I briefly considered designing a board that would GPS discipline the FE-5680A. It’d simply substitute an RS-232 level shifter for the DAC and would speak the protocol to trim the frequency. I haven’t done it, though, because the OH300’s performance specification at low tau is not that far removed from the FE-5680A’s, they’re far smaller and take a lot less power and their supply (at reasonable pricing) is far more certain.
> On Nov 1, 2015, at 6:26 PM, Bob Camp <[email protected]> wrote: > > Hi > > Without being at NIST, there is no real way to report the ADEV on the 3801 > past 100K seconds. They > change ADEV in the 1K to 20K region depending on the state of their internal > filter. They should be below 1x10^-13 > under most conditions once they get past 50K seconds. This all assumes the > unit is working properly and > has been on power for a good long time (many weeks). > > The 5680 ADEV past 1,000 seconds is highly dependent on the state of the > temperature compensation > process. Catch it at a dead zone and they will get into the low parts in > 10^-13 from 5K to 20K seconds. Exactly > where they turn up depends on the temperature environment, aging (how long > they have been on) and the > approach you take to eliminating drift. (ADEV by definition is a “drift > removed” calculation). ADEV under ideal > conditions can be worked out as falling off by square root (tau) with the > starting point being 1 to 2 x10^-11 at 1 second. > At 100 seconds that’s 1 to 2 x10^-12 at 10,000 seconds it could be (but > isn’t) 1 to 2 x10^-13. Since most of the long > term ADEV is a function of the temperature compensation dithering, there is a > lot of variation unit to unit and even within > the same unit under different conditions. > > Bob > >> On Nov 1, 2015, at 6:47 PM, Larry McDavid <[email protected]> wrote: >> >> Would someone familiar with the facts please report the long-term Allen >> Variance of the HP Z3801A and the FE-5680A. >> >> -- >> Best wishes, >> >> Larry McDavid W6FUB >> Anaheim, California (SE of Los Angeles, near Disneyland) >> _______________________________________________ >> 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. > > _______________________________________________ > 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. _______________________________________________ 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.
