Hi
(see below) On Aug 12, 2014, at 3:45 PM, Peter Walker <[email protected]> wrote: > <EWKehren@...> writes: > > > >> > >> Having had more than ten FRK's and M100's on my bench and once locked none > > >> where as low as any thing in E-9. What you should buy is a ubox for $ 16 > > >> shipping included and you will with your counter be able to get better > than > >> 1 E-9 for testing. > >> Bert Kehren > >> > >> > >> In a message dated 8/10/2014 2:30:50 P.M. Eastern Daylight Time, > >> ed_palmer@... writes: > >> > >> How did you make the measurement that showed a 0.8 Hz difference? What > >> are the specs on your Racal reference? When was it calibrated and > >> against what standard? That will tell you how much confidence to put in > >> it's frequency. > >> > >> As others have said, an FRK isn't a primary standard and should be > >> calibrated against a primary standard. However, 0.8 Hz is an error of > >> 8e-8. I don't believe it's possible for a Rb standard to lock and be > >> that far off. > >> > > > >>> > >>> Sorry if dumb questions, but just starting my quest for "time nut" > >> status!!! > >>> > >>> Thanks > >>> > >>> Peter > >>> G0RSQ > >> _______________________________________________ > >> time-nuts mailing list -- time-nuts@... > >> To unsubscribe, go to > >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > >> and follow the instructions there. > >> > >> > > > > Thanks for the interest in my questions. > > > > Unfortunately the FRK has now decided to stop locking, so I need to go back > to fault finding. I suspect the problem is with the "hunting" of the 10Mhz > ref oscillator, as I get a swing of over 300Hz, which I am sure is far to > big. > > The manual is confusing, as sometimes it talks about a swing of +/- 20 Hz, > and other times +/- 1Hz… > At some points it’s talking about the OC-VCXO range, at other points it’s talking about the tune range when locked using the C field. > > > In answer to the questions:- > > > > After the FRK locked up, I connected it to my Racal 1998 counter, which has > > the external reference coming from a Racal 9480 mainframe, which is running > > its back-up internal ovened frequency reference (last calibration was about > > 5 years ago before i acquired it!) (it is this that I want to connect the > > FRK to as its main frequency source). I had a reading which was 0.8 Hz down > > on 10 MHz. I left this for about 1 hour to see if anything changed, but it > > did not. > > I then connected both oscillators to the input channels of my oscilloscope > > (Tek 2465B), triggering of the FRK, and saw a fairly rapid movement on the > > ovened waveform. All this indicates to me one of the references is out by > 0.8Hz. > > > > I understand that rubidium is not a primary frequency standard and the > > output is synthesized, but was hoping that some clever technique was used to > > ensure a certain level of reliability. How far off frequency could one be > > and still lock? Normal accuracy even if it’s been off a long time should be < 10 ppb. If the Cell is off frequency it will lock over a *much* wider range. 10 or 20 ppm is quite possible in this case. > > > > I do have an old "Off Air" standard frequency receiver (similar to a "quartz > > lock") which receives "Droitwich" on 192KHz, which gives a 10MHz output. > > However I am unsure how suitable this is for calibrating my FRK. I will give > > it a try this week, and see how it compares with the FRK. > > > > I also have a Trimble (not a thunderbolt) GPSDO bought from a seller in > > china, which I am yet to set up and try, so maybe this will be my next > > project (but maybe I will be asking the same questions when this is > > running!!!!). > > > > Thanks for the help and comments > > > > Peter > > G0RSQ > > > > _______________________________________________ > 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. Do you have an antenna for the T-Bolt? Do you have power for the T-Bolt? If yes to both, then that’s a simple solution to the problem. Bob _______________________________________________ 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.
