Графични страници
PDF файл
ePub

T

138

ELECTRICAL REVIEW.

spans; although, in this case, the remedy is almost as bad as the disease, in consequence of the serious reduction that would ensue in the insulation resistance. In fact, the more expensive metal does not have fair play in this respect, unless it has a line of poles exclusively for itself.

Again, the rust with which iron wires become coated, sometimes prevents actual metallic contact, while a copper wire preserves a comparatively clean surface; and the low resistance of a copper line allows the current caused by any given contact to rise to a much higher value than would be the case with an iron wire of perhaps treble the resistance. There is no doubt, however, that copper lines will become more reliable when the experience in erecting and jointing them has become more nearly equal to that in the case of iron.

It may thus be fairly concluded that, notwithstanding some slight disappointment which may exist in certain quarters concerning the copper lines at present in use, the question of cost is the only one which will prevent its rapidly supplanting iron, and that the troubles above mentioned will speedily diminish as its employment becomes more general.

Whenever a breakdown occurs of such a magnitude that the telegraph engineers cannot cope with it and restore communication with sufficient rapidity to prevent serious interruption to commercial traffic, the telegraphing section of the public demands the adoption forthwith of an underground system, at any rate, for the important trunk lines, either to entirely replace existing overhead wires, or to serve as a stand-by in case of failure of the latter.

We intend, presently, to discuss this underground proposal; but it is a matter for careful consideration whether the present system cannot be made so reliable as to render the adoption of the more expensive plan unnecessary. As has been remarked, the improvement in overhead construction since the lesson taught by the great storm of 1886 has been enormous, and is still proceeding. It is, in fact, safe to say, that it is possible to prevent such a complete collapse ever again occurring. But there yet remains for consideration the interruptions due to "ordinary" causes, such as the falling of a tree across a line, the slipping of a railway embankment carrying with it a set of poles and wires, or the wrecking of an important pole in a railway accident. The inconvenience attending such accidents can be, to a considerable extent, diminished by arranging that communication between important centres shall be maintained by different routes; or, to use a familiar phrase, by being careful to avoid putting too many eggs into one basket.

For instance, communication between London and Dublin is at present maintained by wires following three distinct routes; one of them going via Swindon and Cardiff to Haverfordwest, and thence by cable to a point near Wexford; a sécond via Chester and the secluded village of Llanfairpwllgwyngyllgogerchwyrndrobwilltysilliogogogoch (relay station), through the Holyhead-Howth cable to Dublin; and the third, which is of copper wire throughout via Worcester and Shrewsbury to Nevin, in Carnarvonshire, thence by cable to Newcastle, Co. Wicklow.

Again, an earthquake even of Javan dimensions could hardly interrupt communication between London and Berlin, unless, indeed, it happened to occur unpleasantly near either of those cities. For the London-Berlin lines also follow three distinct routes -1. By road to Lowestoft, and thence by a cable landing in North Germany, and so on to Berlin. 2. By railway to Lowestoft, and thence by a cable landing at Zandvoort, on the coast of Holland, and viâ Amsterdam to Berlin. 3. From London to Dover, and thence by a Belgian cable via Brussels to Berlin.

By such an arrangement more generally applied an entire stoppage arising from any ordinary or local fault can be prevented.

It should be remarked that, in the case of inland working, the failure of even two-thirds of the lines between two first-class stations need not necessarily be accompanied by any serious delay to the traffic. When everything is going smoothly the number of lines is sufficient to enable the bulk of the messages to be transmitted by hand, each wire working duplex, and averaging, say, 70 messages per hour. When a collapse occurs, the wire or wires still left standing can be immediately switched to automatic high-speed apparatus. This can also be worked on the duplex system, and with a good wire and expert operators a constant stream of ges can be maintained in two directions at the rate of

JANUARY 30, 1891.

about 250 or 300 words per minute. Under normal conditions it is preferable to have a sufficient number of wires to enable the whole of the work to be dealt with by hand, as not only can the messages then be more promptly despatched, but the amount of work per man is greater, the automatic apparatus being simply held in readiness for any

emergency.

