Hello Jim,

I am just a novice here - but, when you say noticeable, can you please tell
us how noticeable?  What is a small downturn vs. a noticeable one?

I just put up an outdoor GPS antenna.  If there is anything I can do to
possibly help, please give me instructions.
I do not have direct access to any Hydrogen masers at the moment.

Thanks,
John W.

On Thu, Oct 9, 2014 at 4:43 PM, Jim Palfreyman jim77742@gmail.com wrote:

Hi

GPS is steered by the Air Force last time I checked.

A really good place to check is the NIST Time and Frequency pages that show both real time and historical data for each GPS sat compared to NIST time:

http://www.nist.gov/pml/div688/grp40/gpsarchive.cfm

Hopefully it’s accessible via that link from a variety of locations.

Since the NIST data is independent of the steering (two different outfits involved) it should not be vulnerable to a “our ground segment broke and we steered everything to match” sort of error.

Bob

On Oct 9, 2014, at 7:43 PM, Jim Palfreyman jim77742@gmail.com wrote:

Le 10 oct. 2014 à 03:09, Bob Camp a écrit :

I remember Jim reported a similar issue back in october last year:

That dates are close enough to make you wonder if it is not part of some cycle.

On 2014-10-09 23:06, mike cook wrote:

From: http://www.usno.navy.mil/USNO/time/gps/gps-info
GPS SYSTEM TIME
GPS system time is given by its Composite Clock (CC).
The CC or "paper" clock consists of all operational Monitor Station and satellite
frequency standards. GPS system time, in turn, is referenced to the Master Clock
(MC) at the USNO and steered to UTC(USNO) from which system time will not deviate
by more than one microsecond. The exact difference is contained in the navigation
message in the form of two constants, A0 and A1, giving the time difference and
rate of system time against UTC(USNO,MC).

Page also gives links to GPS time data ftp://tycho.usno.navy.mil/pub/gps/utcgps30.dat
which shows a 2ns jump in UTC(USNO)-GPS smoothed over 2 days from Oct 7-8, but that
appears normal; the <1ns differences from Oct 2-7 appear anomalous.

Looking at the NIST 10 min data, from Oct 3-8 the gap between GPS samples and NIST
closed about 1.5ns/day, dropping now to about .5ns/day: the graph shows the values
sliding down to the right, and now levelling off about zero.

So are NIST and USNO steering each other?

--
Take care. Thanks, Brian Inglis

No. They don't directly if that's what you mean.  They do regular time
transfers, likely more regularly than most due to accessibility to a common
view and common interest.

The GPS consolation is actually a sub-(sub?)-scale of UTC(USNO_MC) called i
think USNO_OSC?  Have to check.

They have been building out the  timescale ensemble at Schriever
significantly over yhe last 5 or so years. They were up to 26 references
planned last  i spoke with anyone back home - the delta being primarily
H-masers.

I think that's a good move. Despite Timefreq being just  about  2 hours
away, that master clock's survivability  was never certain.  Shipping the
timescale to Colorado Springs from Wisconsin Ave/Bethesda did contain risk,
and they are mitigating it.

I was once told - total anecdote and certainly 15 years ago - that despite
proximity and willingness they rarely exchange views or collaborate
experimentally between Colorado Springs and Boulder. Silly in  my mind but
it is an a active weapons platform and the logistics may not be conducive.

NS

On Friday, October 10, 2014, Brian Inglis Brian.Inglis@systematicsw.ab.ca
wrote:

I'm trying to determine semantics and I was likely mistaken.

They regularly update them both a variety of technologies - TWTCC and
others. Then DC of course sends to COS.

Is it actually possible to phase lock two oscillators together cross the
distance from DC  to  Colorado Springs? (2400 kilometers or so). ?

On Friday, October 10, 2014, Brian Inglis Brian.Inglis@systematicsw.ab.ca
wrote:

Hi

On Oct 10, 2014, at 11:20 PM, Brian Inglis Brian.Inglis@SystematicSw.ab.ca wrote:

That’s a pretty small group

Because USNO is the official keeper of “time” in the US. NIST is the official for frequency.

Consider in all the data that it is coming from fairly normal receivers. They use good survey grade stuff, but it’s a receiver you could buy off the shelf. Pops do happen

All the data you see from NIST and USNO in terms of “what time is it” is the output of some very fancy filtering. They take a weighted set of inputs from a large number of sources. They also do a cross check through BIH to keep in synch worldwide. In the sense that BIH corrections get in the mix, everything is locked to everything.

The GPS master clock is a similar thing. GPS master time (the ground clocks) are fed into a fancy software filter to come up with a local time estimate. That’s massaged to track USNO (more software). The ground estimate is used to look at and steer the satellite clocks through still another software filter.

All of this stuff has crazy long memory in it, so a “once every three months” or even a “once a year” update of some sort is not at all out of the question. In a sense, leap seconds are a “once very rarely” correction to all this stuff.

If there was no notice of a change on the various GPS / NIST / USNO / BIH web sites, I’d look for a paper at an upcoming conference. Depending on who did the bump (if there was one) and who was simply following the herd, that paper could pop up in a lot of different places.

Bob