G54 



Mr. R. Beattie on the Length of tlu 



® 



/ 



© 



break being but slightly affected by secondary actions arising 

 from the production of eddy currents in the core. The 

 experiments were thus carried out under fairly well-defined 

 conditions, especially as the air-gap in the magnetic circuit 

 was kept as large as possible, this hiving the effect of making 

 the reluctance of the magnetic circuit, and therefore the self- 

 inductance of the magnet, approximately constant. 



Uniformity in breaking the circuit was secured by using* 

 the device depicted in tig. 1, which 

 represents merely a simple form of Fig.], 



quick-break switch arranged so that 

 the release of the trigger T allows the 

 indiarubber bands// to pull the rod R, , J l | a t 



and with it the pole P 2 , quickly away 

 from the fixed pole P x . In use this 

 circuit-breaker was placed with its 

 length vertical, and from measurements 

 of the mass of the moving system and ' 

 the tension of the rubber it was calcu- 

 lated that, within the limits of allowable 

 motion, the time taken by the moving- 

 pole to travel a distance I centimetres 

 from the fixed pole would be 



•027 VI 



seconds. As the longest sparks observed 

 in the experiments were about 7 centi- 

 metres in length, their duration was thus 

 '07 second. 



The length of the spark, read on a 

 scale placed behind it, could be deter- 

 mined with considerable accuracy, 

 though sparks from certain materials 

 like lead, tin, bismuth, were more difficult 

 to measure than from others. For each 

 pole-material used, a series of observa- 

 tions were taken with different current- 

 strengths to show the relation ot the spark- 

 current. The variation of the current — from 

 to 15 amperes — was produced by altering the 

 force in the circuit, no change being made 

 which was constantly equal to 3"3 ohms. 



The data thus obtained are given in the accompanying table, 

 and from them the curves in figs. 2-6 have been drawn. As 

 might be expected, these curves exhibit much diversity 



length to the 



1*2 amperes 



electromotive 



in its resistance 



