This paper describes a study to characterise the impact threat to aircraft from runway debris. This work investigated the likelihood of an event exceeding a critical impact energy; it did not address the probability of an event occurring. Runway debris was collected from six airfields and parameters such as size fraction, mass, density, bluntness and material type were characterised. The distribution of the stone mass against frequency was accurately modelled and based on this, an expression was developed for the probability of an impact exceeding a critical level during a single take-off or landing. By considering the probability of a 50J impact from runway debris as a once in a lifetime event, the implication is that the actual threat is significantly more than the currently perceived threat. This study provides the foundations for more comprehensive studies and will give the designers improved understanding of the threat to aircraft from runway debris.

A continuous frequency domain method with roots on the classic Hill’s determinant analysis is presented to approximate the time-varying characteristics of a linear periodic system. The method is particularly useful to derive a time-invariant equivalent form of the time-varying aeroelastic problem of a rotor blade in forward flight. The proposed technique allows methodology usually employed in fixed wing aircraft to obtain closed-loop control laws be extended to rotary wings. The method is first validated solving Mathieu’s equation. Next, the two-degree-of-freedom (flap bending and torsion) problem of rotating beam subject to unsteady and incompressible aerodynamics in forward flight is solved in the laplace domain. As a demonstration of the proposed method, the transfer functions in the ‘s’ plane between a sudden and uniformly distributed input pressure perturbation applied along the beam and the output response of the two elastic degrees of freedom considered are obtained at a set of local sections.

A systematic methodology for the thermodynamic optimisation of civil bypass engines (turbofan or advanced propulsors) is presented, which would be useful for designing air-breathing engines based on “clean-sheet analysis”. The process starts with establishing an optimum specific thrust for the engine based on an economic analysis (installation constraints, noise regulations etc. also need to be considered). The task of the optimisation process is then to find the combination of optimum values of fan pressure ratio, overall pressure ratio, bypass ratio and turbine entry temperature concurrently that maximises overall efficiency at the fixed specific thrust. This procedure is quite different from the usual single-variable parametric performance studies which do not give proper optimum values and may involve large excursion in the value of the specific thrust unacceptable for a particular mission. Additionally, several, simple and explicit, analytical relations are derived here from fundamental principles, which perform well against numerical optimisation performed by a specialist computer program employing iterative and advanced search techniques. The analytical relations accelerate the optimisation process and offer physical insight. Present numerical computations with real gas properties have established new concepts in turbofan optimisation (for example, the existence of an optimum bypass ratio and optimum turbine entry temperature). The question of optimum jet velocity has been addressed. An analytical expression for the optimum jet velocity at a given bypass ratio has been derived which performs well against numerical optimisation results.

In this paper we describe a new 35GHz air-borne radar developed jointly by DERA and Alenia Marconi Systems. This radar is a functional prototype of a more highly configured imaging radar which is designed to meet the needs for all-weather imaging both for future unmanned air vehicles (UAV) and for seekers in the next generation of surface attack weapons.

The radar achieves high range resolution through a combination of pseudo-random phase coding and frequency agility. This is then augmented by synthetic aperture techniques to allow forward looking high resolution 2D imagery to be obtained. The radar has been installed on a fixed wing aircraft and has been used to gather fixed and relocatable target signature data under tactically representative conditions.

The paper begins with a discussion of the requirements for such a radar and then a review of its design and performance is presented. This is followed by examples of target signatures gathered during the initial system flights and a discussion on their exploitation for surveillance, target acquisition and tracking.

Recent interest in aircraft vortex wakes has highlighted the need to identify appropriate algorithms for their simulation. The problem is not well-suited to grid-based techniques, due to the large domain and the need to avoid introducing artificial viscosity. A promising alternative is offered by Lagrangian formulations. This work aims to assess the usefulness of one such approach - the 2D vortex method -via comparison with experimental results. One in-house and two external (from DLR and ONERA) data sets are used. The results show generally good agreement in vortex locations and size, with the calculation performing best when initialised downstream of the strongly-three-dimensional trailing edge region. The neglect of viscosity can lead to some asymmetry in the vortex cores, but, even here, excellent agreement can be obtained when the field is integrated to give circulation (as a function of radius). The 2D vortex method is thus an accurate and efficient tool for wake vortex prediction.

