MEASUREMENT OF PRESSURES 195 



C. Cable signal. In addition to the purely electrical properties of 

 cables, an effect of applied stress on electrical cables must be con- 

 sidered in their use with piezoelectric or other high impedance gauges. 

 This phenomenon, described as cable signal, manifests itself as an elec- 

 trical charge developed on the conductors of a coaxial cable as a result 

 of shock pressures on the cable. The exact origin and nature of the 

 effect are somewhat obscure, but the evidence indicates that static 

 charge is developed on the outer surface of the insulating dielectric, 

 where it is in contact Avith the outer braided shield, as a result of fric- 

 tion between the dielectric and shield. In a low impedance circuit, 

 charge of this kind is quickly neutralized, but a piezoelectric gauge cir- 

 cuit is necessarily one of high direct current resistance which is as effec- 

 tive in retaining such spurious charge as it is in retaining piezoelectric 

 charge. 



The phenomenon is not very reproducible under conditions of use. 

 It has been found, for example, that a given cable exhibits progressively 

 decreasing signal on repeated applications of shock pressures at fairly 

 short intervals, being, so to speak, beaten into submission, but if left 

 unused for several days again becomes as bad as it was at first. Cables 

 with different dielectrics vary greatly as regards cable signal, poly- 

 ethylene cables being particularly bad, and presumably identical pieces 

 of cable frequently differ greatly. Insofar as anything very simple can 

 be said about such a relaxation phenomenon, it is true that the amount 

 of signal is proportional to the length of cable subjected to stress, and 

 that the voltage developed varies inversely as the total circuit capaci- 

 tance (thus leading to the conclusion that a "false piezoelectric" charge 

 is developed). 



The fact that the charge developed increases with length of cable 

 exposed to pressure means that cable signal is most serious when one is 

 interested in measuring slowly changing pressures, particularly when 

 these are preceded by much higher initial pressures. For example, a 

 shock wave with a large initial peak pressure or followed by a gradual 

 decay develops a signal when it encounters the connecting cable from 

 the gauge which increases until the whole of the wave has passed the 

 gauge and met the cable. Release of pressure from sections near 

 the gauge reduces the signal, but this is balanced by the signal from the 

 cable further along. A nearly constant signal is thus developed shortly 

 after the cable is first struck which persists until drained off by circuit 

 leakage or by attenuation of the pressure wave with distance. The 

 cable thus acts to give a sort of integration of the pressure-time curve, 

 and the effect obviously becomes increasingly serious if one attempts to 

 measure relatively weak pressures following a much stronger pressure 

 which has had full effect on the gauge cable. 



The errors resulting from cable signal may be so serious as to spoil 



