374 Dr. Norman Campbell : Experiments on 



with the primary, and by different arrangements of the 

 armatures different couplings could be obtained. The 

 various combinations will be represented by symbols such 

 as 10,000/0 or 8000/4000 ; the first number gives the 

 number of secondary tarns coupled with 160 primary turns 

 in housing A, the second the number of secondary turns 

 uncoupled with the primary in housing B. 



Preliminary Measurements. 



13. The inductances of the circuits were first measured 

 at a frequency of 100 cycles per sec. hy determining the 

 impedance and the ohmic resistance; the primary inductance 

 was determined with a current of 0*5 amp. (which was used 

 in all subsequent measurements), the secondary with 100 volts 

 between the terminals. The inductances were measured with 

 the armatures both in housing; A and in housing B, for most 

 of them were sometimes used in one housing and sometimes 

 in the other. The following results were obtained : — 



Table I. 



L 2 . 



Armature. L 1# ( A ^ 



In A. In B. 



No. 1 000664 henry. 28'6 henry. 



No. 2 0-00691 22-7 23*9 henry. 



No. 3 000688 11'9 138 



No. 4 — 6-32 



No. 5 — 1-68 



An attempt was made to measure the mutual inductance 

 L 21 in Nos. 1, 2, 3, by determining the voltage across the 

 secondary when an alternating current of known magnitude 

 and frequency was passed through the primary. But from 

 the values so obtained the calculated value of the coupling- 

 was greater than 1. This discrepancy is only an expression 

 of the well-known fact that this method of "determining 

 a mutual induction is only applicable strictly when the 

 circuit is non-magnetic, but the failure of the method makes 

 it difficult to find a simple method for determining the 

 mutual induction in the conditions of current and frequency 

 which obtain in the measurement of the self-inductances. 

 Accordingly, no further attempt was made to determine 

 the mutual inductance, but it was clear that the coupling 

 was not very different from unity. 



