[time-nuts] 58503A date code problem
GandalfG8 at aol.com
GandalfG8 at aol.com
Sun Jun 22 04:22:57 EDT 2014
Date: Sat, 21 Jun 2014 13:06:19 -0400
From: Daniel Burch <daniel.burch at ieee.org>
To: time-nuts at febo.com
Subject: [time-nuts] 58503A date code problem
<CAMyugJy7c6LyX7q26zwUM40iEzMFY5ZShrCQ_YGDa5T4VC3sTA at mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
I have a HP/Symm 58530A that has the correct time, but date keeps
defaulting to 1994, Nov, 4 after GPS Lock. The pre-lock is 1996, so I do
see a change when it locks, just to the wrong date.....time is exactly
correct and tracks.
It would seem likely that this could become an ongoing problem.
I don't know what GPS module is used in the 58503A, but last August it was
the turn of the Z3815 thanks to the onboard Furuno GT74.
A pair of 1999 Time and Frequency Solutions GPS8 units I own use the same
Furuno GPS module and suffered the same fate, and some versions of the
Rapco 1804 of similar vintage have also been similarly afflicted. Although in
the case of the 1804 it seems to be a Rapco firmware issue rather than
anything to do with their Trimble SVeeSix GPS modules as it hasn't so far struck
all versions of the same product.
The "problem" is the well known and well documented 1024 week rollover
issue, exacerbated it would seem by some manufacturers, or individual
designers, choosing an arbitrary date for the start of their own 1024 week period,
perhaps either an initial production date or firmware compilation date, and
then not seeing any need to update that as time went by, probably because
they weren't anticipating a near 20 year product life.
With the date chosen unlikely to be public knowledge and perhaps just lost
in time anyway, the "failure" can come as a surprise to all and sundry,
However, if any such manufacturers did have ongoing production of the same
module over an extended period, and had the foresight to update the start
of their 1024 week period at either regular or irregular intervals as
production continued, then we also face the interesting possibilty of
successively failing batches of the same product rather than all showing up in one go.
More information about the time-nuts