[ntp:questions] Re: Using a Rubidium clock with NTP

Richard B. Gilbert rgilbert88 at comcast.net
Thu Oct 7 22:06:30 UTC 2004


Geoff Horner wrote:

>Paul,
>
>I do have a GPS time server on the same network and had considered this, but was worried about the long term drift of the Rb ticks. I image I could easily monitor the difference between the two and edit the fudge factor on the Rb machine's ntp.conf every day or so, but restarting ntpd daily would be very detrimental. I think ntpdc can alter the fudge factor on the fly, but haven't checked it out and am not sure if that's the best idea. Since posting my question, I have found (with TVB and friend's help - many thanks) how to program the DDS in the Rb oscillator. I was thinking it might be best to set the Rb osc internal freq to 14.31818 MHz and substitute the GPS PC xtal for this. 
>
>With all this effort,  I'm starting to think the easiest way to get a good stratum 1 server is to use one of the  miniature OCXO that I've seen, and substitute the 14.31818 MHZ PC xtal.and with one of them and a Motorola M12+ GPS timing receiver, producing good results for little outlay. Maybe next clock....
>Can't wait for the miniature rice grain size Rb clocks to be available. An atomic wristwatch would be nice.
>
>Geoff
>
>
>
>"Paul Croome" <Paul.Croome at softwareag.com> wrote in message news:ea26a3f3.0410062328.6f76f168 at posting.google.com...
>  
>
>>"Geoff Horner" <geoffh at bigfoot.com.au> wrote in message news:<mailman.31.1095599860.26599.questions at ntp.org>...
>>    
>>
>>>Hi all; My first post.
>>>
>>>What is the best way to use a Rubidium oscillator with a PPS output as a st
>>>ratum 1 server?
>>>      
>>>
>>...
>>
>>Geoff,
>>
>>I suggest using your Rb PPS with the NTP PPS driver (type 22).
>>Your Rb oscillator probably doesn't tick synchronously with the TIA/UTC/GPS
>>second, so observe the offset between the Rb ticks and GPS time, then
>>set the time1 fudge factor to compensate.
>>
>>The NTP server that is running the PPS driver also needs another source of time,
>>so that it can number the ticks. Any clock that is within +/- 0.5 second
>>should do.
>>
>>Paul
>>_______________________________________________
>>questions mailing list
>>questions at lists.ntp.isc.org
>>https://lists.ntp.isc.org/mailman/listinfo/questions
>>
>
>  
>
I see Rubidium frequency standards specified with long term stability of 
1*10^(-10).  If I haven't slipped a decimal place somewhere, that's less 
than 100 microseconds per month.  I  don't see why you would need to 
correct it daily.  If you really require time correct to within less 
than 100 usec, maybe you should be thinking about a cesium reference!!!!




More information about the questions mailing list