240 RADIATION BIOLOGY 



with the cell internal resistance. Various methods have been proposed 

 for correcting for the variations in the internal shunt resistance (Rittner, 

 1947). 



Whereas the internal emf is generated almost instantaneously, the cell 

 has an appreciable time constant because of the large shunting capacity 

 presented by the small separation of the electrodes. The time constant 

 decreases as the load resistance is made smaller. For a typical cell 

 several centimeters in diameter, the response falls off rapidly above 

 about 1000 cps with a 50-ohm load resistance. 



The selenium cell has a cosine error that is especially serious when 

 used to measure the flux from a distributed source, such as the lighted 

 walls of a room or an overcast sky. The response should be proportional 

 to the cosine of the angle of incidence, but it is less than this for the 

 selenium cell. Various methods have been developed for correcting this 

 error by the use of specially shaped diffusing plates of opal glass or plastic 

 (Pleijel and Longmore, 1952). 



SENSITIVITY LIMITS OF DETECTORS 



The measurement of medium to high irradiances above 10 /xw cm"'- 

 with thermal detectors or 1 ft-c with a selective detector presents few 

 problems. Relatively simple equipment consisting of a thermocouple 

 and lamp and scale, portable galvanometer or selenium cell, and micro- 

 ammeter can be used. It is at the very low levels of intensity that com- 

 plex electrical instrumentation is required, and consideration must be 

 given to the factors that inherently limit the lowest levels of radiant flux 

 that can be detected. The ultimate limit of sensitivity of a radiation 

 detector is attained when the random fluctuations in the signal, com- 

 monly referred to as "noise," as produced by the detector and the associ- 

 ated amplifier, are equal to the signal. The ability to detect a given 

 radiant flux is measured in terms of the signal/noise ratio. The minimum 

 flux that can be detected is generally considered as that which gives a 

 signal/noise ratio of 1. In the thermal detector the noise arises from 

 random fluctuations in the radiation field, temperature of the receiver, 

 and emf or Johnson noise in the resistance elements of the detector. A 

 detector that is in thermal equilibrium with its surroundings is continually 

 absorbing and reradiating energy. The magnitude of the exchange is 

 proportional to the fourth power of the absolute temperature. Since the 

 receiver is an efficient absorber of all radiant energy from the ultraviolet 

 into the far infrared, there is an appreciable component of exchange at 

 300°K, where the X„, is about 10 n for a Planckian radiator. The various 

 factors that limit the ultimate sensitivity of radiation detectors have been 

 treated theoretically by several authors (Fellgett, 1949; Golay, 1947a; 

 Hornig and O'Keefe, 1947; Jones, 1947). 



Bolometers and thermocouples are electrical resistors that produce a 



