Search

Conductor screens are now in general use on high-voltage power cables. They take several forms, and are especially effective in bringing about an improvement in a.c. electric strength. Good-quality cable paper loaded with a carbon-black material is one form used for conductor-screening purposes, but it can have adverse effects when applied to oil-filled cables. These include an increase in the ionisation factor and a decrease in the impulse electric strength. The paper records the result of comprehensive investigation into many aspects of carbon-black-paper screening, and claims conclusively to establish the behaviour of carbon loading when used in paper for screening purposes.

The main properties required of paper for use in high-voltage dielectrics are concerned with its electric strength and dielectric losses. The effect of such physical properties as thickness, apparent density and air impermeability upon the impulse strength of impregnated paper are studied by both flat-sheet and model test methods, not only for types used in cables and transformers, but also for those used in capacitors. The incorporation of synthetic and other types of fibre into softwood pulp is discussed, and the electric strength is shown to be dependent upon the associated physical properties. Factors affecting dielectric losses, on the other hand, are the chemical constitution and purity of the pulp and water from which the paper is made, as well as its apparent density. The effect on the electrical properties of changes in the pulp and washing treatment is demonstrated, and the importance is shown in the case of capacitor tissues of the presence of conducting particles and other impurities. Methods of assessment of quality are described in detail.

An investigation has been carried out into the effect of ionic and other additives on the power factor of oil-impregnated paper, with the object of relating the power factor of the impregnated paper to the power factor of the impregnant. For power-factor measurements, an improved test cell was used which was designed to operate at high stresses over a wide temperature range. it was observed that there was a variable relationship between the power factor and the volume resistivity of oils depending upon the type of additive. There was no correlation between the power factor of impregnated paper and the power factor of the impregnant, but a definite relationship was found between the power factor of impregnated paper and the volume resistivity of the impregnant, this relationship being independent of the type of additive.

The paper gives the results of an experimental laboratory investigation of model bushings, with the object of predicting the behaviour of full-scale e.h.v. oil-impregnated paper bushings operating at a radial a.c. stress of 3kV r.m.s./mm. One of the aims was to determine whether discharges of up to 5pC in full-scale bushings might damage the insulation during prolonged periods in service, and so reduce the prospective life. Frequency-accelerated life tests were made on over 90 model bushings, some of which were constructed according to accepted manufacturing methods, whilst others contained defects of various kinds. It was found that, while the initial-discharge magnitude is a good indication of air trapped in the windings in a region of high electric stress, it is not a good guide to the prospective life of the bushings. The most critical factor is the ability of the oil to dissolve any trapped gas. If the trapped gas dissolves, bushings working at a stress of 3kV r.m.s./mm will have a low discharge level and a long life, provided they have no constructional defects. If not, the discharge magnitude may fall at first, but will then rise steadily until breakdown occurs after 12–20 years. A further aim of the investigation was to determine the extent to which either permanent damage or temporary impairment of the dielectric properties could be caused by overvoltages in the range 7–16kV r.m.s./mm, applied for periods of 1–50min. The effects of impulse stresses up to 19kV peak/mm were also studied. It was found that a sensitivity of about 0.1pC would be required to detect the discharge resulting from the gas generated during a normal overvoltage test on a well made 400kV bushing, but that much larger quantities of gas produced in this way would dissolve in the oil within a few days, and the bushing would then return to a discharge-free condition, without apparent physical damage. Where a quantity of air was trapped in the winding during manufacture, it was shown that a discharge level of 30pC at working voltage in a 400kV bushing having a moisture content not greater than 1% would not involve a risk of degradation by a single overvoltage test having a duration of 1 min at a stress of 14kV r.m.s./mm.

An independent search through the Heaviside Collection of the Institution of Electrical Engineers for the material discussed in Heaviside's ‘posthumous papers’ is reported.

The online computer control system at Wolvercote Paper Mill, Oxford, was the first of its kind in the UK. An account is given of the events leading to the installation in 1965. The papermaking process is seen as a problem of control and the application of the digital computer to the process is described. The direct, practical approach and the low cost of both hardware and software are considered to be important factors. The benefits from this installation have been considerable. The computer has proved to be a completely reliable tool which can extend the operator's skill. There have, however, been problems in taking measurements accurated and reliable enough to be used in an automatic control system.

Mechanical stresses in a paper-lapped insulation which arise from the bending of a cable around a drum immediately after the lapping process are considered theoretically, and an estimate is obtained for the stress arising from the contact of the insulation with the drum. A known relation between interfacial pressure and lapping tension is rewritten to give the tension explicitly in terms of a prescribed interfacial pressure distribution. Finally, a certain form of internal buckling which could be relevant to the problem is briefly discussed.

