538 



THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1954 



CONTACT SEPARATION 



Fig. 2 — Initiation of the first arc between contacts on break of an inductive 

 circuit. 



portion be corresponds to the sparking potentials in the atmosphere, 

 ab corresponds to the range of small separations, of the order of or less 

 than the mean free path of an electron in the atmosphere, where the arc 

 is initiated by field emission through the influence of surface contamin- 

 tions or films. As was shown in Reference 2, when the cathode surface 

 was carefully cleaned, the constant field line was not obtained and the 

 arc was initiated at the minimum sparking potential of the atmosphere. 

 It occurred on the sides of the contacts along a path much longer than 

 the minimum separation between the contacts.* 



Lines 0-1 and 0-2 represent the voltage rise at the contact with small 

 and large shunt capacities. Points 1 and 2 are the respective first dis- 

 charge points. In the first case, the arc is initiated at a smaller separation 

 and higher field strength without direct influence of the atmosphere. In 

 the second case the arc is initiated at a lower field strength at the spark 

 potential of the atmosphere. f 



The first arc established may or may not be maintained depending on 

 conditions that are discussed in the next Section. When an arc is inter- 



* With Pd contacts a gross field of 20 X 10^ volts/cm was reached between clean 

 contacts without initiating an arc along the shortest gap. According to the Fowler- 

 Nordheim equation a field of about 50 X 10^ volts/cm is required to give the neces- 

 sary initiatory electrons. It is possible, however, that before such a high field is 

 attained a metal bridge is pulled electrostatically' to short the gap. The electro- 

 static stress is roughly given by 0.5 X 10"^^ F"^ Kg/cm^ where F is the field 

 strength in volts/cm. At F = 50 X 10® volts/cm, the stress is 1250 Kg/cm^ which 

 may exceed the yield stress for the contact metal. 



t The first arc may be initiated at an appreciably lower voltage than predicted 

 by the above static consideration. The first break at the contacts usually follows 

 the explosion of molten bridge drawn between the contacts. Thermionic emission 

 can then furnish the initiatory electrons of the arc. This is only possible, however, 

 if the voltage across the contacts exceeds the ionization potential of the metal 

 atoms before e.xcessive cooling of the cathode has occurred. 



