The aim of this paper is to present a method for predicting the aerodynamic characteristics of slender bodies of revolution at small incidence, under flow conditions such that the boundary layer is turbulent. Firstly a panel method based on slender body theory is developed and used to calculate the surface velocity distribution on the body at zero incidence. Secondly this velocity distribution is used in conjunction with an existing boundary layer estimation method to calculate the growth of boundary layer displacement thickness which is added to the body to produce the effective aerodynamic profile. Finally, recourse is again made to slender body theory to calculate the normal force curve slope and centre of pressure position of the effective aerodynamic profile. Comparisons made between predictions and experiment for a number of slender bodies extending from highly boattailed configurations to ogive-cylinders, and covering a large range of boundary layer growth rates, indicate that the method is useful for missile design purposes.

The response of turbulent boundary layer to sudden change in surface roughness have been studied experimentally. Mean velocity measurements have been made in the boundary layer on a flat plate, downstream of a small step change in surface roughness under 3 different pressure gradients. The surface upstream of the step consisted of ‘k* type ‘large roughness’ wall (or ‘small roughness’ wall) and downstream of the step consisted of smooth surface (or ‘small roughness’ wall). Velocity profiles after the step change have been analysed on the basis of the two layer model. The inner region responds very quickly to the new boundary condition while the outer region takes more time to attain equilibrium or a state of local self-preservation. The skin-friction coefficient initially increased after the step change and gradually reached towards a constant value except for a particular roughness combination under adverse pressure gradient wherein the change in the roughness function is gradual over the transition.

This paper describes measurements made in the regions of separated flow associated with three simple sharpedged two-dimensional geometries, a rear-facing step, a front-facing step and a rectangular block. The use of the pulsed-wire anemometer made possible the measurement of the three components of mean velocity and turbulence; earlier techniques, such as the hot-wire anemometer, were not well suited to the accurate determination of these quantities either in regions of continually reversing flows such as the re-circulatory zone or in regions of very high turbulence such as the shear layers bounding these zones. Supplementary measurements of surface pressure and shear stresses are also presented and comparison is made between these shear layers and the plane mixing layer. The work forms the first part of an extended programme for the investigation of bluff body flows but its principal immediate value will probably help in providing data with which to test the validity of mathematical models of turbulence as applied to re-circulating flows.

An experimental investigation has been made to study the interaction between an incident oblique shock wave and a turbulent boundary layer on a solid surface downstream of a porous surface with air injection through the porous surface. The results are clearly of direct interest to the design of cooled turbine blades. However, the boundary-layer profiles in the absence of the shock are similar to those in a boundary layer subjected to an adverse pressure gradient so that the results at the interaction also give some indication of the effects of an isentropic compression upstream of a shock/boundary-layer interaction. The experiments show the surprising result that the shock strength to produce incipient Separation is virtually independent of the shape of the boundary-layer profile upstream of the interaction, although the scale of the interaction increases with increase in the injection rate.

The curvature of an attached shock wave and the flow variable gradients at the tip of a straight edged wedge caused by non-equilibrium effects are determined for a steady plane flow. The specific source of non-equilibrium effects considered here is the dissociation-recombination reaction in a symmetrical diatomic gas. An oxygen like idea dissociating gas is used as an example in calculating the desired quantities. The qualitative behaviour of the exact results is compared with those obtained by using approximate methods.

An inviscid-viscous interaction method has been developed to predict the blade-to-blade flow in axial compressors. The method is primarily intended for high deflection, transonic (but substantially shock-free) blades where the suction surface boundary layer may be separated near the trailing edge. The inviscid flow is calculated by a time marching method and the viscous flow by integral methods for laminar and turbulent boundary layers. A mixing calculation is then carried out to determine the blade deviation angle and pressure loss coefficient. Predictions have been compared with test results for three high deflection, transonic cascades and there is generally good agreement.

