Hi The classic approach is to start with something stable as an input. A TBolt is one example of this. Ideally you would want good phase noise in the sub 10Hz offset range from your base oscillator. Next you PLL up an OCXO in the 60 to 120 MHz range. The idea is to get good phase noise out to a few hundred hertz on the VHF OCXO. Your PLL bandwidth is optimized to give you the best tradeoff between the multiplied noise of your base oscillator and the VHF OCXO.
Past that a lot depends on what you are headed to. If the ultimate goal is microwave rather than UHF, you would go there next. For UHF, you either triple the OCXO or go with another phase locked oscillator. DRO's, SAW oscillators, and transmission line VCO's all have their fans. The low noise floor is the goal and numbers in the 160's to 170's often get mentioned. Weather or not they are achieved .... For 375, the likely OCXO's would be 125 MHz, 93.75 MHz and 75 MHz. If you happen to find 125 MHz, a tripler followed by a narrow band filter might do a good job. 125 and 75 are going to give you a higher phase detector frequency than 93.75, certainly a plus there. Lots of ways to do it. Bob On Mar 17, 2011, at 2:01 PM, Tijd Dingen wrote: > With regard to your mentioning of "a reference is not 10MHz" ... > Suppose one is after a good quality ~ 375 MHz reference with low > jitter and high frequency accuracy, what would be the best bang > for buck time-nut approved way be? > > For now I would settle for a 370 MHz reference, which would be > obtained by multiplying from a 10 MHz source from say a thunberbolt. > But possibly there are more economic ways to go about this... > > > > ----- Original Message ---- > From: Chris Albertson <albertson.ch...@gmail.com> > To: Discussion of precise time and frequency measurement <time-nuts@febo.com> > Sent: Thu, March 17, 2011 6:45:34 PM > Subject: Re: [time-nuts] Thunderbolt vs Motorola M12+T > > On Thu, Mar 17, 2011 at 8:58 AM, Tijd Dingen <tijddin...@yahoo.com> wrote: >> Same here. A GPSDO like the thunderbolt is really nice from a system >> integration point of view. Buy an easy thingy kit on ebay, and you have >> your 10 MHz reference. However, cannot a lot be gained from using a >> more up-to-date design with newer gps engine? >> >> On the one hand we have a thunderbolt which you can view as blackbox >> that has a nice stable 10 MHz as output. On the other hand we have for >> example a Motorola M12+T where you have to do your own oscillator >> disciplining. At this point I am more inclined to go for the thunderbolt >> since it solves a lot of problems. However I feel that with a modern >> gps engine with better pps signal + a good quality XO you could get >> a better 10 MHz signal. Any thought/experience/tips? > > > I've only read reports of experiments and I think in general the > result is that there is an inherent limit to what you can expect from > a GPSDO that is around 10E-12 and that all the GPSDOs if you let them > run, get somewhat close to that limit. What seems to matter the most > is the quality of the VXCO used. So asking if the t-bolt or mt12 is > best may not be the important question as gps may not be not the weak > link, it's the oscillator. That said I thing the quality of the GPS' > PPS determines how long you need to let the GPSDO run. > > I've been thinking recently that if the ultimate goal is a reference > that is not 10MHz you might be best off building a GPDDO that runs at > the desired frequency then trying to lock to a 10MHz. > > > -- > ===== > Chris Albertson > Redondo Beach, California > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
