Crystal Control Applied to the Dynatron Oscillator

The research was suggested by the following two facts: (1) The dynatron oscillator, as compared with the feed-back oscillators, has a much better "frequency stability" (i.e., its frequency is much more independent of changes in the applied electrode voltages); (2) The Pierce types of crystal or magnetostriction oscillators are really controlled feed-back oscillators. One type of magnetostriction controlled, and two types of crystal controlled dynatrons were discovered. Although no stability measurements were made on the magnetostriction dynatron, it was found to have the advantage of simplicity and freedom from parasitic oscillations. Of the two crystal controlled dynatrons, one has the crystal between the screen grid and the plate, the other has the tank circuit of the ordinary dynatron replaced by the crystal. This latter has a better stability than the Pierce types, but it has the disadvantage that it is not self-starting. By inserting in this latter circuit a tuned circuit between the crystal and the plate, it is possible to get a controlled output at n or 1/n times the resonant frequency of the crystal, where n=1, 2, 3, 4, . . . All the new circuits, in common with the simple dynatron, suffer from the fact that at present there is no American-made tube capable of oscillating as a dynatron over a long period of time, because of the abnormal filament emission. Some suggestions are advanced in regard to the design of a suitable tube.