A

an ordinary slide wire or metre bridge, as indicated in the accompanying Fig. 78, which is a diagram of the connections in such case, the lettering being the same

the conditions depicted above, there is either a mistake in the connections, which should be inspected in order to discover and correct it, or the fault may be very near the extremity D, in which case the ends of the loops should be reversed and the operations repeated. Murray's test can also be conducted with the aid of

as in the preceding figure. The slider S is moved along the wire until a point is reached at which a balance is

obtained on the galvanometer G; the same formula then applies, the values of A and C being reckoned in degrees on the attached metre scale, and, if the slide wire be homogeneous, as it should be, very satisfactory results are obtainable by this method.

A rough indication of the locality of a fault can be

FIG. 78.

obtained by this method in the following manner :-A short length (about one yard) of any suitable wire, such as No. 20 gauge, is sweated across the terminals of the loop D H, and treated as the slide wire in Fig. 78, the battery connection being slid along the wire until a balance is obtained. The respective values corresponding to A and C in the preceding test with the bonâ fide metre bridge are then obtained by direct measurement, and inserted in the formula, as before.

I must here digress for a moment to consider the subject of contact and other influences upon this and similar loop tests. To begin with, the extremities of the loop should be directly connected to the bridge terminals if possible, but, if leads be interposed, their equivalent lengths must be determined and allowed for in calculating results.

Thus, if the cable forming the loop be 7-16, and say one yard of No. 16 gauge wire be interposed between the extremity D and the bridge terminal, that connecting lead will be equivalent to 7 yards of the actual loop, and such an allowance must be made in working out the results.

As. regards the connections, if cables of very large sectional area are involved, it is better to employ mercury cups to connect them with the apparatus, the ends being previously well cleaned and amalgamated.

If the ordinary bridge be used in Murray's test, it is very probable that the resistance in A will be greater than the resistance of the loop connected to it, and, in this case, the galvanometer lead must be attached to the extremity of the loop itself on the cable side of the bridge connection, so as to include any possible contact resistance in the bridge circuit. If, on the other hand, the slide wire bridge be employed in the test, the reverse will probably be the case, and the galvanometer may be connected direct to the terminals of the slide wire for a similar reason. In pursuance of the above precaution it is better in using the P.O. bridge to connect the gal vanometer directly across the loop at the points D and H, and employ a separate key in circuit with it rather than run the risk of contact error by using the righthand key of the bridge, as indicated in Fig. 77.

Varley's Loop Test is somewhat similar to Murray's,

but necessitates the use of a Wheatstone bridge, the ordi nary metre bridge being unsuitable. The connections are indicated in Fig. 79, and are similar to those for the

B

www

FIG. 79.

measurement of resistance by the bridge method, with the single exception of the battery connections, the posi tive pole of the latter being connected to earth instead of to the bridge, as usual. The lettering in the figure is the same as in the preceding case.

The proportional values usually unplugged in A and B for this test are 1,000 ohms and 10 ohms respectively; then, the battery key being depressed, C is adjusted until a balance is obtained on the galvanometer G. This value for C having been obtained and noted, the positive pole of the battery is now disconnected from earth and connected instead to the same terminal as the extremity H of the loop, thus completing the ordinary connections for resistance measurement by the bridge method, and a fresh balance is obtained, giving a value R as the conductor resistance of the loop D H. Then, taking L as the length of the loop, the distance of the fault F from the extremity H can be obtained from the formula (A R B C) x = L R (A + B)

To ensure accuracy, the ends of the loop may then be

G

reversed, and the connections being again made as represented in Fig. 79, a fresh balance is obtained, and the distance from the opposite extremity of the loop computed therefrom. The sum of these two distances should, of course, be equal to L, the total length of the loop; if, however, their values vary slightly, the mean of the two may be taken as correct; if a large variation results, the connections and contacts should be examined for possible errors.

Of course, if leads be used between the bridge and the extremities of the loop, their equivalent lengths must be determined and allowed for in calculating results, as mentioned in connection with the preceding test.

Reverting for a moment to Murray's test, better results are frequently obtained by reversing the positions of battery and galvanometer, but this must not be carried into effect if there be any possibility of a polarisation E.M.F. set up at the fault itself, for, in such case, on closing the galvanometer key alone, a deflection would ensue owing to this E.M.F., and the ultimate adjustments would be made to a variable false zero, and, in consequence, no satisfactory balance would be obtained on the galvanometer. In cases where the absence of such an E.M.F. is assured, however, it is often possible to obtain more satisfactory results by reversing the posi tions of battery and galvanometer in this test, as, in such case, the current has not to overcome the resistance due to the fault itself in its passage, but is provided with a free path of comparatively low resistance, the fault resistance being inserted in the galvanometer circuit instead.

Varley's method is unsuitable in cases where the loop is of comparatively low resistance owing to the difficulty of measuring the latter on the ordinary Wheatstone bridge. It is, however, often used when suitable, owing to the similarity of the connections to those for resistance measurement, which will often be arranged as permanencies where a considerable amount of conductor resistance measurement has to be dealt with; Varley's test involves a very slight modification of these connections, and is often adopted for this reason.

It sometimes happens that a conductor is broken without actually severing or damaging the surrounding insu