[time-nuts] GPSDO vs Rb standard

[Bob Camp]

Hi

Some numbers:

Your Rb should hold <10 ppt per month. It's 1 second ADEV should be about 10 ppt as well. (makes it interesting to measure).

Your Rb should hold  < 1 ppt per degree C on average. It will have a nasty glitch (around 1 ppt as it steps over a temperature boundary). Best to keep it stable to < 0.5 C.

For long(er) averaging times the Rb should have an ADEV that goes as the square root of tau. It'll be 10X better at 100 seconds than at 1 second. It *could* be 100X better at 10,000 seconds except for that glitch. It's not going to get beyond 0.2 ppt ADEV without some major help. 

All that said, with simple temperature stabilization, it should let you check out your GPSDO for drift. Within the ADEV imposed limits it'll tell you if things are holding together on a day to day / week to week basis.

Bob


On Sep 18, 2013, at 8:32 PM, Bob Stewart <bob at evoria.net> wrote:

> Bob, I've been mulling over the question of a case, and I was thinking of getting a 2M brick amp and putting it inside, bolted upside down to the heat sink.  Good idea?  Anybody got a busted one?
> 
> Bob
> 
> 
> 
> 
> 
>> ________________________________
>> From: Bob Camp <lists at rtty.us>
>> To: Discussion of precise time and frequency measurement <time-nuts at febo.com> 
>> Sent: Wednesday, September 18, 2013 7:22 PM
>> Subject: Re: [time-nuts] GPSDO vs Rb standard
>> 
>> 
>> Hi
>> 
>> The Rb's have a couple of issues:
>> 
>> 1) Its been years since they were set on frequency and they do have a yearly drift rate
>> 2) They have a temperature coefficient of drift that may be fairly large (0.1 ppb over -30 to +70)
>> 3) They self heat quite a bit, so they do move temperature / need a heatsink
>> 
>> The solution to the temperature issue is a servo controlled fan. The solution to the first drift / accuracy issue is to calibrate them against something else.
>> 
>> Bob
>> 
>> 
>> On Sep 18, 2013, at 7:13 PM, "John C. Westmoreland, P.E." <john at westmorelandengineering.com> wrote:
>> 
>>> Azelio and Fellow Time Nuts,
>>> 
>>> Isn't the GPS 1PPS signal supposed to be 'precise' to within what error?  I
>>> imagine this is in the specs of the specific receiver -
>>> but I was wondering if some of you have actually measured what that is -
>>> and could report the numbers that you have found.
>>> 
>>> Thanks!
>>> John Westmoreland
>>> 
>>> 
>>> 
>>> On Wed, Sep 18, 2013 at 3:07 PM, Azelio Boriani <azelio.boriani at screen.it>wrote:
>>> 
>>>> Instead of using the Rb's PPS, use the GPS receiver's PPS. Maybe you
>>>> will find out that the Rb is slow... you can also check the Rb's PPS
>>>> against the GPS's PPS.
>>>> 
>>>> On Wed, Sep 18, 2013 at 11:47 PM, Bob Stewart <bob at evoria.net> wrote:
>>>>> I hooked the Rb's 1PPS to the trigger input on my old Tek 455, 10MHz
>>>>> GPSDO to Channel A, .05us trace, and turned the lights out so I could
>>>>> see it.  The 10MHz is marching right to left about 1 cycle every 20
>>>>> seconds.  So, can I say that the Rb considers my GPSDO to be too fast by
>>>> about 5ppb?
>>>>> 
>>>>> 
>>>>> Bob
>>>>> _______________________________________________
>>>>> time-nuts mailing list -- time-nuts at 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 at 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 at 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 at 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 at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.