[time-nuts] Re: NIST GPS Data Archive

[Magnus Danielson]

Hi Skip,

First of all, good work and a nice read!

Yes, the TMAS system still is single-frequency common view processing. I 
asked again as I hung out with the NIST folks during WSTS in Denver. For 
what they need, it remains an economical solution. Prices for new 
double-frequency receivers have dropped, so eventually they may make the 
change.

I assume you force-feed a Thunderbolt with the 10 MHz from the maser and 
then just look at the differences, without trying to close the loop.

Can maybe the NIST data be delivered in some standard format so that 
Lady Heather (or other program) could eat that, or should one implement 
it more directly? I would love a core library to be readabtable to 
various useages.

Cheers,
Magnus

On 8/8/22 18:02, Skip Withrow via time-nuts wrote:
> In my quest to get some decent real-time comparison to GPS it was suggested
> to look into the NIST TMAS system.  That is when I discovered the NIST GPS
> Data Archive.  However, my first attempts at correlating my reception
> results with the data archive were miserable failures.
>
> I saw that Mike Lombardi from NIST was on several of the TMAS papers, so
> shot off a quick email to him.  He was very kind and actually gave me a
> call.  He cleared up several questions and the results are looking much
> better.  Thanks!
>
> I was originally using a dual-band L1/L2 receiver in fixed position mode
> with WAAS corrections turned on.  Turns out the NIST archive is done with a
> single-band L1 receiver operating in fixed position mode using only
> receiver computed ionosphere corrections.
>
> I also noted in the archive that the hourly values did not match the
> 10-minute values.  This is because the six 10-minute values are averaged to
> give the hourly value.  AND the 10-minute values are the average of 600
> 1-second readings.
>
> So it was time to put Lady Heather to work.  Using a Trimble Thunderbolt I
> have plotted the 10-minute averages for the last seven days.  So my plot is
> 10-minute averages and the NIST plot is of the hourly averages.  A
> comparison picture is attached.  The Thunderbolt is being driven from the
> MHM-A1 maser which I thought was running about 3 nanoseconds per day slow,
> but it appears to be much closer than that.  In the picture that is
> attached both NIST and LH are plotted at 2ns/div vertical, the horizontal
> scales are close but not exact.
>
> Next task would obviously be to get some automation involved to collect the
> NIST data, Thunderbolt data, and the dual-band data and start crunching
> some numbers.
>
> Regards,
> Skip Withrow
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com