CONTACT PHENOMENA IN TELEPHONE SWITCHING CIRCUITS 47 



reaching a current peak usually ranging from 0.5 to 2 amperes. The 

 first cycle of the oscillation is distorted by the higher resistance of the 

 path to ground caused by the arcing stage in the reclosure. After a 

 few microseconds the contacts are opened a second time by the con- 

 tinued motion. Occasionally they reclose a second time but they 

 usually stay open until the voltage has built up by the continued dis- 

 charge of the load relay inductance to a value between 300 and 350 

 volts. Then a spark occurs at what is usually considered the minimum 

 sparking potential between contacts in air. 



Figures 5 and 6 show the voltage and current of the initial opening 

 and reclosure of the contacts at the start of a "B" transient. The 

 brief arc at initial opening is barely detectable in Fig. 5. Figures 7 and 

 8 show similar voltages and currents at an increased sweep speed. 

 In Fig. 7 the metallic arc established during the reclosure is plainly 

 evidenced by the collapse of the voltage to about 15 volts and its 

 maintenance at this value for about a microsecond before it drops to 

 zero. The effect of the arc in distorting the oscillating discharge of the 

 current from the line wire is evident in Fig. 8. The current oscillation 

 of Fig. 8 may be duplicated merely by charging the line wire to a suit- 

 able voltage through a high resistance and closing the contacts, the 

 far end of the line being grounded through the load relay and a large 

 condenser which replaces the usual battery. 



It is likely that the point discharge precedes the arc on reclosure by 

 such a short time that it cannot ordinarily be resolved. Nevertheless 

 disturbances of the voltage and current are occasionally found which 

 seem to indicate that a discharge path formed and was checked (pos- 

 sibly by melting off the point) without establishing an arc or metallic 

 bridge. Such a disturbance of the rising voltage is indicated in Fig. 9 

 by a high-frequency oscillation about 5 microseconds after the first rise 

 of the voltage trace. Figure 10, which shows the current of the second 

 of two initial reclosures, indicates a similar phenomenon. Five micro- 

 seconds after the rupture of the circuit, shown by the downward 

 deflection of the zero line, a dim line upward records a current surge 

 lasting a fraction of a microsecond and reaching about 3/4 ampere. 

 This surge, however, did not result in the immediate formation of an 

 arc which was established about 5 microseconds later. 



The initial separation of the contacts does not always result in a 

 metallic reclosure. Figure 11 shows the voltage of the early part of the 

 "B" transient. Here the first collapse of the voltage is a sparkover 

 from about — 300 volts which establishes an arc at about — 15 volts. 

 This arc is broken and, as the line is not completely discharged, the 

 voltage between the contacts rises to about + 140 volts; a second arc is 



