ELECTROMOTIVE FORCES OF THE ELECTRIC ARC. 



317 





Top 

 Carbon H( 'Icier 



was decided to make the resistance as small as possible and to allow the temperature 

 of the strips to rise considerably, which necessitated slightly modifying their design. 

 The resistance consists of 12 platinoid strips, each about 170 centims. long, 2 '5 centime, 

 wide, and 0'076 millim. thick. Each strip is folded back on itself and has its ends 

 soldered to two brass blocks let into the top of the frame, and is stretched tight with 

 a tension of about 4 Ibs. by means of a brass spring attached to a small glass tube, 

 about 5 millims. diameter, at the bottom of the loop formed by the strip. Between 

 the up and down sides of the strips is placed a sheet of asbestos millboard, about 

 - 3 - 2 - inch thick, and the strips are pressed together against this by glass rods from 

 side to side of the frame. 



The resistance of each of the 12 strips was roughly adjusted to 0'5 ohm, and the 

 strips could be used in series or parallel by connecting up the brass blocks forming 

 the ends of the strips with copper links and set-screws as required. Owing to the 

 considerable heating of the strips by the current, their resistance depended on the 

 current; thus the resistance of all the strips in series, which was (3 '00 ohms with 

 1 ampere flowing, rose to 6'25 ohms with the current 

 of 10 amperes which was used in many of the experi- 

 ments. For tin's reason, and because it formed a 

 check on the instruments, the resistance of II was 

 determined during each experiment from the known 

 values of the direct current and P.I), between its 

 terminals, and the value so obtained was used in 

 calculating the results. 



The apparent self-induction of 11, including the 

 connection between it and the arc lamp already 

 mentioned, was measured by comparing it with a 

 non-inductive resistance put in place of the arc. 

 This latter resistance (see fig. 4) was made to imitate 

 an arc possessing non-inductive resistance localised 

 between the ends of the carbons. Tt consisted of 

 158 millims. of No. 38 platinoid wire bent back on 

 itself, the two extremities being soldered to the ends, 

 previously copper plated, of two solid " Conradty 

 Noris" carbons; these carbons were held in the 

 carbon holders of the lamp so that the resistance wire 

 occupied the position the arc would when burning. A piece of mica was interposed 

 between the ends of the carbons which served to keep the loop of wire taut. This 

 method of determining the correction to be applied to 11 really converts the test of 

 the arc into a substitution test, for having determined how an ordinary metal 

 resistance behaves when localised between the carbon tips, the behaviour of an arc 

 substituted for it under exactly similar conditions was compared with it. 



Mica _._V-- ' 



FT" 



\Vir<- 



Ikittom 

 Carbon lio 



Fig. 4. 