But much can be done to render the restoration of any line easy and rapid of accomplishment. The first step is to quickly discover the exact locality of breakdown, and in order to do this, it is necessary to lead the wires into a comparatively large number of testing stations, and to give such stations ample facilities for prompt communication with the controlling offices. Much, then, depends upon the lineman, who is responsible for the section in which the fault occurs. The section under his care should be as compact, and all points as easy of access as possible; it should at all times be possible to inform him without delay of any interruption; he should be provided with ample means for dealing with such interruption, and, above all, be intelligent and fertile in resource. Suppose, for instance, a line of a dozen wires to be broken down by any cause, it should be possible for communication to be re-established within a very few hours by the insertion of a length of light cable in the break, and the permanent repair of the line, involving perhaps the setting of a pole, could then be effected at leisure. For this reason, it would probably be found advantageous for each store depôt to be provided with a length of about 100 yards of light cable, containing, say, a dozen wires, each end of each wire being duly numbered and labelled, so as to save the time that would otherwise be taken in identifying them. Many instances, in which serious breakdowns have been promptly met with but meagre resources, could be quoted, and also some cases in which quite the reverse has happened; in fact, few classes of men could be subjected to a sound technical training with such immediate and evident advantage as those responsible for the maintenance of important telegraph lines.

There is another advantage attending the practice of leading wires into a large number of testing stations, and thus dividing them into so many short sections, and that is the facilities which are thereby afforded for effecting "crosses," and so rendering it possible to substitute a good length of an unimportant or idle line for a faulty section of an important one in order to maintain communication by the latter at the expense of the former.

This question of facilities for crossing is an exceedingly important one. It often happens that two wires running on a set of poles are stopped each in a separate locality, and where test boxes admit of this being done, the two faulty sections can be thrown on to one wire, and communication by the other wire restored by utilising the good section of the abandoned wire. As a rule there is not much difficulty with wires belonging to one line, but the principle could be pursued further than it is were the trunk lines to meet more frequently than they do. Doncaster, a comparatively small town in itself, is of considerable importance from a telegraphic point of view, for no less than three important North trunk lines run through the test box there, and such changes as have been indicated can be readily effected; but the stretch from London to Doncaster is a long one, and it would be an advantage if another point about midway could be made similarly available. The desire to erect a line with the minimum underground work caused the new Irish route via Nevin to be taken along second-rate high roads so as to avoid the big towns, but the consequence is that the line only touches other main lines at one point this side of Worcester, viz., at Aylesbury, where it meets the heavily laden line recently erected to Manchester, Liverpool, Scotland and the North of Ireland. It would have been a great boon if the Nevin wires could have been deviated a little, so as to pass through such a town as Oxford, where the test boxes already include two important trunk lines. But to have done this would have involved the introduction of additional underground work, and it must not be forgotten that this would have prevented the accomplishment of the high speeds which have been attained on these wires, viz., 400 words per minute. We have in mind one memorable night when press work was being literally poured into Dublin (and other Irish stations simultaneously) on three separate wires, each working at this enormous rate, so that in two - minutes a good column of matter was rushed through. There

JANUARY 30, 1891.]

ELECTRICAL REVIEW.

are many who condemn high-speed working, but there is one difficulty which such people generally overlook, viz., that to transmit the work by hand on single working circuits, a much larger number of wires would be required, and there is practically no room to erect them. Almost every main road in the country has now been studded with poles, every line of railway has also its array of telegraph poles, and most of these poles have nearly their full number of wires. It is easy, therefore, to see that to meet the ever increasing demands of the telegraphing public, it is essential to increase the carrying capacity of the wires, and this it is to a large extent which has caused the rapid increase in the number of duplex, quadruplex, and multiplex circuits. To enable these systems, together with fast speed Wheatstone, to be worked satisfactorily it is essential that long or numerous underground sections should be avoided. (To be continued.)

PROCEEDINGS OF SOCIETIES.

Liverpool Engineering Society, January 21st, 1891.

THE LIVERPOOL ELECTRIC SUPPLY STATIONS.
BY A. BROMLEY HOLMES, M. Inst. C.E.

THE object of the present paper is to describe briefly the central stations, machinery and appliances now in operation in Liverpool for the supply of electricity for lighting and other purposes under the provisions of the Liverpool Electric Lighting Order of 1889.

Before proceeding to describe the plant in detail a few words as to the past history of the Liverpool Electric Supply Company may be

of interest.

