[time-nuts] Thunderbolt GPSDO
didier at cox.net
Thu Feb 15 22:45:44 EST 2007
This is correct, to the sense that a working GPSDO has sufficient status
information that you know with great probability of being right when the
device operates per specification, particularly when there is a good GPS
signal available from several satellites (3 minimum, 5 is good, more is
The actual performance may be affected by a number of factors which may
not immediately generate an alarm, yet may cause the device to operate
at the limits or maybe even outside the specification for a short time,
and for some people or some applications, that could be a problem.
That's when having more than one unit is helpful.
My original comment about needing 3 or more units was a little bit
tongue-in-cheek, but considering this is the Time-Nuts mailing list,
maybe not completely out of place :-)
That's what a lot of discussions on this mailing list have centered
around recently, and I find this area very interesting and challenging.
So, if you simply want a good reference for your instrumentation, a
stock Thunderbolt and a well placed antenna will be great, and per
specification will get you within 20nS of UTC (at 1 sigma) and within
1e-10 in a second or more (almost 1e-12 in a day) as long as there are
GPS signals. More impressively (compared to other units in a similar
price range), it will drift less than 1uS/hour in holdover mode (no GPS
signal) once stabilized for a day or so (per specification).
I think the Thunderbolt was designed for cellular base station
applications and therefore is a pretty dependable and reliable unit.
Mine has been running 24/7 since November and based on it's data output,
has been consistently within a few nS of what it thinks UTC is. That's
good enough for me.
PS: when you install and power up a Thunderbolt for the first time, it
will do a self-survey to determine it's position. Depending on the
quality of the GPS signal, it may take from one to a few hours. Once
that's done, you will know it's location with great precision. You can
then save it to EEPROM using the Trimble software, which will make
re-acquisition (if you loose power) faster the next time.
John Miles wrote:
> To answer the original question, a Thunderbolt or similar GPS standard is
> MORE than capable of telling you if it is working properly. There is no
> reason to use more than one of them at a single location. The Trimble
> software will tell you how many nanoseconds the 1-PPS output is off, and how
> many parts per billion the 10-MHz output is off, compared to what GPS says
> they should be. And there are numerous diagnostic alarms that are monitored
> It is a plug-and-forget device, except that you do want to set the saved
> location and environment (trees, fixed/mobile, etc.) properties when you
> install it.
> -- john, KE5FX
More information about the time-nuts