538 



THE BELL SYSTEM TECHNICAL JOURNAL, JULY 1951 



These measurements have shown that all of the r's are, to a first approxi- 

 mation, independent of collector voltage so long as the collector voltage 

 is above a few tenths of a volt and so long as the total dissipation is small 

 enough to prevent appreciable heating of the transistor. 



In view of this fact it is perhaps sufficient to show how these quantities 

 vary with emitter current for a moderate fixed value of collector voltage. 

 Figures 7 and 8 show that Yc and r^ are very nearly equal and that they tend 

 to decrease as /« increases. Theoretically Ym. and Yc should both be infinite. 

 The fact that they reach values as low as 10 megohms in this case is a meas- 



28 



24 



20 



O 16 



O 



UJ 



? »2 



-0.2 -0.4 -0.6 -0.8 -1.0 -1.2 -t.4 -1.6 -l.t 

 le IN MILLIAMPERES 



■2.0 



Fig. 7 — The variation of collector resistance with emitter current at a fixed value of 



collector voltage. 



ure of the imperfection in technique of fabricating the transistor. Values as 

 high as 60 megohms have been achieved in the laboratory. 



Figure 9 shows that Yh in this transistor is approximately 240 ohms and 

 is independent of /« . 



Figure 10 shows that Ye decreases with increasing emitter current, ranging 

 from about 500 ohms at 50 microamperes down to about 5 ohms at 5 

 milliamperes. Shockley'* has shown that r « should be given by 



fe = 



9/. 



(1) 



where q is the charge on an electron, k is Boltzman's constant, T is the 

 Kelvin temperature and /« is the emitter current. When the temperature 



