摘要:

This paper is mainly a re-presentation of the magneto-ionic theory for oblique transmission through the ionosphere with special reference to numerical applications, but it is directed particularly to the problem of using ionospheric knowledge to plan short-wave long-distance communication. The quartic equation of the theory is solved graphically by the intersection of a straight line with a certain curve, and a chart is described for the construction of the latter for various values of the electronic density. By the method of stationary phase, simple expressions are obtained for the differential coefficients of the ray path, including the group time, lateral deviation and specific attenuation. Numerical integration along the ray path for a given density distribution is discussed with special reference to the relation of oblique to vertical transmission. Some preliminary results are given, but no detailed numerical survey is included, as this would more fittingly form the subject of a separate paper.

DOI:10.1049/pi-4.1951.0001

出版商:IEE    

年代:1951

数据来源: IET

摘要:

The development of a method of measuring the electric strength of liquid dielectrics with direct current, alternating current and impulse voltages, is described.Experiments onnormalhexane have shown that, when the electrode system is thoroughly cleaned and the liquid carefully filtered, the electric strength is independent of the form and duration of the applied voltage and the dielectric thickness. The electric strength cannot, however, be considered an intrinsic property of the liquid, since with direct current it is dependent on both the material and form of the electrodes.Investigations on a series of liquidnormalparaffin hydrocarbons (pentane to nonane) have shown that the electric strength increases as the chain length of the molecule increases. The cyclic hydrocarbons, cyclohexane and benzene, have approximately the same electric strength asn-hexane.The experimental results are discussed in relation to existing theories of dielectric breakdown in liquids and solids.

DOI:10.1049/pi-4.1951.0002

出版商:IEE    

年代:1951

数据来源: IET

摘要:

The compensation theorem is often used to obtain the approximate effect of small changes in the elements of an electrical network. The paper describes an extension to a continuous system which includes, instead of a set of circuit elements, a boundary-surface exhibiting the property of surface impedance. The change in mutual impedance between two pairs of terminals situated outside this surface, due to a small change of surface impedance, is expressed as a surface integral. This result, which is deduced from an exact expression allied to Huygens's principle, simplifies the evaluation of the effect of imperfectly conducting ground upon the self-impedance of an aerial placed above it, and enables account to be taken of the presence of an earth system. The effect of the earth system upon radiation from an aerial carrying a given current may also be obtained. Two propagation problems are considered, reflection at a land/sea boundary, and reflection at a cliff. An intermediate result is used in the case of cliff reflection, to which the concept of surface impedance is not applicable.

DOI:10.1049/pi-4.1951.0003

出版商:IEE    

年代:1951

数据来源: IET

摘要:

A comprehensive mathematical theory of servo systems in which the difference between input and output quantities is examined discontinuously at the ends of equal intervals of time is developed, and general formulae are obtained from which the behaviour of any particular system may be deduced. These results are applied to obtain the fundamental equations of the two particular types of system which have been developed practically. Two examples of typical elementary systems are analysed. Recommendations for future work in this field are given.

DOI:10.1049/pi-4.1951.0004

出版商:IEE    

年代:1951

数据来源: IET

摘要:

It is shown that for waveforms of simple shapes certain features may be defined:content, epoch, spread, skewness, squatness, etc., which are affected in a very simple way by passage through a linear system (which is regular*at low frequencies). In fact, by proper definition, they are affected only additively by constants which are independent of the applied waveform, and characteristic only of the system. These invariants for the pulses and for the system, correspond to the cumulants used to describe statistical distributions.With due care, these results can be usefully generalized for a.c. wave-packets.

DOI:10.1049/pi-4.1951.0005

出版商:IEE    

年代:1951

数据来源: IET

摘要:

A comparison is made of the output spectra comprising signal and low-frequency noise when a sinusoidal signal plus noise is applied to several types of detector. As typical of the incoherent type of detector the common square-law detector and the so-called linear detector have been studied. The coherent type is represented by the homodyne and commutator detectors. The comparison is made for input noise from several common types of filter.The signal/noise ratio at the output is calculated in terms of that at the input, and curves showing this relationship for the various types of detector are given, both for the case when no further bandwidth limitation is introduced other than the filtering out of the high-frequency components, and also for that when the detector is followed by a narrow-band low-pass filter.It is shown that, provided certain precautions are taken, a very considerable gain in signal/noise ratio may be obtained by using the coherent detector when the input signal/noise ratio is somewhat less than unity. The actual amount of this gain for various circumstances may be read directly from the curves given in the text. This condition has not been of much practical importance until recently, but has now found many applications involving the detection of signals much below noise level by means of post-detector integration.The values of output signal/noise ratio for the various detectors are given for a range of values of the input signal/noise ratio in a series of figures, and the behaviour for large and small values of the latter is compared in Tables 1 and 2.

