292 BELL SYSTEM TECHNICAL JOURNAL 



It is evident from the above examples that careful examinations of 

 trouble records are required before the observed rate of lightning failures 

 can be adequately compared with that obtained from theoretical expectancy 

 curves. If the cable as well as splices and accessories actually have a di- 

 electric strength as assumed in the calculations, it is likely that the average 

 rate of failures due to excessive voltages experienced over a long period will 

 not be any greater than estimated from these curves. 



Based on experience, an average of 4 sheath openings is required to repair 

 damage caused by excessive voltage between the sheath and the cable con- 

 ductors, as compared to about 2 sheath openings when the damage is due 

 mainly to denting and fusing of the sheath, as in the case of full-size, tape- 

 armored cable in low-resistivity territory. Although the damage may be 

 confined to one point, it cannot usually be located by a single sheath opening. 



III. Remedial Measures 

 3.1 General 



From Fig. 12 it is evident that the rate of cable failures to be expected, 

 and hence the need for remedial measures, depend greatly on the earth re- 

 sistivity. Experience has indicated that lightning damage is likely to be 

 encountered even when the surface resistivity is fairly low, provided the 

 resistivity beyond depths of 10 or 20 ft. or so is very high. Considerably 

 less trouble has been experienced where the resistivity below this depth 

 is low, even where the surface resistivity has been high. The lightning 

 stroke may then channel through the surface layer to the good conducting 

 lower layer, so that direct strokes are not experienced as frequently in 

 spite of the high surface resistivity. As a guide m applying protective 

 measures, earth resistivity measurements are usually made along new 

 cable routes. 



The curves given in Fig. 12 may also be used to find the lightning trouble 

 expectancy when extra core insulation, shield wires or both are used. Thus 

 when the insulation strength is doubled the effect is the same as if the 

 sheath resistance is halved. If the shield wires reduce the voltage by a 

 shield factor 77, the effect is the same as if the sheath resistance is multiplied 

 by 77. Considering direct strokes only, curve 2 in Fig. 1 may be used to 

 find the percentage reduction in lightning strokes that will damage the 

 cable, when the stroke current which the cable is able to withstand is in- 

 creased by extra insulation or shield wires. 



3.2 Extra Core Insulation 



One method of reducing failures caused by lightning strokes to buried 

 cables is to increase the insulation between the cable conductors and the 



