410 BELL SYSTEM TECHNICAL JOURNAL 



I'sii OF THF. TU IN Transmission Studies 



In makinj; transmission studies it lias previously been the practice 

 to express the transmission efficiency of limiting subscribers' loops 

 in terms of the resistance of a 22-gauge loop which would have the 

 same total transmitting and receiving loss, thus a 400-ohm loop 

 meant a loop which had the same total transmitting and receiving 

 loss as a loop of 22-gauge ASA cable having a resistance of 400 ohms. 

 At the time of changing from miles to TU, it was decided to abandon 

 this method of expressing limiting loop losses in the Bell System and 

 to express them directly in TU; thus a 5 TU loop means a loop whose 

 total transmitting and receiving loss, taking into account the effici- 

 ency of the subscribers' set, is 5 TU. The following table gives a 

 number of limiting loops expressed in TU and their equivalents in 

 ohms of 22-gauge cable as defined above, assuming the use of the most 

 efficient type of subscriber's set now available. 



Limiting Loops Limiting Loops 



Expressed in TU E.\pressed in Ohms 



3 312 



4 350 



5 387 



6 424 



7 461 



8 499 



9 537 



10 573 



11 610 



Conversion from Miles to TU and Compltation 

 OF Transmission Equivalents 



During the transition period in the adoption of the TU it will 

 freciuenlly be necessary to convert transmission data which are 

 expressed in miles, to TU. This is easily accomplished by multi- 

 plying by a conversion factor and in the case of the transmission 

 efficiencies of subscribers' sets by also correcting for the difference 

 in the reference zero which was brought about for reasons referred 

 to above. Two units both known as miles have been in common use 

 as a measure of transmission; they arc the standard cable mile and 

 the 800-cycle mile. A different conversion factor is required for each. 



The attenuation constant of standard cable for the complex currents 

 used in the transmission of speech varies appreciably with the length 

 of cable considered, since for long lengths the higher frequencies are 

 attenuated to such low values as to have very little effect on the 

 received volume. The best average figure is 0.122, although this 

 value has yet to be determined more precisely by careful laboratory 

 tests. The attenuation corresponding to one 7U for currents of any 



