27-4 Prof. J. A. Fleming on a Method of determining 



In order to put the method to a test, a very careful 

 determination was made by the ballistic method of the 

 hysteresis loss of a sample of the transformer iron strip used. 

 A ring-coil was prepared, and a series of (B-H) cycles of 

 magnetization taken, and a hysteresis loop drawn, and from 

 this loop it was found that for this iron the hysteresis loss was 

 •25 watts per pound per 100 cycles per second, for a maximum 

 value of the induction density equal to 2500 C.Gr.S. units. 



Straight strips of exactly the same iron were then tested 

 by the method here described, which for shortness will be 

 called the " straight coil •" method, and the results of these 

 tests with the iron in different conditions of hardness are 

 shown in the Tables III., IV., and V. (see end). 



It will be seen that the results of the " straight coil " 

 test with the w T attmeter, as given in Table [II., gave exactly 

 the same mean value for the hysteresis loss of the strip 

 when not bent and unbent, as did the calculations made 

 from the ballistic test, viz. : *25 watts per pound per 100~ 

 per maximum induction density of 2500. The difference in 

 the time of taking the tests is, however, immense. A very 

 few minutes suffices to take the wattmeter tests, and the 

 necessary calculations can be made in a comparatively short 

 time. 



Many dozens of tests of irons can be made by the 

 straight coil method in the time occupied by one ballistic 

 test. 



There is another method by which the bifilar wattmeter 

 can be calibrated, but which is. not so generally convenient 

 as that above described. If the shunt and series coils of the 

 wattmeter are joined up in series with each other, so as to 

 form a dynamometer, and if a known current I is passed 

 through these coils, causing a deflexion 6 in the movable 

 coil ; then when the instrument is used as a wattmeter with 

 a total shunt-resistance R (including the added non-inductive 

 resistance) the true power corresponding to a deflexion 9 is 

 PR watts. 



Hence, from a curve showing the deflexion given by 

 various currents, we can at once determine the constant of 

 the instrument, when used as a wattmeter, with given shunt 

 resistance. 



If this standardization is taken with continuous currents, 

 then it is necessary to eliminate the effect of the Earth's field 

 by taking right and left deflexions. The curve connecting 

 square of current strength and the deflexions of the movable 

 coil is nearly a straight line, in the case of the bifilar 

 wattmeter. 