After brief discussion of special vacuum cells developed for this investigation, results are presented showing the effect of adding between 0.1 and 10% moisture content on (a) the dielectric loss tangent and permittivity at frequencies between 50c/s and 100kc/s (b) the dielectric dispersion measured at intervals of 0.1 and 3s and 3 and 300ms (c) the 1 min insulation resistance (d) the 50c/s capacitance increment. In all cases the results were recorded at 10degC intervals between 20°C and 100°C. The effect of the electrode pressure on the electrical properties of paper insulation is discussed and the practical significance of the results and mechanisms giving rise to them are considered.

The equipment described in the paper was developed to record automatically the speed/time curve of an experimental homopolar machine undergoing retardation. These retardation tests were being made to determine tooth-ripple losses, and the highest rates of retardation required were beyond the range within which normal visual observation of a stroboscopic disc is practicable. An electronic binary counter and a punch unit provide a speed/time record on punched paper tape, which may be directly processed in a computer programmed to calculate the tooth-ripple losses. The input to the counter can be made from one or other of two independent detector systems. The first operates on the principle of the stroboscope, interpreted electronically, and is thus essentially a rotational speed-interval detector. As this system presents novel features, it is dealt with in more detail. The alternative system, based on counting revolutions in known time intervals, is utilised in existing commercial instruments. The maximum rate of retardation which this equipment can handle is about 1000 rev/min per second.

In the paper a review is given of the properties of oil-impregnated lapped-plastic-tape dielectrics for use as insulation in extra-high-voltage underground cables. The following subjects are discussed: choice of combinable materials, impregnation technique, results of breakdown measurements as a function of various different parameters, dielectric-loss measurements and thermal conductivity, it appears that a cable insulation with a breakdown strength 80% higher than that of oil-impregnated-paper insulation is technically possible. The new insulation, consisting of film tapes of polyphenylene oxide impregnated with paraffinic oil, has a dissipation factor more than five times smaller than that of conventional oil-impregnated paper. in cables with such an insulation, operating stresses of 15–37kv/mm, depending on tape thickness and pressure, may be applied.

This short paper describes an investigation into the behaviour of a divided-winding-rotor (d.w.r.) synchronous generator, equipped with regulators for controlling the terminal voltage, torque angle and speed, when subjected to unbalanced short circuits at its terminals. The performance of the machine has been compared with that of a conventionally wound rotor arrangement. It has been observed that the d.w.r. arrangement exhibits exceptionally superior transient-performance characteristics compared with the c.w.r. machine, even under unbalanced-fault conditions, which the machine has to encounter more often in practice than balanced faults.

The oxidation stability of inhibited insulating oils is usually assessed in the laboratory under one set of highly accelerated test conditions. The service life of these oils under normal operating conditions is now being shown to be very much less than that predicted by these tests, and in some cases it is no better than that of the base oil. As inhibited oils have always been considered capable of giving a satisfactory performance under much more severe conditions, it is necessary to know why their service performance under normal conditions is poor. This will enable them to be assessed correctly in future, and may suggest how they can be improved if more severe operating conditions arise. Commercially available, laboratory-blended and experimental oils were examined by the assessment procedure described by Wilson.1 Results are compared with service performance and with performance in experimental inductors. Tests were also carried out at higher temperatures and with nitrogen sealing to show the comparable effectiveness of inhibited oils and nitrogen sealing. It is shown that the susceptibility of oils to the effects of copper and oxygen is radically changed by the addition of oxidation inhibitors. The temperature susceptibility is, in general, less affected. Inhibitors are shown to be most effective under high-oxygen conditions and when the copper/oil ratio is high (as found in highly accelerated laboratory tests). The combination of an oxidation inhibitor and a metal passivator, or deactivator, is more successful than an oxidation inhibitor alone, but, even so, these combinations of additives have only limited economic advantage, as they will enable the oil to withstand only marginally more severe conditions. Lack of correlation between oil acidity and resistivity is shown to exist after aging in the case of inhibited oils, passivated oils and oils operating under a blanket of nitrogen.

Although control systems for steering coal cutters through undulating coal seams have been in use since the early 1960s, it is only since Webb's paper (Proceedings of the 3rd UKAC control convention, April 1968) in 1968 that the 3-dimensional dynamic nature of the process has been appreciated. The process model, previously expressed in terms of two independent variables, I and n, is here converted into a model in the single independent variable v. This, for the first time, permits analytical stability studies and proves the fundamental instability of analogue control, which arises from the profound effects of the coal-sensor delay. The technical case for stored-data digital control is therefore one of the main objectives of the paper.

