Page images


to the foregoing, the adjustment and consequent galvanometer balance being, however, effected by moving the contact slider c along the wire in one or the other direction, instead of, as in the previous test, manipulating plugs.

The formula for calculating the result is the same as in the case of the Wheatstone bridge proper, viz.,

a. f, the resistances e and f being represented by the number of divisions on the slide wire scale contained within them, which lengths, as the wire is homogeneous throughout, are proportional to the respective resistances of these sections.

When the resistances which we have to measure are of such magnitude that they cannot be included within the range of the average Wheatstone bridge, they develop into what are technically known as dielectric or insulation resistances, and require to be treated accordingly.

The dielectric, or, as we shall term it, insulation resistance of an object, such, for instance, as a length of wire or cable, is the electrical resistance of the media separating that object from the earth or any contiguous conductor, and it is usually calculated in megohms, the megohm, as already stated, being equivalent to one million ohms.

A totally different system of measurement is necessary in cases where such high resistances are dealt with, and the method most commonly adopted is that of comparison, a method which consists in electrically comparing the resistance to be measured with a standard resistance having a high value.

Such standard resistances are frequently as high as one megohm, and can be constructed in several ways. One type, which is illustrated below, consists of a number of bobbins wound with very fine wire to a resistance of some 100,000 ohms apiece, and connected up in series to the required total of one million ohms.

Another and cheaper form of standard is constructed by Messrs. Johnson and Phillips, of Charlton, Kent, as follows. An oblong strip of plate glass, some 8 ins. long by 4 ins. wide, A, Fig. 38, has two holes drilled through it, at the respective centres a and b, of the two ends. On the ground surface of the glass is drawn with a very hard 6 H plumbago pencil, a zig-zag line some f in. in width, connecting the two holes. This is well rubbed

[graphic][subsumed][ocr errors][ocr errors][ocr errors][subsumed][subsumed][ocr errors]

Standard Resistance Box of one megohm, subdivided into sections

of 100,000 ohms apiece, by Messrs. Nalder Bros. and Co.

into the surface of the glass, and expanded round the holes a and b as shown, to form a sort of lug. Terminal bolts and nuts are then passed through the holes, and, tinfoil washers being interposed between them and the glass, are screwed up very tightly, in order to bring the tinfoil into intimate electrical contact with the particles of plumbago forming the lugs. The combination thus obtained is tested for resistance in the ordinary way,

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

and, if too high, is subjected to more pencilling until the valve is brought approximately to the required dimensions. The whole of the surface carrying the plumbago is then carefully varnished with a thin shellac varnish, and baked in front of a gas flame until fairly hard. Several subsequent coats of varnish are applied and treated in like manner; the plate is then allowed to cool to its normal temperature, and the final test taken to ascertain its finished value. It is then mounted in an ornamental wood base with an ebonite top and insulated terminals, and the final value as a resistance is stamped upon the ebonite in the usual manner. There is, of course, a certain amount of luck attending the manufacture of these standards, which determines the ultimate value obtained, but these are usually sufficiently approximate for all practical purposes, and, if the pencilling and subsequent baking have been carefully carried out, these standards will retain their value for many years without attention, although, of course, they require careful handling.

Having so far described the additional apparatus re quired, we will proceed with

(5) Insulation Resistance Measurement.

The connections for this test are indicated in Fig. 39, where G represents a high resistance Thomson astatic galvanometer of the type already described, with the adjustable 1-9th, 1-99th, and 1-999th shunt s across its terminals (the shunt s may conveniently be arranged on the “Universal" principle). K is a short-circuit key, and Kl a reversing key, which, however, in practice is normally replaced by the Rymer Jones reversing switch, but the key is represented here for the sake of clear

E is a battery which should be of sufficient power to yield an E.M.F. at least double that to which the object under test will be subjected under normal conditions of everyday work. R is the standard megohm or other high resistance, whilst C is the object under test, which is in this case a length of cable.

The method of conducting the test is as follows :—The instruments are first connected as shown in the figure, the cable C being omitted as indicated by the dotted connection. One side of Kl is depressed, thus connecting one pole of the battery through the galvanometer G and the standard resistance R to earth, and the remaining pole direct to earth. Whilst keeping Ki depressed by means of its cam, the short-circuit key K of the galvanometer is gently opened, and the shunt s


manipulated in conjunction with it, in order to obtain the highest readable deflection upon the scale, which deflection we will call d.

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

Having carefully noted d and its attendant shunt, the key K is closed again, and K1 released. The standard resistance R is then disconnected, and replaced by one end of the cable under test, the opposite extremity being left free. The key Ki is now again depressed, and the time noted. K is then carefully manipulated as before, until a readable deflection dl is obtained, which, in the case of cables, should be noted at the end of a minute from the time of depressing K1 in order to allow a suitable interval for what is known as electrification," a term which will be described presently. In most cases, where a careful test is desired, successive readings of dl are obtained at intervals of half a minute for the space of some four or five minutes, or longer. During this period, the deflection should steadily decrease, quickly at first, and then more slowly, until its fall becomes almost imperceptible.

This property of electrification is one which the cable possesses of, as it were, absorbing a small portion of the testing current into itself, but the exact action which takes place is not thoroughly understood. This absorption is very rapid when the current is first applied, but


becomes slower as the cable approaches saturation. It also varies considerably with the temperature and material composing the insulating medium, being more marked at high than low temperatures. In the case of gutta-percha insulation, the electrification amounts to from 2 to 5 per cent. of the total deflection between the first and second minutes following the application of the current, whilst, with india-rubber insulation, on the other hand, the electrification may reach as high a value as 50 per cent. between the first and fifth minutes.

In order to ensure a steady electrification, as also to guard against surface leakage, the ends of the cable under test require to be carefully prepared before testing. To this end, the braid and tape, with which cables and wires are usually provided as an external finish and protection, are laid back for some two or three inches at either extremity of the cable, and the rubber or gutta-percha thus exposed is cleaned and scraped so as to expose a fresh dry surface to the air ; in the case of rubber-covered cables this clean surface may be further ensured by paring the rubber down after the manner of sharpening a cedar pencil, with a keen knife.

The ends thus prepared are immersed for a few seconds in hot molten paraffin wax, which has been previously purified. This treatment serves to provide the extre mities of the cable with a moisture-resisting film, which prevents surface leakage, and tends to secure the true results desired in the test.

The method of working out the results from the readings obtained is as follows:-First obtain the true value of the observed deflections by multiplying them by the fraction

which constitutes what is known as the

G + s


"multiplying power” of the shunt, G and s being the respective resistances of the galvanometer and shunt in ohms. It amounts, in short, to multiplying the deflections obtained with the 1-9th shunt by 10, with the 1-99th by 100, and with the 1-999th by 1,000 respectively, for, let G equal 9,000 ohms, then if s be the 1-9th shunt, it will be equivalent to 1,000 ohms, and

9,000 + 1,000 10,000

1,000 1,000

G + 8

= 10.


« PreviousContinue »