Sec. 1-0] MFA'HASICAL IS ITT TltASSDUVKliS 135 



per cent of the tube diameter is recommended; for the Hush-mounted 

 electrode the width should be about one-half the approach distance, 

 or y in Fig. ( 1-6)13. 



A theoretical treatment of the induction process is given by 

 Shockley 1 and b} r Bouwcamp and de Brujin. 2 The polarity of the 

 peak charge and of the current is determined by the sign of the charge 

 of the ion cloud. The approach of a predominantly negative charge 

 causes the flow of positive charges from ground to the pickup elec- 

 trode P (arrow). The flow direction of positive charges is from the 

 positive to the negative pole; hence the grid of the following amplifier 

 becomes negative during the approach of the negative cloud, and 

 positive thereafter, as shown in curve B, Fig. (1-6)13. The cloud may 

 travel as much as 2 to 3 in. distance from the electrode. The reference 

 point for the determination of the time difference At is measured from 

 the point of maximum charge or of zero current and can be deter- 

 mined, in the latter case, to within 0.2 msec. The ion cloud will 

 diffuse as it moves downstream, but for drift distances up to 12 in., 

 the ionization will remain localized within a cylinder of l^in. 

 diameter at a drift velocity of 1 50 mph. Moderate lengthening of the 

 discharge time, that is, of the time necessary to produce the ion 

 cloud, does not change the waveform at the pickup electrode appre- 

 ciably, compared to a point charge. 



Gas velocities as low as 20 mph have been measured with this 

 method. The method is not applicable for smaller velocities (because 

 of radial and axial diffusion and of recombination of the ions) but has 

 been used successfully in supersonic flow up to Mach 2 (twice the 

 velocity of sound). Path lengths in the order of 1 to 10 in. have been 

 used. The accuracy of the method at high velocity is difficult to 

 evaluate because precision methods for comparison do not exist; 

 at best the error is estimated with a few per cent. Semicontinuous 

 operation can be obtained by triggering the gas discharge from the 

 pickup electrode via a feedback loop and observing the recurrence 

 frequency, which depends upon the ion transit time. A system of 

 this kind is described by Mellen. 3 



For references see W. M. Cady, Physical Measurements in Gas Dynamics 

 and Combustion, in R. W. Ladenburg (ed.), "High Speed Aerodynamics and 

 Jet Propulsion," vol. 9, Princeton University Press, Princeton, N.J., 1954. 



1 W. Shockley, ./. Appl. Physics, 9, 635 (1938). 



2 C. F. Bouwcamp and X. G. de Brujin, J. Appl. Physics, 17, 562 (1945), and 

 Errata, 19, 105 (1947). 



3 G. L. Mellen, Electronics, 23, 80 (February, 1950). 