Synchronous-generator excitation supply systems that are based on a.c. exciters operating in conjunction with rectifiers, as is now widely the practice, give rise to particular modes of generator operation that do not occur when d.c. exciters are used. In the case of a d.c. exciter, the field current can reverse, if it is required so to change by stator operating conditions, whereas an a.c. exciter and a rectifier system provide an essentially unidirectional form of supply. A preclusion of field-current reversals by the rectifiers introduces a constraint into the interaction between the rotor and stator circuits of the generator for which the excitation system provides a supply. For a wide range of generator operating conditions, the constraint is inactive, and so no adverse effects arise, but credible circumstances can combine to cause the constraint to become active. When this is the case, the transition of the constraint from inactive to active is accompanied by a pulse of induced voltage in the field system. The paper reports a study of field-system overvoltages that might be generated in this way for the particular conditions encountered in synchronising. A mathematical model suitable for use in computer studies in this area of applied analysis in power systems is summarised, and the validity of the model is checked by comparing computer study results with those available from fullscale tests on a 120 MW turbogenerator unit. Numerous synchronising studies have been based on this model, and representative results from them are collected together in the paper, together with the synchronising conditions that can give rise to overvoltages. When synchronising is achieved within the close limits typical of British practice, the hazard of induced field voltages of substantial magnitude should be avoided altogether, but the continuously recurring nature of synchronising generator units with a parent system suggests a general study of the conditions. It is a general study of this kind to which the paper is devoted.

Progessively reducing the measuring time of high-speed primary protection emphasises its dynamic response, both in the initial development and design of protection, and in its subsequent application. Of growing importance is the interaction between the distinguishable transient responses of the primary system protected, the transducers from which relaying signals are derived, and the measuring processes of the protection itself. To study and to evaluate this interaction requires considerable detail in modelling each primary-system and protectionsystem element that contributes to it. The paper reports an extensive computer simulation that has been developed to meet these requirements. Marshalled in a large overlay scheme, the simulation includes detailed models for the primary system protected, (including synchronous-generator units, transformers and transmission circuits), capacitor voltage transformers, current transformers, the transducer-protection interface, and the measuring andtiming sequences of the protection. The paper provides an account of the development of the simulation and of the analytical techniques on which it is based.

The paper describes several circuits, which include subsystems switched with alternating polarity. Practical applications are shown. The analysis of such circuits is presented in a companion paper (Proc. IEE, 1974, 121, (2), pp.(88–90). The circuits considered in the paper are for a high-power oscillator, for power-factor control by means of a switched coil, and for load-power regulation by means of a switched capacitor.

The paper presents a new method for the realisation of optimal linear-quadratic regulators, for linear time-invariant systems, using the concept of proportional-integral derivative (p.i.d.) feedback. It is shown that an optimum complete-state-feedback regulator-system response can be alternatively obtained with incomplete feedback, by introducing the derivative-control action. Considering also the integral feedback, needed to cancel the unknown input disturbances, the method can be viewed as a design method for a class of optimal multivariable p.i.d. controllers with a prescribed closed-loop system response. The paper presents an analytical derivation and discusses the properties of such controllers. An illustrative numerical example is also included.

The paper is concerned with the stability of closed-loop control systems incorporating controlled rectifiers. A suitable small-signal model for a convertor is developed which takes into account the discrete behaviour of the controlled rectifier. This model permits the use of standard sampled-data control-system theory in the stability analysis of the system. Contrary to usual practice, the a.c. system reactance is not neglected in the modelling. It is shown that the commutation angle is an important factor in the determination of stability boundaries. Finally, again-correction factor is employed to allow for the effect of the ripple under finite-commutation-angle conditions. Discrete Laplace transformation (z transformation) is used throughout the analysis together with root-locus techniques. Stability boundaries derived from the new model and other existing models are compared, and it is shown that experimental results are in good agreement with the theory developed in this paper.

A previous paper outlined the principles of a new method for dynamically varying the shape of polarised quadrilateral relay characteristics for the protection of e.h.v. transmission lines. The present paper extends these methods to the protection of double-circuit lines. Consideration is given to the compensation of earth-fault relays, and it is shown that compensation schemes which ignore the effect of mutual coupling between circuits are desirable. Digital-computer application studies of a typical double-circuit 400kV transmission line of earthed construction indicate that, from the point of view of accuracy and fault coverage, there is little advantage in the use of variable devices in zone 1. However, they enable a greater proportion of the protected feeder to be covered with minimum fault-detection time. Emphasis is placed on the response of relays connected to healthy conductors, and the possibility of avoiding unwanted tripping emerges. A high degree of immunity to maloperation during heavy-circuit loading and power swings is achieved.

