Showing papers in "Journal of the Institution of Electrical Engineers in 1937"

The properties of a dielectric containing semiconducting particles of various shapes

Abstract: It is pointed out that, although dielectric losses in certain materials are frequently attributed to the presence of particles of conducting impurity, any discussion as to the type and magnitude of loss produced by a given quantity and disposition of impurity is often extremely vague or entirely lacking. A description is given of some experiments with a suspension of water droplets in wax, the results of which did not agree with the predictions of Wagner's theory except when the suspension was kept at a temperature near to the melting point of the wax. The behaviour of a model inhomogeneous material containing spheroidal particles is investigated analytically and it is concluded that a minute amount of conducting impurity in the form of fine needles could produce a serious loss at low frequencies, although the effect of the same quantity of impurity in spherical form would be negligible.

The performance of noise meters in terms of the primary standard

Abstract: The paper discusses the performance of noise meters, such as are used for engineering measurements, in terms of the primary standard of “Equivalent Loudness” which has recently been formulated. The unit of equivalent loudness is the British Standard Phon. The technique for the basic determination of the equivalent loudness of a noise is discussed, and the Phon values for a number of specific noises of importance in engineering are given. The fundamentally important equal-loudness relations for pure tones of differing frequency are considered, and the results of some direct determinations in the free field are given. The performance of subjective noise meters is then examined for noises of various types. A meter using two telephone receivers, which offers some advantage over the more usual types, is described. The performance of the objective type of meter is also examined. The measurement of noise associated with irregular sound fields is briefly discussed.

Nickel-iron alloys of high permeability, with special reference to mumetal

Abstract: The paper summarizes the development of high-permeability alloys from early researches to the commercial products of the present day.The difficulties encountered during manufacture, and the precautions necessary in these metallurgical processes, are described.Stress is laid on the correct methods of application to ensure that the magnetic properties of the alloys are utilized to full advantage, and the need for heat treatment is emphasized . Propositions are put forward to explain the mechanism of this heat-treatment process, and the practical procedure is given.Data on curves show the electrical and magnetic properties of the alloys.The paper finally discusses the practical applications of high-permeability materials, giving examples of commercial instruments and indicating cases in which their extended use would be advantageous.

Thermal fluctuations in complex networks

Abstract: This paper is concerned with the thermal fluctuations generated in networks containing reactance as well as resistance. Earlier work is reviewed, and it emerges that Moullin and Ellis were unable to interpret Nyquist's equation for a network at uniform temperature by their method of representation. An alternative method is suggested in which each element of resistance is replaced by a fluctuation generator in series with the element of resistance. The fluctuation voltage appearing between any two points in a linear network with any temperature distribution is then evaluated. This general expression is shown to be in agreement with Nyquist's result if the temperature is uniform. It is also shown that the relevant values of resistance in these formulae are the equivalent power-loss resistances and not the metallic resistances.There follows a description of experimental tests which are regarded as satisfactory verification of the method of representation adopted and of the values of resistance used.The noise/signal ratio existing between any two points in the generalized network is then discussed, and is shown to be incapable of exceeding the value obtaining in the arm in which the signal is introduced. General recommendations for optimum noise/signal ratio are given.A few special cases are then considered. It is shown that in radio reception a resonant circuit is preferable to a bandpass filter as a coupling between aerial and amplifier. In the former case an analysis of the optimum coupling between aerial and circuit is given which takes account of the fluctuations generated in the amplifier itself. A simple practical method of obtaining the optimum conditions is outlined.Fluctuations in long cables are briefly discussed. It is shown that temperature conditions in the vicinity of the receiving end only need be considered. The use of transformers or resonant circuits to couple the cable to an amplifier is governed by the same recommendations as for aerialamplifier coupling.

