NOVEMBER 15, 1889.]

THE ZIPERNOWSKY ELECTRIC TRAMWAY WITH A VERTICAL RAIL.

[FROM A CORRESPONDENT.]

THE constantly increasing intercourse in the larger towns naturally makes the want of suitable means of transit easily available to the public to be more distinctly felt. The most suitable means for meeting this demand are at present indisputably the tramways, since they render it possible to effect transit of the masses most conveniently and most suitably. Certain hindrances, however, have hitherto stood in the way of their universal introduction and development, as we will now proceed to show.

The normal rail, almost universally employed for tramway purposes, makes their introduction often a difficult task, and, in many cases, prevents them from adapting themselves to existing requirements. Further, the laying of a line with normal rails on the level of the existing streets is not only a relatively expensive work, but it involves taking up and relaying the

547

tramways with vertical rails. The principle of the line is that the wheels upon which the weight of the car rests run on a double slit-rail, beneath which there is a channel of masonry; into this channel there extend strong arms in fixed connection with the car and resting by means of guide-rollers upon rails laid down on both sides of the channel, and thus giving the car the necessary stability.

The execution of this principle is shown in the figures 1, 2, and 3. The body of the car, K, with its framework, R, rests by means of the cross-bearer, T, upon the running wheels, L, L', which are fixed obliquely, converging downwards in order to transfer the horizontal pressure (arising from unsymmetrical loading, and from oscillations of the car) to the double rail, s, s (running rail).

From the frame of the car the strong arms, A, extend through the slit, Z, down in to the channel, N. These arms, which are firmly fixed to the frame, support at their lower end guide wheels, or rollers, F, which rest against the side rails, E, E', and thus keep the car up

pavement for the average width of three metres along the whole length of the line, and increases the cost of working as it necessitates the repairs of the pavement for so wide a track.

As a further defect of the tram lines with a normal horizontal track we must note on the one hand the want of elasticity in the movement, and, on the other, the relatively too great demand for space, which renders it impossible to carry such tramways along lines of road into which it is necessary to use curves of a small radius or which are too narrow to admit of communication with tramcars of ordinary dimensions over and above the usual traffic. In consequence of these defects the tram lines are compelled to leave many important channels of traffic untouched and to traverse only the wider streets.

To meet some of these defects narrower tramways have been proposed, which, however, introduce new difficulties. Thus the stability of the narrower cars is reduced, which makes itself unpleasantly felt where there is a depression on one side of the track, a state of things which frequently occurs.

These disadvantages seem to be almost entirely obviated or reduced to a minimum by the system of

right. If, e.g., the load on the right side of the car is greater the wheels on the left side will press against the left rail, and conversely.

Both the main rails, S, S, and the guide rails, E, E', rest on sleepers, B, which are laid in the earth at suitable intervals. Between these sleepers the channel is made in a suitable manner, for instance, of bricks, coated with cement.

The line thus described differs therefore in its outward construction very decidedly from the common lines with horizontal rails, since in the former there is only one line of rails on the level of the street, whilst the second rail lies beneath in the channel above described. The plane connecting these two rails is vertical, whence this system is called a tram line with a perpendicular track.

According as the line is arranged for working by means of cables, chains, or electric power, the channel serves to receive the cable, &c., or the electric leads. At suitable intervals there are placed refuse collectors which open into the sewer. The cars, without any alteration in their make or in the arrangement of the line, can be propelled at will by horse-power or by other motors.

E

R

R

N'

M

M

rail being utilised to introduce the contact arrangement into the channel.

The regulation of the speed of travelling is effected by inserting variable resistances; stoppages are produced either by means of a friction brake, or by shortcircuiting the armature of the motor by a suitable resistance or by reversing the field magnet current of the electro-motor, in consequence of which it seeks to

any mechanism which, further, would be rapidly ren dered useless by mud and dust.

In the line here described, the execution of automatic points is attended with much less difficulty, as all the movable parts are in a protected position beneath the level of the street.

The principle of these points is characterised by the simultaneous application of an upper tongue for the

NOVEMBER 15, 1889.)

running rail and a lower tongue for the guide rails. The upper tongue, on account of the pressure of the cars or the incidental pressure of cars crossing the points diagonally, rests upon bolts which are thrust forward automatically when the points are turned. The lower tongue is carried on rollers or gliding surfaces.

A special way of carrying out this principle with selfacting displacement is shown in figs. 4 to 9. Fig. 4 is the ground plan, fig. 5 a longitudinal section, and figs. 6, 7, 8 and 9 transverse sections in different parts.

549

tongues to the right into the position shown by dots, when the car leaves the points in the direction of the arrow, 1.

When the car has passed the points, the weight, Q (fig. 9), or a suitable spring pushes both tongues to the left into its original position, so that a car coming from 2 or 3 can proceed straight on.

Whilst the tongue, 0, applies itself laterally to the running rails, S, S, the flat tongue, U, pushes itself under the guide rails, E, E. In most cases the proportion between the radius of the curve and the distance of the guide rails from each other is such that the lower tongue, U, could not be made pointed, or, at least, they would be much too thin at the point. The rollers, F, are broad enough to be supported by the lower tongue.

