NnvFMni'.R I, 1904.] 



THE INDIA RUBBER WORLD 



43 



INSULATING MATERIALS IN HIGH TENSION CABLES.* 



Two insulating materials are now principally competing 

 in the field of high tension cables — vulcanized rubber 

 and paper impregnated with rosin and oil mixtures. 

 Paper insulation has made great progress in the last 

 few years. The utility of using good manila paper, laid on in 

 thin and regular layers, without wrinkles and crumpling, has 

 been recognized, and also the utility of having it properly des- 

 sicated, at a moderate temperature, in a vacuum, and impreg- 

 nated with a compound of rosin, or wax, or asphalt, with min- 

 eral, or castor, or linseed, or some other oil, that does not be- 

 come brittle or pulverize with age. But rubber also has made 

 progress ; and if some feared formerly that it would decay with 

 age, it is now certain that first class rubber cables, well vulcan- 

 ized, and removed from the infiuence of brush discharges in 

 the air, or not alternately dry and wet, will last indefinitely. 



Rubber has a dielectric strength much higher than impreg- 

 nated paper. Testing good rubber cables in such lengths as to 

 include the inevitable irregularities of manufacture, with ten- 

 sions progressively increasing and subjected to dielectric strain 

 at least one hour, we can easily obtain for the rubber a dielec- 

 tric strength of 12 to 15 kilovolts per millimeter. Paper in the 

 same conditions would only stand 8 to 10 kilovolts per milli- 

 meter. These numbers represent as good an average as we 

 can reach in normal manufacturing; it is not rare to find 20 to 

 30 per cent, more, or even higher percentages, but we cannot 

 reckon upon these. The higher dielectric strength of rubber 

 brings us to the conclusion that the use of rubber for very high 

 tension will extend more and more. 



* * * 



A CAUSE of inferiority of the rubber is the lesser homoge- 

 neity of Its products. It is not uncommon to find that two 

 cables, manufactured in the same manner, with the same qual- 

 ity of rubber, afford a very different resistance to perforation — 

 a difference, say, of 30 to 40 or 50 per cent. Paper cables are 

 more homogeneous. The figures relative to dielectric strength 

 given above are the result of a great number of tests made by 

 the author on cables of various mikers. They do not take ac- 

 count of some exceptionally high strengths; I found some 

 pieces of rubber cable to withstand 20 to 25 kilovolts per milli- 

 meter. The elasticity of rubber gives it a great superiority 

 over paper. A paper cable with large thickness of paper can 

 not be easily bent, especially in cold weather, owing to crack- 

 ing ; on the other hand, the manufacture of concentric, or 

 stranded, multiple core cables is simpler in the case of paper 

 cables, for the insulating material can be uniformly distributed 

 in the interspaces among the conductors, which remain buried 

 in the insulator, which is not possible with rubber. 



The great success of paper cables is a consequence of their 

 lower price. But very high tensions require such a greater 

 thickness of paper, that the cost of the paper added to the 

 extra price for the larger quantity of lead, steel, tape, etc., per- 

 mits the rubber to win in the competition. 



The problem of manufacturing high tension cables would be 

 simpler if the gradient of the potential within the body of the 



•The matter presented herewith consists of a series of paragraphs selected 

 from an exhaustive paper presented at the International Electrical Congress, at 

 St. Louis, by Sienor Emmanuel Jona, chief electrician of the establishmeiu of 

 Pirelli & Co. (Milan). ;tnd a delegate to the congress from the Associazione Klet- 

 trotecnia Italiana. There is not sp.tce in these pages for all the data introduced 

 by the author as the basis for liis conclusions ; but without these the paper, 

 though incomplete, will not fail to be of interest, in connection with the relative 

 merit of rubber and paper as insulating materials.— The Editor. 



insulator was constant. Suppose a 38 square millimeter cable 

 insulated to 14.5 millimeters outer diameter, and working at 

 25,000 volts. The layer near the copper supports a strain of 

 5000 volts per millimeter, while near the lead the stress is only 

 1200 volts per millimeter. Should the stress be constant 

 throughout, each layer of i millimeter would support a strain 

 of 2270 volts, and the cable would be much safer. We could 

 then also diminish the thickness of the insulation to, say, 5 

 millimeters, letting every layer work at 5000 volts. 



* * • 



Without claiming to get an absolutely constant gradient, 

 we can, therefore, try to have the potential better distributed 

 along the radius of the insulation, and at the same time use in 

 the proper place materials having greater dielectric strength, 

 by making the insulating layers of different materials specially 

 chosen. This method I studied and applied to the manufac- 

 ture of high tension cables, as early as 1S98. Such cables, con- 

 sisting of conductors first insulated with several layers of rub- 

 ber, on which were wound layers of paper or jute, were patent- 

 ed by Messrs. Pirelli & Co. [Milan, Italy], March, 1900. A 

 cable of this kind was working at 25,000 volts, during the I^aris 

 exhibition of 1900. 



The specific inductive capacity of paper cables varies from 3 

 to 4, according to the type of paper and mixture adopted. The 

 inductive capacity of paper is about 2 ; that of rosin 2 to 3, ac- 

 cording to its origin ; and mixtures of rosin, oil, paraffin, ozo- 

 kerite, and other materials, have a capacity of 3 to 4, or even 

 more. For example, lubricating oil 55 parts, rosin 560, paraffin 

 224, ozokerite 160, has a standard inductive capacity of 3.6; 

 oxydized linseed oil 90. rosin 370, Arkangel pitch 70, have 4.4 ; 

 Arkangel pitch itself has 59; a mixture with Gallipot, instead 

 of rosin — for example. Gallipot 600, Arkangel pitch 1 10 and lin- 

 seed oil 130 — has 4.8 ; a mixture of lubricating oil 9, rosin 52, 

 black ozokerite 23, white ozokerite 16, has only 3 55. 



It appears from these figures that it is possible to have a large 

 range of inductive capacity with paper cables. But as they are 

 impregnated in mass, the entire mass has the same standard in- 

 ductive capacity unless we change the type of paper, by usingi 

 for example, paper loaded with some materials, as suggested 

 very ingeniously by Mr. 0"Gorman. 



* » * 



Ox the contrary, it is easy to use different rubbers having 

 varied standard inductive capacity, for rubber is put on in suc- 

 cessive layers which can be quite different one from another, 

 and which have no tendency to mingle together, either during 

 or after manufacture. The cables I alluded to are manufac- 

 tured with layers of various qualities of rubber in the inner part 

 of the insulation ; but as soon as the gradient of potential be- 

 comes so diminished as to allow the use of paper, the insula- 

 tion is continued with paper, and after the paper with jute, if 

 the gradient is sufficiently low to allow the use of jute. The 

 rubber insulation is generally first vulcanized and the conductor 

 tested in water, as usual, before adding the outer layers of 

 paper and jute. 



Pure vulcanized rubber has an inductive capacity something 

 like three as an average ; but it is very easy to " load " the 

 rubber with large quantities of extraneous materials, which, 

 without sensibly lessening its specific dielectric strength, aug- 

 ment the capacity very much. A rubber with 58 per cent. 

 Para, 2 per cent, sulphur, 26 per cent, talc and 14 per cent. 