Most of the modern distance relays are designed to avoid overreaching due to the transient d.c. component of the fault current, whereas a more likely source of transients in e.h.v. systems is the oscillatory discharge of the system charging current into the fault. Until now attempts have not been made to reproduce these transients in the laboratory. This paper describes an analogue and an accurate digital simulation of these harmonic transients. The dynamic behaviour of a typical polarised mho-type relay is analysed, and results are presented. The paper also advocates the use of active filters for filtering the harmonics associated with e.h.v. system, and hence, to improve the speed of response and accuracy of the protective relays.

Nonhomogeneous multiconductor transmission systems occur in a number of instances in practice. Examples are crossbonded cable systems and regularly transported overhead-line systems. Recursive methods of analysis fail because of the amount of computation involved. In the paper the theoretical equations for the multiconductor chain matrix are developed. It is proved that the eigenvalues of the chain matrix occur in reciprocal pairs and that the eigenvectors of the reciprocals are simply related to those of the eigenvalues. This reduces computation and avoids numerical instability in evaluating widely separated eigenvalues. These properties have not previously been published and may have applications in a number of fields as well as transmission lines. It is then noted that, provided the system response is only needed at a comparatively small number of points, computation is much more efficient than would be the case for recursive methods. In conclusion it is noted that this paper is the last in a series in which the equivalence between the theorems of simple and multiconductor systems is established by means of matrix functions.

The paper describes and examines a technique which permits the power-flow problem in a power system to be analysed probabilistically instead of using normal deterministic methods. All the nodal loads and generation are defined as random variables and the power flow in each line is computed in terms of a probability density function. The expected values and standard deviation of each power flow are also calculated, and, in addition, the overall balance of power in the system is determined in terms of a density function. The purpose of this analysis is to account for the errors and statistical variations known to exist in the operation and planning of systems within one solution. This enables the power-flow problem to be treated objectively and allows quantitative assessment of reliability and security. The paper compares the results obtained probabilistically with those that would be obtained deterministically, and shows the much wider range of information gained in this type of analysis.

Earlier papers established the theory of the electric differential and certain simple practical tests for finding the performance parameters. The present paper is a logical development, and describes the manner in which a multiunit system of 3 N machines, providing that they can be arranged in three identical groups, can be connected to a simple double induction-motor frequency-changer arrangement, so that the supply system is presented with a balanced load.

Dynamic-response evaluations of high-speed primary-protection systems are assuming an increasing importance as protection measuring times are progressively reduced, and these require very considerable detail of representation, not only for protection systems, but for the primary transducers from which they operate, and for the section of the primary system protected. Within this composite simulation, it is the latter area, that of the primary system for which the protection system under investigation provides coverage, to which the paper is devoted. In particular, the paper develops a phase-variable analysis scheme for synchronous generation in a form that allows it to be used directly in conjunction with electromagnetic transient analysis for the network system interconnecting generation. This provides a rational and unified scheme of analysis which combines fast-solution properties with considerable flexibility in application.

The subject is introduced by referring to the development of rural electrification since the early 1920s, and describes the British rural electrification programme 1953–63, which ensured that by the latter date 85% of farms in Great Britain had a public supply of electricity. After reviewing the state of farming today, and pointing out that the scale of operations is still increasing while the labour content is decreasing the review makes the point that the viability of farming in these circumstances is due to the intensification of livestock keeping and crop production, which has enabled the productivity of farms to be vastly increased. The characteristics of the electrical load on the farms are next described, and an estimate is made of the probable electricity consumptions and connected loads for various types of farming enterprise. Details are then given of the electrical equipment involved in the production of milk especially for milk cooling and for water heating to maintain hygiene. The various applications of electricity associated with pig, poultry and beef production are quantified, and especially in connection with pig and poultry production some indication is given of the principles involved in ventilation systems. The paper then describes the equipment required for the preparation of livestock feed and for automatic feeding of farm animals. Crops are next dealt with. First the conservation of green crops with an indication of the electrical potential for grass drying and hay drying, and a detailed description of the methods used for the latter. Types of grain drier and the storage facilities required, together with an indication of the development of these are given. The Section on crops is concluded by a description of methods of vegetable conditioning and storage, and particularly that required for potatoes and onions. After describing the need for observing safe conditions on the farm, both for electrical machinery and associated wiring as well as mobile equipment, the paper is concluded by suggesting possible future developments, based on current research work, which may become viable within the near future.

