Effusion cooling consists in forcing a gas under pressure through a porous material thereby absorbing heat from the material and forming a heat insulating layer on the exposed surface. The internal heat transfer between the porous material and coolant is considered and the heat transfer coefficient obtained from experiment. An approximate analysis for the heat insulating effect based on a heat balance method is derived in detail and applied to experiments with porous plugs set into the side of a duct carrying hot gases, and to porcras cylinders swept by hot gases. It has been found that this analysis applies reasonably accurately to the results of these experiments and of most of the published data. The manufacture of porous materials is discussed briefly and a representative list of commercially available materials is included. The application of effusion and sweat cooling to the blading and combustion chamber linings for gas turbines, rocket motors and the outside skin of flying vehicles is considered.

An X-ray back reflection technique has been used to measure the surface residual stresses resulting from the cold and hot forming of certain high strength aluminium alloys. The alloys examined were to specifications DTD 683, DTD 687 and B.S.S. L65, and the residual stresses have been related to the residual strain in bending of these alloys. The results show that, apart from the degree of straining, the residual stresses are affected by the heat treatment (cold or hot quench) and the amount (if any) of controlled stretching that the alloy has had between solution treatment and precipitation.

Flutter calculations were made on a fixed root model wing carrying a localised mass. Modes were calculated for the wing for certain conditions of restraint at the localised mass position, the choice of modes being based on observation of the behaviour of wind tunnel models under fluttering conditions. The general agreement between the calculations and experimental results was good and justified the extension of the calculations to conditions of localised mass loading that were not covered by experiment. The extended calculations show that there is an optimum spanwise position for a localised mass from the flutter point of view and, that this optimum position is related to the nodal line position in the overtone torsion mode of the bare wing. For values of localised mass parameters likely to occur in practice, flutter speeds can be estimated from calculations in a very limited number of modes, the choice of modes depending on the localised mass loading.

Work on rocket propulsion was begun by the de Havilland Engine Co. in 1946, almost exactly a year after the end of the Second World War. For about the first year activities were confined largely to a general appreciation of rocket problems and techniques and to an assessment of the experience which had been obtained elsewhere. At that time “elsewhere” meant almost exclusively Germany, from where many technical reports were then becoming available.

In 1946 the policy in this country regarding guided weapons was, to say the least, obscure and the part which the liquid propellant rocket engine was to play in that field, even more obscure. It was not surprising, therefore, that the de Havilland Engine Co., being already well established in the field of aircraft propulsion, should very soon tend to concentrate its rocket propulsion efforts to manned aircraft applications. They were encouraged in this policy by the successful German use of the Me. 163 and of rocket take-off units fitted to the He.lll, Ju.88 and other aircraft. In particular, it was foreseen that the defence requirements of the United Kingdom might necessitate at an early date the use of a fast climbing rocket-propelled fighter.

Inertial navigation is a completely new approach to the navigational problem initiated to a very large extent by the military requirement for a blind navigational system which could not be jammed by, nor give actual assistance to, any potential enemy. It relies for its operation on principles which were understood by Newton, but the practical realisation of a working system has become possible only by the intensive development in both the design and manufacturing techniques of gyroscopes and accelerometers. This development work has, to a considerable extent, been stimulated by Dr. Draper of the Massachusetts Institute of Technology and, although he may not be the inventor of inertial navigation as such, there is little doubt that but for the work carried out under his direction, inertial navigation would still be little more than a dream.

Inertial navigation depends basically on the simple fact that if we can measure the acceleration vector of a body in any convenient axis system, then by carrying out the suitable integration processes we can determine its path in space, its position and its distance from its initial position.

This note is concerned with the evaluation of the finite sum

where the m quantities Zi are arbitrary and where

is the binomial coefficient in a series of n terms. Summations of this kind appear in the evaluation of matrix products involving the reciprocal of order n of a segment of a Hilbert matrix or of a generalised Hilbert matrix. Such products are found, for example, in the representation of a lifting aerodynamical surface by a set of horseshoe vortices.

Let the m factors (r+Zi) be multiplied out, and let Pl be the sum ∑Zi, P2 the sum of the products taken in pairs, and so on; then

In connection with the note by Henry Barrow in the November 1958 JOURNAL (page 830) it may be of interest to call attention to a little known reference which describes an extensive series of experiments. A 190 mm. pipe with 10, 35 and 100 mm. inserts was studied up to a Reynolds number of 106. Besides the fully developed profiles some interesting measurements showing the development of the flow in the inlet are included.

The general neglect of these results which constitute an appreciable fraction of the available data for this type of flow is surprising.