A review of air–to–ground weapon aiming is given, with emphasis placed on the use of the head–up display (HUD), the main cockpit instrument used for accurate weapon aiming over the last 35 years. Nevertheless, the HUD is only of use for the aiming of forward–firing weapons. More advanced weapons have an off–axis capability and their aiming is greatly facilitated by the use of a helmet-mounted sight (HMS) or helmet–mounted display (HMD). The surface–to–air threat and the rules of engagement, particularly in operations other than war, place high demands on the aircrew and the weapon aiming system, both to stand off from the target and to have a high degree of confidence that it is the target. The requirement to perform an accurate in–flight transfer alignment of the weapon places further demands upon the aircrew. Timely and accurate target data, digitally received, plus an on–board targeting system which can automatically search for and recognise a target, are of great utility in the final stages prior to weapon release. The Defence Evaluation and Research Agency is performing research in these areas.

Several afterbody flows, involving shock-boundary-layer interaction, are used to evaluate recent developments in a realizable low-Reynolds-number, second-moment closure of turbulence. The model considered is a compressibility-adapted variant of the recent incompressible-flow form of Craft and Launder. This includes a tensorially cubic model for the influential pressure-strain process, ϕij, which satisfies the two-component-turbulence limit at the wall, is directly applicable to low-Reynolds-number flow regions and does not rely on or use surface-topography parameters, such as wall-normal distance or direction. Improved predictions for afterbody flows are demonstrated, relative to existing low-Reynolds-number two-equation models and the most elaborate form of Reynolds-stress closure incorporating a linear approximation for the pressure-strain process.

For the purpose of the design and certification of inflight icing protection systems for transport and general aviation aircraft, the eventual re-definition/expansion of the icing environment of FAR 25/JAR 25, Appendix C is under consideration. Such a re-definition will be aided by gathering as much inflight icing event data as reasonably possible, from widely-different geographic locations. The results of a 12-month pilot programme of icing event data gathering are presented. Using non-instrumented turboprop aircraft flying upon mid-altitude routine air transport operations, the programme has gathered observational data from across the British Isles and central France. By observing a number of metrics, notably windscreen lower-corner ice impingement limits, against an opposing corner vortex-flow, supported by wing leading edge impingement limits, the observed icing events have been classified as ‘small’, ‘medium’ or ‘large’ droplet. Using the guidance of droplet trajectory modelling, MVD values for the three droplet size bins have been conjectured to be 15, 40 and 80mm. Hence, the ‘large’ droplet category would be in exceedance of FAR/JAR 25, Appendix C.

Data sets of 117 winter-season and 55 summer-season icing events have been statistically analysed. As defined above, the data sets include 11 winter and five summer large droplet icing encounters. Icing events included ‘sandpaper’ icing from short-duration ‘large’ droplets, and a singular ridge formation icing event in ‘large’ droplet. The frequency of ‘large’ droplet icing events amounted to 1 in 20 flight hours in winter and 1 in 35 flight hours in summer. These figures reflect ‘large’ droplet icing encounter probabilities perhaps substantially greater than previously considered. The ‘large’ droplet events were quite localised, mean scale-size being about 6nm.

Extensions to Maskell’s original correction method, developed over several years, are consolidated and designated ‘Maskell III’. The procedures were applied to dedicated tests on a family of flat-plate wing models in a small, low-speed wind tunnel at NRC. Test conditions included angles of attack from -10° to 110° and models of up to 16% of tunnel area. Off-centre tests were included with model-to-wall distances down to 0.72 chords. Corrected lift and drag data correlated well between models of different sizes and at different offsets from the tunnel centreline. Comparisons are made with corrections using the pressure-signature and two-variable methods, emphasising post-stall conditions. These showed that the ‘Maskell III’ procedures, which require minimal input, correlated as well as the other methods for most model sizes and positions in the tunnel.

An unstructured flow solver with adaptive mesh refinement and multigrid acceleration is developed to efficiently compute two-dimensional inviscid and viscous steady flows about complex configurations. High resolution is achieved by using the upwind scheme coupled with adaptive mesh refinement. An aspect-ratio adaptive multigrid method is developed and applied to effectively accelerate the solution convergence of the explicit time-marching in the near wall regions with high aspect mesh ratios. Numerical examples are presented for configurations and conditions ranging from transonic to low speed flows to demonstrate accuracy, speed, and robustness of the method.

This paper presents a μt-limited explicit algebraic stress model for shock-wave-turbulent boundary-layer interactions based on an investigation of various two-equation turbulence models. The results of the κ-ω model, the shear stress transport (SST) model, and a quadratic explicit algebraic stress model are presented and analysed for the Delery channel bump, and the Bachalo and Johnson axisym-metric bump test cases. While the κ-ω model failed to give good predictions, the SST model proved reasonably successful in predicting strong interaction problems. The non-linear quadratic explicit algebraic stress model gave improved results (when compared with the original κ-ω model) for the channel bump test cases, but was not as good as the SST model. Inspired by this investigation, a model formulation is proposed in which Bradshaw's assumption for eddy-viscosity limiting (as used in the SST model) is applied to a quadratic explicit algebraic stress model. The QSST model further improves the SST model for strong shock-boundary-layer interactions.