At the date of formation of the Company in January, 1883, there was not a single installation of incandescent lamps in Liverpool, and the first building lighted by this means, in July, 1883, was the restaurant in Eberle Street, where the current was generated by a gas engine and dynamo on the premises, and the light supplied at a fixed annual charge.

About the same date the Company acquired a site and constructed its first central station in Rose Street, off Lime Street. This station was started in December, 1883, to supply electric light to the Grand and Adelphi Hotels and to other consumers in the neighbourhood, and continued without intermission in daily operation from that date till November, 1890, when the mains were connected to one of the newer stations. The capacity of the Rose Street station was under 1,000 lamps.

In 1887 a larger station was built, and commenced work in December of that year, in Tithebarn Street. This now forms part of the Highfield Street station.

In September, 1888, the Harrington Street station was completed and at work.

In 1889 the Company's new works, offices and central station were built in Highfield Street, and the supply of electricity began in October of that year.

The Company's latest station in Oldham Place commenced work in October, 1890,

[ocr errors]

The following figures show the rapid extension of electric lighting in Liverpool, the number of lamps of sixteen candle power (or their equivalent) connected with the Company's stations being as follows:On January 1st, 1888 ... 977 sixteen candle-power lamps. 3,330

Do.

Do.

Do.

1889

1890

1891

do.

[blocks in formation]

do.

do.

do.

[merged small][merged small][merged small][merged small][merged small][ocr errors][merged small]

Total 15,500 The above figures give the number of 16 candle-power lamps that the machinery is actually capable of supplying direct at one time; and as experience has shown that never more than 75 per cent. of the lamps connected to the supply mains are in use at any one hour, the above plant would have a working margin of 25 per cent. spare power for the number of lamps stated, without taking into account the storage of electricity always available from the accumulators.

The system in use is a direct current continuous supply at a pressure of 110 volts, all the mains being in simple parallel. It may be convenient to give a detailed description of the plant before describing the method of working.

The boilers are 28 feet long and 7 feet 6 inches in diameter, and are made of mild steel. The shell plates and end plates are g-inch thick, and the ends are strengthened by gusset stays. The boilers have two flues 3 feet diameter, tapered to 2 feet 6 inches diameter at the back end, and made of mild steel plates 7-inch thick, and each

139

flue is fitted with six steel Galloway tubes. The working pressure is from 110 to 120 lbs. per square inch.

Of the 10 boilers in use seven are fitted with Proctor's mechanical stoker, and the remaining three with Viccars's mechanical stoker.

The water supply is taken into the storage tanks from the town mains, and, where practicable, a duplicate service has been carried into the station from a second main.

Two feed pumps (or a feed pump and injectors) are provided at each station, and a complete double system of feed pipes, with separate check valves, on each boiler.

Feed water heaters are used, and an arrangement of valves, by which, in case of any breakdown of a heater, the water can be sent direct to the boiler or pumps without passing through the heater.

The steam and exhaust pipes are divided into sections, so that any necessary repairs may be effected without stopping the station. The engines are Willans central-valve compound engines of two sizes, ten of them being of the I.I. size, with cylinders 14 inches and 20 inches diameter and 9-inch stroke, capable of indicating 180 horsepower, the remaining three engines being of the G.G. size, with cylinders 10 inches and 14 inches diameter and 6-inch stroke, indicating 60 horse-power.

All the engines, which are single acting, have two cranks placed on opposite sides of the crank shaft. Eack line of pistons is connected to its corresponding crank by a pair of connecting rods, between which work the eccentric and eccentric rod which actuate the valve gear. The eccentric is forged solid on the crank pin.

Piston valves are used, working inside the hollow piston rod which connects the pistons.

All the moving parts are in constant compression and subject to a downward thrust.

At the top of the engine is the steam chest, common to both lines of cylinders. Underneath the steam chest are the high pressure cylinders; next, the low pressure cylinders; and underneath these latter the guide cylinders, which act as air cushions.

A sight feed lubricator is provided on the steam chest, and a grease cup on top of each line of cylinders for giving a flush of oil if needed. The chamber in which the cranks work is filled with water up to the level of crank shaft. A small proportion of castor oil is added to the water, and is churned up with it by the movement of the cranks and effectually lubricates the internal bearings.

