Submultiple Generator

SUBMULTIPLE GENERATOR

Lower frequencies can be obtained from a source of high frequency by what was called a "submultiple generator" in the mid 1920's to 1930's. It was a well-known fact that two oscillating triode circuits had a tendency to pull into synchronization with each other, even with weak coupling; with stronger coupling the effect was greater. Now, if an oscillating triode, was weakly coupled from plate to grid so that it was more or less unstable, had its bias so arranged that the plate current was full of harmonics, the circuit would synchronize itself with another having the frequency of one of its harmonics. Thus suppose that this unstable oscillator was adjusted for about 10,000 cycles and that the 100,000-cycle power from a crystal oscillator was impressed on its plate; then the tenth harmonic of the unstable oscillator would synchronize exactly with the impressed 100,000-cycle power. The plate circuit of the unstable generator, tuned for 10,000 cycles, would then oscillate at exactly 10,000 cycles, to the same degree of constancy as the 100,000-cycle crystal held its frequency. The two circuits so synchronized (100,000 cycles to 10,000 cycles, the latter with a pronounced tenth harmonic, of course) would stay together as well as if they were mechanically geared.

Now, another unstable generator may have its plate circuit tuned for 1,000 cycles, and its bias adjusted to give a plate current rich in harmonics; then if the 10,000-cycle generator was coupled to its plate it would synchronize with the10,000-cycle power, by the action of its tenth harmonic. Thus, there was obtained 1,000-cycle power, the accuracy of the frequency being the same as that of the 100,000-cycle crystal which was acting as the driver of the two submultiple generators.

Reference

Electron Tubes by John H. Morecroft, ScD. 1933