[time-nuts] A couple of answers to recent questions
w3iwi at toad.net
Tue Aug 9 16:33:12 EDT 2005
Relative timing like that should certainly be doable using the
and right setup. People that do this uses Geodesic receivers such
Ashtech Z12 and a good antenna.
With the relative proximity of only a few km away, the common view
correlate quite well.
Using terms like "common view" and "Ashtech Z12" you will Google
alot of usefull information. You should harvest the archives of
NIST, PTTI etc.
for good articles on the topic.
I didn't suggest the use of high quality, dual frequency, expensive
receivers (like the Z12) because Shaun made the statement that his
area of interest was only a few km in scale. Over that size network,
nearly all of the errors due to the ionosphere (which forces the need
for two-frequencies), troposphere & satellite orbit errors cancel out
as a common mode source (i.e. the stations see identical errors). Only
differential multipath, cable stability and mechanical integrity of
the antenna are of much concern. Note that I also made the presumption
that what he really needs is RELATIVE timing. BTW, I know of several
other examples of the use of GPS to achieve "network" timing at <10
nsec over scales of tens of km.
The otherday I asked about the Jupoiter T receivers and Brooke mentioned the re
sults of his experience with it, I was curious what he meant when he spoke of t
he "clock zero-beat times (TvB calls them "hanging bridges")" .Brooke, can you
or Tom direct me to somewhere where I might get a better understanding of this?
Take a look at figures 9 & 10 in
ftp://ftp.cnssys.com/pub/ION-GPS2000/ion-time.pdf and you will see
examples of the receiver clock going thru zero-beat.
For the ONCORE & M12+ receivers, the explanation for the zero-beat,
hanging bridge phenomena (as well as the underlying sawtooth error) is
seen in ftp://ftp.cnssys.com/pub/ivs-tow-2005/tow-time2005.ppt in
For the Jupiter-T, the effect is the same but the reason is slightly
different. The Collins/Rockwell/Conexxant chipset generates its 1PPS
signal from a programmable numerically controlled oscillator (NCO)
programmed with the differential frequency error for the next pulse
(instead of the long counter for the ONCORE where the integral time
error is controlled). The manufacturer does not keep track of the
differential "seed" (or equivalently the constant of integration) for
the NCO. The result is that the Jupiter-T does not have a serial data
output with the pulse error as does the ONCORE. Aside from that
difference, the Jupiter-T is as good a timing receiver as any of the
More information about the time-nuts