228 THE BELL SYSTEM TECHNICAL JOURNAL, APRIL 1951 



garding this extension was obtained by observations on circuits covering a 

 range in the various factors affecting transmission, and a count made of 

 the number of repetitions which were requested per unit time by the users 

 in carrying on normal telephone conversations. This repetition-rate method 

 of measuring performance of telephone circuits bridged the gap between 

 the laboratory and the plant, and established a relation between physical 

 and subjective measurements in the laboratory and subjective results in 

 service. '*°''^^ In fact, the rating method derived from the repetition count 

 observations was needed to prove that the effect of the reduced sidetone of 

 the anti-sidetone circuit was sufficient to offset the additional circuit losses, 

 complexity and cost of that circuit and so to justify its general use. 



The development of the idealized transmission system and of the devices 

 for measuring the electrical and acoustic inputs and outputs of telephone 

 instruments, and the carrying out of the articulation and repetition rate 

 measurements, together with the analyses of their results and deduction of 

 relationships, required a large amount of activity for about fifteen years 

 from the time of Arnold's concept, to cover the scope outhned here. Sub- 

 sequent work has been directed to refining the devices and the results. 



This work of physical and subjective measurement produced the knowl- 

 edge of the performance characteristics which telephone transmitters and 

 receivers should have and also the way to specify and analyze their per- 

 formance — in other words, what to strive for in the development of new 

 designs and how to determine the degree to which it has been attained. 



Design Theory 



The work which was done in developing the transn^itters and receivers 

 for the idealized transmission system promoted an evolution of the theory 

 of the vibratory elements of such devices, including the effects of the as- 

 sociated air chambers. From this came large advances in the theoretical 

 understanding of electro-acoustic converters, as exemplified by the book of 

 Crandall "Theory of Vibrating Systems and Sound"^^ ^nd pubhcations by 

 Wegel,^^ Wente^^' ^ and others. 



One further concept was necessary to bring the design theory on instru- 

 ments to its present level. As has been discussed earlier, the analysis of 

 telephone circuits from the electrical standpoint made extensive application 

 of the idea of the equivalent network. The new concept involved two steps: 

 One was that the theory of electro-acoustic devices could be reduced to the 

 simplicity of electrical network theory by using electrical analogs for the 

 vibrating system. This was well brought out by R. L. Wegel in his paper 

 of 1921«. The second step, promoted by H. C. Harrison,^^' *^ E. L. Norton 

 and others, was that mechanical wave transmission systems could be de- 

 signed as analogs of electric circuits. 



