748 



BELL SYSTEM TECHNICAL JOURNAL 



It should be understood that the attenuation of an open-wire pair 

 varies from time to time over a wide range of values, and therefore it is 

 not to be expected that the values of attenuation measured at any par- 

 ticular time will necessarily coincide with the theoretical values. It 

 should also be understood that the attenuation measured on an actual 

 pair never bears the perfectly smooth relation to frequency which is 

 shown on the standard attenuation curves, but exhibits irregularities 

 varying in magnitude according to the irregularities existing on the line. 

 Thus the curve of attenuation as measured on a very well transposed 

 open-wire pair," which is delineated in Fig. 12, represents about as 



.20 



UJ 



LlJ 

 CL 



m 



Q 

 I 



z 

 o 



< 



D 



z 



u 

 I- 

 t- 

 < 



.15 



.10 



05 





10 20 30 



FREQUENCY -KILOCYCLES 



40 



50 



Fig. 12 — Attenuation measured on a well-transposed 128-mil open-wire pair with 



8-inch spacing. 



smooth an attenuation curve as it is possible to obtain on an open- wire 

 circuit. The attenuation values shown on this curve are somewhat 

 lower than the standard values for similar pairs. This is doubtless 

 explained by the fact that the insulators on this particular pair were 

 new, and the further fact that the measurements were made in winter 

 when the temperature was low. 



An illustration of the significance of the attenuation data given above 

 may be of interest. One of the longest carrier telephone systems now in 

 service extends from Davenport, Iowa to Sacramento, California, a 



" Methods of measuring the attenuation, impedance, and crosstalk are discussed 

 in "High-Frequency Measurements of Communication Lines," by H. A. Affel and 

 J. T. O'Leary, ^. I. E. E. Trans., Vol. 46, 1927, pp. 504-513. 



