18 



ARMATURE CONSTRUCTION 



This should be done for each of the samples I. , II., III., and IV., 

 and by taking the average of the four values of the instrument 

 constant thus obtained, the mean constant for the instrument 

 may be derived, and entered at the bottom of Table III. 



TABLE III. 



TABLE FOR RECORDING RESULTS OF DETERMINATION OF CONSTANT 

 OF EWING TESTER. 



Having thus calibrated the instrument, we are in a position to 

 determine the actual hysteresis loss in watts per kg. for any other 

 sample of iron whose quality we desire to determine. After con- 

 ducting a series of tests on any of the samples, the instrument 

 should be again standardised and its constant re-determined, using 

 another table such as Table III. The mean between the constants 

 determined before and after the tests should be taken, and this 

 value employed in calculating the results of tests on the samples 

 tested. The results of tests on various samples should be recorded 

 in tables, for which Table IV. is a convenient model : 



