226 Mr. W. M. Mordey. On Slow Changes [Jan. 17, 



Siemens electro-dynamometer. These were instruments in regular 

 use for manufacturing purposes and used in fixed positions. The 

 readings of watts and volts are reliable within about 0'5 per cent. 

 The current readings are less accurate, as the range of the instrument 

 was not very suitable. The source of current was a 37-kilowatt 

 alternator of the author's type, working at 100 periods per second, 

 and at a high E.M.F. The low E.M.F. required for the tests, from 

 20 to 60 volts, was given by the secondary of a transformer, the 

 primary of which was connected to the alternator. Variations of 

 E.M.F. were obtained by adjustment of the field excitation of the 

 alternator. 



There are irregularities in the tabulated readings which are 

 probably not entirely due to errors of observation, but to differences 

 of the conditions under which the tests were made. For example, the 

 iron when removed from the oven was allowed to cool, but the exact 

 temperature was not taken, and it certainly was not the same in all 

 cases. In August it would probably be higher than in December. 

 And although the alternator used to supply the testing current was 

 the same in all cases and was run at the same periodicity, there were 

 differences of condition of the circuit and load which may have had a 

 slight influence on the results. 



Sample No. 1. This consisted of a block built up of stampings 

 arranged as in Fig. 1, which shows section and plan, the outside 

 dimensions of the plates being 16J ins. by 11 ins. The sheets were 

 0'014 in. (0*354 mm.) thick, varying slightly; they were built up to 

 a certain thickness with 100 stampings separated by paper, the total 

 weight being about 55 Ibs. The winding is shown at C. 



Samples Nos. 2 and 3 were made up exactly the same as No. 1, 

 but were taken from different supplies of iron. 



They were occasionally removed from the oven, allowed to cool, and 

 then returned to the oven after a test had been taken of the loss of 

 energy with a given magnetisation. 



The results obtained from these three samples are given respec- 

 tively in Tables 1, 2, and 3, and graphically in Diagrams 1, 2, and 3. 

 For ease of comparison the watts are plotted as percentages in all 

 the diagrams. 



The magnetising coil in each of these cases consisted of 100 turns 

 of copper strip 0*16 in. by 0'09 in., having a resistance of 0'15 ohm, 

 or so low as to make the C 8 R loss negligible, as compared with the 

 loss in the iron. For example : with the maximum current used, 

 07 ampere, C 2 Ji = 0'0735 watt, the total loss being 29'3 watts. 



In the foregoing examples the iron magnetic circuit is not quite 

 continuous : it is interrupted in part by butt joints and in part by 

 lap joints. 



Sample No. 5. In order to eliminate any errors that might con- 



