Understanding Feedback Control in Klystrons

[artis]

I was puzzled when I thought about what happens to a klystron when it's output is fed back to it's input.

This doesn't apply to magnetrons because they cannot be driven like amplifiers so they are oscillators by definition.
I've dealt with audio amplifiers mostly and there almost always feedback is used whether positive or negative.

But I do not know how it is with high frequency tubes like klystrons , how much feedback you can use and still have an amplifier and when it becomes an oscillator that then resonates at some frequency determined by the geometry of the tube or it's cavities.
At lower frequencies in audio amplifiers for example feedback is decreased in amplitude by passing it through resistors and then fed back into the input stage. But I am not familiar with how feedback is decreased in amplitude in RF applications.

What are the feedback control mechanisms for a klystron or other RF tube for example?Another question I am thinking about is how one would couple a wire loop from a toroidal (or any other shaped core) to a waveguide?

 

[Baluncore]

Any amplifier can become an oscillator by arranging feedback with a loop gain of unity at only the required frequency. There are too many methods of adjusting the frequency to list.
If you want to make an oscillator from a klystron, select a reflex klystron.

Improving an amplifier, by adjusting the feedback, is fraught with a high probability of oscillation. The named oscillator eponyms, are failed amplifier designers.