Sec. 12.4] IONIZATION CHAMBERS 155 



the vibrating reed a reverse voltage just sufficient to balance the signal 

 voltage. This feature provides a considerable advantage in taking measure- 

 ments since the slope of the recorded trace is proportional to the ionization 

 rate. Further, the incidence of an alpha particle in the chamber produces 

 an abrupt deflection in the slope of the line, thus allowing the background of 

 alpha particles to be fully accounted for separately in each measurement 

 without resorting to statistical methods. 



The vibrating-reed electrometer is less susceptible to microphonics and to 

 those factors which in other forms of electrometers frequently cause long 

 period drift and high noise level [6]. It is one of the most useful instruments 

 for detecting small currents because it combines very high sensitivity with 

 great stability and can be used under the most adverse physical conditions. 



e. Alternating-current Amplifiers. Heavily ionizing particles such as 

 protons, alpha particles, and fission fragments produce sufficient ionization 

 to be detected in ionization chambers with a-c pulse amplifiers and counted 

 by conventional recording circuits. The voltage developed across a resist- 

 ance of 10 8 to 10 10 ohms is impressed on the grid of a vacuum tube operated 

 usually as a cathode follower, i.e., with a cathode resistor of 10,000 to 50,000 

 ohms and the output signal taken from the cathode. This is followed by 

 one or two stages of amplification and possibly a cathode follower output 

 stage leading to the counting circuits. Considerable precautions must be 

 taken with preamplifiers of this sort to ensure good insulation of the grid 

 circuit of the first tube and generally to provide insulation and shielding 

 against vibration and external electromagnetic fields. Miniature tubes 

 of the types 6AK5, 9001, 9002, 945, etc., have proved to be highly satis- 

 factory for this application [7]. 



12.4. Statistics of Measurements. The probable error of measurements 

 with an ionization chamber recording discrete events is estimated by the 

 same procedure used for other counting methods and is given in detail in 

 the section on statistics for Geiger-Miiller counters. Chambers for detecting 

 beta, gamma, and cosmic radiation, on the other hand, require the measure- 

 ment of a continuous variable which may be either scale division per second 

 or millivolts per second. The procedure is precisely the same, however, 

 when the terms are redefined. 



Assuming only cosmic rays and alpha particles from the chamber walls 

 to be responsible for the observed background, the standard deviation of a 

 single measurement of the background charge taken over a time t b is 



c b = — 7= (a 2 A -f- b-B) y * ion pairs/sec 



where a = average number of ion pairs produced in the chamber per alpha 

 particle 



