26 ARMATURE CONSTRUCTION 



Sheets of this latter quality are stated by these authors to be 

 obtained by treating steels containing a small percentage of 

 aluminium or silicon (2 to 4 per cent.). This superior material 

 can only be obtained at a very substantially higher price as com- 

 pared with the ordinary material. It is, moreover, not so necessary 

 in armature construction to obtain an exceedingly low iron loss, 

 as it is in transformer construction. 



The property of high specific resistance, while it does not, of 

 course, affect the hysteresis loss, affects the other component of 

 the iron loss, namely, the loss due to Foucault currents ; and hence 

 progress in this direction is of great importance. 



The writers above alluded to, in describing the conditions con- 

 trolling French practice, state the customary composition of 

 armature plates to be an extra-soft steel containing about 0'03 

 per cent, of carbon, and often small quantities of silicon or alu- 

 minium. It would appear that, in France, sheet iron for armature 

 construction is most generally delivered in plates 0'80 metres wide 

 by 1*65 metres long. Plates as wide as 1*2 metres may be 

 obtained, but the price will be some 30 per cent, to 35 per cent, 

 higher than that of plates of the former dimensions. 



Eef erring to plates of a thickness of 0'4 mm. (apparently the 

 most customary thickness employed in France), Brunswick and 

 Aliamet state that Class I. l plates of a width of 0'85 metre cost 

 about 15 per ton, as against about 20 per ton for widths of T20 

 metres. 



Allusion has been made above to the circumstance that 

 instruments, such as those of Ewing and of Blondel, only indicate 

 the order of magnitude of one component of the iron loss, giving 

 no indication of the other component loss due to Foucault currents. 

 The quality of the iron as regards Foucault current loss may be 

 readily ascertained by measuring the resistance of strips of deter- 

 mined dimensions. It is desirable to have as high a specific 

 resistance as practicable. 



In Germany, it has been deemed preferable to employ methods 

 of measurement of the quality of sheet iron which give the total 

 iron loss i.e. methods which do not distinguish between the 

 hysteresis loss and the Foucault current loss. The Kichter 

 apparatus, already described on p. 16, is an instance of this 

 practice. Another instance is afforded by the Epstein apparatus, 

 which, in fact, is the apparatus to which specific allusion is made 



1 Class II. plates at present cost from two to three times as much as 

 Class I. plates. 



