Yeah, I mean you gotta have negative feedback as a component of the positive feedback loop, in order to ensure that the fed-back signal is of the correct (pos) phase and doesn't hit the supply rails once the oscillator gets up to speed. Naturally it needs to allow the oscillations to build up from the initial switch on. That's why it was such an elegant solution. Somebody above said it was a thermistor rather than a lamp and I believe he's dead right. I also recall that there was only one type which really did the trick beautifully and that was sadly discontinued years ago. :(
The alternative lamp filament method is just as effective but it needs more drive from the oscillator circuit for this to work (that's the method used by the Thorens TD125 oscillator - you can even see the lamp filament glowing a dull yellow!).
I couldn't find a suitable thermistor when I was making a test sinewave oscillator (1, 10 and 18.5 KHz spot frequencies at 0 and
-20dbmW balanced/ unbalanced outputs) 30 odd years ago so I used a 6v
40mA indicator lamp. The only other candidate I had a ready supply of were 12v 20mA lamps but I think the higher voltage requirement precluded its use in a circuit powered from a PP3 battery[1]).
As it was, I needed a dark room and night adapted vision to detect the very dim orange filament glow in that PP3 powered oscillator. The
6v 40mA lamp is right on the edge of usability as a feedback gain stabiliser in this case.
[1] I also made a mains powered variable frequency Wein bridge oscillator where I may have taken advantage of the higher DC voltage bias rails to make use of the 12v 20mW lamp. It's been many years now since I last saw the circuit diagram I drew up for this project so I really can't recall which lamp type I'd elected to use.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.