Ultra-Short-Wave Refraction and Diffraction

Abstract: The work of G. N. Watson on the propagation of electric waves over a spherical earth has been extended by the author to take account of the finite resistivity of the earth, the effect of which is of great importance in ultra-short-wave transmission. The work is in such a form that the field intensities above the earth can be computed numerically. The effect of refraction in the earth's atmosphere can also be taken into account. The results for a range of wavelength between 2 and 10 m and for heights up to 4 000 m and distances up to 400 km, are published in the paper in a set of curves, the general properties of which are discussed. The effect of atmospheric refraction is considered, and a comparison between observation and theory is made, in which good agreement is obtained, on the average, when neglecting refraction. Major changes may, however, be produced occasionally by refraction.

International comparison of impulse-voltage tests

Abstract: A number of impulse-voltage tests have been made under the auspices of the International Electrotechnical Commission in certain high-voltage laboratories in Europe. The tests were devised to ascertain the extent of agreement which can be reached on the measurement of impulse voltages at the present time, and to act as a guide to the Commission in framing a specification for impulse-voltage testing of electrical apparatus. The dispersion of results obtained on the measurement of the sparkover voltage of a simple test gap in eight European laboratories is at present very wide, being as great as ± 8%, though good agreement exists between the average values so obtained and the average values for similar tests recently published in the U.S.A. The tests on line insulators also reveal a wide diversity of results, but it has been established that insulators which have successfully withstood power-frequency voltage tests may sometimes be fractured by impulse voltages, and sufficient information is now available to warrant the introduction of impulse-voltage tests on line insulators by the I.E.C.

The effects of impulse voltages on transformer windings

Abstract: The effects of the application of impulse voltages of varied wave-shapes to transformers are analysed theoretically and experimentally. Single and multiple transients have been considered, and oscillographic tests have been made on model transformers of different winding characteristics, and on both single-phase and 3-phase transformers. The constants of the transformers important for determining the voltage distribution within the windings due to Ihe impact of a surge have been calculated and measured, and the effect of altering these constants has been examined. In general, good agreement has been obtained between theory and experiment. The transference of surges from one winding to another has been briefly considered, and the testing of transformers with impulse voltages has been discussed in the light of the foregoing analytical study.

The transmission of alternating-current power with small eddy-current losses

Abstract: A method is outlined of designing single-phase conductors with small eddy-current losses. Experimental results are given verifying the theory and the accuracy of the formula for calculating the eddy-current losses.

Restriking voltage, and its import in circuit-breaker operation

Abstract: Restriking voltage and its principal significance are described, and the problem of relating these to circuitbreaker operation in power systems is stated. An instrument for showing inherent restriking-voltage characteristics is described in some detail, reasons being given for the leading features adopted; and the method of operation is outlined. The scope of possibilities in system investigations and in circuit-breaker development is indicated. A short appendix deals with design of the instrument.

Layout and rupturing capacity of protective devices in motor circuits

Abstract: The paper deals with some of the practical considerations that will determine the decisions on technical points taken by engineers who are responsible for the installation of motor control gear and sub-distribution switchgear, for isolating and protecting motor circuits.The functions of isolation and protection are shared by motor control gear and sub-distribution switchgear, especially when motors function in groups. Various grouping arrangements are compared.The use of 3 300 volts in motor circuits is discussed, and a reference is made to the use of 3 300-volt contactors.The increasing use of contactors emphasizes the importance of further study of the process of making and breaking circuit by means of contactors, which can be examined in more detail by a demonstration with slow-motion films.After a note on thermal overload trips, the rupturing capacity of starters is discussed with particular reference to the short-circuit values which can be fed from the transformers and supply cables usual in l.t. circuits.The rupturing capacities of rewireable and cartridge fuses are discussed, with particular reference to the time of operation of the latter on severe short-circuits.The rupturing capacity of l.t. circuit breakers is shown to involve problems akin to those which have been studied in h.t. switchgear design.Finally, the problem of the anticipation of short-circuits by leakage indication is stated in terms of practical application.

