Abstract :
The paper puts forward the view that domestic electricity supplies are heavily subsidized by other business of the Electricity Boards and that, in the interests of overall economy, drastic revision of present methods of l.v. distribution design is necessary. It is asserted that recent legislation and researches into domestic-load characteristics now provide sufficiently accurate data on which to design networks for new housing estates. An interpretation of the results of this research is given, and the application of the data to l.v. network planning is explained. In order to compare alternative l.v. distribution networks to supply a given estate, it is necessary to formulate a theoretical symmetrical network closely approximate to the mean site condition to which analytical treatment can be applied. Such a theoretical network is postulated and the closeness of its approximation to site conditions is indicated. Formulae for mean losses and maximum voltage-drop are derived from the theoretical network. It is explained that, since designs of theoretical maximum economy are unattainable, because of the physical limitations of underground cables, the nearest practical approach derives from loading cables to their maximum permissible thermal rating. The costs of designs on this basis, as applied to the theoretical network, are given for each of seven standard sizes of l.v. cable. Whilst it does not appear possible to generalize where h.v. feeder and substation costs are concerned, their effect on the overall costs of a specific case which recurs in practice is examined. It is found that maximum practicable overall economy is obtained when 0.06-in2 l. v. distributor cables are used, with the consequent number of substations of an economical design and the necessary h.v. feeder cables. Further economies designed to limit initial capital expenditure are mentioned. The degree of supply continuity obtained from the suggested methods is examined, and it is found to be economic- ally unsound to provide anything approaching 100% standby coverage on the l.v. network. The application of the suggested method to a housing estate now being built is described, with a comparison of the costs of three alternative types of distribution, and the degree of accuracy of costs calculated by the use of the theoretical network is demonstrated. In conclusion, it is contended that, with the methods illustrated, networks capable of supplying satisfactorily ultimate loads of 3 kW per consumer (after-diversity maximum demand) can be provided at a cost of about ?25 per consumer, excluding service connections. This contrasts with a cost of ?40 per consumer for designs hitherto accepted as standard, catering for loads of little more than 1.5 kW per consumer.
