On Mar 20, 4:32 pm, "Anthony Fremont" wrote:



Well there sure isn't much talk about it out there. Material I find is
like, "here's a schematic, pick a coil and cap and your done. No one seems
to care what the result looks like. Seems like you can make a reasonably
decent wave _and_ still have the oscillator start reliably.

Some people worry a whole lot about it. What they worry about,
typically, is phase noise first and stability second. Lack of
harmonics in the waveform are generally lower on the list. That's
because filtering out harmonics is relatively easy, compared with
cleaning up phase noise and stability problems. You can find lots of
articles on minimizing phase noise, but there's also quite a bit of
trade-secret sorts of knowledge whose owners aren't particularly
interested in sharing, understandably. It's far from trivial to get
the phase noise and spurious performance you'll find in the good
commercial signal generators. Good as they are, though, I know of
none that's good enough to be used without harmonic filtering to do
low-level harmonic distortion measurements.

Cheers,
Tom