Modern factors affecting electricity costs and charges

Abstract: The paper contains a brief review of the thesis presented in 1933 by Mr. Kennedy and Miss Noakes in their paper before The Institution. These authors concluded that the efficiency of electricity distribution had not improved appreciably during the period 1921?31, and their hypothesis has been accepted in this paper, which consists of a review of modern factors which may be contributing to the lack of increase in the efficiency of distribution. A brief comparison is made between rural and urban areas and it is concluded that rural areas quickly become remunerative and that they are not likely to have affected the efficiency of the supply industry as a whole. The rate of increase of domestic consumption is analysed for the whole country and it is shown that the average consumption per consumer using electricity for lighting, heating, and cooking, has not increased during the period 1927?35 and that it may have actually decreased during that period, owing to the connection of large numbers of domestic consumers who use only a small quantity of electricity for the minimum purposes, and. who are not subsequently increasing their annual consumption. In order to examine the causes of this lack of development in domestic supplies an analysis is made of the average charges of each undertaking for domestic and power purposes, and the relationship between charges and consumption is also considered. The differentiation between the charges for power and domestic supplies is shown to be irrational if domestic development is envisaged on broad and progressive lines, and it is suggested that domestic supplies may be subsidizing power supplies in many instances. Finally, the factors mentioned above are shown to be likely to affect seriously the trend of supply development in the immediate future, and the conclusion is reached that the very necessary expansion of development must be controlled and directed in a more co-ordinated manner. It is suggested that the expansion of demand must be preceded by a reduction of charges for domestic supply and a halt in the downward trend of charges for power. The alteration in charges must be accompanied by a greater unification of tariffs and charges which may ultimately be achieved by voluntary arrangements between undertakings but which can only be achieved quickly by means of a single executive authority which would have the exclusive right to direct and initiate the unification of tariffs and charges. Three reasons are advanced in the paper to account for the mal-development of the domestic supplies, viz.:? (1) Prices for domestic supply are generally too high to permit of the extensive use of electricity, and reductions in price tend to follow an increased demand, instead of preceding: isfe., As a result the domestic load continues to be restricted, the fixed charges per unit sold remain high, and a condition of expanding demand with consequent falling costs becomes impossible of achievement. This is particularly true for undertakings supplying less than 10 million units per annum. (2), Electricity for power purposes is sold, in many cases, at charges which are unremunerative and which are considerably below the equivalent cost of production by private plant. The effect of this is to place on domestic charges a burden which prohibits the domestic load from increasing to the point at which its inherently high diversity makes it profitable. This applies chiefly in the case of the larger undertakings. (3) The absence of a unified tariff basis and a unified scale of charges for electricity causes anomalies and inequities which have a psychological effect in deterring consumption.

Modern receiving valves. Design and manufacture

Abstract: The authors discuss the main features in the geometrical design of the types of valve in common use to-day and the various factors, mechanical and chemical, which impose limitations in manufacture. The minimum tolerances to which it is possible to reproduce characteristics are indicated. The paper includes a brief historical survey of the recent improvements in thermionic emitters and gives details of the precautions necessary in the production of modern highly efficient oxide-coated cathodes and insulated heaters. Pumping and activation processes are described, and the main factors affecting the life of a valve are discussed. The last section of the paper deals with some of the limitations encountered in the use of valves, such as hum, microphony, noise, and frequency limitation, and the methods of minimizing these factors are given. The authors conclude with some observations on possible future developments.

Modern systems of multi-channel telephony on cables

Abstract: Following a brief reference to progress in the application of carrier telephony to circuits in cables in this and other countries, an account is given of the carrier system operating between Bristol and Plymouth. This system affords 12 telephone channels per pair in the range 12 to 60 kc. Separate cables are provided for the two directions of transmission, and negative-feedback repeaters are provided at intervals of approximately 20 miles. When fully equipped the system will provide 228 circuits. The account of this system is followed by a description of the 4-tube underground coaxial cable which has been installed between London and Birmingham and of the modulating and repeater equipment that has been developed to operate 320 carrier telephone channels in the range 0.5–2.1 Mc. Repeaters are provided at intervals of approximately 7 miles, separate tubes being used for the two directions of transmission.Applications to television and the probable line of development of multi-channel and wide-band systems are discussed.

