[time-nuts] I thought GPS repeated every 12 hours (-2 minutes)

Tom Van Baak tvb at LeapSecond.com
Tue May 24 09:03:34 UTC 2016


Hi Magnus,

> Hi Skip and Tom,
> Yes... almost. The thing is that the GPS orbits is a few minutes shy of 12 hours

Right. I think that's why he picked the subject line: "12 hours (-2 minutes)".

Since you're interested in this level of detail, there are papers about GPS orbits, repeat times, sidereal time, and orbital maneuvers:

http://www.kristinelarson.net/wp-content/uploads/2015/10/ig0806_gnss-solutions.pdf
http://www.kristinelarson.net/wp-content/uploads/2015/10/gpsrep.pdf

These details turn out to be more important in the geodetic community than the T&F community. We tend to average a lot, but they have embraced high-rate kinematic GPS receivers as zero-drift seismometers. Some more papers:

http://xenon.colorado.edu/igs5_revised.pdf
ftp://ftp.ngs.noaa.gov/pub/abilich/papers/Bilich2008_Denali.pdf

This issue of "just a bit less than 12 hours" caught my eye because I ran across it in ADEV plots. See:

http://leapsecond.com/pages/sidereal/index.htm
http://leapsecond.com/pages/sidereal/sv.htm
http://leapsecond.com/pages/sidereal/14years.htm

The 14years.htm page nicely shows the occasional "station keeping" orbital maneuvers of each GPS SV.

/tvb


----- Original Message ----- 
From: "Magnus Danielson" <magnus at rubidium.dyndns.org>
To: <time-nuts at febo.com>
Cc: <magnus at rubidium.se>
Sent: Monday, May 23, 2016 11:57 PM
Subject: Re: [time-nuts] I thought GPS repeated every 12 hours (-2 minutes)


> Hi Skip and Tom,
> 
> Yes... almost. The thing is that the GPS orbits is a few minutes shy of 
> 12 hours, since they is aligned to sidereal time, so the pattern shift 
> on the sky and it takes half a year to repeat exactly... if it where not 
> for orbital changes.
> 
> Cheers,
> Magnus
> 
> On 05/24/2016 02:01 AM, Tom Van Baak wrote:
>> Hi Skip,
>>
>>> Any help in understanding this behavior?  Thanks in advance.
>>
>> Yes, GPS satellites do repeat every ~12 hours in orbit around the mass of the earth -- but -- you and the earth turns 180 degrees during those 12 hours. So you're no longer where you should be when the 1st repeat occurs. Instead you have to wait yet another 12 hours for the earth to get back to the place where you were, in time to see the 2nd repeat. Now when you hear "get back, get back, get back to where you once belonged", you'll think of GPS satellites instead of the Beatles.
>>
>> So the LH plots are correct. Here's another take:
>>
>> 1) Say it's 6 PM MDT in Denver at lat/lon +39/-104 and you see a pattern of N satellites in the sky.
>> 2) Tomorrow morning at 6 AM MDT that same pattern will be in the sky -- not for you -- but for some guy at 6 PM lost in Inner Mongolia at lat/lon +39/+104.
>> 3) Tomorrow evening at 6 PM MDT that same pattern will again be in the sky, this time for you in Denver.
>>
>> /tvb
>>
>> ----- Original Message -----
>> From: "Skip Withrow" <skip.withrow at gmail.com>
>> To: "time-nuts" <time-nuts at febo.com>
>> Sent: Monday, May 23, 2016 3:43 PM
>> Subject: [time-nuts] I thought GPS repeated every 12 hours (-2 minutes)
>>
>>
>>> Hello Nuts,
>>>
>>> I am attaching a capture from Lady Heather of a 3-day run.  You can see the
>>> temperature vary by 7C over each day.  The TB is being run open loop and
>>> another GPSDO 10MHz input to the unit instead of the unit's oscillator.
>>>
>>> I expected the purple line to repeat every 12 hours based on the GPS
>>> constellation being the same (which maybe it kind of does), but there is
>>> definitely a 24 hour repeat.  What is really weird is that the number of
>>> satellites that LH sees also repeats on a 24 hour cycle, not 12 (bottom
>>> trace).
>>>
>>> Any help in understanding this behavior?  Thanks in advance.
>>>
>>> Skip Withrow
>>>
>>
>




More information about the Time-nuts_lists.febo.com mailing list