of all pressure data reflects improvements in the Glasgow University BVI facility and in the method of reducing and presenting data. The data collected during oblique interactions is found qualitatively and quantitatively similar to that collected in corresponding parallel interactions, for interactions within ± 30° of parallel. Details of the pressure data are examined in the light of understanding gained from parallel BVI experimentation. The study highlights the effects of three dimensional flow interactions not accounted for in the nondimensionalisation traditionally utilised in the analysis of parallel blade-vortex interactions.

A unified analytical method is presented for evaluating the generic buckling behaviours of simply-supported thick orthotropic cylindrical shells subjected to both axial compression and lateral pressure. The thick shell theory and Flugge type equations are employed. The equilibrium equations of shells become particularly simple in the chosen space by considering an appropriate transformation and utilizing some previous material constant definitions due to Brunelle. Hence, comprehensive solutions are found to this problem so that the curves presented in the text are generic rather than specific. The analysis is for the restricted case of a single homogeneous layer. The effects of various parameters on buckling loads are studied. The results indicate that the buckling coefficient increases with increasing generalised rigidity ratio D* but decreases with increasing generalised Poisson's ratio ε. The results of a reduced set of equations agree with those obtained by previous investigators.

The experimental results of a parallel Blade-Vortex Interaction (BVI) study are presented. The quality of all pressure data reflects improvements in the Glasgow University BVI facility and in the method of reducing and presenting data. Pressure histories clearly track the passage of a vortex-associated disturbance across the aerofoil. Tracking this disturbance permits an examination of vortex convection velocities, and also allows for a correlation between vortex trailing edge passage and force and moment perturbations.

A general-purpose low turbulence wind tunnel was constructed using the design method of Bradshaw. Sound absorbent material was used in every four corners to decrease sound intensity produced by a fan. The longitudinal component of turbulence intensity at the centre of the closed working section is less than 0.02% of the mean velocity in the speed range between 18 m/s arid 53 m/s. The mean velocity variations across the working section are within ±0.1% of the mean velocity.

Performance measurements have been done at representative tunnel cross sections to clarify the behaviour of flow in the tunnel. This work differs from previous studies in the sense that emphasis is placed not only on velocity distributions, but also on turbulence intensity distributions at several cross sections of the tunnel.

The critical Reynolds number for a flat plate at zero incidence, Rec = 3.5 x 106, measured in a stream of the very low turbulence intensity of 0.016%, is larger than that reported by Schubauer and Skramstad.

The application of the unstructured mesh system FLITE3D.0 to the numerical analysis of the inviscid Euler flow past installed aero-engine nacelle is described. The main features of the system are detailed and the approach is demonstrated by simulating the flow over a Rolls-Royce model of a twin-engined civil transport aircraft with long-cowl nacelle powerplants. The computed pressure distributions are compared with the measurements produced by transonic wind tunnel experiments.

The fatal accident rate in the world commercial fleet of large jet transports has been on a near plateau for more than 15 years since the absorption of the major advances made with the last generation of technology and operational systems.

Notwithstanding continuing technical advances, the increasing size of the world fleet demands more significant improvements if the actual number of accidents is to be reduced.

This paper promotes the view that a further overall increase in the safety of air transport from the high levels currently achieved is unlikely to come about by normal evolution. It suggests that what will be required in future is planned improvements, quantified by statistical safety assessment methods. Such assessment should encompass the totality of an integrated Air Transport System, as distinct from the current practice of assessing the Aircraft Airworthiness, Air Traffic Management, Flight Operations Management and Air Navigation systems in isolation. The setting of new targets for overall safety should derive from an integrated systems approach.

A relevant history of systems statistical safety assessment is reviewed, and certain aspects of cockpit automation and systems which may contribute to the future advancement of overall safety are considered.

