Underground alternating-current network: Distribution for central station systems

1. Certain experimental arrangements for an underground alternating current distribution system are covered in the paper. 2. Low costs and characteristics satisfactory for general utilization of alternating current are contrasted with difficulty of obtaining reliability. The advantage of adopting system characteristics which will make each service suitable for lighting, appliances, and motors is indicated. There is described experience with a combined light and power network on one set of mains. Effect of voltage variation on incandescent lamp illumination is discussed. An appendix covers tests on this subject. 3. Reliability in service of standard distribution materials is considered. Tests on cables are described which indicate that for underground distribution conditions, low-voltage cables will eliminate arcing faults while high-voltage cable will not do so. An appendix of arcing tests is attached. 4. Certain factors are discussed relative to size of mains and location of transformers necessary to make the low voltage network clear its own faults. 5. There is a description of an experimental system consisting of several radial high-voltage feeders, the distribution transformers of all these being connected on the low-voltage side to a common low-voltage cable network. 6. The reactance of the transformer circuits is almost three times that of standard distribution transformers. 7. The only protective devices used are automatic circuit breakers installed in the low-voltage cables between the distribution transformers and the network. The switches open on a reversal of energy from the network and close when the transformers are in a condition to supply power to the network. The devices are sensitive enough to open on transformer magnetization energy. 8. Great reliability is obtained for the network as a switch failure an a high-voltage fault is the equivalent of a short-circuit on the secondary network. 9. Ability to switch automatically all transformers by controlling the supply end of the feeder to which they are connected allows a higher all-day efficiency due to the saving of iron loss at periods of light load. This feature also makes it easy to work on high-voltage equipment. 10. A description of the equipment used in this experimental installation and operating results of switch equipment from date of installation in April 1922 to March 1924 are given. 11. It is believed that with low-voltage a-c. underground networks arranged as described, the reliability is the same as the reliability of the sources supplying the various radial high-voltage feeders. The possibility in the future of a protected network arrangement of the supply system is mentioned.