Magnetic Hysteresis loss in Straight Iron Strips. 267 



of different lengths and sections, and the results are collected 

 and represented in the form of curves in the diagrams in 

 Plate V. 



The upper curves in each case in the figs. 1, 2, and 3 in 

 Plate V. represent the variation of B along the half-length of 

 the bar, and the lower curves the corresponding curves for 

 the variation of B 1 ' 6 . The first pair of diagrams represent the 

 result of varying the maximum induction density in the centre 

 of a bundle of strips of iron of the same length and section. 



The second pair of diagrams give the curves showing the 

 same results for bundles of iron strips of varying lengths but 

 the same section. 



The third pair for various samples of iron strips of the same 

 length but different section. 



It is clear from these diagrams therefore that such a bar or 

 bundle of iron strips placed in an originally uniform magnetic 

 field has a non-uniform induction density produced in it at its 

 various points, which is symmetrical about the centre ; and a 

 non-uniform hysteresis loss per unit of volume also symmetrical 

 about the centre. The maximum values of each being at the 

 centre of the bar. 



If we take the true mean value of the varying distributed 

 hysteresis loss per unit of volume, it is obvious that at some 

 point in the half-length of the bar there must exist an actual 

 induction density B x such that the true mean hysteresis loss 

 in the whole bar is proportional to Bx 1 ' 6 . Let this value of 

 the induction density be called the effective value, and the 

 point in the bar at which it occurs the effective point. Then 

 the effective value Bj must be equal to the l"6th root of the 

 mean of the l*6th powers of the actual induction densities 

 taken at equidistant points all along the bar, or 



where m . B 1 ' 6 stands for the mean ordinate of the curve 

 representing the varying values of B 1 ' 6 all along the half- 

 length of the bar. 



If then this effective value B x of the induction density is 

 calculated for each of the samples of iron strip used, and the 

 position at which the actual induction has this effective value 

 is marked off on the half-length of the bar (this position is 

 represented by the cross on the curves given in Plate V.), we 

 have noticed the remarkable fact that whatever the length or 

 section of the iron strip may he, the point in the bar at which the 

 actual induction density has a value equal to the effective value, 

 always comes at the same proportioned distance from the centre 