Fig. 2 shows the tongue, U, as running on the rollers, B; the pointed tongue, o, lies on the bolts, R, R1, which are pushed from the right and the left under the tongue, 0, by the rods, T, T,, and the levers, H, H, (Fig. 8), according to the movement of the tongue.

In the construction shown in the figures, the points are set automatically by the guide rollers, F, or any other part of the car projecting into the channel, N. This is effected by means of the setting lever, D. the traction rod, G, and the lever, K, which push both

FIG. 9.

A peculiar advantage of this construction of points is that they can be set gradually and without any shock. This is effected both by the form and the length of the lever, D, and by the circumstance that the acting point of the force can traverse a distance almost equal to the entire width of the channel.

We will now sum up the chief advantages of this construction of the cars and of the line, which are as follows:

1. The car does not run upon two rails separated from each other by the normal breath of the tramway but upon a single twofold line, whence the surface of the road is interfered with to a slight extent only.

2. Hence the pavement is spared to a great degree, since in repairing the line or in altering its position the road has not to be taken up for more than a metre in width.

3. If electric motors are used a better and safer insulation than has hitherto been practicable is ensured in consequence of the greater depth of the channel.

4. As the upper bearing rails with the lower guide rails and sleepers form a pipe-like conduction for the whole length of the track, the upper part of the structure attains a degree of stability and elasticity with which no other existing construction can compete.

5. No subsidence of the line on either side is possible, and hence one of the chief causes of unsteady driving is removed.

6. As the stability of this construction does not depend on the breadth of the car, it becomes possible to use narrower cars, and hence this kind of tramway may be used in the narrowest streets without hindrance to the ordinary traffic.

7. The simple application of self-acting points. From the above descriptions and figures it is easily seen that an electric tramway constructed on this principle possesses all the attributes needful for creating a means of traffic adequate to the requirements of large towns in a degree not attained by any previous system, and culculated to a great degree to displace the omnibus traffic.

The New Electric Light Works at Wolverhampton. -The erection of the new works of the Electric Construction Corporation, Limited, has commenced. They will be of a most extensive character, and complete in every detail.

THE BLAVIER METHOD FOR LOCALISING FAULTS IN SUBMARINE CABLES.

IT is well-known that the Blavier method of testing suffers from the disadvantage of exposing the fault to a higher potential when the distant end of the cable is insulated than when it is earthed. The consequence is that the current through the fault to earth is of different strength in the two cases, and that the resistance of the fault which is a function of the current must vary between the two measurements. This difficulty can partly be overcome by using a smaller electromotive force when the cable is freed than when it is put to earth, and the ratio between the two battery powers can be found approximately by calculation. But there are other sources of disturbance produced by the earthing of the cable, such as the shunt circuit formed by the length beyond the fault, and as these errors cannot be compensated for, it is only in some few cases that a localisation by Blavier's formula can be relied upon. Refuge is then, as a rule, taken to the so-called "overlap methods" of testing from both ends of the cable, viz., localisation by the two measurements with the distant ends insulated, or by the two measurements with the distant ends earthed. Both these, but especially the latter, have been developed into tests of great practical merit and accuracy by Messrs. A. E. Kennelly and J. Anderson.

There are, however, cases when, if possible, it is desirable to localise a fault by measurements from one end only, and to be enabled to do this with as much accuracy as possible, it occurred to Mr. Jordan, one of the acting electricians of the Great Northern Telegraph Company in the Far East, that the following modification of the ordinary Blavier method might be made, in order to remove some of its disadvantages as mentioned above.

The diagram shows Mr. Jordan's arrangement. Box I. is an ordinary Wheatstone bridge with resistance coils, and Box II. an ordinary resistance box from 1 to 10,000 ohms. G is a Thomson's reflecting galvanometer, A is a commutator, and to the same and the bridge is connected a cable, in which there is a fault, y, with the lengths, and z, on either side thereof.

The modus operandi is as follows :-First, the resistance is measured in the usual manner with the other end of the cable earthed and with no plug in A, and balance is obtained by unplugging a resistance, R., y z in Box I.; then R2 = x + y + z'

The moment it is observed that the cable is freed at the other end, a plug is inserted in A, R, is left unaltered, and the resistance in Box II. is adjusted until the spot Rz (x + y) of light again covers zero; then =R2, from Rz+x+y

[NOVEMBER 15, 1839.

From this it is manifest that when the cable is m sulated the current is shunted by the resistance, E, E the same way as z is a shunt when the cable is earthed and the result is that the fault in the two cases is acted upon by currents of more equal strength than in th ordinary Blavier test; and further, that it is possite to take the tests so quickly and nearly instantaneously after each other, by having adjusted the resistance R2 and R3, in accordance with the two conditions, the the resistance of the fault may be assumed to reman approximately constant during both measurements.

