236 



BELL SYSTEM TECHNICAL JOURNAL 



tance of a particular phantom on a line having only that phantom 

 present. This capacitance is the ratio of balanced charge (on each 

 side of the phantom) to the balanced impressed voltage. The phantom 

 capacitance may be readily estimated from the potential coefficients. 

 For example, if the phantom involves pairs 1-2 and 5-6 the phantom 

 capacitance is very nearly: 



C = 



Pn + ^22 + ^55 + ^66 + 2^12 + 2^56 " 2^i5 — 2/)25 — 2pi& — 2/?2G 



If the disturbing circuit is pair 1-2 and the disturbed circuit is 

 pair 3-4: 



T = 



Plh + Pu — p2h — p2i 



2{pn — pn) 



Tpb = ~~ {plZ + p2S + Pib + Pi6 — Pu — p2i — p3b — Pse)- 



These computations of indirect coefficients are necessarily laborious. 

 They can be simplified to some extent by ignoring phantoms for which 

 either Tap or Tpb is zero or small. For example, the voltage transfer 

 coefficient is zero for pair 1-2 to such phantoms as 1-2 and 11-12, 

 11-12 and 21-22, etc. 



In the following table are given comparisons of far-end crosstalk 

 coefficients as measured in a 40-wire line and as computed by the 

 methods discussed above. The spacing of the various wires and 

 crossarms is indicated by Fig. 27A. The measured values are for 

 40 wires and the computed values are for 10, 20 and 30 wires. It 

 will be seen that a considerable number of wires must be taken into 

 account in the computations in order to obtain a fair check with the 

 coefficient measured for a heavy line. 



Far-End Crosstalk Per Mile Per Kilocycle 