The governor is fitted directly on to the end of the crank shaft and actuates an equilibrium throttle valve, fixed at the entrance of steam chest.

The bed-plate of engine is extended to carry the dynamo.

Each engine is fitted with a cylindrical steam separator, which ensures a supply of dry steam.

The dynamos have magnets of the single-inverted horse-shoe type, with massive wrought-iron limbs. The armature shafts are coupled directly on to the crank shafts of the engines.

The dynamos for the smaller engines give an output of 300 ampères, and an E.M.F. of from 110 to 140 volts, at from 450 to 500 revolutions per minute.

The larger dynamos give 900 ampères, and from 115 to 135 volts, at 375 revolutions per minute.

Some of the dynamos are compound wound, but the later patterns are plain shunt wound, and all have a variable resistance attached to them, by which the current through the shunt circuit can be varied and the E.M.F. regulated at will.

Two of the dynamos were manufactured by Crompton & Co., and the rest by Siemens Brothers. In the Crompton dynamos the resistance of the armature is 0028 ohm and of the shunt coils 11 ohms. The armature is built up of charcoal iron discs keyed directly on to a steel shaft 5 inches diameter. The core is 36 inches long and 18 inches diameter. The effort is transmitted from the shaft to the discs by the feather, and from the discs to the winding by means of delta metal driving teeth fixed in holes drilled in the core. The winding consists of eighty turns of pressed stranded bars, measuring 6 inch by 63 inch in section. These bars are made slightly taper, so that they may lie evenly against each other. The armature bars are connected by copper end connections, turned right and left on what is termed the butterfly plan. The commutator is made of cast copper sections, and is 9 inches in diameter and 8 inches long.

In the Siemens dynamos the resistance of the armature is '0024 ohm and of the magnet coils 6 ohms.

The armature is constructed with a core of thin sheet-iron discs threaded upon a round steel shaft. The core is insulated and surrounded with 144 insulated bars, each bar consisting of a number of wires stranded together and pressed into a square shape. The insulated bars are connected to one another and to the commutator, as usual in drum armatures. The armature is bound with bands of steel wire to resist the centrifugal force generated whilst revolving.

The accumulators have lead containing boxes, each about 4 feet long by 12 inches wide and 13 inches deep, supported on heavy wood slabs resting on glass insulators.

The cells are of two kinds, manufactured respectively by the Electric Construction Corporation (formerly known as the Electrical Power Storage Company), and by the Crompton-Howell Electrical Storage Company. The Construction Company's cells are of the well-known L. type, E.P.S. pattern, manufactured as follows:—

The grids are cast in iron chills, the negatives of lead-antimony alloy, and the positives of lead. Both kinds of grids are pasted with red lead mixed into a paste with dilute sulphuric acid, and are then dried and put into the forming tanks. Charging is commenced with a current density of two ampères per positive plate, increasing up to four ampères, the total number of ampère hours being about 100 each for the positive plates, which are then removed, and their places taken by blind positives, with which the completion of the forming of the negatives, which require about 200 ampère hours per plate, is continued. When formed, the plates are taken out of the tanks and

140

ELECTRICAL REVIEW.

burned together in sections as required. Each central station cell has three sections of twenty-seven plates each, or eighty-one plates in all, each plate measuring 9 inches by 9 inches by ths inch thick.

The maximum safe rate of discharge of the cells is 250 ampères, and the storage capacity about 1,700 ampère hours at ordinary rates of discharge.

The plates of the Crompton-Howell cells are composed of porous lead, the structure of the plates being entirely crystalline and presenting a large surface to the electrolyte. The active material is formed electrically right through the plate and does not readily fall away, but adheres firmly to the lead.

The plates are mounted on celluloid racks resting on the bottom of the cells, and are kept apart at the top by celluloid combs. The plates are not connected into sections, but the positives of one cell and the negatives of the next are burned on to a common lead bar, supported on insulators between each cell.

Each cell has fifty-one plates 8 inches by 8 inches by 4-inch thick, and may safely be discharged at the rate of 400 ampères, having a storage capacity of 800 ampère hours when discharged at the normal rate of say 200 ampères.

All the distributing mains of the Liverpool Electric Supply Company are laid underground, and have been manufactured and laid by the Callender Company.