An experimental investigation of the theory of eddy currents in laminated cores of rectangular section

Abstract: It is shown that a complete verification of the well-known theory of eddy-current shielding in plane sheets can be obtained by measuring the variation of the impedance with frequency of a coil wound on a laminated core consisting of non-magnetic material. The verification of the theory is straightforward when the screening effect is small, because laminations thin enough to simulate the conditions of the plane sheet can be used, whereas much thicker laminations are necessary when the screening effect is large, and the simple theory must be extended to include the ratio of lamination thickness to lamination width. This extension is derived from an analysis of the eddy currents in an elliptical core, which, as the measurements show, applies to a rectangular core when the screening effect is large. When the screening effect is small, however, the results of the analysis of the elliptical core no longer apply to the rectangular one, and the difficulties of obtaining a mathematical solution in this case are indicated. With magnetic laminations the simple theory no longer holds in practice, and Peterson and Wrathall have shown that it is necessary to modify this theory to take into account the presence of a surface layer. The modifications are examined in an Appendix, and the results obtained indicate that they apply when the screening effect is negligible but not when it is large. Nevertheless, the equations obtained by Peterson and Wrathall are shown to provide a means of evaluating the thickness of the surface layer.

The use of protective multiple earthing and earth-leakage circuit-breakers in rural areas

Abstract: This paper describes investigations made to ascertain to what extent consumers in rural areas are protected from risk of shock with methods of earthing at present in use, and what changes in practice might be desirable, with special reference to protective multiple earthing. The investigation shows that present safeguards are inadequate. Recommendations are made which it is believed will lead to material improvement, but further experience and experiment will be necessary before the matter can be put on an entirely satisfactory footing. It has been realized for some time that in rural areas fed by overhead lines where no water mains exist the protection afforded by ?ordinary earthing? is not entirely satisfactory. With this in mind an investigation was undertaken by the E.R.A. with a view to finding means for improving the situation. Alternative methods of protection are provided by earthleakage circuit breakers, or the use of the neutral to carry the fault current. This latter method, known throughout this paper as ?protective multiple earthing,? is subject to the disadvantage that the frameworks of apparatus connected to the neutral rise to the same voltage above earth as the neutral conductor. It was found from field tests, however, that the voltagerise of the neutral did not exceed a value dangerous to human beings under normal operation, and that even animals would be protected if the statutory voltage-drop were not exceeded. No interference with the satisfactory working of the telephone system was observed during any of the tests. Other possible sources of trouble due to the use of the protective multiple earthing system have been considered in some detail. An investigation has also been made to find the tripping voltage of earth-leakage circuit breakers which have been in use for a period of about 2 years. The tests were made by Mr. E. Fawssett in the electricity supply area of the Dumfriesshire County Council. The paper concludes with the major deductions reached from the practical tests, some conclusions respecting protective multiple earthing in general, and, finally, recommendations respecting existing and new installations and certain proposed conditions which should be complied with before protective multiple earthing is permitted. The Director of the E.R.A. will welcome comments from any who may have occasion to make use of this paper.

The effect of change of temperature on the strength of permanent magnets, with special reference to modern magnet steels

Abstract: A survey is made of the literature and it is found that a magnet can be put into the steady condition, as regards magnetic strength, by mechanical shocks or by cyclic heating, and then it has a definite strength at each temperature, the connection between the two generally giving rise to a negative coefficient. The majority of the experimenters have assumed a linear relationship and have determined the strength/temperature coefficient from measurements at two temperatures only. A summary of published work shows that the coefficient depends upon the linear dimensions of the magnet and can be reduced by increase of the length/breadth ratio, by increase of the degree of drawing, by heat treatment, or, in the case of carbon steels, by increase of the carbon content. In the present paper a description is given of a rotating search-coil and ballistic-galvanometer arrangement for testing cobalt steel, aluminium-nickel, and aluminium-nickel-cobalt alloy magnets over the temperature range −60° C. to 100° C An electrically heated oil bath was used for the range 20° C. to 100° C, whilst solid carbon dioxide-dissolved in acetone wasemployed for the range 20° C. to −60° C. When the magnets had attained the reversible state it was found that the magnetic strength Hτ could be connected with the temperature τ(°C.) by an equation of the typeHτ = H0(1 + aτ + bτ2), where a and b were negative constants and were determined for each magnet. The total permanent loss of strength up to the attainment of the steady state was also measured. It was found that, in the case of the cobalt-steel magnets, the coefficient a was reduced by increase of the cobalt content; the total loss was similarly affected, but not to such a marked extent. In the case of 15% Co steels and the Al-Ni and Al-Ni-Co alloys, a and the loss were affected by the temperature at which magnetization was performed. The other cobalt magnets were magnetized at ordinary temperatures.

