252 Mr Peake, A Novel Instrument for Illustrating 



rigidly attached to the turn-table, while the other, the B scale, is 

 fastened to the back of the stand. A strip of looking-glass is 

 provided to avoid parallax in reading the position of the index ; 

 this index consists of a fine wire at the end of the aluminium 

 pointer, stretching across both scales. A damping vane is attached 

 to the end of the rod supporting the control weight, and generally 

 is immersed in an oil dash-pot. 



Method of Operation. 



The stand is levelled by means of the levelling screw in the 

 base ; this is best done by putting the cradle in position without 

 any iron in it, and then turning the levelling screw until the index 

 points to the zero of the B scale ; the zero of the H scale should 

 then be brought into line with the pointer, and the instrument is 

 then ready to receive the specimen. After the insertion of the 

 sample in the cradle, the latter is replaced in position, and if the 

 sample is not magnetised the index will again be over the zero of 

 both scales, the plane of the specimen will be vertical, and normal 

 to the lines of force due to the permanent magnets. Then by 

 gently turning the turn-table, avoiding jerks which would cause 

 the pointer to oscillate, the specimen may be subjected to any 

 desired cycle of magnetising force, and the corresponding readings 

 on the B scale may be observed. The whole time required to 

 take the observations for a complete hysteresis cycle is but a few 

 minutes. 



Demagnetising of Sample. 



To demagnetise a sample of iron that may have been magnet- 

 ised previous to testing, all that is necessary is to place it in 

 position with the oil dash-pot removed, and start it swinging ; this 

 demagnetises the sample by reversals, and it soon comes to rest, 

 owing to the damping action of the hysteresis loss. 



It was found by experiment that the BH curve obtained by 

 slowly increasing H after demagnetisation was the same as that 

 obtained by reversing several times successive values of H, and 

 hence, the former method being much more rapid, it was the one 

 used in obtaining the "reversal curve." 



Calibration of Instrument. 



The calibration of the scales of the instrument has not been 

 carried out with any great refinement, as the instrument is in- 

 tended for illustrating principles, and is not suitable for deter- 

 mining quantitatively the actual hysteresis loss in the samples 

 tested ; nevertheless it was thought advisable to obtain approxi- 

 mately the value of the scale divisions in ordinary units, and also 

 to determine the reduction in the magnetising force, resulting 

 from the actual production of magnetic induction in the specimen. 