The conductors are of stranded-copper wire, with a conductivity of 98 per cent.

The largest feeding mains are composed of ninety-one No. 9 wires, and have a sectional area of 1:57 square inches. The conductor is covered with two layers of jute yarn and boiled in refined bitumen. It is next wrapped with parchment tape and covered with a lead casing. The lead is protected by an outside covering of rough jute yarn treated with asphalte and tar.

The ordinary feeders are made of sixty-one No. 11 wires with an area of 6 of a square inch, and the distributors of thirty-seven No. 13 wires one quarter of a square inch in area. In both these mains the copper conductor is covered by a solid sheath of bitite or vulcanised bitumen put on under heavy pressure at one operation. This core is next taped and compounded, and finally braided with hemp yarn and passed through a bath of asphalte compound.

The mains are laid in cast-iron troughs 4-inch thick and about 6 feet long, with socket joints. When the troughs are placed in position about 4-inch of refined bitumen in a molten state is run in, and, before setting, spacing bridges of wood are placed in it about 18 inches apart. These bridges support the cables and hold them in place, clear of the sides and bottom of the trough and of each other. Bitumen is then run in so as to entirely cover the cables and fill up the iron troughs to within -inch of the top. The troughs are then finished by a covering of Portland cement concrete about one inch thick. Strong cast-iron covers are in some cases substituted for the concrete.

The service lines from the distributing mains to the consumers' premises are made of nineteen No. 14 wires, insulated and covered with lead pipe, and are run in wood grooved casing, filled in with bitumen and protected with a hard-wood cover strip.

The connections of the various lengths of mains to each other, and of the feeders to the distributors, are made by sweating copper lug pieces on to the ends of the cables and by bolting copper connecting bars to the lugs.

The distributing mains are jointed in a similar, way, the connecting bars being fitted with terminal screws for the attachment of fusible tin strips, to which the service lines are attached.

The connections described above are made in cast-iron joint boxes, measuring inside 16 inches by 12 inches by 12 inches, provided with socket pieces to receive the ends of the cast-iron troughs in which the mains are laid. Each box has a strong cast-iron cover held down by a central bronze bolt and crossbar. A watertight joint is secured by means of a soft Indiarubber washer between the cover and the box.

All joint boxes are below the level of the pavement, and the surface of the footway is completed by an iron frame and surface plate, by means of which access is afforded to the joint box. In many cases the surface plates are formed in cement to match the adjoining paying.

The general arrangement of mains is as follows:-A pair of distributing mains are laid under the footways on each side of each street, and are connected together to form a complete distributing network, to which are connected all the service lines to consumers' premises. For convenience in working the network is divided into sections. The supply of current is brought to selected distributing points in this network by heavy feeding mains, which run direct from the generating stations.

The feeding mains are so proportioned relatively to the distance from the station of each distributing point and the number of lamps to be supplied from it that the percentage of loss in pressure in each feeding main may be as nearly as possible the same; in other words, if the lamps were turned on and off uniformly all over the area of supply, a constant pressure might be maintained throughout the network by varying the E.M.F. of the dynamos at central station in exact proportion to the amount of current supplied.

As a matter of practice, however, the hours of demand vary in different streets and districts, and it is found necessary not only to vary the station pressure as a whole, but also to regulate the pressure for each set of feeders individually, The following figures will give some idea of the extent of the mains already laid :

The district served by the mains extends 1,300 yards (measured straight on map) from Lime Street to bottom of Brunswick Street, 1,000 yards from Old Hall Street to Head Post Office in Canning Place, and 1,550 yards from Highfield Street to top of Bold Street and Wood Street.

[blocks in formation]

[JANUARY 30, 1891.

[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small]

The weight of copper in the mains is 210 tons. The number of joint boxes, service boxes, and pressure wire boxes is 630.

The number of separate service lines is 275, but many of these serve for the suppply of several consumers.

Pressure wires are run in specially-laid cast-iron pipes from the various distributing points to the central stations, and are connected to instruments, by means of which the actual pressure at each point is indicated to the engineer in charge and is continuously and automatically registered.

The main switchboard in each station is fitted with an ammeter and registering pressure indicator for each pair of feeders.

These street mains are connected up to massive terminals on the switchboard fitted with safety cut-out strips, which would fuse in the event of any serious short circuit on the main.

