110 



♦ KNOWLEDGE ♦ 



[Auo. 8, 1884. 



conjunction with another, declaring that " The strength of 

 an electro magnet is proportional to the number of turns of 

 •wire." Regarding these two laws together, it is clear tliat 

 the thick wire coil has the greater effect, because, while 

 there are twice the number of turns of the thin as com- 

 pared with the thick wire, there may be said to be 

 three times the current circulating through the thick as 

 compai'ed with the thin wire. Were the two currents 

 made equal by modifying the battery power, then the 

 thin wire would be twice as effective as the other. On 

 the other hand, if while the conditions of the thin coil 

 remained constant, the thick wire could by any means 

 be made to encircle the core twice as often (the current 

 being kept constant) then tlie threefold curreut circu- 

 lating in the thick wire would exert three times the 

 magnetising itifluence upon the core. 



Even these laws, however, are not always applicable, 

 because there is a limit to which iron or any other sub- 

 stance is capable of being magnetised, and the fact of there 

 being such a limit implies that the above laws may onlj' 

 V>e regarded as applicalile when the effect to be produced is 

 considerably below this limit, which is called the " saturation 

 point." 



It was just now stated the magnetic strength varies in 

 proportion to the number of turns. Supposing that the 

 iron core were covered, and that no room could be found 

 for another turn, how arc we to increase that number 1 It 

 might be done by coiling a wire over tlie previously wound 

 coil, but what about the direction 1 It is evident that if 

 we send a current round the core through a left-handed 

 helix (as in Fig. 1), so as to make the end N a north-pole, 

 and then .send a current through an outer coil in such a 

 manner as would induce a south-pole at N.then no polarity 

 will be manifested unless the inducing force of one helix 

 exceeds that of the other, in which case tlie magnetism 

 induced will be a measure of the difierence between these 

 forces. 



It is, however, very easy to calculate the inductive effect 

 that would be produced by any particulai cod of wire. 

 Suppose, for example, that we are dealing with the arrange- 

 ment depicted in Fig. 1, and that we wind a second left- 

 handed helix over the core, then join a wire from the S 

 end of the inner coil to the N end of the outer one, so 

 that the current is made to travel in the direction indicated 

 by the arrows through each coil in succession. Neglecting 

 the .slight increase in the length of wire involved in wind- 

 ing one coil over another, we shall get an external circuit 

 having twice the resistance of the iuner coil or (with the 

 two cells). 



_E 4 



R + r ~ 2-f 2 - ^ 



The current is thus 1 ampere as compared with 1 3, but 

 as it passes twice as often round the core, it may be said 

 to have an inducing power of 2 as compared with l.'; 

 exerted by the single coil. Here, then, au advantage is 

 clearly gained. 



Supposing, however, that instead of connecting the two 

 coils in series, we join them in parallel circuit — that is, by 

 connecting the two N ends together, and likewi.se the two 

 S ends. Then the current will divide at N, and, the 

 resistance of the coils being identical, the current will 

 divide equally, to re-unite at S. But the joittt resistance 

 of two equal wires is only half that of one of them, or their 

 conductivity (the reciprocal of resistance) will be doubled. 

 This principle has been more than once enlarged upon, so 

 it need not further take up time here. The simple equa- 

 tion becomes 



_i =16. 



i + 2 



1 6 amperes do not, however, go through each coil, but the 

 current halving itself at N becomes only IS in each. Mani- 

 festly, then, a curreut of 8 ampere through a certain 

 number of turns can only have the same eflect as a current 

 of ro ampijres through half that number of turns. 



In other words, the eflect jiroduced is the same as would 

 result from a single coil having the same number of turns 

 as before, but with twice the conductivity or half the 

 resistance. This could not be obtained with the same kind 

 of wire, becau.se to have half the conductivity it would 

 require twice the weight of metal. Of course, it would be 

 possible to gain this end by using rectangular wire whose 

 thickness in one direction is equal to the diameter of the 

 original round wire. But even then the trouble involved 

 would not be recompensed by the gain in inducing power, 

 which, as pointed out, is only 16 as compared with lo. 



Under the circumstances, then, the best eflect is jiroduced 

 when the current is made to traverse the coils successively. 

 But, on the other hand, to wind a number of individual 

 layers, and then to connect them by longitudinal lengths of 

 wire is out of the question, more especially as the desired 

 effect may be more easily jiroduced without such an 

 arrangement. 



Fige. 1 and 2. 



By examining Figs. 1 and 2, the method to be employed 

 may be readily made clear. Let us start with a layer 

 wound, as shown in Fig. 1 — that is as a left-handed helix — 

 and let the current travel from left to right, or from N. to S. 

 Now it has been stated (and Fig. 2 will illustrate it) that a 

 current passing through a right-handed helix induced a 

 north at its exit. Suppose, then, that instead of travelling 

 from left to right, the current pursues the opposite direc- 

 tion, or from right to left. Then the end marked N 

 becomes a south pole, and the end marked S a north pole. 

 The effect thus produced is the same as that resulting from 

 left to right through a left-handed helix. If a right-handed 

 helix be wound over a leftrhanded one, and the adjacent 

 extremities, say on the right-hand side, be connected to- 

 gether, then the current will jiass from left to right through 

 a lef1>handed, and from right to left through a right-handed, 

 helix, when the inducing effect of each coil will harmonise, 

 and a north pole will be produced on the left-hand side, and 

 a south pole on the other (as in Fig. 1). 



It is scarcely necessary to say that the resistance of a 

 wire is not affected by the direction of winding, consequently 

 the current flowing will be 



^=1 



•2 + 2 



which (as may be gathered from what has been said above) 

 will produce a magnetising force of 2 as compared with 1 '3 

 resulting from either of the coils used singly. If the 

 student takes a piece of wire or .string in his hand and 

 winds it round a rod so as to make a left handed helix, and 

 when he reaches the end of the rod continue the winding 

 back to the other end without bending the wire back on 

 itself, he will find that the second layer becomes a right- 

 handed helix. If, furthermore, on finishing the second layer, 

 he winds back again for a third layer, he will find that this 

 is a left-handed helix. So, continuously winding, he gene- 

 rates left and right-handed helices alternately. Every 



