87*1 THE BELL SYSTEM TECHNICAL JOURNAL, JULY 1956 



due to their spread in velocity, the horizontal sweeping plates are 

 mounted just as close to the aperture as is deemed practical. The ob- 

 served velocity spreads in the beam were such as to give less than 0.2 

 radians error in phase under the worst conditions. 



The horizontal deflecting plates were driven synchronously with a 

 sub-harmonic of the RF input to the helix, and the resulting deflection 

 served to separate electrons according to phase in the final display. 



Placing the focusing lens after the deflection plates results in a con- 

 siderable reduction in deflection sensitivity. However, undesirable mag- 

 nification of the pinhole aperture dictated that the lens could not be 

 close to it, and it was important to initiate the deflection as early as 

 possible. The lens consists of three discs, the center one being biased to 

 about 800 volts above the mean voltage of the rest of the system. 



Immediately after the lens there are two iron pole pieces and two insu- 

 lated electric deflection plates which extend parallel to the beam for IJ^ 

 inches. The pole pieces provide a dc magnetic field up to about 20 gausses 

 induced by small coils outside of the envelope, and the electric deflection 

 plates are biased with up to a corresponding 50 volts dc polarized to 

 oppose the magnetic deflection of the beam. The electric and magnetic 

 fields are adjusted so that the normal unmodulated electron beam tra- 

 verses the region with no deflection and strikes the center of the fluores- 

 cent screen. In the crossed field region 



1= W^2^Fo. (26) 



Electrons having greater or lesser velocity are deflected parallel to the 

 electric field, and give a corresponding deflection from the center of the 

 fluorescent screen. 



To get a display in which the various elements are not hopelessly en- 

 tangled, it was necessary to sweep the trace in an initial ellipse at a 



subharmonic rate. The sweep voltage was applied to the horizontal de- : 



flection plates, with just a little applied to the vertical plates through a , 



phase shifter. The relative phase of any part of the trace was measured k 



from the ellipse, and the velocity sensitivity was calibrated by observing | 



the ellipse deflection as a function of the dc beam potential, as shown f 



in Fig. 6(a). There is a small error due to the sensitivity of deflection to ( 



velocity, and due to distortion of the ellipse by fringing fields. ( 



In order to measure velocity and current density in the displayed pat- ;; 



tern, the fluorescent screen was photographed, and the negati^'e pro- h 



jected in a microcomparator. It was assumed that with the small ciu'rents P 



used, the light intensity was proportional to current, and the film i 



linearity was calibrated by making exposures of several different dura- j 

 tions. The trace density was measured with a densitometer, sweeping 