It should be mentioned that Mr. Jordan was wel acquainted with Mr. H. Kingsford's method (Journal the Society of Telegraph-Engineers, December 1 1885), when he devised the above arrangement, and i may add that Mr. Jordan has tried his test practical on several occasions, and invariably proved it to b superior and to give better results than the old method P. CHR. DRESING.

Copehagen, November 8th, 1889.

SIR WILLIAM THOMSON'S CHAIN OF ELECTROSTATIC VOLTMETERS.

By ANDREW W. MEIKLE, M.A.

(Continued from page 526.)

The sensibility required for the different ranges of the multicellular voltmeter is obtained by varying the number of cells and vanes, the cross section of the suspending wire, and also the length of wire used At first a little difficulty was experienced in obtaining constancy in the wire suspension, and a little trouble was consequently given by having to reset the zero froz time to time. This difficulty was caused most probably by the straining of the wire in the process of drawing. so that some little time elapsed before it came to a normal condition. A very carefully drawn iridioplatinum wire is now used, and is subjected to a preliminary tempering by heating, longitudinal tension, and torsion. The results with this suspension have been most satisfactory, and our standard instruments which are constantly in use now give no trouble from changes of zero.

Standardising of the Multicellular Voltmeters.

For low potentials the instruments are graduated by joining their terminals to those of a standard antiinductive resistance through which a current measured by one of Sir William Thomson's standard Centiampere Balances is passed. The arrangement is shown diagramatically in fig. 4. A rheostat of platinoid wire is joined up in the circuit in order to vary the cur rent to any desired value. The standard resistance

used is wound with platinoid wire, and its temperature variation is very nearly 4th per cent. per degree Centigrade so that any uncertainty in the estimation of the temperature may be neglected. The Centi-ampere Balance is the standard used in the Physical Laboratory of Glasgow University, and its accuracy is certainly withinth per cent. These particulars are given simply to show that uncertainties in the standardising apparatus have been as far as possible removed. The multicellular is joined up to the terminals, T, T ́, of the standard resistance, and shows the difference of potential

NOVEMBER 15, 1889.]

=

between them, which at any one time is equal to C × R, where C current in ampères through the resistance as shown on the balance, and R = value in ohms of the standard resistance.

The number of ohms in the standard resistance is varied from time to time, so that the deflection on the balance may be sufficiently large to enable any possible error in the reading of the position of the weight to be so extremely small as to be neglected. By this method and using the battery power at our disposal, it is possible to push the graduation of the instruments up to 140 volts with great accuracy.

For ranges above this point a Volta pile is used to augment the potential, as shown in fig. 5. This pile consists of zinc-copper couples fixed to a vulcanite frame in such a manner that when their extremities are dipped in water the liquid is drawn up by capillary action and a chain of little cells is formed, each of which gives about 0-9 volt. These Volta piles are made up in sets of 200 couples, and the stand containing each set measures about 50 by 10 centimetres. We have found them give excellent results, and can, by those we have, obtain over 1,000 volts.

The zinc end of the pile is joined to the positive terminal, T, of the standard resistance, and a voltmeter, A, formerly graduated by the method explained above, is joined up to the same terminal and to a point in the pile such that it gives a convenient reading on a sensitive part of its scale-say, for example, 100 volts. Another voltmeter, B, also graduated to 140 volts, and whose graduation it is wished to extend, is then joined to the other terminal, T', of the standard resistance and the same point of the pile as A. The Multicellular, B, then measures the difference of potential between the terminals of the standard resistance, plus the difference of potential due to the portion of the pile measured by the Multicellular, A. By this means the graduation of Multicellular, B, may be pushed up to 240 volts. By substituting Multicellular B for A, and using it to measure a higher number of volts, a third instrument can be graduated up to 380 volts, and so on. By this

V.P

551

These voltmeters have now been in use in the Physical Laboratory of Glasgow University for some time, and have proved a great convenience as, for many purposes, they save a large expenditure of time which would be taken up in making temperature corrections, calculations, &c., were other voltmeters used. We have been able, by using them, to determine by direct experiment, how far other voltmeters could be relied upon to give accurate measurements on alternating circuit under varying conditions as to period of alternation, temperature, &c. The instruments in conjunction with the Volta pile have also proved very useful in standardising the electrometers used in the investigation of atmospheric electricity by Messrs. Maclean and Goto, on which a preliminary Note was communicated by Sir William Thomson to the British Association.

X

N

V. P, volta pile. FIG. 6.

means a chain of standards has been obtained giving readings on sensitive portions of their scales throughout the entire range, so that instruments can now be graduated by direct comparison, but two of a lower grade, X and Y, joined in series, as shown in fig. 6, are always used in the graduation of a higher range instrument, so that their indications may be amply sensitive as compared with the one being standardised. These standards, of course, undergo periodical checkings, and are constantly being compared one with another to make certain that they have not been injured in any way since their calibration.

Electrostatic balance. FIG. 8.

the glass tube being filled with paraffin to prevent the lodgement of moisture and give great resistance to disruptive discharge. A sheath formed by a short piece of glass tube pulls up over the terminal, T, and protects it from being touched by accident. The slate base plate