INSTRUMENTS FOR THE NEW TELEPHONE SETS 353 



3. These improvements have been accomplished without sacrificing 

 simpHcity of design or introducing features which complicate 

 manufacture of the receiver or increase the maintenance 

 required. 



Response. — An equivalent electrical circuit for the receiver and a 

 typical closed coupler response curve are shown on Fig. 8. Referring 

 to this figure it will be noted that there are two meshes in the circuit 

 which contain mass, stiffness and resistance and which control the 

 motion of the diaphragm. One of these meshes consists of the acous- 

 tical resistance, Wi^i, coupled to the diaphragm, maSafo, by the stiffness, 

 5i, of the cavity between the diaphragm and the plate which surrounds 

 the pole tips. Included in this mesh is the stiffness, s^, of the cavity 

 in the handset handle or receiver shell. The other mesh is composed 

 of a cap grid, mzYz, and the load, Si, coupled to the diaphragm by means 

 of the cavity stiffness, S3. The grid of the receiver unit proper is 

 provided for mechanical protection only and has holes large enough to 

 have no reaction on response. The mass of the resilient screen is 

 small and is lumped with the diaphragm mass, mo. The electrical 

 portion of the circuit consisting of the winding, RiLi, and the equiva- 

 lent eddy current circuit, R^Li, is coupled to the mechanical and 

 acoustical portion by means of the force factor M''. 



The response computed from the equivalent circuit for a number of 

 frequencies is included on Fig. 8. The agreement between this curve 

 and the measured curve is excellent and makes it possible to predeter- 

 mine the response of the receiver with a high degree of accuracy, and to 

 evaluate the effect on the overall response of the receiver of changes in 

 the constants of the component parts. This type of analysis also has 

 been invaluable as an aid in establishing the causes of variations in 

 response which have been observed during the development and pro- 

 duction of the receiver. A measured response curve of a receiver of 

 the earlier type has been added to Fig. 8 for convenience of reference. 

 The improvement in uniformity and range of response is obvious. It 

 will be noted that large gains have been effected for frequencies in the 

 range from 1500 to 3000 cycles per second. 



The response of the receiver to a square topped wave affords an excel- 

 lent measure of frequency distortion. Oscillographic records of the out- 

 put of typical receivers of the new and earlier types are shown on Fig. 9 

 for a frequency of approximately 50 cycles per second. The distorting 

 effect of diaphragm resonance is so obvious as to require no comment 

 beyond pointing out that for accurate reproduction of square waves 

 uniform response for an infinite frequency range is required and that 

 the slight rounding of the corners of the wave as reproduced by the 



