TRANSISTOR PULSE REGENERATIVE AMPLIFIERS 1107 



repetition period and, in such a case, the clock signal is made available 

 in four phases. 



Although the clock signal may have any one of a number of forms, a 

 sine or a square wave are the most common forms. Usually a sine wave 

 is preferred because it is simpler to distribute to a large number of 

 amplifiers. Exceptions occur in cases where exceptionally precise timing 

 is necessary, or the use of a square wave requires considerably less clock 

 power. In the following discussion of where to synchronize, a square wave 

 will do as well or better than the assumed sine wave. With either signal 

 it is desirable to keep the clock power to a minimum. 



If the synchronizing circuit is to be effective, the clock signal must be 

 capable of accomplishing the following actions: 



a. It must be able to hold the transistor in the "off" state in the pres- 

 ence of trigger current in order to control turn-on. 



b. At the turn-on time it must rapidly inject the trigger current into 

 the transistor. 



c. At the turn-off time it must alter the conditions in the feedback 

 loop in such a manner that the transistor turns off promptly. 



In other words the synchronizing circuit must act like an inhibit logic 

 circuit with the clock signal appearing as the inhibit signal during the 

 interdigital period. 



It is recognized that there are many amplifier configurations and 

 several ways to synchronize each configuration. Generally it is preferable 

 to synchronize at only one input terminal of the transistor or at only one 

 point in the feedback circuit. A relatively complete discussion can be 

 given with the aid of the following four examples. 



A circuit that employs negative resistance feedback, such as in Fig. 4, 

 requires a relatively large amount of clock power for synchronization. 

 Because a capacitor (C2) is required on the emitter foi regeneration, ^ 

 the clock signal must be applied to the base of the transistor to control 

 turn-on accurately. As far as turn-off is concerned, another clock signal 

 might be applied to the emitter or to the current gate in the feedback 

 circuit. However, this would result in additional components, a second 

 clock signal 180° out of phase with the base clock signal, and approxi- 

 mately the same required clock power as if the base clock signal alone 

 were used. Turn-off at the emitter is impractical due to the negative re- 

 sistance characteristic. The power that the clock signal on the base must 

 furnish is made up of two parts. One part is the average standby power 

 that is absorbed every time the clock voltage is positive. It is composed 

 of the Ico power supplied to the transistor plus the power dissipated in 

 in Rl and R2. R2 and D2 serve to reduce the clock current in Rl and 



