136 Mr. W. L. Upson : Observatioas 



Arcs in Coal Gas. 



Ou-C. The arc was maintained with much more difficulty 

 than in air, and required larger currents. Great quan- 

 tities of carbon were deposited on the copper so quickly 

 that it was impossible to measure accurately the arc- 

 length. The arc was small and dull in appearance. 



Cu-Cu. At about 28 volts and 8 amperes with an arc- 

 length of approximately 01 in., a discharge took place, 

 which in a moment went out. It could not be made con- 

 tinuous even with the smallest gap between the electrodes. 

 On continuing the discharge for ten or fifteen minutes the 

 arc suddenly 'started and remained steady to a length of 

 0'05 in. The current dropped considerably below 8 am- 

 peres, the arc continuing. The current was raised to 

 11 amperes and the arc burned quietly at 0*1 in. length, 

 while a heavy deposit formed on both poles much as in 

 the C-C arc in H. What really happens seems to be 

 this : — With clean copper poles no arc will exist at cur- 

 rents up to 15 amperes and higher, as in hydrogen. But 

 with a continued spark discharge carbon is separated from 

 the coal-gas and deposited upon the poles. When there 

 is sufficient carbon, the arc starts, and we really have a 

 carbon-carbon arc. There was some copper deposit 

 mixed with the carbon on the electrodes, but the latter 

 were only very slightly burnt by the arc. 



In general, it may be inferred that coal-gas is unsuitable 

 for investigation on the actions of arcs between metal elec- 

 trodes, for in a moment the carbon deposit makes the arc 

 virtually one between carbon electrodes. The arc itself 

 resembles those in hydrogen in brilliancy. 



Maximum Arc Lengths. 



When the arc took the form of a spark-discharge, as with 

 various metal-metal arcs in hydrogen, a maximum possible 

 length of 0*05 in. was reached for all such arcs, provided 

 both terminals were not of the same metal. This limit of 

 0*05 in. was with 110 volts supply and current up to 

 15 amperes. When the terminals are of the same metal 

 the results vary. Copper and iron give no definite length 

 of arc, and aluminium reaches its limit at 0'05 in. When 

 carbon is one of the electrodes we have always a continuous 

 arc. If carbon is negative there is a constant limit of 0'07 in. 



