INDUCTIVE LOADING FOR TELEPHONE FACILITIES 153 



ance. This can be understood from the fact that in the (low-impedance) 

 non-loaded circuit, the series dissipation losses which are proportional to 

 the square of the line current are ordinarily very large relative to the di- 

 electric dissipation (i.e. shunt) losses which are proportional to the square 

 of the line potential. When the line impedance is increased a suitable amount 

 by the loading, the decrease in series losses is much greater than the increase 

 in shunt losses. In commercial practice, economic considerations generally 

 prevent the use of high loading impedances which would result in the shunt 

 losses becoming as great or greater than the series losses. 



In situations where voice frequency attenuation improvement is the 

 principal objective, the unit transmission loss can usually be reduced to the 

 order of one-third to one-fourth of the non-loaded value. The loss reduction 

 is less than this at low voice frequencies and more at high frequencies, 

 resulting in a much more uniform transmission of the important frequencies 

 that are required for intelligibility and naturalness. In certain situations 

 which will be discussed later a lower ratio of attenuation reduction is ac- 

 cepted in order to obtain other, more important, transmission advantages. 



Pioneering ^ Developments 



General 



A full account of the pioneering research and development work would 

 take much more space than is available in a review devoted primarily to 

 the evolution of the loading art. The present account is therefore limited 

 to a brief description of the first loading systems and apparatus standards 

 that resulted from the pioneering work.^^^ 



Although the success of the 1899 laboratory investigations, and the Bell 

 System's 1900 experimental installations on exchange cables and on open- 

 wire lines, quickly built up a substantial demand for loading, the com- 

 mercial applications had to be deferred pending the development of 

 satisfactory types of loading coils. Then there followed a series of what 

 should be considered as trial installations of different types of loading, 

 tailored to the specific needs of particular projects. Analyses of the per- 

 formance characteristics of these installations, supplemented by continuing 

 experimental work ui the laboratory and by engineering cost-studies, re- 

 sulted in the establishment of a series of standard cable loading systems for 

 general use late in 1904. The commercial development of satisfactory open- 

 wire loading encountered many even more complicated problems than 

 those involved in cable loading, and in consequence the standard loading 

 for 104-mil lines did not evolve until 1905. This same type of loading be- 



(b) A more complete description of these standards is given in Bibliography Reference 

 (6). Reference (7) is also of interest. Reference (1) gives some details of Campbell s early 

 work. 



