A weird idea bouncing around in my head. It goes like this: I can use a pot and diode to acts as a variable cap. Why not use a pot and coil to act as a variable inductance. We have magnetic amplifiers which are basically variable inductors controlled by a signal applied to a control winding. On to the sims.
A simple 2N2222 oscillator. This one does not seem "strong enough". I think we need more output to drive our DBM.
The lambda diode oscillator is much stronger. Could be a good choice.
Ok , now we need a control winding and an out put winding.
The sim supports the theory.
The sim supported the theory using Andy's oscillator. I wound a core with 3 coils, 1 turn, 3 turn and 16 turn and got the values I was looking for.
Time for a build when I get some bench time?