418 On a Method of finding Faults inJnsulated Wires. 



practice be put right in an hour or two, involving no more waste 

 than the insulator, which can be held between the fingers, and 

 without even cutting the conductor. The position of the fault 

 can be reduced into a length represented by C D by cleaning 

 and drying the surfaces on each side of it. 



It is obvious if C D represent a joint, it may be tested by this 

 method with great ease. In fact so delicate is this test when 

 used for joints in this way, that if A and B each be 2 or 3 

 miles in length, and C D, a length of 10 inches, be slightly heated 

 above the portions on A and B, the difference in insulation will 

 be readily perceptible on the electrometer when this portion is 

 connected to earth. The limits to the delicacy of the test are 

 only reached when A and B are absolutely insulated. 



A very interesting lecture-experiment may be formed from 

 these circumstances. 



Let AF represent a length of insulated wire (a series of insu- 

 lated Leyden jars would answer as well). Clean the outer surface 

 at B,C,D, and E ; charge the core on an electrometer, connecting 

 the portions A B, B C, C D, D E, and E F to earth. Remove 



a j i 1 | f 



B C D E 



the earth-wires and finally the battery wire. We can if we wish 

 remove the whole of the charge from any of these sections with- 

 out affecting the charge in any of the other sections. 



Suppose we wish to remove the charge from the section D E ; 

 we first connect the earth-wire to the surface between D and E, 

 and touch A or the electrometer if still left on, when the needle 

 will instantly fall to zero ; we remove the earth- wire and after- 

 wards test the other sections, when we shall find that their 

 charges have been unaltered ; and if the spaces have been well 

 cleaned and dried, we shall find that, although the conductor is 

 continuous, no charge will flow into D E ; the charge in E F can 

 be noted on the electrometer ; and although a section D E be- 

 tween it and the electrometer be empty, it will not communicate 

 a particle of its charge to it. 



In the second case, where the bulk would prevent the whole 

 from being insulated, we should continue to coil the core upon 

 an insulated drum until the fault disappeared — that is, until it 

 was coiled on the drum. This is a useful method when dealing 

 with " served core " at a cable factory. 



When a galvanometer is used, the corrections are as for ordi- 

 nary insulation-tests. The galvanometer is short-circuited ; and 

 A and B, whilst connected to earth, are charged as before. The 

 earth-wires are removed and the short-circuit key opened. The 

 drums are after a few moments connected to earth alternately, 



