Page images

Clark's Fall of Potential Method is illustrated in Fig. 74, where B D is a length of cable having a fault at the point C. A B is a length of good cable of resistance

[blocks in formation]

R, joined in series with it. E is a battery of constant E.M.F. connected between the point A and earth, and G is a galvanometer of the d’Arsonval type in series with a resistance r, such that their combined resistance is appreciably greater than that of the line under test. G is first connected between the points A and B, D being insulated, and the deflection due to the fall of potential from A to B noted; we will call it d. G is then disconnected, and connected instead as shown in the figure between the point B and earth, and the resulting deflection dl is also noted. The galvanometer G is then removed, and set up under exactly the same conditions as to scale distance, etc., at the point D, and connected between the point D and earth, the third deflection d2 being noted, then the resistance between the point B

R dl d2 and the fault, x =

d It is quite possible, with a D'Arsonval galvanometer, to shift it in this manner without altering its constant, but, if such an instrument be not available, or if the extremities B and D are so far apart that independent tests with various apparatus have to be taken, involving the use of two distinct galvanometers, then the test is rendered somewhat more complicated, as the instruments must afterwards be connected in series with a high resistance and standard cell, and their “constants” noted in terms of the E.M.F. of the standard cell, thus providing a means of comparing the respective deflections d, di, and d2, on the same basis, and reducing them all to the values they would have attained had they all been produced upon the same instrument, as described above.

Siemens' Equal Potential Method is somewhat similar to the foregoing, and is illustrated in Fig. 75, where A B is a line having a fault at C. The mode of procedure is as follows :—The battery E, of constant E.M.F.. is connected as shown between the extremity A and earth, the point B being insulated. A galvanometer (preferably D'Arsonval) is then connected between A and earth, and the deflection d noted. The galvanometer is next connected between B and earth, and a second deflection di, due to the potential at B, and consequently at the point C, is obtained. The battery is then disconnected from A and connected instead between B and earth, the same pole being connected to the line as in the first instance; then, the galvanometer being con


Fig. 75.

nected between the point A and earth, the resistance r in the battery circuit is adjusted until the second deflection dl is reproduced. The galvanometer is then again connected between the point B and earth, and a third deflection d2 is obtained; then, taking L as the total length of the line A B, the distance x between A and

d - dl the fault at C, x = L

(d2 - dl) + (il - dl) Of course, if separate galvanometers, or a single instrument with a variable constant be employed in making this test, the same remarks as to the reduction of the values of d to a common comparative basis apply as in the preceding case.

Siemens' Equilibrium Method is illustrated in Fig. 76, and its principle consists in so arranging two independent electromotive forces, one at either end of the line, that the potential at the fault is zero, and no current in consequence will flow to earth at that point.

A B, Fig. 76, represents the line under test with a fault at the point C. E and El are two batteries with their opposite poles connected to the extremities of the line through the resistances r, rl, r2, and r3 respectively, of which rl and r2 are equal fixed values, and r, r3 adjustable. Two galvanometers giving the same constant," or one instrument with a non-variable constant, such as a D'Arsonval, which can be moved from one extremity to another of the line, are then connected across the resistances rl and r2 respectively or in turn, high resistances being placed in series with the instrument or instruments, in order, as before, to make the total resistance high as compared with that of the line. The

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

resistance r, or r3, is then varied until the same deflection is obtained in either case. The galvanometer is then disconnected from rl and r2, and connected instead between the junction of r and rl and earth, and the point A and earth, thus obtaining two distinct deflec

dl tions, d and dl respectively, then x = rl

d - di Of general fault localisation systems, however, those known as the “Loop" methods are by far the most satisfactory in that they are, within certain limits, independent of the resistance and E.M.F. set up at the fault itself. It is necessary, however, for the conduct of such tests that both extremities of the cable or line under test, or even two extremities of two separate cables or lines looped together at their far extremities, be available at the testing point for connection to the apparatus. Thus, if both ends of the faulty section of a certain cable be not available, it may be looped at its distant extremity with a return cable running parallel to it, or even with a totally independent cable, so long as two extremities of the complete loop are available for connection to the testing apparatus.

There are two distinct methods of loop testing, known respectively as Murray's and Varley's. We will deal in the first instance with

Murray's Loop Test.—The connections for this test with the ordinary P.O. pattern of Wheatstone bridge are shown in Fig. 77, where A, B, and C are the proportional and adjustable arms respectively of a P.O. bridge, D H. the loop of cable under test with a fault at the point F; E the testing battery of sufficient E.M.F. to send a current through the resistance of the fault, and G the galvanometer. The values in B are all plugged up so that the galvanometer connection can be made through the key, as shown, which key, under the usual conditions of usage, acts as the battery key. The modus operandi consists in unplugging a suitable value in A, which value is roughly determined by experiment, and then adjusting C until a balance is obtained on the galvanometer. Then the distance of the fault from the extremity D,


where L is equal to the total length A + C of the loop. The value thus obtained for x will be in terms of the same unit of length as L. If it be impossible to obtain a balance on the galvanometer G under

[ocr errors]

= L

« PreviousContinue »