[time-nuts] Navsync CW12 - Great News! New Firmware Load!

Ed Palmer ed_palmer at sasktel.net
Fri Jan 20 18:23:49 EST 2012


You may remember my rant a few weeks ago regarding the 10 MHz output
of the Navsync CW12 GPS receiver.  The issue was that it was, by
design, slightly off frequency.  Further testing showed that the
offset was also temperature-dependent.  Although this doesn't affect
its usability for it's intended applications, the 10 MHz output
wasn't appropriate for typical Time-Nuts applications.

Connor-Winfield (Navsync's parent company) contacted me and offered
a firmware load that turns the CW12 into a CW12-wi.  The 10 MHz output
changes from a frequency-locked-loop with an offset to a phase-locked-
loop with no offset.  This load also replaces both 8-channel and
12-Channel OnCore receivers.  This isn't a custom Time-Nuts version.
It was developed for a particular customer and just happens to resolve
the current issue. If you have the NMEA version of the CW12, you can
still apply this firmware, but this load emulates the Motorola binary
commands only.  You will lose all NMEA capabilities and messages.
Probably the biggest thing you'll lose is the ability to change the
frequency of the 10 MHz output.

I am pleased to report that this firmware has resolved the frequency
offset issue.

I tested both the 1 PPS and 10 MHz outputs and found no drift,
just the random wander typical of any GPS receiver.  The jitter
is about the same as I measured previously.  Overlapping Allan
Deviation is comparable to other GPS receivers.  e.g.
http://www.leapsecond.com/pages/3gps/gps-adev-mdev.gif .

Testing the 10 MHz output with an HP 5372A Time Interval Analyzer in
Histogram Time Interval mode showed that the algorithm has been
completely changed.  Measuring the period of 50M cycles shows two
normal distributions, one centered at 94.0 ns (~20M readings) and
the other at 104.4 ns (~30M readings).  Apparently, this load
'brackets' the 100 ns point by sending short and long cycles on
an almost one-to-one basis that average to 100ns.  See
http://www.febo.com/pipermail/time-nuts/2011-February/054307.html
for similar tests that I did on the existing firmware.

Conner-Winfield has asked me not to distribute the software
publicly so if you have a CW12 and want the new load, please
contact me off-list and I will send you a link to the file.
It includes the firmware, flash program, instructions on how
to perform the upgrade, and a note on a  WinOnCore incompatibility
and work-around.  The upgrade requires soldering a zero ohm
resistor or a piece of wire onto the CW12.

If anyone wants to buy a new CW12-wi, I've been told that you
can order it directly from Connor-Winfield.  I don't have any
pricing information.

I would like to thank Conner-Winfield for making this firmware
available to the Time-Nuts community and I would like to apologize
for any offence caused by my rant.

Ed





More information about the time-nuts mailing list