142 K. E. Guthe — Measurement of Self-Inductance by 



For great sensitiveness we need a rather large condenser. The 

 absorption, as was also mentioned in Rowland's paper, has a 

 very disturbing influence. In a former paper* I pointed out 

 also the variation of the apparent resistance. The Stanley 

 condensers are remarkably free from these faults. 



The work with the two methods referred to showed indeed 

 that they are superior to others and especially Rowland's 

 method 6 gave good satisfaction. In this the capacity and the 

 movable coil (r) in parallel with a non-inductive resistance (R /7 ) 

 are placed in series with a parallel arrangement of the self- 

 induction and stationary coil (R) with another non-inductive 

 resistance (R'). 



£=(R + B')<R'+r). 



On account of the large self-induction of the coils of the 

 electrodynamometer, which is always measured in addition to 

 ^, difficulties arose, when we attempted to measure coils with 

 small self-inductance. The third method I devised enables us 

 to measure these very accurately. It is a simple application of 

 the above mentioned rule. 



I use a two-phase generator, consisting of a stationary 

 Gramme-ring, inside of which a two-pole field-magnet rotates. 

 This generator has been designed by Professor Carhart and is 

 described by Carhart and Patterson. f It gives two currents 

 differing in phase by 90°. The E. M. F. of the generator can 

 easily be adjusted to any desired value up to 60 volts by vary- 

 ing the current through the movable field magnets. One of 

 the currents we send through the stationary, the other one 

 through the movable coil of the electrodynamometer. By 

 inserting resistance in series with the stationary coil, a balance 

 is easily obtained. We connect then the coil, whose induc- 

 tance we wish to measure, in series with one of the branches 

 and add resistance, until the balance is again obtained. By 

 using a standard induction we can determine the increase in 

 resistance necessary to balance a given increase of self-induc- 

 tance. For any given frequency, slight variations of which do 

 not appreciably affect the result, the formula is 



L i: L 2 ::R i: R 2 



To show the delicacy of the method I add two sets of observa- 

 tions, taken on different days. 



* Electrical Engineer, Sept. 16th, 189*7. 



f Electrical Measurements, p. 114, or Phys. Rev., iii, p. 141, 1895. 



