[time-nuts] KS-24361 REF-0 standalone

paul swed paulswedb at gmail.com
Sat Oct 3 12:09:49 EDT 2015


Dan it looks like the thread has re-developed.
As my early test of your simple solution demonstrated. Better PPS better
stability (Used a TBolt to test this) in the form of semi-short term
jitter. The beauty of Dans solution is the user can select the quality and
cost of the receiver used. Anything can work.
I used the 6M also because of the $11 cost and at the time as a alpha
tester there could have been issues. Reality, the 6M was on the shelf.
Regards
Paul
WB8TSL



On Sat, Oct 3, 2015 at 9:47 AM, Bob Camp <kb8tq at n1k.org> wrote:

> Hi
>
> If indeed you are into a professional system, where holdover matters  then
> a lashup on a bunch of surplus gear likely isn’t going to measure up. Based
> on about 10,000 previous posts on the list I’d say that holdover does not
> appear
> to matter to Time Nuts. It would be difficult to find a thread where
> holdover performance
> was the main topic. The focus (rightly) seems to be to keep the box locked
> to
> GPS all the time. That approach (if you can do it) will always give you
> the best
> timing.
>
> The “not a T” uBlox units *do* put out “TRAIM" information in the words you
> can get at. The data is valid and it can be used if you wish to put the
> unit
> in holdover. I’m sure it works better with a well known position, but that
> it not
> a requisite.
>
> TRAIM was a really big deal back in the mid 90’s. That’s when Motorola put
> it on
> the Oncore’s and did a lot of publicity on the topic. They saw it as a way
> to differentiate
> their product in the marketplace. Much of our view of TRAIM is slanted by
> the
> Motorola information on the topic.
>
> Here’s one way to look at TRAIM:
>
> If you are in position hold, you can get timing off of a single satellite.
> TRAIM
> (just like any estimator) looks at the single input it has and says “that
> must be
> correct”. Give it two inputs and it can look at the solution for each and
> decide if
> they are close enough (your TRAIM threshold) to be correct. As you get up
> to 8 or
> 12 inputs, the right answer may be to throw away the single one that is
> (say) a microsecond
> off from the rest. Somewhere between 3 satellites and 6 satellites, the “I
> need
> a solid position” thing becomes less of a factor.
>
> There are other ways to put the device into holdover. One is to simply
> look at the
> number of satellites. If you are locked on to less than 4 sats, go into
> holdover. It’s not
> elegant, but it does work. In the case of a lash up, just translate the “I
> have <= 3 sats” info into
> “my TRAIM is junk”. Instant holdover.
>
> Bob
>
>
> > On Oct 2, 2015, at 10:44 PM, Bill Hawkins <bill at iaxs.net> wrote:
> >
> > Actually, the Lucent software uses RAIM, and reports the value in its
> > status message. If the position appears to have drifted off, or there
> > aren't enough satellites to calculate the position, the software
> > declares the oscillators to be free-wheeling, an expression meaning that
> > the oscillators are free from discipline and are now drifting.
> >
> > So yes, the positioning aspects matter.
> >
> > Disclaimer: I haven't studied RAIM (or TRAIM) enough to know exactly
> > what goes on, but that's the behavior I've observed.
> >
> > Bill Hawkins
> >
> >
> > -----Original Message-----
> > From: time-nuts [mailto:time-nuts-bounces at febo.com] On Behalf Of Gregory
> > Beat
> > Sent: Friday, October 02, 2015 6:07 PM
> > To: time-nuts at febo.com
> > Subject: Re: [time-nuts] KS-24361 REF-0 standalone
> >
> > Dan -
> >
> > I have been following your experimentation with the surplus Lucent
> > KS-24361 REF-0 module, to transform it into a standalone GPSDO.
> >
> > The original usage of the classic Oncore UT+ GPS receiver for KS-24361
> > REF-1, by Symmetricom / Datum for Lucent, was deliberate.
> > For usage at a cellular data/telecom site, the focus was on the timing
> > and frequency discipline from the GPS satellite transmission, rather
> > than the position or dead reckoning aspects -- used by smartphones,
> > automobiles, and other GPS applications on the market.
> > ===
> > A couple of comments.
> > While I can appreciate being economical (main criteria) and selecting
> > the NEO-6M receiver, I believe that a u-Blox timing specific module
> > (like LEA-6T) would be more desirable in this application.
> >
> > In addition, the u-Blox 6-series is the trailing edge of product support
> > (market demand dictates its continuance), while the 7 and 8-series are
> > their current modules (largely for the cellular / mobile industry
> > (smartphones or cell sites themselves)
> >
> > u-Blox 6-series Timing Application Note (using the LEA-6T)
> > https://www.u-blox.com/sites/default/files/products/documents/Timing_App
> > Note_%28GPS.G6-X-11007%29.pdf
> >
> > IF you successfully adopt the u-Blox module to correctly "mimic" the
> > Oncore UT+ GPS receiver command suite, THEN you open up a larger
> > audience of "time-nuts" and Frequency Standard users (HP Z3801A
> > frequency standard universe) as a receiver alternative.
> > http://www.realhamradio.com/GPS_Frequency_Standard.htm
> >
> > These users may desire a "newer" GPS receiver that has more channels
> > (8-channel); latest generation receiver; access to the newest GPS
> > constellations.
> > TAPR might be interested in sponsoring, as a kit/module, if a wider
> > audience existed.
> >
> > The Heol Designs N024 receiver (France) accomplished this replacement
> > role for the Trimble ACE II/III GPS receiver used in the
> > Symmetricom/Datum TymServe TS2100.
> > Their solution resolved shortcomings in the mid-1990 Trimble receiver
> > design and giving this Symmetricom NTP server, time IRIG-B time code
> > generator, and 10 MHz reference appliance a new lease on life (no longer
> > a door stop).
> > http://www.heoldesign.com/index.php?module=products&action=catalog&cat=1
> > 4&id=54
> >
> > w9gb
> >
> >
> > _______________________________________________
> > 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.
>


More information about the time-nuts mailing list