PEIRCE. — CHANGES IN INDUCTANCES OF ELECTRIC CIRCUITS. 559 



I 



<D 







are represented as stretched over four small pulleys to keep them taut. 

 I have found that this model made of three weighted roller skates, 

 moving over a level table top, and connected in 

 the manner indicated by cords passing around /"^ /^"^ 

 such small cheap pulleys as are obtainable at any \.I>' \Ly 

 ironmonger's shop, may be made to work ex- 

 tremely well. The effects of sudden changes of 

 inductance can be directly observed by dropping 

 suitable masses into the skates as they move. In 

 Figure 14, which illustrates the same problem, the 

 mass of Q is 31, and those of P and Q are Zi — 31, 

 1/2 — 3J, respectively. Q should have no vane. 



Scores of other models, more or less simple of 

 construction, can easily be devised. It is to be 

 noticed, however, that in some of the models which 

 have been used to illustrate this problem, the 

 masses representative of some of the combinations 

 of the inductances would need to be negative if 

 they were to correspond to cases which occasion- 

 ally arise in electrical engineering. 



If either of the two neighboring circuits con- 

 tains no battery, the corresponding value of E in Figure 18. 

 equations (24) is to be put equal to zero. Fig- 

 ure 19 is drawn for the case of an induction coil without iron and 

 with no cell in the secondary circuit. The self-inductances of the two 

 circuits are equal. The dotted curve, P, shows the form of the current 

 induced in the secondary circuit when the primary circuit, which has 

 been carrying a steady current, is suddenly broken. If, after a few 

 seconds, the primary circuit containing its battery be closed again, the 

 current in the secondary circuit will have the general form of either Q 

 or S. Q, R, and S are drawn for mutual inductances respectively half 

 as great, nine tenths as great, and equal to, the inductance of either 

 circuit. P is drawn for 31= L/2, and corresponds, therefore, to Q ; 

 the areas V and W are equal. Curves like P corresponding to R and 

 S could be found by exaggerating all of P's ordinates in the ratio 9/5, 

 or the ratio 2. 



Figures 20, 21, 22, 23, and 24 illustrate some phenomena which are 

 frequently encountered in the practical use of neighboring inductive 

 circuits. The curves have been drawn to scale for certain numerical 

 values of the resistances, and the inductances so chosen as to make 

 the results typical. There is no iron in either circuit, and only one 

 circuit, the primary, contains a battery. 