All the dynamos in a station are arranged to work in parallel on to a pair of positive and negative collecting conductors, and are switched in and out of circuit as the variations in load require.

Each negative feeding main is connected from its cut-out to the station main negative conductor, and each positive feeding main is connected through an ammeter to a main regulating switch, which, when in its normal position, completes contact with the station main positive conductor. By moving the regulating switch it is possible to introduce a number of accumulators, one after another, in the circuit of any desired feeder. These accumulators are arranged in opposition to the dynamo current, and consequently, the pressure in the feeder is reduced by two volts for each cell put into circuit by the regulating switch.

This at first sight may appear a wasteful method of regulation, but in practice the load is found to rise so uniformly over the whole area of supply that the amount of energy absorbed in regulation is practically unimportant.

The battery of accumulators at each station is capable of supplying 500 lamps at a time, thus enabling the machinery to be stopped at from 10 to 12 o'clock at night and in some cases entirely on Sunday, and also largely in the day time in summer. The discharge from the accumulators is registered by a recording ammeter. By a special arrangement of dividing the accumulators into sections, devised by the author, they are charged from the dynamos supplying the town at the ordinary working pressure, and by a little care it is in this way possible to so regulate the load on the engines and dynamos as to run them at their highest efficiency.

The very great advantage of this will be appreciated from an inspection of the efficiency diagram of one of the 900 ampère sets of plant, from which it is seen that the electrical efficiency (that is to say, the ratio of the external electrical output of the dynamo to the indicated horse-power in the cylinders) is as follows:

[merged small][ocr errors][merged small][ocr errors][ocr errors][ocr errors][merged small][ocr errors][merged small][ocr errors][merged small][ocr errors][merged small][merged small][ocr errors]

Continuous tests of the insulation resistance of the mains are made at each station by means of a special apparatus devised by the author for this purpose. If two lamps be connected in series between the positive and negative mains, and a point between the lamps be connected to the earth, the lamps will be found to burn with a variation in brilliancy proportional to the relative insulation resistance of the mains to which each lamp is attached. If two voltmeters be substituted for the lamps their readings will indicate the same ratio. In the apparatus referred to, a registering voltmeter is used to indicate this ratio continuously. By a suitable arrangement of switches, an ammeter with one terminal connected to earth can be momentarily connected to either main, and the insulation resistance in ohms of the main to which the instrument is not connected is readily arrived at by dividing the station pressure in volts by the number of ampères indicated by the ammeter. For example, if the station pressure be 120 volts and the ammeter reading 3 ampères, the resistance of the opposite main would be 40 ohms.

The ammeter used for testing must be accurate at low readings, and therefore is only made to read up to 5 ampères. In the event of any dead earth fault on the mains, the instrument might be destroyed by the sudden rush of current, and to prevent such an accident a magnetic cut-out is placed in series with the ammeter. When more than 5 ampères passes through the cut-out it breaks contact, leaving in circuit a resistance coil of 25 ohms, so that the current through the instrument, with a station pressure of 125 volts, cannot exceed 5 ampères. By this contrivance it is possible to measure faults of very low resistance.

The Hookham electric meter used by the company consists of a simple electromotor doing work against the retarding influence of Foucault currents generated in a copper disc rotating between the poles of a permanent magnet. The armature coils are built up on the copper brake disc, and the axis carries a worm which actuates a train of counting gear. The meter is calibrated so that the dial reads directly in Board of Trade units. The speed of rotation (subject to a small frictional error at small loads) is proportional to

the current.

Diagrams are exhibited which show the very variable demands made on central station plant at different hours of the day and at various seasons of the year. From these it will be seen that the number of

JANUARY 30, 1891.]

ELECTRICAL REVIEW.

Board of Trade units per lamp per month varies from 6.3 units in January to 1-2 units in June. Also, that in December the output reaches 4,060 ampères, whilst in June it does not exceed 500 ampères. This paper has been limited to the description of the central station plant and appliances of the Liverpool Electric Supply Company, and to the method of working simply from an engineering point of view, and no attempt has been made to deal with commercial details either of capital expenditure or working costs, which may be more usefully brought before this society, after longer experience, at a future date.

Manchester Association of Engineers, Saturday,
January 24th, 1891.