In a practical situation, the control engineer may have experience with his system and have some insight into its dynamic behaviour. This paper suggests a theoretical method for making use of this background knowledge in the feedback-control design process by manipulating the frequency-response information available in the plant transfer-function matrix H(s) into a form suitable for physical interpretation. The approach used is that of dyadic expansion of H(s). The technique represents an extension of the dyadic-approximation method used in the analysis of nuclear-reactor spatial control systems to include a description of interactions in a general case in terms of the matrix H(s)H−1(s). However, the representation given in this paper is exact. As the aim is to enable frequency-response data to be interpreted physically, the method described does not generate a complete design technique. An example indicates, however, that, if used with physical intuition, the procedure can provide guidelines to practical controller structures for highly interacting systems.

The paper is intended for the general nonspecialised reader, but sufficient detailed information has been included for those whose work is closely associated with cables. Introductory Sections cover cable standards and codes of practice in current use, together with a brief historical review. Conductors are dealt with separately, and it is described how, for economic reasons, aluminium has now almost wholly superseded copper for power-cable conductors. Similarly, rating practice is treated separately, rating parameters being stated and discussion centred on performance of cables in practical situations. For the purpose of the paper, cables have been classified as (a) pressure assisted 132 kV to 33 kV, (b) 33 kV solid type, (c) 11 kV, (d) m.v./l.v. (i.e. 415/240 V) and (e) pilot/telephone. In each classification covered, cable construction, jointing practice and service performance are fully detailed together with technical reasons for selecting particular types of cable. Installation and construction practice is described and discussed covering topics such as legislation and codes of practice, depth of burial, protection against damage, selection of routes, multiple service trenches, identification of services, cable tunnels and duct systems, submarine cables and river crossings. Earthing and bonding is discussed with particular reference to single-point and crossbonded systems. Testing procedure is described, together with methods of fault location. Cables versus overhead lines are dealt with briefly to emphasise that the overhead system is not used in towns and cities. Comparative costs and fault statistics are quoted. Life expectancy of cables is discussed in general terms, and an attempt is made to assess what life expectancy one might expect for various, types. Finally, the paper discusses possible future developments, particularly regarding plastic-insulated cables.

Capacitors may be included in the secondary circuits of current transformers, to effect filtering or tuning or to compensate for the capacitance of equipment being protected. While, in the past, considerable attention has been given to the behaviour of current transformers feeding resistive and inductive burdens, very little has been given to the performance obtained when there are capacitors in the burden. This paper is devoted to this purpose. Initially, the steady-state behaviour obtained when a sinusoidal current is flowing in the transformer primary winding is examined, with both purely capacitive and series RC burdens connected to the secondary winding. The types of output obtained when core saturation occurs, and the core-flux saturation levels needed to avoid saturation, are determined. The interesting feature revealed is that it is possible, under certain conditions, to have three possible secondary currents for a given primary current, with fixed circuit parameters. This effect is due to ferroresonance, and it is clearly important, as it could seriously affect the performance of associated protective equipment. The later part of the paper deals with the behaviour, during transient conditions on the system, of transformers with capacitive burdens. It is shown that the presence of capacitance in the burden of a transformer causes its core flux to contain a component which changes linearly with time. Such a component causes the inevitable onset of saturation in the event of a system fault which persists for a few cycles, and thus it makes it impossible to design transformers which should be suitable for protective equipment such as that used in association with busbars.

The H plane stripline circulator consists of a ferrite disc in contact with air or some other dielectric on either side of the TEM line. The purpose of the paper is to give some experimental results which have been obtained with such a device. These results include the bandwidth and operating frequency of the device as a function of the ferrite filling factor. An approximate theory of the device is also included. The method used in the paper consists of replacing the inhomogeneous line by a homogeneous one on the assumption of TEM propagation.

Voltage impact excitation of a solid iron core creates eddy currents in the core, retarding the rise of flux. An analysis of the problem involves the simultaneous solution of partial and ordinary differential equations. The boundary condition for the partial differential equation contains an unknown quantity, which is variable with respect to time in the ordinary differential equation associated with the physical system. Such situations are commonly termed field-circuit problems. The paper presents the method of eigenvalues for solving linear field-circuit problems. The eigenfunctions are mutually orthogonal, enabling an exact fit of the boundary condition by a linear combination of these eigenfunctions. The final equations, when applied to a practical voltage-impact-excitation problem, show favourable agreement with experimental results for current and flux transients. A special and interesting feature of the paper is the predetermination of contours of constant magnetising force and eddy-current-density distribution within the rectangular section during the transient period. The contours, obtained rigorously for the first time, give insight into the phenomena involved in the magnetisation of a rectangular solid iron core.