The micro-gap switch

Abstract: Alternating-current circuits that can be classed as noninductive may be broken by a switch in which the separation between the sparking contacts is extremely small. In these micro-gap switches the separation of the contacts on a 250-volt circuit is of the order of 5 thousandths of an inch, and, by the use of a back stop, is not allowed to be more than that distance. The current is interrupted, not by stretching the arc at break, but by the mutual repulsion of .electrons that enter the gap from the contacts. Oscillograms are given which show that, when the contacts in these switches begin to open at any point in a half-wave, the current breaks at the next zero point and does not restrike. If broken at zero it persists for the next halfwave. This type of switch is most suitable for the thermostatic control of electric heating systems.

Control rooms and control equipment of the grid system

Abstract: The general problem of control of the grid is outlined. Reference is made to the location and layout of the control buildings. Details are given of the communication system rented from the Post Office and of the special requirements with regard to its use in conjunction with the Central Electricity Board's system. The equipment of the control rooms, consisting of telephones and automatic indicating equipment, control desks, system diagram, frequency and time control instruments, is described. Reference is made to the general features of the telephone system and automatic indicating gear. Particulars are given of the various systems of remote metering employed. Concentration of essential equipment in a convenient form on the control desk is explained, with special reference to the desk at Bristol. Types of system diagram and frequency-control equipment are discussed.

Continuous extrusion of lead cable-sheaths

Abstract: The paper deals primarily with a new development in methods of producing lead cable-sheaths, in which the pressure required to extrude the lead is obtained by means of a motordriven screw member instead of by the ram of a hydraulic press which has hitherto been universal practice. The author's experiments were commenced in 1929 when the first lead pipe was extruded by a continuous process, and the development has continued steadily up to the present day, and to the stage where commercial continuous lead-extrusion machines are now being delivered. After reviewing in detail the disadvantages of former methods in order to explain the reasons for the development, the paper describes the construction and operation of the continuous lead-extrusion machine, and from considerations of the physical and metallurgical properties of the sheath demonstrates the improvements in the product achieved. As the advantages attendant on the use of the continuous lead-extrusion machine are associated to some extent with the methods employed in melting the lead, and in the handling of the cable as it comes from the machine, special attention has been given to these features alongside the development of the machine itself. The paper describes improved lead handling and melting equipment and automatic reeling mechanism for he finished cable.

The Response of Rectifiers to Fluctuation Voltages

Abstract: This paper is concerned with the response of square-law and linear rectifiers to fluctuation voltages. Analyses of the square-law rectifier and of the linear rectifier in the presence of a dominant harmonic voltage are given. The analyses include the limitations imposed on the output by pre-rectification and post-rectification amplifiers, and are tested experimentally. The behaviour of linear rectifiers in the absence of a dominant harmonic voltage is examined by experimental methods.It is shown that, unless precautions are taken, rectifying stages emphasize the background of fluctuation noise. The reception of telegraphic and telephonic transmission is considered and the precautions relevant to various conditions are discussed. “Note corrected” reception is shown to be preferable to “uncorrected” reception, both in respect of fluctuation interference and in respect of distortion due to non-linearity of the rectifier characteristic.

