of the Coefficients of Induction. 461 



smaller magnetizing forces than those used above, and there- 

 fore for fields very much weaker than that due to the earth's 

 magnetism. 



Coefficients for large Currents. 



8. The coefficient of self-induction of a coil with an iron 

 magnetic circuit can far more easily be obtained for strong 

 magnetic forces when the magnetism is maintained by an 

 independent magnetic field than when the magnetism of the 

 iron has to be excited by the current through the coil itself. 

 In a transformer, for instance, where there are two coils 

 identically situated with respect to the iron circuit, the 

 coefficients of self and mutual induction are in the ratio 



n\ : n\ : n 1 w 2 , 



where the ratio of the turns on the first coil to that on the 

 second is n x : n 2 . This will be true whatever the state of 

 magnetization. If, therefore, one coil is used to excite the 

 magnetism, and the other to test it, the values of the coeffi- 

 cients can be obtained without the least difficulty. Again, it 

 is easy to obtain by Maxwell's method the value of L for the 

 armature of a Series dynamo when the field-magnet coils are 

 used independently to excite magnetism in the armature. 

 Experiments were made in February 1887, at Prof. Ayrtoir's 

 suggestion, to determine the coefficient of self-induction of the 

 armature of a Gramme dynamo of the A pattern. The 

 numbers obtained by Mr. S. Watney and the writer were as 

 follow : — 



Amperes round 

 field-magnets. 



Self-induction, 

 in secohms. 



00 

 61 



151 



24 



29 



•0218 

 •0179 

 •0135 

 •0122 

 •0117 



9. When, however, it is desired to determine the self-induction 

 of the field-magnet coils of a dynamo when a strong current 

 is flowing through them, several difficulties present themselves. 

 If strong currents are to be maintained the resistances must 

 be small, unless a large amount of power is to be wasted and 

 unless very high electromotive forces are procurable. On the 

 other hand, the self-induction to be measured is very large. 

 The time-constant, which determines the rate at which the 

 magnetism will change, will therefore be large ; and unless a 



