摘要:

The speed of operation of a parallel adder tends to be limited by the time taken to propagate a ‘carry’ from one end to the other. Three methods have been used to reduce this limitation: special circuits, improved logic, in which carries are not propagated entirely serially, and detection of the end of carry propagation instead of waiting the maximum time.The existing literature is based either on particular transistor logic units, or on units with somewhat artificial properties. The paper considers some existing designs and some new refinements, and gives circuits for constructing them with crossed-film cryotrons. Comparisons are then made of the speed and cost of the various designs.A design of cryotron adder which is often described consists of a long ladder network. The propagation of carries in this network is found to be extremely slow. This result may be useful in guiding the design of other cryotron applications, such as special stores. The experimental analogue described in the paper and used to obtain exact speeds for a complicated network may have other applications in cryotron circuit design.

DOI:10.1049/piee.1963.0138

出版商:IEE    

年代:1963

数据来源: IET

摘要:

Measurements of the attenuations produced by air-filled rectangular waveguides of various materials have been made at frequencies of 35, 70 and 140Gc/s. Some effects of corrosion, annealing waveguides in a hydrogen atmosphere, acid etching of internal surfaces and electroplating have been determined. Internal surface finishes of many waveguide samples have been observed and d.c. resistivities of the materials have been measured.Measured losses are all well in excess of the theoretical values as predicted by formulae for perfectly smooth walls (e.g. up to 2.5 times in the case of copper at frequencies of 70 and 140 Gc/s). The discrepancies between calculated and experimental values increase with frequency and cannot be wholly accounted for at millimetre wavelengths by surface roughness. Attenuations can be reduced considerably by annealing or etching waveguide surfaces.Losses in waveguides increase after exposure to the atmosphere, and the corrosion characteristics of tellurium-copper, which has been recommended13for millimetre wavelengths, appear to be no better than those of pure copper or standard silver.Electroplating of certain waveguide components for millimetre wavelengths may be worth while.

DOI:10.1049/piee.1963.0139

出版商:IEE    

年代:1963

数据来源: IET

摘要:

The purpose of the paper is to deduce the overall statistical distribution of signal/noise ratio over a 2-hop line-of-sight radio link when the fading statistics for the individual hops are assumed to be known. A theoretical model is proposed that reproduces the essential features of the situation. The results deduced from the model are intended to have particular reference to double over-water hops. The model is also adapted to allow ideal diversity reception to be assumed on both sections of the link.The general results are applied to the Rayleigh fading distribution, and in this case the overall distributions are obtained in closed form together with much simpler asymptotic forms which are adequate for most system planning applications. It is seen that, if the asymptotic expression is used instead of the full expression to determine the median signal/noise ratio for the 2-hop link (without diversity), a maximum error of 1. 5 dB can occur.A parameterkappears in the distributions that can be adjusted to allow for inequalities in path loss, feeder losses, aerial gains and receiver temperatures.

DOI:10.1049/piee.1963.0140

出版商:IEE    

年代:1963

数据来源: IET

摘要:

When a transmitting apertureA0and a receiving apertureB0are coupled in the Fraunhofer region, the ratio of received to transmitted power is given by the Friis formula. It is well known that the Fraunhofer gainsGa0andGb0ofA0andB0are constant and are determined independently.In the paper, the product of the gainsGaGbin the Fresnel region is determined as an integral over both apertures, assuming the gainsGaandGbforA0andB0. It is pointed out thatGaGbcannot generally be separated into the individual factors, but a formula similar to that of Friis still holds in the Fresnel region. The author proposes to define the ‘gain-product reduction factor’,γaγb, as the ratioGaGbGa0Gb0. The individual gainGaorGbmay be determined only whenA0andB0are identical and the illuminations of both apertures are the same. Then the ‘gain reduction factor’,γ, is defined as the ratioGa/Ga0.Assuming uniform amplitude and phase of illumination overA0andB0,γaγbandγare determined for circular and rectangular apertures. For circular apertures, the same factors are determined for a set of illuminations which approximate to commonly used tapers, and the effect of defocusing the primary feed in lenses and dishes is also discussed.

DOI:10.1049/piee.1963.0141

出版商:IEE    

年代:1963

数据来源: IET

摘要:

Distributed networks are described briefly, and their limitations discussed. Suitable locations of poles and zeros of lumped networks are proposed, giving a frequency response approximating to an equal-ripple characteristic. The design of several networks covering three decades of frequency, and with deviations from the ideal of approximately 0.25 dB, is described in detail.

