Hi

I would like to draw your attention to an article which I posted here.

Actually, I came across that circuit more than 10 years ago. Unfortunately, I don't have any reference to it now,
I have tested the circuit with various BJTs like BC546/7/8, 2N3904, 2N2222A etc. and found them to be working without any problem.

The question which has been bugging my mind all these years, is that from where that Magical Constant 50,000 come from in the formula:

Frequency = 50,000 / Inductance OR Inductance = 50,000 / Frequency

 

time constant τ = L / R

two times τ = 2L / R

frequency f = 1 / τ = R / 2L

If R = 1 Ω and L = 1 μH

f = 1 / 2 MHz = 500,000 Hz

We don't know the switching criteria of that oscillator and the value of R. The variable resistor is there for calibration.
You can calibrate it with any value inductor and use whatever multiplication factor results. You don't have to use 50,000.
The recorded frequency multiplied by the inductance (in μH) should always give you the same number.