The flow-fields associated with single and twin jets impinging in cross-flows have been studied. Parameters which affect the position of the ground vortex have been investigated: cross-flow-to-nozzle velocity ratio, cross-flow boundary layer thickness, nozzle height, nozzle pressure ratio, vector angle and nozzle splay with both fixed and moving ground planes. Results show that the ground vortex moves away from the nozzle centre-line as the ratio of cross-flow velocity to nozzle exit velocity is decreased; the rate of change of position, however, depends on other parameters (including the precise definition of nozzle exit velocity). Increasing nozzle height above the ground appears to cause little consistent variation in vortex position for a single jet but a marked forward movement of ground sheet separation point in the case of twin jets. For all cases there is an increase in penetration with increasing nozzle pressure ratio up to choking, with the subsequent behaviour dependent on the definition of nozzle equivalent velocity and cross-flow velocity ratio. The effect of the moving ground plane is to reduce vortex penetration significantly; this suggests that a moving ground plane simulation (or moving model) is essential when testing design configurations in ground effect in wind-tunnels. It is also shown that rig design can produce a blockage effect which moves the ground vortex significantly and can change other apparent parametric effects. Self-similarity laws are proposed for the ground vortex and the wall jet.

The effect of camber on the aerodynamic characteristics of an infinitely thin aerofoil approaching the ground is studied using a numerical model that accounts for the unsteady nature of the flow. The wake is computed as a part of the solution and the image technique is used to account for ground effects. The results indicate that relative deviations of lift and moment coefficients, from corresponding values far from ground, decrease as the camber ratio or angle of attack increases. Moving the location of maximum camber backward has a similar effect. Meanlines of NACA 4-digit series showed smaller deviations compared to NACA 6-digit series meanlines. Increasing camber ratio or moving maximum camber backward causes the wake trajectory to be lower. The steady ground-effect approach is inaccurate and may give erroneous results for plates with high camber ratios at small angles of attack, especially near the ground.

Numerical solutions of the Reynolds-averaged Navier-Stokes equations are presented for the shock/boundary-layer interaction which occurs with the attached, near-adiabatic, turbulent flow of air over a two-dimensional compression corner of 11° at a Mach number of 2·5.

It is shown that for this near-adiabatic flow, the isenthalpic and non-isenthalpic Navier-Stokes equations produce solutions which are very similar in respect of the velocity,pressure and density fields.

In the first part of the work, two different upstream, or approaching boundary-layer profiles are considered: one with a normal full velocity profile and the other with a less full velocity profile corresponding to a boundary layer that has developed in an adverse pressure gradient. Several turbulence models are compared using the isenthalpic form of the equations (because the computations are more economical). It is found that agreement with previous experiments (such as in the location of the shock and density profile predictions) is improved in the vicinity of the interaction using the non-equilibrium Johnson-King model compared to the equilibrium models of Baldwin-Lomax and Cebeci-Smith.

The second part of the work examines the effects of turbulent Prandtl number (Prt) modelling for this flow using the non-isenthalpic form of the Navier-Stokes equations. Three models for the turbulent Prandtl number are tested. It is found that modelling Prt affects the velocity, pressure and density fields only very slightly, and that the main impact is on the prediction of temperatures very close to the wall. The adiabatic wall temperature variation through the interaction has been measured experimentally and compared with the predictions. The best agreement was obtained using Wassel and Catton's model for Pr, rather than the familiar assumption that Prt = 0·9 everywhere in the flow.

Results are presented of an investigation into the interference effects associated with end plates of a type commonly used for the simulation of infinitely long (‘two-dimensional’) cylinder flow conditions around models confined with the working section of a wind tunnel. Experiments were performed with smooth circular cylinders of low area blockage in steady, low-turbulence, uniform airflow at a Reynolds number of 4·4 x 104 and involved the measurement of vortex shedding frequencies in the wake of the body, together with mean and fluctuating surface pressures, from which coefficients of local mean drag, r.m.s. lift and r.m.s. drag were calculated, and the use of a surface flow visualisation technique. The results reveal the aerodynamic characteristics of the disturbance induced by an end plate within the interference region which extends to 3·5 diameters from each plate.