Telephone Plant in the United States

The sheath of a typical long cable is about 2-1 in. in outside diameter. The conductors placed in such a cable are usually either 51 mils in diameter, weighing 41 lb. to the wire mile or 36 mils in diameter, weighing lb. In the construction of this type of cable, two wires, each insulated with dry paper, are first twisted together to form a "pair." Then two of these pairs are twisted together to form what is termed a "quad." Each quad furnishes three talking paths :—two "physical" circuits and one "phantom" circuit, the latter resulting from a combination of the two physical circuits. Approximately 14o of these quads formed together and enclosed in a metallic sheath make a full-sized cable which, in a typical case, furnishes about 30o telephone circuits and an equal, or even greater, number of telegraph cir cuits. The cable is attached to a steel wire strand of high tensile strength strung on short stocky poles. Nothing short of a storm of most unusual severity can interrupt service with construction of this type. In passing through cities, the cable is usually placed in underground ducts. Loading coils are connected in each tele phone circuit a little over a mile apart and, at intervals of about 5o miles, "repeater" stations for amplifying the speech currents are placed along the route.

Many times during the period of 6o years that have elapsed since the invention of the telephone, scientific research and en gineering development have not only found the way out of diffi cult situations which threatened to hamper the growth of tele phone service but, in advance of immediate needs, they have often created new instrumentalities which, in turn, have led to further extensions and improvements of the service.

Cables.

Telephone cables not only provide the sole means for placing a large number of telephone circuits compactly either overhead or underground, but they enable the lines to be carried across the beds of rivers and other waterways. In the cables first employed, single wires with a ground return were used and the wires were insulated with either gutta-percha or rubber. It was soon found, however, that such cables were not well suited for telephone purposes. After experiments with many kinds of in sulating materials it was found that dry paper was the best for telephone cables. This discovery and the concurrent development of means for encasing the paper insulated wires in a metallic sheath or tube which protected them from moisture laid the foundation for the modern telephone cable. Research and de velopment covering many materials and processes have made it possible gradually to increase the number of pairs of wires, in a full-sized cable for subscribers' lines, from a maximum of 5o in the year 1888 to 1,800 in 1928. This progress has resulted in a greatly increased effectiveness of utilization of the space in under ground ducts in addition to other marked advantages, including a reduction in the cost per mile of cable pair (including ducts) to less than one-tenth of its cost in 1888. The increase in the

number in pairs in a single cable has been accompanied by a re duction n the size of the wires, those used in the 1,800-pair cable, for weighing only about 4 lb. per wire mile, while those i used in the 50-pair cable, of 1888, weighed about 25 lb. per wire mile. Such fine wires could not be used had it not been for im provements in many other parts of the telephone plant, such as transmitters, receivers and coils. Plate II., fig. 4 shows a 1,2m pair subscribers' cable with its wires fanned out. The design, manufacture, laying and splicing of cables, as well as the choice of a type to meet a given set of conditions, have required a large amount of engineering study. A typical problem has been the prevention of "cross-talk," or overhearing, between the pairs necessarily placed close together in a large cable. This has been accomplished by twisting the wires together according to various plans.

Loading.

For some time prior to the year 1900, it had been known to those technically skilled in the art of telephony that the transmission efficiency of long telephone circuits could be improved by increasing their electrical property known as the "uniformly distributed inductance." This knowledge did not, however, lead to practically beneficial results because no one was able to suggest any feasible method of increasing this property of a telephone circuit without bringing in, at the same time, diffi culties of one kind or another which were fatal. Numerous in vestigators sought unsuccessfully to simulate the beneficial effect of increasing the uniformly distributed inductance by introducing, into the circuit, inductance concentrated, or lumped, in the form of coils. It was reserved for Professor Pupin of Columbia Uni versity to discover, in 1900, that the proper spacing of the coils along the telephone circuit was the key to the solution and it was from failure to establish this that earlier workers had failed.

The term "loading" comes from the mechanical analogy of a "loaded" string, the study of which was utilized in developing the theory of electrical "loading." When a series of small weights was placed at intervals on a string it was found that mechanical waves, produced in the string, did not die out as quickly as when the string was not weighted. The mechanical loading reduced the dying out of the waves in the string and the analogous electrical loading reduces the attenuation of the waves of speech current in the telephone circuit. It makes the line a better path for elec trical waves. To indicate its practical value, it can be stated that, under favourable circumstances, a loaded cable circuit is as good a conductor for telephonic currents as would be a non-loaded circuit of conductors weighing about eight times as much.