OPEN-WIRE CROSSTALK 203 



the complexity of the transposition design problem. Uniform seg- 

 ments are therefore used except in special circumstances. 



Wire Spacing Irregularities 



In the past there has been a tendency to permit wire spacing irregu- 

 larities in order to reduce the cost of construction and maintenance. 

 For example, "H fixture" crossarms formerly had special wire spacing 

 to permit the two poles to pass between pairs of wires and thus reduce 

 the length of the arms. Another example is that of resetting a pole 

 with a rotted base and reducing the spacing between crossarms to get 

 clearance between wires and ground. The development of repeatered 

 circuits and carrier current operation has increased the seriousness of 

 the crosstalk resulting from such irregularities and made such practices 

 generally undesirable. 



There are, of course, unavoidable irregularities in wire spacing due 

 to variations in dimensions of crossarms, insulators and pole line 

 hardware and warping of crossarms. Corners and hills are other 

 causes since the crossarms at a corner and the poles on a hill are not 

 at right-angles to the direction of the wires. The most important 

 unavoidable spacing irregularity is, however, due to variations in wire 

 sag. Of recent years, limits have been set on wire sag deviations to 

 insure that this effect is properly limited during construction. The 

 main criterion adopted has been the difference in sag of the two wires 

 of a pair. This difference is a rough measure of the crosstalk increment 

 due to variations of the sag from normal. The crosstalk between 

 two pairs in a given span will be abnormal if the two pairs have different 

 sags even if there is no difference in sag for the two wires of a pair. 

 The crosstalk is usually more nearly normal, however, than in the 

 case of two pairs having the same average sag but different sags for 

 the two wires of a pair. As far as practicable, all pairs are sagged 

 alike in a given span. 



The crosstalk between two pairs due to sag differences is computed 

 much like that due to pole spacing irregularities. The change in 

 crosstalk due to a known pole spacing deviation may, however, be 

 computed from the crosstalk coefficient while the change in crosstalk 

 due to a sag deviation is not related to the crosstalk coefficient in any 

 simple way. Two methods have been used to obtain constants for 

 calculation. 



With the first method, crosstalk measurements were made on a 

 long line (about 100 miles) having small pole spacing and type un- 

 balance crosstalk. The r.m.s. of a number of crosstalk measurements 

 was determined for each particular type of pair combination, for 



