> ------------------------------ > > Message: 2 > Date: Sat, 6 Jan 2018 09:19:31 -0500 > From: Bob kb8tq <[email protected]> > To: Discussion of precise time and frequency measurement > <[email protected]> > Subject: Re: [time-nuts] AM vs PM noise of signal sources > Message-ID: <[email protected]> > Content-Type: text/plain; charset=utf-8 > > Hi > > The key point missing is the fact that any real oscillator must have > a limiter > in the loop. Otherwise it will “create one” by going over the max > output of this or > that amplifier. To the degree that the limiter has issues (limits > poorly) you will get > AM noise.
Hmm. Not strictly true. One can also use an AGC loop, like a wein bridge oscillator. That said, some kind of softish limiter is commonly used. Joe Gwinn > On a practical basis, loop Q is as significant as resonator Q . The various > elements in the loop degrade the total Q by a significant amount. > Getting 25 to > 50% of the resonator Q is “doing well” with his or that common > circuit. Yes, there > are even more layers past this …. > > Bob > >> On Jan 6, 2018, at 1:53 AM, donald collie <[email protected]> wrote: >> >> So to be lowest noise, an oscillator should have the highest Q resonator >> possible in its feedback loop, operate in class "A" [for maximum >> linearity], and utilise active amplifier device(s) that contribute the >> least noise [both amplitude, or 1/f], and phase. This latter implies >> operating the active device at maximum output level [ie signal to noise]. >> The quality of the power supply effects the amplifier SNR, so in the >> persuit of superlative oscillator phase noise, the power supply should be >> as good as possible. >> Resistors in the oscillator carrying DC make 1/f noise - the best in this >> respect are the metal type, I think - so use metal resistors or WW. >> What are the other conciderations that come into the design, for lowest >> noise of the oscillator itself >> Split, then >> lump...;-).................................................Cheers, de : Don >> ZL4GX >> > End of time-nuts Digest, Vol 162, Issue 10 > ****************************************** _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