DOI:10.1049/pi-4.1951.0006

出版商:IEE    

年代:1951

数据来源: IET

摘要:

The magnetic vector potential of current-carrying conductors in a rectangular slot is the same as that due to an infinite number of images of these conductors from all the four walls of the slot. A solution of this problem can thus be obtained in the form of a Fourier series, periodic in both thex- andy-directions, the constants of the solution being determined by the boundary conditions, i.e. by the dimensions of the slot and of the conductors, and by the currents passing through them. Expressions for the magnetic vector potential due to any conductor arrangement can thus be developed, and these lend themselves to considerable simplification without undue loss of accuracy. Lines of magnetic flux density correspond to constant values of the vector potential, and this yields a ready method for plotting magnetic-flux maps. The results of this investigation can, with certain assumptions, be directly applied to transformer reactance problems and also to the equivalent problem of heat conduction of windings embedded in the slots of electrical machines. Formulae are developed for the reactance and forces in concentric-winding arrangements, showing the effects of slight asymmetries.

DOI:10.1049/pi-4.1951.0007

出版商:IEE    

年代:1951

数据来源: IET

摘要:

The analogue representation of the properties (in terms of complex frequency λ) of image-parameter transfer functions of the Zobel type by current-flow in an electrolytic tank provides, in the first place, a convenient means of designing such filters, and it has the important consequence that the properties of these current-flow representations are found to lead to a process for the solution of the approximation problem in the insertion-loss theory of filter design. The paper is in two parts, of which Part 1 is devoted to the representation of image-parameter transfer functions by the electrolytic tank. The electrolytic tank itself is first considered to be of the double-layer form described in an earlier paper;2for the accurate representation of functions of the filtering type, however, it is found more appropriate to work in a transformed version of the λ-plane in which the entire plane is represented in a finite rectangle—the electrolytic tank is then of particularly simple construction. In Part 2 of the paper, the use of the electrolytic tank as an aid to filter design on an insertion-loss basis is described. A transformation is developed, by means of which the approximation problem can be solved for a low-pass-filter specification of insertion loss, the poles and zeros of the insertion modulus function of the filter being accurately determined by simple measurements on the electrolytic tank. The transformation itself follows from the properties of image-parameter functions considered in Part 1. Detailed design procedures are given, including provision for the effects of inherent dissipation in the reactance elements of the filter. It is also found possible to extend the method of design to a general, unsymmetrical band-pass filter, and to apply the basic principles to certain root-finding operations involved in Darlington's method of network realization. Detailed examples and experimental results are given throughout the paper.

DOI:10.1049/pi-4.1951.0008

出版商:IEE    

年代:1951

数据来源: IET

摘要:

Butterworth's two formulae for the resistance of short coils with a finite number of turns and long coils with an infinite number of turns are merged into a single formula valid for coils of any length and having any number of turns. Numerical values of the functions appearing in the formula are given in tables. The formula is shown to give results in reasonable agreement with the experimental figures of Medhurst and Hickman.

DOI:10.1049/pi-4.1951.0009

出版商:IEE    

年代:1951

数据来源: IET

摘要:

Electrical discharges at 2800 Mc/s are produced in a nosed-in cavity type of resonator by energizing it from a pulsed magnetron. Individual pulses of various durations in the range 0.25 microsec to 2.5 microsec are employed. Electrons are provided in mid gap by the short ultra-violet radiation from an auxiliary d.c. spark actuated from the circuit used to trigger the power pulse. Descriptions and illustrations are given of the form of the resulting discharge, which in typical cases begins in mid gap and advances towards the electrodes, crossing a 1-cm gap in times of the order of 1 microsec. The discharges in different gases at atmospheric pressure show wide differences in appearance.Oscillograms of the envelope of the field in the resonator show a very rapid collapse when the discharge occurs; in certain gases, notably hydrogen, a striking consistency of behaviour is exhibited.Studies of the absorption coefficient of the active radiation from the auxiliary gap agree with the supposition that it lies in the short ultra-violet region, and it is found possible to calculate the variation of sparking probability with distance of the irradiating spark from the resonator. Confirmatory experiments with pulsed d.c. sparks support the argument.

DOI:10.1049/pi-4.1951.0010

出版商:IEE    

年代:1951

数据来源: IET