The results of wind tunnel tests carried out on a semi-span ‘lambda’ planform model with leading and trailing edges swept at 40° are presented. The planform investigated has been specified as having features likely to be encountered on future ‘ow observable’ aircraft. Balance data has been acquired to angles of attack beyond stall for both forward and reverse flow orientations. Dynamic calibration of surface pressure tappings has allowed both time-averaged and unsteady pressures to be measured over the entire upper surface of the wing. Unsteady pressure time histories have been analysed in the frequency domain to identify their spatial resolution, dominant frequencies and associated magnitudes.

In forward flow conditions, maximum buffet is observed between the primary attachment of the leading edge vortex system and the vortex core. The dominant buffet frequency is shown to be dependant on the chordwise location, incidence and free stream velocity as expected. A second coherent buffet frequency has been identified in the region of high buffet over a wide range of operating conditions. Inboard of this region a third dominant frequency has been observed. This is of an intermediate frequency and varies in a similar manner. A simple mechanism has been proposed to explain both the spanwise RMS distribution and the presence of a number of coherent buffet frequencies.

The development of more powerful and efficient aero-engines requires ways of increasing the torque transmitted by shafts, whilst also restricting their dimensions and weight. Thin-walled designs can assist this objective, but their use is limited by their torsional collapse behaviour. Of particular interest are conditions leading to buckling instability. The paper investigates the factors influencing this behaviour in order to provide the basis for an improved analysis method applicable to typical gas turbine aero-engine components.

The Riks finite element algorithm has been successfully applied to both plain shafts and shafts with holes. In the former case, it is shown that the perfect cylindrical geometry must be given an initial perturbation in order to give accurate predictions. The perturbation imposed is obtained by scaling the mode shape from an eigenvalue solution so that the maximum radial deformation is a percentage of the wall thickness. The predictions for both plain and holed shafts have been validated experimentally.

The first part of this survey dealt mainly with the development of the earliest ideas on air resistance and their application to flight. That part of the survey paused at the achievement of powered flight during the early years of the twentieth century. Even then, the main problem remained as that of aerodynamics, a subject only vaguely understood by those aviation pioneers endeavouring to get off the ground. For them theoretical guidelines, as yet, could offer little help. One of the earliest of those very few analytically based results to achieve general acceptance was the statement that resistance is proportional to the product of fluid density, a characteristic body area and the square of the flow's speed. This result had appeared during the seventeenth century in Newton's work and had received some experimental confirmation. Newton's concurrent ‘rare medium’ concept, viewing fluid flow as a stream of disconnected particles, gained credence at that time since its results agreed with the above resistance proportionality. Yet this concept became mistakenly applied to airflow so as to predict plate resistances proportional to the square of the sine of the plates' incidence angles, a misleading result which caused confusion for at least the next century.

For an augmented aircraft, it is shown that filter error method leads to improved identifiability of aerodynamic parameters from measurements that are affected by the feedback from the flight controller. Simulated lateral-directional data of an advanced fighter aircraft, obtained from a nonlinear six-degree-of-freedom simulator, are used for the purpose of analysis. The aircraft stability and control derivatives are estimated using filter error method and compared to the estimates obtained from output error method to ascertain the degree of improvement achieved in the identified derivatives. Identification results with output error method show that feedback causes a discernible offset in the estimated values of the derivatives relative to the reference parameter values. The filter error method helps to reduce this offset and thereby yield estimates that match better with the reference parameter values.

The goal of life extending control (LEC) is to enhance the service life of complex mechanical systems, such as aircraft, spacecraft, and energy conversion devices, without any significant loss of performance, and can be achieved by making a trade-off between dynamic performance and structural durability. This paper presents the concept and a design methodology for robust life extending control of aircraft structures that are typically subjected to cyclic mechanical stresses. The controller design procedure relies on the specifications of flight performance and allowable fatigue crack damage at critical points of aircraft structures that serve as indicators of the effective service life. As an example, an aeroelastic model of the aircraft wings has been formulated and is incorporated into a nonlinear rigid-body model of the flight-dynamics. The H∞-based structured singular value (μ) synthesis method has been used to design robust life extending controllers based on a linearised model of the aircraft and a (nonlinear) state-space model of fatigue crack growth. The results of simulation experiments show significant savings in fatigue life of the wings while retaining the dynamic performance of the aircraft.

