OPEN-WIRE CROSSTALK 55 



The latter expression divided by F is the frequency times the far-end 

 type unbalance for the length 2D. If one of the circuits were trans- 

 posed at point B the crosstalk in length AB would be cancelled by 

 that in length BC. The sign of the interaction crosstalk between the 

 two lengths would be reversed and the expression would become 



27 



If neither circuit were transposed at B, the far-end crosstalk would 

 be that for a non-transposed length of 2D or: 



1 _ ,-470 



fk'—^ . 



27 



The frequency times the type unbalance values for the cases of one 

 transposition and no transpositions are the same (in magnitude) as 

 those derived for near-end crosstalk which were plotted (neglecting 

 attenuation) as curves and P on Fig. 15A. 



If both circuits are transposed at B the near-end type unbalance 

 remains the same as if there were no transpositions. The far-end type 

 unbalance is radically altered, however. This is evident if the above 

 equation is compared with that for the case of both circuits transposed. 



This process of computing type unbalances may be extended from 

 two equal lengths to any number of equal lengths. It is necessary to 

 consider the interaction crosstalk between each length of the disturbing 

 circuit and each preceding length of the disturbed circuit. The rela- 

 tive propagation distances through the various interaction crosstalk 

 couplings must be taken account of. 



Computations of far-end type unbalances are greatly simplified by 

 assuming the same propagation constants for the disturbing, disturbed 

 and tertiary circuits and by neglecting attenuation within a trans- 

 position section as in the case of near-end crosstalk. Since the tertiary 

 circuit may be composed of any combination of wires on the line or of 

 these wires and ground return, the propagation constant for a tertiary 

 circuit may be somewhat different from that for the disturbing and 

 disturbed circuits. This is particularly true of earth-return circuits, 

 but these are of little practical importance due to their relatively high 

 attenuation. All circuits not involving the earth have somewhere 

 near the same speed of propagation but the tertiary circuits may 

 differ greatly in attenuation constants. 



For practical reasons a fair balance against crosstalk must be 

 obtained in each transposition section (eight miles or less) and, as in 

 the case of near-end crosstalk, type unbalances are calculated for the 



