Transformer Indicator Diagrams. 175 



determined, C l is deflected through 6 1? The mutual induct- 

 ance of p l and s was •0000485 henry, while that of p 2 and s 

 was "000291 henry, which bear the same ratio to each other 

 as n^ to n 2 , that is as 1 to 6. 



The points marked F in fig. 1 represent the terminals of 

 the single loop of" insulated wire wound round and close to 

 the transformer core. When the flax-wave is being obtained 

 these are connected direct to the commutator, and thence 

 through a suitable resistance to the galvanometer. 



By means of the key K either the flux leads or those of 

 any of the secondaries of the different mutual inductances 

 can be connected with the commutator so as to obtain the 

 corresponding wave-trace. 



A glance at fig. 1 will show that the true primary current 

 is less than the measured Cj by the small current by which E : 

 is determined. To correct for this (if correction is necessary) 

 we subtract from each galvanometer Q^ reading an easily 

 determinable fraction of the equi-phase E L galvanometer- 

 reading. A similar correction has to be applied to the C 2 

 readings ; but in this case the corrections have to be added. 

 The galvanometer-readings for » 1 C 1 + n 2 C 2 have also to be 

 similarly corrected, and in this case the corrections are of 

 great importance as the C\ correction is affected by the 

 factor ??! and the C 2 correction by the factor n. 2> and both 

 corrections are approximately in the same phase. 



5. For the present paper three separate sets of connected 

 transformer quantities were determined, and they will be 

 given below by their harmonic expressions and also repre- 

 sented by curves or indicator-diagrams. 



These are for the transformer : — 



(1) At no load. 



(2) At (q.p.) full non-inductive load. 



(3) At (q.p.) full inductive load. 



The method by which the wave-tracer galvanometer-readings 

 are reduced to absolute measure and the harmonic expression 

 for the periodic quantity deduced from them has been fully 

 explained in former papers *. 



In the present experiments, for each of the periodic quan- 

 tities determined, 30 wave-tracer galvanometer-readings were 

 taken, each differing from the next in order by 12°. The full 

 wave was thus covered. Corresponding deflexions in each 

 half of the wave were added, that is the 1st and 16th, the 



* "Ware-Tracer and Analyser," Phil. Mag. Nov. 1903. "Variation 

 of Magnetic Hysteresis with Frequency," Phil. Mag. Jan. 1905. " Expe- 

 ditious Practical Method ot* Harmonic Analysis," Phil. Mag. Jan. 190o\ 



