1034 THE BELL SYSTEM TECHNICAL JOURNAL, SEPTEMBER 1954 



Fig. 6 — Oscilloscope trace of the induced voltage in the secondary winding 

 while a sfiuare current pulse 8m sec long was applied to the primary. The voltage 

 spike at the beginning is not completely understood, but is thought to be con- 

 nected with the fact that the domain wall spikes associated with imperfections 

 do not pull back on the domain wall until it has moved a little distance. In this 

 sample, once this initial peak is over, the wall velocity comes to its steady state 

 value and is not much disturbed by imperfections, so that the observer need do 

 no averaging. Ih this case, the wall was moving with a velocity of 3590 cm/sec. 

 At this velocity, the wall was unable to reverse the magnetization in 8/i sec, so the 

 signal ends as the magnetic field goes to zero at the end of the pulse. After the 

 pressure on the wall due to the magnetic field stops, the domain wall spikes asso- 

 ciated with imperfections pull the wall back slightly, giving rise to the voltage 

 spike in the opposite direction. 



The applied field due to the prunary pulse is deduced from the cur- 

 rent in the primary winding (measured by observing the voltage across 

 the series resistor) using the solenoid formula H = ^NI. To obtain the 

 relation between wall velocity and induced \oltage per secondary turn 

 we have: 



Volts/turn = {d^/dt) X 10~* = ^'K^L{^z/ M)w^-,n X 10~^ (l) 



where {Az/ M) is equal to the domain wall velocity v, and it\vau is the 

 width of the wall between the boundaries of the sample in the direction 

 perpendicular to the direction of magnetization. It is in deriving (1), 

 of course, that we use our detailed knowledge of the domain pattern in 

 the sample. 



We are thus able to obtain a ^•alue of the domain wall \-elocity v for 

 each value of the applied field H. These data are the results of the 

 experiment. 



The value of M^ used in (1) is the measured value (322.5 cgs units/cc) 

 at room temperature. The values used at other temperatures ha^^e been 

 deduced on the assumption that Me varies the same way with tempera- 

 ture in this material as it does in magnetite as measured by Weiss and 