Prompted by claims made in the popular aviation literature, analysis is used in an attempt to estimate the turning performance of certain fighter aircraft flown during the Battle of Britain. In this it is necessary to provide values for such aerodynamic quantities as the drag coefficient at zero lift and these are estimated from published flight data.

The interaction between vortices, sound and combustion can lead to self-excited oscillations of such large amplitudes that structural damage is done. These occur because any small unsteadiness in the rate of combustion is a source of sound, generating pressure and velocity fluctuations. However, the velocity fluctuations perturb the flame, thereby altering the instantaneous rate of heat release. Instability is then possible because while acoustic waves perturb the combustion, the unsteady combustion generates yet more sound! Combustion oscillations can occur in afterburners and at idle in conventional aeroengine combustors. Lean premixed, prevapourised technology has tremendous potential to reduce NOx emissions, but is proving highly susceptible to self-excited oscillations. An overview of the physics of the interaction between vortices, sound and flames is presented, and illustrated by examples of instability in generic premixed ducted flames and in aeroengine combustors. The potential for both passive and active control is discussed.

Aircraft noise disturbs a significant number of people who live close to airports. It is tolerated due to the economic and social benefits those aircraft bring to society as a whole. However, the levels that are tolerated from individual aircraft must reduce, to compensate for increases in numbers of aircraft movements and the desire for an ever-improving environment for all of society's members. The issue is how to achieve that reduction without destroying social and economic benefits of air travel.

With respect to wind tunnel aerodynamic load measurement, an internal strain gauge balance (often referred to as a sting balance) is essentially a compact load cell designed to fit within a cavity of the aerodynamic body and form a link between the model and a fixed ground point via a sting support system. The structure of an internal strain gauge balance is designed to incorporate a series of planar surfaces such that the deflection of each surface is predominantly induced by a unique aerodynamic load. Strain gauges, mounted on groups of surfaces in a Wheatstone bridge arrangement produce output signals proportional to the applied aerodynamic loads. A strain gauge balance is calibrated by applying known loads, measuring the bridge outputs and then formulating an equation which relates the two variables together. Although calibration techniques are well established, reservations have been recently expressed concerning the ability of the associated calibration equation to satisfactorily model the response of the balance when subjected to a six component aerodynamic loading. This generally accepted calibration equation (referred to here as the traditional equation) results in a quadratic approximation to the behaviour of the output signals with applied loads, whereas a more appropriate variation would be cubic. Other limitations of the traditional calibration equation are that the behaviour of the balance to two simultaneously applied loads is based upon limited combinations of the two applied loads, and that the acquisition of the required loads from the strain gauge signals is frequently based upon an approximate matrix inversion method. The proposed calibration equation, described within this paper, models the behaviour of the sting balance to the third order, takes account of all possible combinations of two simultaneously applied loads, and avoids the use of an approximate matrix inversion when deriving the desired aerodynamic loading from the signal outputs. It is also shown that the proposed method may be used to determine the interaction of all possible combinations of up to three simultaneously applied loads.

This paper presents a comparison of turbulent skin-friction coefficients measured by floating element balances in zero pressure gradient conditions for a wide range of Mach numbers and Reynolds numbers. From these comparisons it is clear that although a large number of measurements have been made it is impossible to use these measurements to find the skin-friction coefficient for a given Mach number and Reynolds number to an accuracy of better than ±2 to 3%. This estimate of accuracy applies only to the range of Mach numbers and Reynolds within which there are a reasonable number of measurements to compare. Outside this range, say for Reynolds numbers based on momentum thickness greater than 40,000 and Mach numbers greater than 2 the uncertainty is greater.

A ten-year engineering research project on a super/hypersonic transport propulsion system (HYPR) initiated by AIST (Agency of Industrial Science and Technology) in 1989 has been successfully completed. The HYPR project was aimed at establishing the technology base for a propulsion system of supersonic/hypersonic transport (SST/HST) aircraft that will fly as fast as Mach 5. The project proved the feasibility of a combined cycle engine (CCE) for SST/HST by developing and running the world's first CCE , a combination of a variable cycle turbofan engine (VCE) and a methane-fuelled ramjet. The development of the CCE began with a variety of component tests, followed by the development of the high temperature core engine (HTCE) and VCE. This step-by-step approach led to steady and solid success.