DYNAMOS: THE CONSIDERATION OF THE CHIEF
FEATURES WHICH REGULATE THEIR APPLICATION.
By A. B. BLACKBURN, Assoc. M. Inst. C.E., Assistant Manager, Messrs.
Mather & Platt's Electrical Department.

THE Continuous current dynamo, which can alone engage attention tonight, consists in its most usual form, of a fixed magnet and a rotating armature, separated by a narrow air space. Under the influence of the magnet, the air space between the polar surfaces and the armature core experiences a change of condition, which constitutes what is known as a magnetic field, and the whole activity of the machine is centred in the armature conductors as they revolve in this field. A difference of electrical pressure is constantly maintained along their length, and when their ends are connected through the brushes to the external circuit, so as to form a closed system, a current of electricity is established. The pressure is proportional to the strength of field and to the rate at which the conductors cut it.

If iron filings be sprinkled on a surface placed in front of a magnet, they arrange themselves in lines, having definite directions and more or less densely distributed, in different parts of the field, fig. 1,

141

stress, but in the conductor, as its name implies, the field breaks down, and the power, which has to be continually supplied to build it up again as the conductor moves into a fresh position through the field, is conducted away, to reappear as current energy in the circuit, where it can be either reconverted into mechanical energy by acting on another magnetic field, stored up as latent chemical energy, or dissipated in light and heat.

The effect may be popularly likened to that of a pump, employed to circulate water in a closed system of pipes, and it is no more necessary to suppose that the dynamo creates electricity than that the pump creates water. The electricity may be regarded as lying dormant in the system, until set in motion by the pressure generated in the revolving armature, when it makes its presence felt by giving rise to what we call an electric current. In the pump, work is done in lifting the water against an opposing head and in overcoming friction, and the measure of the work done in a given time, is the product of the head of water into the quantity. In the dynamo, work has to be done in order to drive current through the circuit against the opposing pressure or resistance, and the rate at which it is done, is measured by the product of the pressure into the current. The unit employed for reckoning the pressure is called the volt, and for the current the ampère, and for their product the watt, of which 746 are equal to one horse-power. The other quantity with which we are concerned is the resistance, which is the name employed to express the opposition to the passage of the current, offered in greater or less degree by all substances; its unit is called the ohm. The product of resistance into current is equal to the pressure, so that loss of pressure, due to resistance between any two points in the circuit, is proportional to the current flowing:

[blocks in formation]
[graphic]
[graphic][subsumed][subsumed][subsumed][subsumed][subsumed][subsumed][merged small]

FIG. 1.-MAGNETIC FIELD.

whence Faraday derived the conception of defining its strength as proportional to the lines contained in a given area. In a dynamo the field is produced by means of an electro-magnet, the construction of which will be familiar to all. The current round the magnet is said to induce the lines of force. or to give rise to an induction, and these lines in turn induce a current in the moving conductors.

The effect which a magnet has on a compass needle, and the powerful way in which it attracts iron from a distance, are well marked evidences that the induction can be transmitted across space, and it is impossible to resist the conclusion, that this can only occur through the agency of a strain of some sort, produced in the intervening medium or dielectric, following the course of a closed circuit running through the iron and crossing from pole to pole, as delineated by the lines of force. Now a field identical in its properties with this magnetic field, surrounds every conductor along which an electric current is flowing, the lines of force being circular, with the conductor threading through them. Such a field is essential for the propagation of current, and whatever affects the strength of one, also affects the other, consequently by interlinking a magnetic with a conducting circuit, as is done in a dynamo, and giving them a relative movement so as to vary the number of lines enclosed, a displacement takes place which is in the nature of a flux. The field owes its properties to the dielectric being able to withstand the application of the

over 30,000 lines per square centimetre, but as shown in the curve, fig. 2, where the relation between magnetising force and induction is plotted, there is a falling off in the permeability as the magnetisation increases. In iron it is probable that every molecule is already a magnet, but as far as external action is concerned, the resultant force is nil, which Professor Ewing has recently shown there is reason to believe, is because the molecules under their mutual attraction form groups, in each of which the magnetic circuit is closed on itself. When, however, the iron is subjected to a magnetising force, such as is produced by passing an electric current round it, these groups are broken up, and the molecules tend to set with their axes parallel, so as to form a continuous magnetic circuit through the whole length, with poles at the bounding surfaces of the extremities. The rotation of the molecules is not, however, proportionate to the strength of the current, and to make them assume their ultimate positions, with poles at right angles to its direction, would require an enormously strong current, hence the falling off in permeability