This paper presents the results of a study into wake vortex-related safety issues associated with simultaneous rotary and fixed-wing aircraft operations at busy airports. An analysis conducted using both simple analytic models and high fidelity FLIGHTLAB simulations has considered a helicopter located in the hover above the landing point and during approach and landing. Tip vortices from a Boeing 747 are shed and are assumed to be travelling in a horizontal plane by the time they reach the helicopter. A worst case scenario is simulated, with no vortex ageing or attenuation due to ground effect. Both simple and high fidelity simulations indicate that the rates of climb and descent induced by the vortex tails can be significant. When the helicopter flies through the vortex core, moderately large transient excursions in attitude occur within a few seconds. When the helicopter does not pass through or close to the core, while the flight path perturbations are still significant, the attitude response is shown to be significantly reduced. The predicted cyclic control power required to counteract the vortex-induced hub moments is about 40% of full control, compared with more than 100% for an ‘equivalent’ fixed-wing aircraft. The control power required in the vertical, collective, axis to overcome the downdraught and updraught in the vortex tails can be as high as 15–20%. While the authors have not tried to make judgements as to whether pilots would find the transients manageable, an approach to quantifying the extent of the hazard has been suggested using the failure transients criteria from the handling qualities performance standard ADS–33. Combined with analysis of the likelihood of such occurrences at particular airports, such response criteria offer a rational approach to developing safety cases for simultaneous operations.

The flow in many wind tunnel experiments is affected by the presence of test section walls. The resulting interference can be minimised by correcting the measured model pressures, or by influencing the model flow directly with the use of ventilated or adaptive test section walls. The objective behind the latter technique is to guide the flow in the test section to achieve low interference (i.e. free flow) condition at the model. The most successful technique of flexible, adaptive walls is still restricted to small research wind tunnels due to its mechanical complexity. However, a very promising alternative is the use of adaptive slots in the test section walls. This concept combines the method of passive slotted walls, as they are already implemented in many large wind tunnels, and flexible walls. Additionally, this technique presents the opportunity of full 3D adaptations because the slots can be situated in all four test section walls.

This paper presents preliminary experimental results and the latest numerical calculations on the effectiveness of adaptive slots. The experiments were conducted under high subsonic flow conditions in the new slotted test section of the transonic wind tunnel at TU Berlin’s Aeronautical Institute (ILR).

The numerical results presented are focussed on the 2D slot adaptation of a 2D-model (CAST7 aerofoil) and the 3D slot adaptation of a body of revolution (3D-ETB). In addition, basic studies were made of the flows associated with a single slot on one wall and a bump on the other.

The numerical and the first experimental investigations have shown the potential of adaptive slots to reduce wall interferences effectively. The adaptation accuracy of the investigated slot configurations deviated not more than 3% from the reference case (2D-wall adaptation).

The shock-structure and the related acoustic field of underexpanded jets undergoes significant changes as the Mach number Mj is increased. The present investigation is carried out to study the effect of Mach number on an underexpanded 2:1 elliptic-slot jet. Experimental data are presented for fully expanded Mach numbers ranging from 1.3 to 2.0. It is observed that the ‘cross-over’ point at the end of the first cell at low Mach numbers gets replaced by a normal shock at a highly underexpanded condition resulting in the formation of a ‘barrel’ shock along the minor-axis side with a ‘bulb’ shock formed along the major-axis side. The above change in shock structure is accompanied by a related change in the acoustic field. The amplitude of fundamental frequency along the minor-axis side grows with Mj but falls beyond Mj = 1.75. Along the major-axis side, however, the fundamental frequency does not exist at low Mach numbers. It appears at Mj = 1.75 but then falls at Mj = 2.0. The related azimuthal directivity of overall noise levels (OASPL) shows significant changes with Mj.

The effect of multigrid acceleration implemented within an upwind-biased Euler method for hovering rotor flows is presented. The requirement to capture the vortical wake development over several turns means a long numerical integration time is required for hovering rotors, and the solution (wake) away from the blade is significant. Furthermore, the flow in the region near the blade root is effectively incompressible. Hence, the solution evolution and convergence is different to a fixed wing case where convergence depends primarily on propagating errors away from the surface as quickly as possible, and multigrid acceleration is shown to be less effective for hovering rotor flows. It is found that a simple V-cycle is the most effective, smoothing in the decreasing mesh density direction only, with a relaxed trilinear prolongation operator. Results are presented for multigrid computations with 2, 3, 4, and 5 mesh levels, and a CPU reduction of approximately 80% is demonstrated for five mesh levels.

