366 BELL SYSTEM TECHNICAL JOURNAL 



repeaters. The magnitude of the over-all project was so great that the con- 

 struction work and the manufacturing effort had to be undertaken in several 

 steps, spread over a period of years. 



The first step was to lay a new underground conduit between Wilmington 

 and Washington via Baltimore, and install a composite coarse-gauge loaded 

 cable between Philadelphia and Washington, thus closing the gap between 

 New York and Washington. It was an economically fortunate coincidence 

 that when it became necessary to engineer the cable and the loading, a good 

 start had been made on the development of duplex cable and of cable phan- 

 tom group loading. The status of these developments, however, was such 

 that it was very far from a certainty that long-distance loaded duplex cable 

 facilities could be made satisfactory from the crosstalk standpoint. On the 

 other hand, the economic stakes were very great; if the crosstalk could be 

 kept within tolerable limits, the complement of coarse-gauge circuits would 

 be nearly 50% larger than if a coarse-gauge non-quadded cable should be 

 used, and the transmission equivalents obtainable in the loaded phantoms 

 would be appreciably lower than those on loaded non-quadded pairs using 

 the same size conductors. The increment cost in getting the extra comple- 

 ment of higher grade circuits consisted principally of the cost of the phan- 

 tom loading coils and the cable capacitance unbalance adjustments. This 

 was judged to be small in proportion to the potential value of the circuits. 

 The decision to proceed with the development of the duplex cable and phan- 

 tom group loading was made in the spring of 1910, while the quadded cable 

 for the Boston-Neponset project was under production, and prior to the 

 start of production of the new phantom loading apparatus. 



A fundamental transmission objective was to obtain facilities which 

 upon completion of the project would meet talking standards desirable for 

 Boston-New York and New York- Washington connections, with a few 

 circuits suitable for emergency use between Boston and Washington in 

 case the open- wire lines should go down. These requirements called for 

 10-gauge conductors. Also, there were requirements for all-cable circuits 

 connecting intermediate points, which called for 13-gauge or 16-gauge con- 

 ductors. Since the cable route intersected the existing open-wire routes at 

 many points, patches could be made for the protection of the long distance 

 service normally handled in open- wire. 



Philadelphia-Washington Section: The requirements above sununarized 

 resulted in the design of a cable having 7 quads of 10 B&S gauge conductors, 

 18 quads of 13-gauge conductors, 6 non-quadded 13-gauge pairs in the in- 

 terstices of the layer of 10-gauge quads, and 18 non-quadded 16-gauge 

 pairs in the interstices of the 13-gauge quads. All quads were of the 

 multiple-twin t3^e. The loading cost study resulted in a decision to use a 

 new weight of loading, medium-hea^y, with coils installed at intervals of 



