242 RADIATION BIOLOGY 



MEASURING INSTRUMENTS 



The classical measuring instrument is the galvanometer, and it is still 

 one of the most sensitive and useful. Cartwright ( 1940a, b) has discussed 

 the theoretical limits of resolution of galvanometers and the various 

 means of attainment. The principal limitation of the galvanometer is 

 the delicate nature of its mechanical system and its sensitivity to vibra- 

 tion. The development of technics for converting d-c signals into a-c 

 signals has made it possible to attain theoretical limits of sensitivity with 

 instruments which are rugged and simple to use and which yield precision 

 superior to that of the galvanometer. 



The d-c amplifier is useful for the measurement of the voltage output 

 of high-impedance detectors, such as the vacuum photocell and the photo- 

 multiplier. However, it is not possible to separate the small voltage 

 fluctuations of the heater and polarizing voltages of the amplifier from 

 the d-c signal. Therefore the d-c amplifier is seldom reliable for detecting 

 voltage below 1 mv, and it is less suitable than the a-c amplifiers for driv- 

 ing recorders because of the slow but constant drift in the output voltage. 



In the a-c amplifier, slow fluctuations in the power-supply voltages are 

 readily separated from the a-c signals that are to be amplified. As a 

 result, the amplified a-c signal is relatively free of drift. The output is 

 usually rectified and converted to direct current for the operation of d-c 

 meters and recorders. If the amplifier has a large amount of inverse 

 feedback, its amplification can be made very independent of supply- 

 voltage fluctuations and tube characteristics, and the output made a 

 linear function of the amphfied signal. 



Numerous methods have been developed for producing a-c signals. 

 The ultimate objective is to make the over-all system as nearly inde- 

 pendent as possible of fluctuations in all extraneous factors associated 

 with the measurement, such as variations in stray flux, temperature, and 

 power-supply voltages. One of the earlier steps taken to eliminate fluc- 

 tuating factors was to apply alternating current to the bolometer bridge 

 (Moon, 1935). Later advances came with the development of fast detec- 

 tors which made possible the modulation or chopping of the radiant flux 

 at the source so that the detector would receive a d-c component from 

 the stray flux and an a-c component from the source. Only the a-c 

 component is amphfied, and slow random changes in the stray flux do 

 not appear in the output. This system is commonly used in commercial 

 recording infrared spectrophotometers. 



When the radiant flux is modulated at constant frequency, a mechani- 

 cal chopping device is required unless the source is modulated electrically 

 from an a-c supply. Consequently it is often more convenient to con- 

 vert the d-c signal of the detector into alternating current for amplifi- 

 cation. The alternating current can then be rectified for operating d-c 