It has been shown recently that asymmetrical fractional-slot windings can be devised to draw balanced currents. The leakage reactance for the three phases of these windings may, however, no longer be equal. The paper discusses the effects of the asymmetry of the windings upon one component of leakage, namely the slot leakage. It is shown that, by careful selection of the coil pitch for the winding, equal slot leakages are possible for the three phases. The paper also suggests how the slot leakage reactance may be determined for other coil pitches.

In ferromagnetic circuits, a problem of interest is the predetermination of the time required for complete magnetisation of the material under impact excitation. The time of magnetisation gives an indication of the rapidity of response. Rigorous solutions of the problem are available in the literature for constant permeability and sections of simple geometry. For a nonlinear case of variable permeability, it is well known that an analytical description of the behaviour is extremely difficult. The problem of nonlinear magnetisation has applications in electrical machines, pulse transformers, magnetic memories and magnetic shift registers.The paper presents nonlinear analysis of the problem for a step m.m.f. excitation, using Maxwell's equations. For easy handling by practical design engineers, assumptions are made to obtain the final equations in a simple form. The results based on these equations are then compared with those of the painstaking graphical solution of the same problem. The favourable agreement justifies the assumptions made in obtaining the solution of the complex nonlinear problem. It is also shown in the paper that, for quick approximate computations, the author's previously reported equivalent circuit can be successfully used.

The objects of the experimental programme, which is the subject of the paper, were to assess the technical feasibility and economic return that could be expected from applying an automatic self-optimising control system to coal-fired steam-generating plant. The work centred on the control of combustion, this being directly exposed to disturbances arising from variability of the fuel quality. A hill-climbing control system was used which maximised combustion efficiency by manipulating the air/fuel ratio. As efficiency could not be measured directly, an inferential method based on changes in boiler pressure or turbine output was adopted. Two control strategies, utilising, respectively, sinusoidal and step perturbations of air flow, were investigated. The paper contains estimates of the disturbance spectrum, based on measurements of both coal-quality and boiler-output fluctuations, and shows this to be equivalent to the addition of coloured noise to the system input. The methods of obtaining optimum discrimination against this noise for each of the two strategies are discussed. Experimental data obtained by using the control system to track movements of the peak of the combustion hill are given, a range of 2% excess oxygen being observed. It is concluded that, although the factors governing this movement have not been fully identified, the fineness of grinding is of great importance in determining the optimum combustion conditions. It is also concluded that this form of control is feasible and could lead to an improvement in efficiency of about 0.25%.

The literature of the crossbonded a.c. cable systems is sparse and is mainly of Japanese and American origin. Recent work in the UK has indicated the need for precise numerical data on sheath conditions of these systems. The paper gives the numerical magnitudes of steady-state voltages and overvoltages to be expected on these systems, since they do not appear to have been formulated comprehensively in the literature. The sheath voltages can be found by identical methods for any cable system, whether it be transposed, crossbonded or normally laid. The system of calculation in the paper is thus part of a method for determining overall cable behaviour in terms of sheath voltages, currents and losses, and for either single or double circuits. Attention is drawn to the complexity of calculations for many modern systems with crossbonding, transposition and unequal lay lengths, from which it follows that more powerful mathematical methods, allied to fast large-store-capacity computers, become necessary. An economical system of notation, avoiding the use of multiple suffixes, is used, even though 3 × 3 × 3 × 2 × 2 variables must be permuted. Furthermore, in any lay (or section in the more complex cases) it is necessary to locate a particular sheath relative to a particular conductor, as well as locating each one geographically in the cable trench at any particular junction in a barrel of lays and sections. Results are presented for all the practical cases, together with single and double 3-phase circuits representing the theoretical limits. All results are given to the same base of induced voltages per 1000ft per 100A in the reference conductor; comparison of results is thus facilitated.

Distortion in transistor amplifiers is a very complex problem not amenable to easy solution. In this paper, an equation is found for the complete collector characteristic of a planar transistor which permits calculation of l.f. distortion when operating in each of the nonlinear regions, i.e. saturation, central and breakdown regions. Distortion due to the nonlinearity of input impedance is also calculated. Means of cancelling distortion are shown. Very good experimental verification is observed. Although the theory presented is valid only at low frequencies, the transistors used were v.h.f. types with short-circuit current gain-bandwidth products between 80 and 400 Mc/s. For reasons discussed in the paper, the results should be valid up to approximately the 1-5 Mc/s region.