This paper reports on part of a larger study examining the applicability of fuzzy logic to the higher level supervisory control functions of unmanned air vehicles and details the work so far in testing the feasibility of a candidate self-organising scheme. The candidate algorithm consists of a regular fuzzy control system whose rule set is improved over time as a function of the deviation of the plant from some desired performance. Historically, virtually all work in this area has examined the application of the algorithm to the classic regulator problem, and focussed on achieving both better performance in that role and establishing tuning guidelines based on a more systematic approach. The application discussed in this paper is assessing the algorithm to ascertain its possible utility in the higher level supervisory capacities. The candidate control problem was one of station-keeping of one air vehicle behind another.

The past has been interesting, but what of the future? As Fig. 1 shows, extrapolation is an attractive tool, with pitfalls for the incautious - most often exemplified in aerospace by the prediction that by 2015 the entire US defence budget would fund one aeroplane, or more recently by the belief that avionics will expand into infallibility and reduce costs to invisibility. The first extrapolation is premature, and the latter two as near as makes no difference wrong.

In the case of the single affordable US defence aircraft the extrapolation probably correctly estimated the 2015 US defence budget, but erred in expressing the US Department of Defence requirement in terms of aircraft rather than damage-effects on target. The subsequent step change in avionics miniaturisation undermined the extrapolation and now looks likely to enable the DoD’s damage-effect requirements to be attained using products that lie on a different cost-effectiveness trend line.

Ion thrusters are well suited for near-Earth and deep space missions because of their exceptionally high specific impulse. Those already flying are also proving to be very reliable. The flow of neutral and charged particles within the different components of these devices is very complex and although they have been studied for several years, little of the detail of their internal operation is properly understood. Numerical simulations of these flows can potentially yield an improved understanding of the physical phenomena involved and in turn this should assist in the optimising of new designs and in achieving higher specific impulses. The external plume, which can potentially cause damage to other spacecraft components such as solar panels, has been the object of some studies but the flow within the internal chambers where many complex phenomena are suspected to occur has been seriously neglected. For this reason the present fully kinetic neutral and charged particle simulations, which take account of the ionisation processes and the applied magnetic field, have been conducted. The studies have been made on one specific electrostatic ion thruster - the T5 model produced by the Space Department at the Defence and Evaluation Research Agency (DERA, Farnborough, UK). Two regions within the thruster have been concentrated on, namely the exit area of the hollow cathode and the main chamber.

The coupling of independent structural dynamic and inviscid aerodynamic models, in the time domain, is considered. The accuracy and CPU requirements of the two common approaches, namely ‘weak’ and ‘strong’ coupling procedures, are investigated. It is found that the strong coupling scheme is more accurate than the weak coupling approach, and only for large real time-steps is the strong coupling scheme more expensive. The computational method developed is used to perform transonic aeroelastic and aeroservoelastic calculations in the time domain, and used to compute stability (flutter) boundaries of 2D wing sections. A control law is implemented within the aeroelastic solver to investigate active means of flutter suppression via control surface motion. Comparisons of open and closed loop calculations show that the control law can successfully suppress the flutter and results in a significant increase in the allowable speed index in the transonic regime. The effect of structural nonlinearity, in the form of hinge axis backlash is also investigated. The effect is found to be destabilising, but the control law is shown to still alleviate the destabilising effect.

This paper presents the development and validation of a parallel unsteady flow and aeroelasticity code for large-scale numerical models used in turbo machinery applications. The work is based on an existing unstructured Navier-Stokes solver developed over the past ten years by the Aeroelasticity Research Group at Imperial College Vibration University Technology Centre. The single-process multiple-data paradigm was adopted for the parallelisation of the solver and several validation cases were considered. The computational mesh was divided into several sub-sections using a domain decomposition technique. The performance and numerical accuracy of the parallel solver was validated across several computer platforms for various problem sizes. In cases where the solution could be obtained on a single CPU, the serial and parallel versions of the code were found to produce identical results. Studies on up to 32 CPUs showed varying levels of parallelisation efficiency, an almost linear speed-up being obtained in some cases. Finally, an industrial configuration, a 17 blade row turbine with a 47 million point mesh, was discussed to illustrate the potential of the proposed large-scale modelling methodology.