Abstract :
In a previous paper* by one of the authors, the state of knowledge then existing upon the subject of telephone interference was described. The present paper deals with the results of investigations in Great Britain and elsewhere since completed, and particularly with the series of investigations in which the authors have been concerned. These have been carried out during the past few years under the aegis of the E.R.A., in co-operation with communication and power interests. Consideration of the subject may be divided roughly under two heads: disturbance from harmonics during normal operation of the power system, and danger due to excessive induction at the fundamental frequency under transitory fault conditions. The former gives rise to noise and, with regard to its disturbing effect, may be calculated or measured in terms of an e.m.f. at 800 cycles per sec. In the same way the total harmonic output of electrical plant may be expressed by its telephone interference factor. For modern installations this quantity can usually be kept below 1 per cent. Although mercury-arc rectifiers (fully discussed in connection with experiments on traction systems) exceed this value, the telephone interference factor can be reduced to less than 1 per cent by appropriate smoothing equipment. Both disturbance and danger are generally due to induction from earth currents, a problem which is treated in detail. The magnitude of the currents is investigated for various cases, and it is shown that the induction can be calculated if the earth resistivity is known. Alternative methods of measuring this are available. The screening effect of conductors in parallel (such as telephone-cable and power-cable sheaths, earth wires, and rails) must be taken into account, and formul? and experimental measurements are given for these. Brief mention is made of cases of induction at close separation. Although further study of certain aspects is still needed, it is concluded that theory and experience ar- e available for the approximate calculation and predetermination of the amount of interference likely to occur in any given case. Possible methods of prevention are indicated in various sections of the paper.