The operating costs of a transmitting station can be substantially reduced if the equipment is automatically tuned and all services are directly controlled from a central point. This is due to the reduction of operating staff covering three shifts, and to the saving in ‘off the air’ time made possible by rapid change of frequency.The paper describes a range of transmitters with output powers up to 85kW which have completely automatic control of tuning, coupling and gain. The transmitters are of the linear-amplifier type, and hence are suitable for all forms of service used in h.f. communication, when associated with the appropriate signal source. In particular, the paper describes the automatic control system of the transmitter.Various aspects of performance are discussed, including the design requirements for low distortion and for the reduction of spurious and harmonic emissions.If the saving in operating costs is to be fully realized, a high standard of reliability is necessary. The types of components used are described. They include variable vacuum capacitors, variable inductors, silicon rectifiers and vapour-cooled valves.

The investigation presented in this paper refers to the development of load-flow techniques in order to improve the accuracy and convergence rates of the standard a.c. programs, when the system contains h.v.d.c. links. In the first part of the paper, an improved simulation of the d.c. link is described. The improvement includes routines for constant-power, as well as constant-current control, checks of control instability, direct interpolation of optimum tap-change positions, and the use of matrix algebra for the interface between the d.c. and a.c. load conditions. In the second part, a program is developed which uses the d.c. routines in conjunction with several standard a.c. load-flow methods, and includes several alternative iterative loops. The program is finally applied to a practical system, and the results of the different combinations are compared.

The paper describes experimental results obtained from transient tests on a laboratory universal machine, connected to operate as a 3-phase alternator. During these tests, which were carried out in order to measure the parameters, severe harmonic distortion was found in the armature-current waveform under conditions of sudden symmetrical short circuit. This appeared to be mainly second and third harmonic together with a very low-frequency subharmonic. The paper gives an analysis of the machine with computed values of transient current and shows that, with the compromise design used to achieve a measure of generalisation, such harmonics are to be expected. It also shows mathematically that the usual transient d.c. component in the armature current may become a transient, very-low-frequency oscillation, depending on the machine parameters. These effects observed on a `universal machine? were also found during tests carried out on a second generalised machine of different design. It was found that damping and field time constant have considerable effect on the transient harmonic distortion in the short-circuit armature current.

Improvements in computer store cycle and computing times have called for an increase in the rate of transfer of information beween the computer and its peripheral devices. This paper describes a system which can control the simultaneous demands of a number of transferring peripheral devices, to make up for the large differential between the information transfer rates of modern computer stores and existing peripheral devices. The system can control the transfer of information between 64 transferring peripheral devices, where the individual character transfer rate does not exceed 2.5kc/s, and a main store whose cycle time is 6μs. The system has been designed to minimise the amount of buffer storage, between each peripheral device and the main store, without limiting the variety of peripheral operations. The paper also describes how this system has been extended to control a smaller number of simultaneously transferring high-transfer-rate devices, where the combined character rate does not exceed 1.2 Mc/s. The peripheral transfer control works in conjunction with a fixed, or director, programme which enables the peripheral devices to be divided amongst several time-shared programmes.

The definition contained in the 14th edition of the IEE Wiring Regulations has been extended to include a wider range of electrical accessory, such as would be used for lighting, heating or power requirements in a domestic or similar type of wiring installation. No attempt has been made to cover the complete range of electrical accessories, such as those used for an industrial installation or the larger type of commercial or specialised wiring installation. The first part of the paper is devoted to a brief survey of electrical accessories that have been in use for wiring installations over the past 80 years. This is followed by a more detailed approach, dealing with accessories for specific purposes, having regard to the development of the use of electricity for performing everyday tasks. Apart from the basic design requirements to satisfy an appropriate British Standard, the factors primarily considered are those which are of interest to the user or installer of the accessory. These include safety in operation, adequate protection and enclosure of wiring connections, facility for making such connections, and general external appearance. The paper concludes with some suggestions for possible future design considerations, some of which may materially affect long-accepted practice in the use of electrical accessories.

Engineering or economic problems can be modelled by equivalent electrical networks and conveniently solved by modern advanced techniques. This paper describes a particular electrical model made of impedances, transformers, and current and voltage sources, used to represent the general economic problem which could be solved by the powerful technique of diakoptics shown in a companion paper. The modelling technique is applied to a gas-dispatch problem and shows, by example, the analogy of orthogonal transformation with linear programming.

