244 Mr. R. H. M. Bosanquet on Practical Electricity. 



Serrin lamp. 



Revolutions about 900 per minute. 

 Current about 20 amperes. 



Difference of tension between lamp-terminals about 40 

 volts. 

 These are the highest values obtained under favourable cir- 

 cumstances. 



According to these numbers the resistance of the arc is 

 about 2 ohms; that of the machine and connexions is about 1/5. 

 This corresponds to 800 volt-amperes, or rather more than 

 1 horse-power, in the light itself (1 horse-power = 746 volt- 

 amperes). 



We observe here that little more than half the power ex- 

 pended is developed in the lamp. 



Circuits of High Resistance. 



Suppose I take a single British incandescent lamp (150 

 ohms cold, and about 80 ohms hot), the machine only just 

 raises it to a red glow, though a speed of as much as 2000 

 revolutions per minute be employed. The current developed 

 is much less than 1 ampere. 



Here the high resistance kills down the current, and pre- 

 vents the proper excitation of the field-magnets. Hence the 

 advantage in this case of machines in which the magnets form 

 a shunt circuit of high resistance. It is easy to excite the 

 Gramme machine by putting the magnets into a shunt 

 circuit with a suitable resistance; but if a resistance-wire be 

 employed, the power developed in heating it is wasted. I 

 have therefore in many cases adopted the plan of putting 

 useful work into the shunt-circuit of the magnets. This may 

 sometimes with advantage consist of a number of incandes- 

 cent lamps in parallel circuit. Substituting for this a certain 

 number of accumulator-cells, we have the origin of a useful 

 method which I call charging in balanced circuit, to which I 

 shall return. 



Galvanometers. 



The galvanometers I employ were made in the laboratory. 

 They are of a simple character, but quite sufficient for practical 

 purposes. There are three of them. 



Tension Galvanometer. — A circular wooden channel about 

 •11 m. radius, wound with wire of which 1 metre = roughly 

 10 ohms. Resistance = 2900 ohms. This had its constant 

 determined by reading the current from 9 Daniell's cells, 

 which are taken as representing 10 volts. Resistance of 

 Daniell's cells determined and allowed for. For high ten- 