This paper reports an experimental investigation into the impingement of three jets from a convergent, conically divergent nozzle on to three cones of apex angles 120°, 90° and 60°. The exit Mach number of the nozzle was 2.2 and the jets were produced by operating with ratios of nozzle lip pressure to ambient pressure of 1, 1.2 and 2. The cones were arranged symmetrically in the jets at nozzle to apex distances of 0, 1 and 2 times the nozzle exit diameter. Surface pressures and shadowgraph pictures are presented. The most striking feature of the flows is the shock pattern produced by the interaction between the cone shock and the jet shock. This pattern can take a wide variety of forms depending on the structure of the free jet and strongly influences the form of the surface pressure distribution. For the most part, the flows can be explained on the basis of inviscid behaviour.

The effects of slanting the base of a slender axisymmetric cylinder (length/diameter ratio of 9), aligned with the flow, was studied experimentally. The body was equipped with interchangeable rear ends covering a range of slant angles between 0° (vertical) and 70°. It was found that the base slant has a very dramatic effect on body drag, particularly in a relatively narrow range of slant angles where the drag coefficient exhibits a large local maximum (over-shoot). Detailed study of the flow showed that the drag overshoot is related to the existence of two very different Separation patterns on the slanted base. One pattern is similar to that found behind axisymmetric bodies with no base slant, and its main feature is the presence of a closed Separation region adjacent to the base. The other pattern is highly three-dimensional with two streamwise vortices forming along the sides of the slanted base. This pattern sets in very abruptly at a “critical” slant angle α ∼ 47°. Drag force measurements showed that, at first, the drag coefficient slowly increases with the slant angle, but then jumps suddenly upwards to more than double its baseline value (from CD = 0.24 to CD = 0.625) at the critical angle. At angles higher than that CD decreases again, and at 70° it is about equal to the baseline value. Further effects of the slant angle are the generation of a large side force and a significant increase in near-wake flow periodicity.

A design procedure has been developed to generate aerofoil shapes for prescribed pressure distributions in an incompressible viscous attached flow. It is based on the method of singularities, originally proposed by Chen and later modified by Kennedy and Marsden, for inviscid flows. The classical approach of adding the displacement thickness of the boundary layer and wake to the aerofoil contour is used to account for viscous effects. Several numerical examples are worked out and are compared with the inviscid flow results. Significant changes in aerofoil contours due to viscous effects are observed and these are discussed.

A simulation of the steady-state and dynamic performance of a supersonic powerplant intake is described. A lumped-parameter form of the equations of motion for a compressible fluid is developed. The resulting equations, together with steady-state characteristics, are used to model the performance of a two-dimensional supersonic intake. The model is implemented on the NGTE hybrid computer. The performance, both steady-state and transient, of the resulting real-time simulation has been investigated. Step and frequency responses at two high Mach number flight conditions have been examined and compared with test data. The favourable comparisons obtained demonstrate the validity of the chosen technique.

Significant new results are presented to show to what extent a simplified theory for transonic flow may be used. Solutions are obtained by classical techniques and compared with experiment. Results are given for two-dimensional and three-dimensional, steady and unsteady flow. The effects of flow separation and improvements in Bernoulli’s equation and the surface boundary condition are also briefly discussed.

This paper analyses the unsteady lift produced on a thin aerofoil by an arbitrary three-dimensional gust. In particular, it evaluates the effects of α, the angle of incidence of the aerofoil to the mean stream. A thin aerofoil theory which includes the effects of distortion of the turbulence by the mean flow field of the aerofoil is applied to the case of an aerofoil in isotropic turbulence. Some computed results are compared with measurements made on a NACA 0015 aerofoil in grid turbulence in a wind tunnel. It is shown that for isotropic turbulence the effect of mean incidence on the lift spectrum is of order α2, and is significant at low frequencies for the length scale ratios tested.

Jets issuing from rectangular slots cut in thin plates exhibit some unusual features, including unequal spreading rates in the spanwise and transverse directions, the appearance of velocity peaks near the “ends” of the jet and changing rates of centreline velocity decay in the downstream direction. This study examines the effects of aspect ratio on such flows. The flow field has been investigated using both total head tubes and hot wire anemometry. The results are presented in the form of three-dimensional plots of total pressure and contour maps of constant velocity, streamwise turbulence intensity and the Reynolds shear stresses. The decay of mean velocity and stream-wise turbulence intensity along the centreline are presented. The rates of spanwise spreading and the location of the velocity peaks at various downstream stations are discussed. If the aspect ratio is small enough, spanwise peaks in the mean velocity distribution are suppressed.