3D Industries (United Kingdom)

Positive-immittance convertors (p.i.c.s) with a purely reactive conversion ratio are used to simulate inductors in LCnetworks. This approach, in addition to all the advantages and disadvantages of multigyrator-capacitor networks, results in simulated inductors whose magnitudes are controlled by resistor networks. Consequently the capacitors directly involved in the simulation of the inductors may all be made equal to some convenient magnitude.

A method is described for detecting metastable transitions which take place with the release of kinetic energy in the field-free region between the source and electrostatic analyser of a doublefocusing mass spectrometer, and a procedure is given for evaluating the kinetic energy release. Values are given for a number of transitions and are in agreement with those obtained by other methods where comparison is possible. The variation of peak shape with accelerating voltage is ascribed to the varying efficiency with which product ions are collected.

Sectionalisation is a principle of exchange design which requires the switching network to be divided into a number of sections, each with its own supervisory and selection apparatus. Each call has access to some or all sections and connection is completed by means of either intersectional or intrasectional links. A central control equipment determines which section is best able to deal with each call. It is shown that this enables a sectionalised exchange to use fewer crosspoints than the equivalent unsectionalised switching network.The paper describes the trunking of an electronically controlled reed-relay exchange system designed on these principles and gives details of the statistical studies conducted in the course of its development. Typical congestion formulas for different types of call are given. Traffic simulations with a digital computer confirmed that the theory is sufficiently accurate to serve as a basis for exchange design.

The paper describes the first phase of a collaborative investigation into the practicability of the synthetic testing of circuit breakers in which current and voltage are supplied from independent sources. Two basic forms, current injection and voltage injection, are discussed, and reasons given for the choice of the method of current injection suggested by the ERA. A statistical approach has been made to the investigation, and the results of a very large number of tests using air-blast circuit breakers on a single-phase circuit, with one loop of current, are given.The tests, which cover a range of powers from 33 to 1200MVA, have been made with the objects of establishing the validity of the method and determining the relative severity of the direct and synthetic tests. Analysis of the results shows that, provided particular attention is given to the voltage and current waveforms around the current-zero period, the method is valid, and there is no significant difference between the seventies of the direct and synthetic tests.Mention is made of both practical and analytical studies of the application of the basic synthetic circuit to the various circuit arrangements and test duties normally associated with direct testing, and it is concluded that the proving of circuit breakers with low-arcing-voltage characteristics is a practical proposition, and that validity will also hold for powers above the limit of 1200MVA so far investigated.

The paper describes equipment manufactured for the Central Electricity Research Laboratory for the purpose of measuring and continuously recording the corona power loss in an experimental transmission line. The measurement is made by means of a novel high-voltage bridge circuit. The balance of this bridge is recorded on a chart, and can also be fed into a data-logging equipment. The loss can be measured over a very wide range, from 200W to 50kW, and from 200 to 500kV, with an overall accuracy of about 1%.

An investigation has been carried out into the problems of measurement of the initial permeability curves and hysteresis ioops of samples of high-permeability sheet steel, particularly oriented 3% silicon iron. Experiments were carried out to determine the variation of permeability from point to point in a sheet, the uniformity of flux in samples under test, and the errors in ballistic galvanometers due to prolonged impulses. A new precision permeameter was constructed, particular features being a fluxgate magnetometer, to ensure correct compensation, and pushbutton-controlled relays, for switching the magnetisation and compensation currents. A quantitative analysis of the m.m.f. required for compensation is made, and the need for correct compensation for all measurements in such a permeameter is demonstrated.

A programme channel requires greater bandwidth than a normal telephone speech channel. It can be accommodated in a time-division multiplex (t.d.m.) transmission system by using the pulse trains of several speech channels. The paper shows that, when these pulse trains are not equally spaced, distortion components are produced in the output signal.In general, the distortion components are at the channel pulse-repetition frequency and its harmonics, together with upper and lower sideband components. To obtain a sufficiently low distortion level, the permissible departures from equal pulse spacing are only of the order of ±1° or less. Similar distortion is also introduced if the pulses are not identical in height and shape. To obtain sufficiently low distortion level, the differences in height and length of the pulses used for a programme channel may only be of the order of 1%.Although the analysis is made for t.d.m. systems using pulse-amplitude modulation (p.a.m.), its results are shown to be applicable to some systems using other forms of modulation, such as pulse-code modulation (p.c.m.). It is concluded that p.c.m. systems are unsuitable for programme transmission unless the total number of time slots per frame (including synchronising pulses) is an exact multiple of the number of speech channels required to constitute a programme channel.

A new RC active synthesis method is proposed, the active units used being negative-impedance convertors (n.i.c.s.). Any driving-point immittance, specified as a rational function with real coefficients, may be realised by an RC active ladder structure. These results can also be adapted to the realisation of arbitrary (rational) transfer functions. For specific filter functions, the ladder-network elements are derived directly in terms of the rational-function coefficients. It is shown how, in specific cases, constraints may be imposed on the spread of element values.

A programme channel requires greater bandwidth than a normal telephone speech channel. It can be accommodated in a time-division multiplex (t.d.m.) transmission system by using the pulse trains of several speech channels. The paper shows that, when these pulse trains are not equally spaced, distortion components are produced in the output signal.In general, the distortion components are at the channel pulse-repetition frequency and its harmonics, together with upper and lower sideband components. To obtain a sufficiently low distortion level, the permissible departures from equal pulse spacing are only of the order of ±1° or less. Similar distortion is also introduced if the pulses are not identical in height and shape. To obtain sufficiently low distortion level, the differences in height and length of the pulses used for a programme channel may only be of the order of 1%.Although the analysis is made for t.d.m. systems using pulse-amplitude modulation (p.a.m.), its results are shown to be applicable to some systems using other forms of modulation, such as pulse-code modulation (p.c.m.). It is concluded that p.c.m. systems are unsuitable for programme transmission unless the total number of time slots per frame (including synchronising pulses) is an exact multiple of the number of speech channels required to constitute a programme channel.