MEASUREMENTS IN MULTIPAIRED CABLES 653 



pirical methods of capacitance-space analysis do not provide any insight 

 as to whether these deviations are due to anomaUes in the mechanical prop- 

 erties or in the dielectric properties of the insulation, or both. 



With respect to some electrical and physical characteristics, it is evident 

 that there is a close analogy between the cable pair and an "ideal" balanced 

 shielded pair consisting of two straight and parallel soUd cylindrical con- 

 ductors enclosed in a cylindrical conducting shield, with the center line 

 of the pair coinciding with the axis of the shield. A cross-section of such a 

 circuit is shown in Fig. 1. The conductors are insulated from one another and 

 from the shield by a homogeneous dielectric. 



Rigorous mathematical expressions for the capacitance and inductance 

 of the ideal pair in terms of its dimensions and dielectric constant have been 



Cmu-fc — Ct2 + -g" 

 Fig. 1 — A balanced, shielded pair. 



derived by the Mathematical Research Group of the Laboratories and 

 others. Measured values of the capacitance and inductance of a given cable 

 pair, when substituted into these expressions, determine a set of dimensions 

 and a dielectric constant value which describe an ideal pair having the same 

 capacitance and inductance as the cable pair. The extent to which these 

 idealized values represent actual cable conditions depends on the accuracy 

 of the assumed equivalence of the two structures. 



The purpose of this paper is to describe a few simple but direct experi- 

 ments, the results of which demonstrate that these idealized parameters are 

 closely representative of actual cable conditions in so far as interaxial spacing 

 and dielectric constant are concerned. Thus emerges a simple technique, 

 based on easy-to-make low-frequency measurements, for quantitative evalu- 

 ation of these two important cable pair parameters. AppUcations of the 