The metering of mercury-arc rectifier supplies and outputs

Abstract: A preliminary survey is made of the difficulties involved in metering (a) active energy, wattless kVA, total kVA, and power factor, in the input circuits of mercury-arc rectifiers, owing to the presence of harmonics in current and voltage;and (b) energy output from mercury-arc rectifiers when considerable ripple exists in the current and voltageThe performance of induction-type and motor-type integrating meters is then examined experimentally on rectifier circuits, and the observed errors are quoted Good agreement is found between the observed errors and those expected from theory and the known variation of the meter characteristics with frequency, both in the case of energy and in the case of kVA measurement Similar tests are described on motor- and mercury-type integrating meters on the output side of rectifiersIt is concluded that ac power measurements can be made with satisfactory accuracy by induction- and motor-type meters No entirely satisfactory solution is found, however, for power-factor and kVA measurement Finally, the necessary precautions to be taken with dc integrating meters on the output side are dealt with

Measurements of the high-frequency resistance of single-layer solenoids

Abstract: The paper describes an attempt to determine experimentally the accuracy of Butterworth's formulae for the highfrequency resistance of single-layer solenoids. The investigation has been carried out on 4-in. diameter air-spaced coils of 10, 30, and 50 turns over the frequency range from 2 × 105 to 2.025 × 106 cycles per sec. The coils were wound with bare wire of diameter 1.63 mm. (No. 16 S.W.G.), with a distance of 2.60 mm. between the centres of adjacent wires. The method of measurement described was based on the use, for each coil shape, of a series of identical coils wound with copper, aluminium, brass, german silver, and eureka wire, respectively. There is some uncertainty in the interpretation of the results, but, in spite of this, it is evident that of the two formulae with which a comparison is made, one of them is correct to within about 4 per cent for the 30- and 50-turn coils over the whole frequency range considered. This formula is quite inapplicable to the 10-turn coils; however, for these the second formula gives reasonably good agreement with the experimental results.

Single-Phase Electrification on the Pennsylvania Railroad

Abstract: This paper describes the steps leading up to the present electrification of the Pennsylvania Railroad, the system used, the extent, and the operating results. The 4-track main line is now electrified between New York, Philadelphia, Wilmington, Baltimore, and Washington, as are also the commutation lines around Philadelphia and New York. Since this paper was prepared, work has started on the extension of the electrification to include the connecting lines for both passenger and freight service west to Harrisburg, Pennsylvania. The chief parts of lines in the new electrification are the main line from Paoli, just west of Philadelphia, through Lancaster to Harrisburg; the low-grade freight line from Morrisville, Pennsylvania, near Trenton, New Jersey, via Columbia to Enola Yards, near Harrisburg; the freight line from Columbia following the Susquehanna River to Perryville, Maryland; the freight line from Monmouth Junction to South Amboy, New Jersey, with necessary yards, connecting branches, and equipment. The new work will involve the electrification of 315 miles of line and 773 miles of track. Upon its completion the Pennsylvania Railroad system will have 41 per cent of the total electrically-operated standard railroad track in the entire United States.

A short-wave Adcock direction-finder

Abstract: This paper gives full details of the construction and performance of a short-wave aural Adcock direction-finder of the coupled type with a working wavelength range of from 35 to 70 metres (frequency 8.6 to 4.3 Mc.). It describes how the design of the instrument is based on the principles brought to light in an earlier investigation. It deals with its design in respect of polarization error, pick-up factor, and instrumental error, and describes experiments by which these properties were measured. The values obtained for these were as follows: standard-wave error, 3° to 8°; pick-up factor, 4.5 to 6 metres; instrumental error, 2° to 3° at maximum. The apparatus is considered to constitute a satisfactory direction-finder from a practical point of view, though it will be possible in future models, one of which is already under construction, to take advantage of the experience gained to obtain an instrument with lower instrumental and polarization errors and higher pick-up factor.