In the paper, the method of dimensional analysis is applied to determine the flashover characteristics of polluted insulators. Expressions deduced for the flashover voltage in a d.c. circuit under simplified conditions are found to agree with the analytical findings of previous authors. The work is then extended to flashover characteristics in a.c. systems, where the dynamic properties of arcs are introduced. It is found that the same functional dependence of the flashover voltage on pollution and insulator characteristics prevails as for d.c. conditions, although the magnitude of the flashover voltage may be different in the two cases. Finally, an expression for the velocity of arc propagation along the polluted insulator surface is deduced with due consideration of insulator, layer and circuit parameters. Of special interest is the dependence of the ultimate arc velocity on the short-circuit current of the test circuit, as well as the expression for the minimum voltage necessary to start arc motion. Whenever possible, the analyses included in the paper are compared with previous field and/or laboratory experience.

Earlier papers have discussed the air-gap flux-density distribution in inductor alternators and have verified the theory by measurement. It is the aim of this paper to show that, with a knowledge of flux-density distribution, it is possible to predict the load performance of these machines from the open-circuit curve—an easily calculable starting point.First, the measured open-circuit voltages are compared with those calculated from previous knowledge of the flux-density wave, which in turn has earlier been compared with that calculated from air-gap geometry. By taking the correct components of armature reaction, it is shown that the voltage at any load point can be obtained by combining these with the open-circuit flux distribution, to obtain a resultant flux distribution, and hence voltage. This involves the intersection of generalised curves, expressing the vector-diagram geometry, with the special curve for the particular machine, expressing the open-circuit-voltage/field-current relation. Examples of this method are given both for the experimental machine and for a much larger commercial alternator, and are shown to predict the load curves under widely varying conditions.

The paper considers the conditions necessary for a transistor to realize both good steady-state performance and high switching speed when operating as a low-level switch. It is found that, when functioning in this manner, the device is not adequately represented by the familiar simple model in terms of two independent charge stores, or by the equivalent definition of switching behaviour by charge control parameters. A more complete device model is proposed in which the distributed nature of the base charge is taken into account to the extent of representing the property that charge stored in relatively remote regions of the base may have quite a different time dependence from that of the charge immediately between the emitter and collector and under their direct control. Methods of measuring the relevant parameters of this modified device representation, which involves four independent charge stores, are given.When switching a low-level transistor switch to the ‘off’ state a tendency to switch partially ‘on’ again is observed. This effect, believed to be due to delayed in-flow of remote base charge, is explained by the proposed device model but not by the simple two-charge-store representation. It is shown that a base switching network may be designed by means of which such remote charge flow is closely compensated by an appropriate base drive current; high switch-off speed may then be achieved. An analytical design of the switching network is presented though its experimental derivation is more convenient in practice. By suitable design, the switching speed of the voltage-driven transistor switch may be approached without excessive transients in a circuit which, for good definition of d.c. conditions, is essentially a base-current-driven switch in the steady state.Although transistors of uniform-base type are discussed throughout the paper, the approach to the characterization of switching performance and the operation of the low-level switch at high speed is also applicable to devices of graded base type.

The paper describes work that was undertaken to try to explain the anomalies that exist in the interpretation of the power-loss measurements made on electrical sheet steel. Wattmeter and bridge methods of making power-loss measurements are considered; the bridge method, which was judged to be the most suitable means of making the measurements recorded subsequently in the paper, is described, and it is shown that power-loss measurements made by this bridge circuit and a wattmeter agree to within 0.2%. A series of measurements taken by both Epstein and Lloyd-Fisher squares are presented, and the ways by which the results of these measurements can be varied are indicated; these observations lead to the conclusion that the Lloyd-Fisher/Epstein anomaly is part of a much wider problem.The flux waveforms that occur in the various parts of a stack of laminations are shown, and the many forms of distortion and flux flow between strips are considered; they are discounted as the main sources of the anomalies. The values of the power loss due to eddy-current formation, when electrical sheet steel is subjected to an alternating flux normal to its plane, are given, and these show that the performance of the steel in this respect is similar to that of any other metal. Considerations of this result lead to the view that the Lloyd-Fisher/Epstein anomaly and the anomaly factor may be due to a lateral flow of flux, and experimental results which support this view are presented. Finally it is pointed out that the Polivanov theory, and extensions of this theory, are inadequate in their present form because they do not enable an explanation to be given of the experimental results recorded in the paper.