[time-nuts] worth salvaging ?

David Martindale dave.martindale at gmail.com
Sun Jan 30 00:03:21 UTC 2011

On Sat, Jan 29, 2011 at 10:14 AM, Mark J. Blair <nf6x at nf6x.net> wrote:

> In a receiver that doesn't have a recent almanac, and particularly in an older receiver that takes a very simple approach to downloading ephemeris and almanac information, initial acquisition could take a long time. It'll need to do a slow full-sky search for its first satellite, and older receivers couldn't do that nearly as quickly as newer ones can. Once it gets that first bird, it may sit there downloading ephemeris and almanac data for at least 12.5 minutes before it does anything else. With an old receiver from that era, give it at least a half hour of good open-sky conditions before you begin to suspect that it's dead.

It all depends on the receiver firmware.  I remember when the GPS week
rolled over; I was in the process of driving from a conference in
California back to home in British Columbia, with a GPS receiver or
two in the car tracking our progress.  We stopped for a meal, powering
the GPS receivers off.  When we returned to the car, one of the
receivers (a Garmin) would not re-acquire satellites no matter how
long I gave it.  I later figured out that the week rollover had
happened that afternoon, while we were driving, and the first restart
after the rollover failed.

But it apparently wasn't just a case of having to do a cold start to
get a new almanac, because I couldn't get that receiver to work again
myself.  In a little while, Garmin released a small utility that you
ran on a PC connected to the GPS via its serial port, and it reset
something that allowed the GPS to do a successful cold start.  I
remember an rumour that it simply reset the saved date and time far
enough away that the receiver dumped all its old almanac entries,
forcing it to do a cold start that worked.  This suggests that it had
been keeping its pre-rollover almanac data and tried to use it, but
there was a bug in the calculation that resulted in mispredicting what
satellites should be visible where.  But that's just a guess - Garmin
never said exactly what was wrong, or what they did to fix it.

> Back to the original topic now: That OCXO may seem mundane by time-nutty standards, but I'd certainly consider it to be worth salvaging. It could have all sorts of applications for radio stuff, portable test equipment, and even time-nutty stuff in an application that wants to be smaller and more portable than a Rb standard or full GPSDO.

Yes.  If you offered them for postage cost, you'd probably find a
bunch of takers.  You wouldn't even have to unsolder them - let the
recipient do that.  It seems a waste to discard a perfectly good OCXO
- it would make a fine upgrade to an inexpensive frequency counter
(many of which just have a bare crystal, not even a TCXO).

More information about the time-nuts mailing list