Mutual impedance of circuits with return in a horizontally stratified earth

Abstract: During recent years considerable attention has been given to investigation of the mutual impedance between two circuits with earth return. The problem is of importance. On a knowledge of this coupling depends the predetermination of the telephone interference which may arise from earth-fault currents or residual currents of harmonic frequency in power and traction systems. Theoretical formulae, which have been confirmed by field experiments, are available for its calculation where the earth can be considered to be homogeneous and of uniform conductivity. Experience has shown, however, that this condition is not always fulfilled. Where it is not, the assumption of an earth structure consisting of two horizontal layers of different conductivity would in nearly all cases give the accuracy required for engineering purposes. Curves have been prepared to facilitate calculation of the mutual impedance in the case of stratification of this type. They give an equivalent conductivity which, when introduced into the formulae relating to a homogeneous earth, will yield a mutual impedance equal in magnitude to that resulting from the given stratified conditions.

The examination and recording of the human electrocardiogram by means of the cathode-ray oscillograph

Abstract: The cathode-ray oscillograph, having no inherent inertia and yet being comparatively indestructible, is an instrument peculiarly suitable for everyday electrocardiography. A fluorescent screen of very long afterglow enables the electrocardiogram to be examined directly at the patient's bedside, which is a new facility.The design of the amplifier has to be adapted to the peculiarities of the heart-action voltages and electrical constants of the patient; these are considered in some detail, and practical methods for the examination' of the adequacy of the amplifier are discussed.The provision of a high-voltage supply suitable for any eventuality is considered, together with methods of obviating interference, especially from alternating-current mains.

Constant temperature: a study of principles in electric thermostat design; and a mains-operated isothermal chamber constant to one-thousandth of a degree centigrade

Abstract: The paper is concerned with the design of apparatus for maintaining a chamber at a nearly constant temperature, above the ambient temperature, despite fluctuations of its environment. After brief discussion of thermostats in general, attention is confined to electrically-operated devices in which the temperature-sensitive element enters into a bridge configuration: departure of the temperature from the assigned working value, by throwing the bridge out of balance, effects a corrective change in the heat supply. By analysis and experiment published elsewhere, the author has shown that in such systems hunting, necessarily present if the heat supply is controlled in discrete quantities, will occur also if there is a continuous relation between temperature and heat supply, provided that the control sensitivity exceeds a critical threshold value. The bearing of this result on the design of fine temperature-regulating apparatus is examined, and the frequency and amplitude of hunting are considered. The relations of the factors determining the residual inconstancy of temperature are analysed. Amongst the conclusions reached are (roughly stated): (a) hunting does not depend on the heat capacity of the chamber; (b) the residual inconstancy does not depend on thoroughness of thermal insulation of the chamber; (c) there is a single figure of merit measuring the intrinsic effectiveness in resisting changes of ambient temperature and of supply voltage; (d) of these two factors, the former is of less practical importance than the latter; (e) the effect of temperature-difference between the chamber and the temperature-sensitive element must not be ignored; (f) identification of the temperature-sensitive resistor with the resistor carrying the controlled current, advocated by someprevious workers, is wrong in principle. The author's apparatus is described: it is the outcome of an attempt to obtain the smallest possible inconstancy consistent with derivation of electric supplies from the a.c. mains alone, and is in accord with the principles reached in the foregoing study. Difficulties experienced in measuring small temperature-changes are reported, and the significance of readings of mercury thermometers is assessed. Finally, it is shown that the author's design could be modified for large size and high temperature. A term not in common use has, for the sake of brevity, been employed throughout the paper: a ?millideg.? (to be pronounced as written) stands for a thousandth of a degree Centigrade.

The gas-impregnated cable

Abstract: In Part (1) the various types recently put forward to solve the high-voltage cable problem are discussed. The principle of the gas cable is stated, together with the advantages to be expected. Part (2) describes very briefly the experimental work on the basis of which the cables were made, and gives reasons for the choice of paper, gas, operating pressure, etc. Part (3) is devoted to the construction and performance of the actual cables. Examples are given which show their extreme stability, and also the accuracy with which their performance can be predicted. A series of typical designs are tabulated, and the advantages anticipated in Part (1) are shown to be realized. Part (4) deals with the design of gas-cable joints and sealing ends, and the methods used to maintain dryness during their making. The gas feeding arrangements and other accessory apparatus are also described.