If the path traversed by the lines of force for the whole of their length was through air, as would be the case with coils enclosing no iron, their number would be enormously reduced, and they would be scattered. The coils necessarily take up a good deal of room, so the length of the magnetic circuit has to be considerable, but with iron present, the force is handed on with nearly as full intensity as if all the coils were packed into a thin belt close to the armature, and the length of the circuit proportionately reduced; so that in the dynamo the part performed by the iron is one of concentration.

All armatures are built on one or other of two fundamental types,

REVIEW

the drum, fig. 3, in which the conductors are all wound externally, and the ring, fig. 4, in which they envelope both inside and outside of the iron core. As the connections for each are analogous, it will suffice in tracing the action to select the ring winding as an example. A core is built up of thin iron rings, electrically separated from each other by means of such material as paper, this is done to prevent heating due to currents which if it were composed of a solid mass, would circulate through the length. By means of a non-magnetic spider frame, this core is secured to the rotating shaft. On it are wound the insulated conductors, in the form of a closed spiral, the

[merged small][graphic]
[graphic]

FIG. 3-DRUM ARMATURE WINDING.

external part of every turn of which cuts the magnetic circuit twice in a revolution, the impluses of each turn being added. The ends of each section of the spiral are connected to a pair of commutator bars, on which the brushes rest, and by which communication with the external circuit is established.

The commutator is built of bars of copper insulated from each other, and the brushes are set to bear on diametrically opposite sides; each bar in turn is brought into action twice in a revolution, and conveys the current alternately in a positive and then in a negative direction, as it passes under the brushes.

The direction of the current in the conductors is shown on fig. 5, by means of arrows, and the commutator bars, which for sake of clearness are reduced in number, are shown with their connections, the sections being defined by the portions contained between contiguous bars. It will be seen that the currents in each half converge on the brushes, and as each section belongs first to the right and then to the left, two impluses are given to it in each complete revolution. As the armature revolves, the sections now belonging to the side marked south, pass to the side marked north, and below in

[merged small][merged small][graphic][subsumed][subsumed]
[graphic][merged small]

the reverse order. The lines of force, which cross between the poles and the armature, are cut by the conductors on the right from below, and on the left from above, hence on either side the currents will flow in opposite directions round the spiral, the two streams entering the brush in the same direction.

These currents surrounding the iron core necessarily magnetise it, and the polarity which they impart differs in direction to that of the field magnets. The two sets of lines of force are superimposed, with the result that a fresh distribution of field is produced, fig. 6, and the line dividing the core into two halves, across the diameter at which the lines of force enter from one side and emerge from the other, which is called the neutral line, is now oblique to the plane of symmetry. Changes in the strength of the current cause the position of the neutral line to change. The brushes to work without sparking have to be set a little in advance of it, and when its position changes, the brushes have to follow suit. This occasions the need for attention in working the machine, as there is always excessive wear of the commutator when the brushes are improperly set, and an accurate comprehension of what takes place is of material assistance in the management. In fig. 5 the section shown is just on the boundary between the lines of force entering the core on one side and emerging on the other, where the field is nil, so that comparatively little difference of pressure exists between contiguous bars, and the section has just been closed on itself by the two commutator bars being bridged by the brush. Up to this point the section has been carrying the same amount of current as all the other sections on the same side which are in series with it, but now being momentarily shortcircuited, the strength and direction of current will depend on

FIG. 6.-ARMATURE REACTION.

self-induction and the resistance of the section. The amount involved varies as the square of the current strength, therefore with heavy load a stronger reversing field is required. It, however, unfortunately happens that the obliquity of the field due to the armature reaction, increases with current in a direction which leaves the brush in a weaker field, whereas a stronger one is wanted, hence when load is added, it is always necessary to advance the brushes in the direction of rotation, into a stronger field. If they are not placed far enough

[graphic][subsumed][subsumed][subsumed][subsumed][merged small][merged small]
« ПредишнаНапред »