Based on this success, a new project was started in 1999, titled "Research and development of environmentally compatible propulsion system for next generation supersonic transport (ESPR)."

This paper introduces the various achievements during the ten years of the HYPR project, and the prospects of the upcoming ESPR project.

The stresses around the passenger door, emergency door and window of an aircraft, Airbus A300B, are considered in this paper. By idealising these opening as rectangular holes with rounded corners in an infinite plate subjected to inplane loads and considering the presence of adjacent holes the stresses are determined using the complex variable method along with Schwarz Alternating Method. The results of the analytical solution are found to be in good agreement with those of the finite element analysis using ANSYS 5.0 software within 5%.

Current weaknesses in modelling the turbulent stresses arguably provide the main obstacle to relying on CFD predictions of complex aerodynamic flows. The paper reviews some recent strategies for obtaining more reliable models with especial focus on cubic non-linear eddy viscosity models and stress-transport schemes which comply with the two-component limit. Applications are shown for a broad range of test flows. Finally, important remaining weaknesses are considered.

'Life extension’ is the theme of this paper. However, unless there is a clear understanding of what ‘life’ means it is not readily obvious as to the appropriate means to adopt for its ‘extension'. This paper will explore some of the background to the meaning of life and the means being adopted to extend it for fixed wing civil transport aircraft. The focus will be on the safety of airframe structures from a regulatory viewpoint.

A comparative review is made of the evolution, on both sides of the Atlantic, of civil aviation requirements for the fatigue, failsafe and damage tolerance design of structures. Definitions of life are discussed in the context of recent developments of ageing aircraft structural programmes. These definitions are accompanied by an extensive reference list.

Example results are presented from a recent research study which compares the performance of a number of fracture mechanics analysis methods with conventional fatigue endurance approaches to the prediction of lives. The work was carried in a way directly representative of the current aircraft design environment. It was undertaken by three separate aircraft manufacturers and applied to a range of aircraft structural features that were not necessarily originally designed to damage tolerant principles.

The inertia characteristics of symmetrically finned projectiles are shown to be invariant of roll angle. This special property is used to derive ‘despun’ equations of motion so avoiding the rapid rates of change associated with the use of body fixed axes. By exploiting the special inertia and aerodynamic characteristics of symmetrically finned spinning projectiles it is possible to reduce the computer solution time required for long down range calculations. By combining the system with the use of quaternions for attitude determination, firings at high elevation can also be accommodated. A speed up by an order of magnitude is observed for the example of a 40mm shell.

We report our findings on the aerodynamic characteristics of thin slab delta wings at hypersonic speeds. We have tested wings with three sweeps (76°,70° and 65°) and four t/c ratios (0·04, 0·053, 0·066 and 0·10) using a special thin three-component balance. To date, the wing with tic of 0·040 is the thinnest wing ever tested without lee-side body at any hypersonic Mach number. The study was carried out at Mach number of 8·2, Reynolds number of 2·13 x 106 and incidence of -4° to 12°. In general, for any given t/c ratio, the increase in sweep decreased the CN whereas increased CM and CA. A possible explanation, for the effect of sweep, is given using the pressure data available in the literature. We also observed that for the sweeps considered, the t/c does not have any effect on CN and CL but increasing t/c increased CM and CA, reduced (L/D)max and moved the Xcp upstream. Blunt slab delta wings are shown to possess a unique property, wherein the CD increases with increase in sweep at higher sweep angles. This property also reflects in the (L/D) characteristics. Furthermore, the slab delta wings does not exhibit the classical CLα behaviour of sharp delta wings given in the literature. An empirical correlation has been developed to correlate CAo over a wide range of Mach number, t/c and sweep. Linearised theory has been shown to be useful in predicting the (L/D)max at Mach numbers as high as 20, for a wide range of t/c and sweep back angle.

Elastic buckling design curves are presented for flat isotropic rectangular plates, which are either isolated or form part of a larger continuous plate structure. The design curves illustrate the effect of introducing combinations of elastic rptational restraints to the edges of simply supported isolated plates, which are subject to uniform in-plane compression or shear. Results for infinitely long and/or wide plates, with rectangular bays, are superimposed for comparison.