DOI:10.1049/piee.1963.0143

出版商:IEE    

年代:1963

数据来源: IET

DOI:10.1049/piee.1963.0144

出版商:IEE    

年代:1963

数据来源: IET

摘要:

Early use of sodium light for the crossbars of the Calvert runway approach lighting system and experiments at the Royal Air Force station at Gutersloh during the Berlin Airlift gave rise to a strong preference among pilots for this type of light. However, controlled dimming is not practicable with the 140W lamp and its use was limited. When the linear-type lamp became available in 1959, the Air Ministry raised the question of its suitability, and as a result a special form of the lamp was developed in which a lamp oven was used instead of the vacuum jacket to maintain the arc-tube temperature.After experiments with other methods, a system of thyratron brightness control was chosen, and equipment was made up for trial at Stansted Airport, Essex, in 1961. The projectors each gave 20kcd intensity over ±4° vertically on their direction of aim, with a wide fan horizontally. Six stages of dimming were provided, each reducing the intensity to one-third that of the previous stage. It was possible to flash any or all of the lamps at any desired rate from 30 to 240 per minute.The trials confirmed the eminent suitability of the linear sodium lamp for approach lighting on all-weather runways, providing as it does a non-dazzling, distinctive, high- and low-intensity all-purpose light, with brilliancy control adjustable within an acceptable time response to suit the prevailing meteorological visibility. Two aerodromes are now to be equipped in Scotland.The technical and economic advantages of the development are of great interest to airport authorities throughout the world, not only in latitudes subject to fog and mist but also wherever dust storms and other disturbances impair a pilot's vision.

DOI:10.1049/piee.1963.0145

出版商:IEE    

年代:1963

数据来源: IET

摘要:

The electric field near bundle conductors can be determined by replacing each conductor of the bundle by two filament conductors suitably displaced from each other and symmetrically placed with respect to the centre of the conductor they represent. The equipotential surface taking the place of the conductor will approach more closely a true circle than the corresponding surface arising from replacing each conductor by a single filament as described in a previous paper by the author. When the ratior/Rof the radiusrof each conductor to the radiusRof the bundle circle is small, say equal to 0.1 for a bundle of two conductors, the single-filament representation will be sufficiently accurate in calculating the electric field. For higher ratios ofr/R, however, the twin-filament representation will yield better results. The position of the twin filaments relative to the original conductor is so arranged that, at the point of maximum surface gradient, the equipotential surface representing the conductor exactly coincides with the conductor.The paper deals first with the evaluation of the electric field outside the bundle conductors as well as the field at various points on the conductor surface. It is followed by applying the field outside the bundle conductors to the case of a triode and also to a 3-core belted-type cable. In the triode, the amplification factor as calculated by the method given in the paper is checked against that by Ollendorff for different values of screening fractionS. The field inside a 3-core cable leads to the determination of the thermal resistance of the cable, and the result compares favourably with that given by Simons.

DOI:10.1049/piee.1963.0146

出版商:IEE    

年代:1963

数据来源: IET

摘要:

The paper describes the results of field tests, carried out on a section of the Central Electricity Generating Board's system during the summer of 1961, to measure the behaviour of interconnected generating sets, including a 200 MW unit, under the influence of transient faults and to compare with predicted performance by different methods of analysis. The results stressed that a rapid clearance of close-up 3-phase faults is essential with modern generators to prevent transient instability, but the dynamic severity of the fault attenuates rapidly away from the point of its application. They also confirmed previous findings, namely that the conventional method of analysis by network analyser or digital computer predicts the amplitude of generator-rotor first swing with good accuracy, and thereafter gives a pessimistic indication of system-recovery. Micro-models and analogue computers give a greatly improved prediction of system post-fault recovery.

DOI:10.1049/piee.1963.0147

出版商:IEE    

年代:1963

数据来源: IET

摘要:

The paper gives a general introduction to a new technique of using silicon controlled rectifiers in ring counters for polyphase sine-wave generation based on the experimental observation that such waveforms occur across diametrically opposite output terminals of a multistage ring counter having an optimum network geometry. A specific application of this-principle in the form of a d.c. to 3-phase a.c. sinusoidal inverter is then described in detail, the design and evaluation data of a prototype 30 V d.c. to 240 V, 60 VA per phase unit operating at 400c/s being presented as an example of practical design.

DOI:10.1049/piee.1963.0148

出版商:IEE    

年代:1963

数据来源: IET