NTP? (accurate clocks)

[Jim Lux]

At 06:26 PM 10/11/2001 -0400, you wrote:
>On Thu, 11 Oct 2001, Greg Lindahl wrote:
>
> > Now if you want GOOD time synchronization (to within a microsecond)
> > that's a *much* harder problem...
>
>Yes, it starts to become very difficult and expensive right in there
>somewhere.  The so called "atomic clocks" one can now buy in stores like
>Wal Mart for $50 or so are autosync'd to WWVB at NIST in Colorado.
>Unfortunately, Colorado is about 1000 miles away, which is around 3
>microseconds right off the top that is difficult to compensate for.
>Also, this station broadcasts at only 60 kHz which means that time
>resolution is likely to be more like milliseconds or worse. Network
>jitter and so forth complicate net-based time sync connections, although
>one can get to millisecond accuracy on a good day (or over several good
>days).
>Beowulf mailing list, Beowulf at beowulf.org
>To change your subscription (digest mode or unsubscribe) visit 
>http://www.beowulf.org/mailman/listinfo/beowulf

I have built a variety of PC based systems that need to have very accurate 
absolute timing in UT terms(to record satellite signals that you are later 
going to compare against other records).  We used a variety of GPS based 
systems for this.

Your basic run of the mill GPS receiver has a 1pps output typically good to 
around 100 nanoseconds (absolute).  A variety of NTP and NTP server 
programs can take this input on one of the RS232 status lines  (like 
carrier detect or ring indicator) of a PC and distribute the time 
appropriately (limited mostly by the uncertainty in interrupt latency)

For better performance, there are a number of GPS disciplined clocks around 
for a few thousand bucks that can maintain accuracy to better than a part 
in 1E11 (which is awfully good...). They have a temperature compensated 
quartz oscillator that is "disciplined" by the GPS (which has some 
variability for a bunch of esoteric reasons).

There is also an inexpensive hacker way to go about it, if you are willing 
to do a bit of soldering.  The "totally accurate clock" kit is essentially 
an interface to an OEM GPS receiver that does the interface to the  PC and 
a hardware counter.  There is also a $50 kit to discipline a surplus VCXO 
(Voltage controlled crystal oscillator).

OEM Motorola receivers and antennas are available from 
http://www.synergy-gps.com/ among other suppliers.
Trimble and Garmin also sell appropriate products.

The summary is: For around $300 you can probably have atomic clock 
accuracy, at least over time spans of a day.  This is more than accurate 
enough to measure relativistic changes in satellite signals, the tectonic 
movement of the plate upon which your house sits, etc.

If you want very very accurate timing, things like antennas, choke rings, 
etc. become important to deal with multipath and things like shifts in the 
antenna phase center as a function of look angle.


Jim Lux
Spacecraft Telecommunications Equipment Section
Jet Propulsion Laboratory
4800 Oak Grove Road, Mail Stop 161-213
Pasadena CA 91109

818/354-2075, fax 818/393-6875