Jim Thompson wrote:
On Tue, 24 Feb 2004 15:29:43 GMT, "Rick Karlquist N6RK"
wrote:


You're probably thinking of the oscillator that Marv Frerking
called a "grounded-base oscillator". I have seen it called other
names as well. Basically, what you do is first build an LC
(ie no xtal) Colpitts oscillator and tune it to the crystal frequency
you want to eventually use.

[snip]

Sounds like an oscillator that I've used since the '60's for my G-jobs
(you know, the ones that *have* to work, 'cause they're for me .

See "XtalSeriesOsc.pdf" on the S.E.D/Schematics page of my website.

I've never been able to get any custom IC customers to use it, since
it takes three pins, but it works, period, no messy matching issues,
even handles overtone modes.

...Jim Thompson

This pdf schematic looks very similar, except for the crystal and emitter
part, to a circuit described in "Experimental Methods in RF Design" (p4.13).

The circuit is described as "worth building......to observe first hand just
what a noisy oscillator will sound like in a receiver".

Earlier in the same chapter it appears in figure 4.13 as a type of negative
resistance one port oscillator.

I can only assume that the changes and crystal (in the circuit shown in the
pdf) solve the problem of the "noisy" LC only configuration.

I mention it as I built it up last night and took it into work today in order
to have a look at the output on a spectrum analyser. Output was quite low at
-27dBm. Sadly the HP kit couldn't measure phase noise directly, and I didn't
have a good crystal oscillator to check it against.

We were uncertain regarding the configuration too, but my colleague worked out
that it had severe voltage limiting features and predicted the output swing
quite accurately before it was measured on a scope.

Last night I tried it with a number of inductors from the junk box and it
oscillated quite readily from 114MHz down to 5MHz. I quite like the use of a
non tapped L and only a single C. Shame about the phase noise :-(.

regards...

--Gary