518 



NA 7URE 



\April 2-1, l8 76 



Siemens and Dr. C. William Siemens, who early in the 

 history of submarine telegraphy communicated their re- 

 searches on the subject to the British Association at the 

 Oxford meeting of i860. The principle of these early 

 methods still remains the principle of the methods employed 

 by Sir W. Thomson in his testing of the Direct United 

 States Cable at Ballinskelligs Bay Station in September, 

 1875, and upon which he has reported to the manufacturers 

 of the cable, Messrs. Siemens Brothers. It is the purport 

 of this paper to describe these tests and the results 

 obtained. 



To those who may be unacquainted with the route of 

 che Direct United States cable, it will be necessary to 

 explain that the course taken is from BaUinskelligs Bay, 

 on the west coast of Ireland, to Torbay, in Nova Scotia, 

 whence it agam passes to Rye Beach, in New Hampshire, 

 America. 



The construction of the cable, which was decided upon 

 by the company acting under the advice of Dr. William 

 Siemens, their scientific consultant, is as follows : — The 

 cable from Ireland to Nova Scotia consists of a con- 

 ductor formed of a strand of twelve copper wires weighing 

 400 lbs. per nautical mile. This conductor is surrounded 

 with four coatings of gutta-percha and gutta-percha- com- 

 pound weighing 360 lbs. per nautical mile, so that the 

 total weight of the " core," as it is technically termed, is 

 760 lbs. per knot. It was specified that the core should 

 have an insulation resistance per nautical mile equal to 

 1 60 millions of mercury units ; tests, however, checked 

 and taken under the direction of Mr. von Chauvin, the 

 manager and electrician to the Company, show that no 

 length of core was passed that did not insulate to nearly 

 double this extent, or to 300 million units per knot, the tests 

 being taken after twenty-four hours' immersion of the 

 core in water at 75° F. The " core " is " served " or enve- 

 loped in jute yarn, and is then sheathed or covered with 

 iron wires of a diameter best suited to the position of the 

 cable. Thus for the deep-sea, 1,630 knots of the cable 

 are sheathed with ten strands of wire and hemp, each 

 strand consisting of a homogeneous iron wire surrounded 

 with five strands of Manilla hemp, each strand being 

 passed through a compound of pitch, tar, and india- 

 rubber. Each of the iron wires has an average breaking 

 strain of 53 tons per square inch, and is of 0*099 inch 

 diameter. The cable termed medium cable is sheathed 

 with fifteen wires of ©•148 inch diameter v.ith proper 

 sewings of yarn, while for the shore ends, where there is 

 considered to be more friction or wear, this medium 

 cable is again surrounded with iron sheathings of twelve 

 strands of iron wires, each strand consisting of three iron 

 wires of o"23o of an inch diameter. 



The cable from Nova Scotia to New Hampshire con- 

 sists of a strand conductor of seven copper wires weighing 

 107 lbs. per knot, covered with three coatings of gutta- 

 percha and compound weighing 150 lbs. per Icnot, and is 

 also sheathed with iron wires. 



The non-electrical reader who may choose to wade 

 through detail that must be somewhat technical will 

 perhaps find help in considering the conducting wire as 

 representing a line of flow or force, such that if two of 

 these lines be directed into a galvanometer or current- 

 measurer in opposite directions, that having the greatest 

 head or greatest force will preponderate, v/hile no indi- 

 cation will be found on the instrument when the forces 

 are equal ; also that from a known force giving through 

 a known resistance a certain instrumental measure, any 

 unknown resistance may be reduced when its instrumental 

 measure is ascertained. 



Testing the Resistance of the Copper Conductor. — 

 Electrical measurements upon a long submerged cable 

 differ from measurements made in the laboratory as 

 described in text-books in one very important particular — 

 that of earth-currents. Earth-currents are the bete-noir 

 of the electrician, who not infrequently finds them so far 



masters of the field that his chance of obtaining accurate 

 measures is a poor one. Fortunately, earth-currents do 

 not have so much influence upon the working of a cable 

 as they have upon the testing, and more fortunately still 

 these currents do not always exist, so it is possible to ob- 

 tain measures during a tranquil period. On the Direct 

 United States' Cable, Sir William Thomson found these 

 currents to be equal in value at a period of greatest 

 strength to that from about eighteen cells of the testing- 

 battery — the Irish end being positive generally to the 

 Nova Scotian end. Under such conditions. Sir W. Thom- 

 son employed the simple deflection-method of measuring 

 the conductor-resistance, which he takes to be " ihe only 

 proper method for measuring copper-resistance in a sub- 

 merged cable." In the following description of the 

 method and its results, it will be seen that the method 

 consists in applying together with a measuring instrument 

 an electric force which yields a certain measure through 

 the unknown resistance of the cable ; a known resistance 

 (7,300 units) is then substituted for the resistance of the 

 cable, and the latter determined by proportion. The 

 principle of this method is applicable not only to the 

 measurement of the copper-resistance, but is that also of 

 the ordinary method of measuring insulation-resistance, 

 a higher known resistance being used in order more rea- 

 dily to effect comparison with the unknown and much 

 greater insulation-resistance. The actual operation during 

 the period of testing is thus described : — 



*" The insulation-galvanometer quickened three- or four- 

 fold by a magnetic adjustment, and, with a shunt ot 

 twenty Siemens' units on its coil, was put in circuit 

 between line, battery, and earth, and the deflection was 

 observed and recorded every ten seconds. As was to be 

 expected, large and rapid variations of the deflections 

 were continually taking place on account of earth-currents. 

 The direction of the earth-current was from east to west 

 the whole time, as was shown by the 'copper' current 

 being always greater, and the ' zinc ' current less, than the 

 true mean concluded from the observations. It increased 

 gradually (but with some slight backward pulsation) from 

 the beginning- when its amount was that due to a differ- 

 ence of potentials between the Ballinskelligs and Torbay 

 earths equal to 17 of a cell — till the end, when it was 

 more than five times as strong, and corresponded to nine 

 cells ; the Irish earth positive relatively to the Nova Scotian 

 earth the whole time. To measure the copper-resistance ^ 

 a time of comparative tranquillity was chosen, a reading -'M 

 taken, and then as quickly as possible the galvano- ^ 

 meter short-circuited, the battery reversed, the galvano- 

 meter circuit reopened, and a fresh reading taken. Half 

 the space travelled by the spot of light from the first 

 reading to the second is taken, as being the deflection 

 which would be produced by the battery applied in either 

 direction were there no earth-currents. This was done 

 seven times, and the half ranges were as follows : — 235, 

 231, 2295-, 234^, 231, 235, 230 — mean 232"3. I found that 

 the same battery applied in the two directions through 

 the galvanometer, and 7,300 Siemens' units gave 232 

 divisions on one side of zero, and 233 on the other — mean 

 232'5. Hence the copper-resistance to be inferred from 

 the observations is — 



7300 X ~ — -, or 7306 Siemens' units." 

 232-3 



As the cable in question is 2,420 nautical miles in length, 



we have '-^ — = 3*02 Siemens' units per knot. 

 2420 



Insulation TVj/.— The ordinary method of testing the 

 insulation-resistance of a cable consists, as has been said, 

 in obtaining upon the galvanometer or measuring instru- 

 ment a certain measure with a known resistance, and a 

 measure with the unknown resistance, the electric force ' 

 being constant during the two measurements. From j 

 these two measures the unknown resistance is determined. i 



