A NEW CURRENT WEIGHER, ETC. 481 



each coil consists of a whole number of turns, there was no necessity to rotate the 

 cylinder from the time the drilling of the first radial hole was commenced to the 

 completion of the last. 



An estimate was made of the accuracy with which the number of turns is known. 

 On the fixed cyclinders there are 90 turns to each coil and the diameter is about 

 33'0 centims. From observations on the radial holes, the number of turns is con- 

 sidered to be correct within 2 parts in 1,0.00,000. 



Between the inner orifices of the passages b and c, fig. 8, a short V-groove e, % inch 

 deep, was cut, and between the corresponding apertures of a and d a groove f f, 

 -Jfi inch deep, was made ; in these grooves portions of the leads of the coils were laid. 



The copper wire with which the four coils were wound was supplied by the 

 London Electric Wire Company, Limited, on bobbins of the same diameter as the 

 cylinders. It is hard-drawn bare No. 24 S.W.G., and has a conductivity such that 

 1 metre weighing 1 gramme has a resistance of 0'149 ohm at 15 0- 5 C. The mean 

 diameter of the wire is 0'559 millim. ; this is the average of several hundreds of 

 measurements, the maximum variation being 1 per cent. 



As a guide in winding, an arm was fixed to the saddle and tool carriage of the 

 lathe which supported the bobbin and a small grooved brass pulley over which the 

 wire passed on its way to the cylinder. At the commencement the pulley was set in 

 position for a straight feed and the tool carriage was placed in gear with the leading 

 screw. On the axle carrying the bobbin a grooved pulley was fixed, and around this 

 a rope passed ; one end of the rope was attached to a spring balance fixed to the 

 lathe saddle, and the other end was tied to a heavy weight which just swung clear of 

 the floor. The effective load on the wire during the winding of the coils was 10 Ibs., 

 which resulted in an extension of 0'16 per cent., the limit of elasticity not being 

 exceeded. YOUNG'S modulus for the material of*the wire was experimentally deter- 

 mined as ri 6 x 10 l> (C.G.S. units). The coefficient of linear expansion of copper is 

 l'7xlO~ 4 ; hence for an increase in temperature of 80 C. the expansion is 0'14 per 

 cent. When the coils were immersed in paraffin wax the temperature of the copper 

 was very nearly 100 C., but the wire appeared to be quite taut on the cylinders. 

 The reason for this is apparent. 



In the case of the fixed cylinders the winding was commenced by threading a free end 

 of the wire from the outside through the hole c and back through b, fig. 8. When a 

 few centimetres had been pulled through the nipple, it was passed through the slit 

 therein and pressed back towards the surface of the marble; it was then given a 

 couple of turns about the nipple and soldered to it. The wire was afterwards pulled 

 taut and the necessary bends made to commence the winding. The position of these 

 bends was estimated beforehand and the wire in the vicinity softened over the flame 

 of a spirit lamp. During the winding the cylinder was rotated very slowly and 

 stopped after each revolution for a couple of measurements to be made of the diameter 

 of the wire. On its way from the bobbin to the marble the wire passed between two 



VOL. ccvii. A. 3 Q 



