382 the bell system technical journal, april 1951 



The Relation between Vacuum Tube and Transistor Properties 



It is the purpose of this section to show that the properties of a transistor 

 are related to those of a vacuum tube triode through an interchange of 

 current and voltage, and that transistor currents behave Hke vacuum tube 

 voltages and vice versa. The discussion is aimed particularly at the large- 

 signal properties of the two devices and is restricted to the frequency range 

 in which static characteristics are sufficient to determine circuit perform- 

 ance. 



Consider first the grid-cathode input terminals of a triode as compared 

 to the emitter-base input terminals of a transistor. With respect to these 

 terminals each device behaves as a diode rectifier the properties of which 

 are relatively unaffected by biases appHed to the third electrode (plate or 

 collector). The grid conducts when biased in the forward direction and fails 

 to conduct when biased in the reverse direction. A similar statement can be 

 made about the emitter. Furthermore, either device behaves as a low im- 

 pedance when biased in the forward direction and as a relatively high im- 

 pedance when biased in the reverse direction. 



The difference between the emitter circuit and the grid circuit comes about 

 in the following way: The vacuuSi tube is most effective as an amplifier 

 when the grid is biased in the reverse direction, while the transistor is most 

 effective when the emitter is biased in the forward direction. With respect 

 to these input terminals, then, the essential difference between the two de- 

 vices amounts to the difference between "forward" and "reverse". But this, 

 in turn, amounts to an interchange of current and voltage. 



Whatever quahtative statements can be made about emitter current 

 and voltage can also be made about grid voltage and current, respectively. 

 For example, the grid is normally given a moderate voltage bias at which 

 the grid current is essentially zero, while the emitter is normally given a 

 moderate current bias at which the emitter voltage is essentially zero. 

 Furthermore, the principal non-h*nearity in the grid circuit occurs when the 

 grid voltage is allowed to swing through zero with the result that grid cur- 

 rent begins to rise, while the principal non-linearity in the emitter circuit 

 occurs when the emitter current is allowed to swing through zero with the 

 result that emitter voltage begins to increase. 



The comparison between the plate-cathode output circuit of the triode 

 and the collector-base output circuit of the transistor is somewhat compli- 

 cated by the effects of grid and emitter biases. Consider first the situation 

 in which zero bias is applied to the input circuits {vg = and ie = 0) . 

 In this case, both the plate and the collector behave like diode rectifiers, 

 conducting readily when biased in the forward direction and conducting 

 relatively poorly when biased in the reverse direction. When input biases 



