On Wed 14 Jan. Brian Curtis wrote -
> In the Sydney Sun-Herald on sunday 11th they had an article
> on the millenium bug and they referred to Global Position
> Systems and in a short one liner they made a statement thet
> the GPS system had a glitch 132 days before the 1/1/2000.
> That was all they said.
>
> What are the consequences of this glitch for those of us who use
the GPS to atlas ?
>
> Brian Curtis
Below is a thread from my other passion - soaring (or gliding, as some
people call it). It is probably more than you bargained for, but it
explains the 130 day thing.
Regards
Terry Gould.
# # # # # # # # # # # # # # # # # # # # # # # # #
From:
To:
Cc:
Subject: GPS Millenium Problems
Date: Wednesday, 11 September 1996 8:34AM
I recently put the following post into rec.aviation.soaring and also
into sci.geo.sattelite-nav:
An article in New Scientist (10th Aug 1996) suggests possible problems
for GPSunits, with the millenium date change over.
The GPS units run their own clock which is kept accurate by the satelite
signals. Position sensing depends on differences between the time
signals transmitted by the sats and the GPS receivers clock.
Aparently the GPS clocks all start on the 5th January 1980 and should
roll over 20 years later. The problem is that the counters in use are
binary and count 20 years as 2^10 weeks (7168 days). As a result the
clocks will roll over 130 days early on 21 August 1999. The resultant
difference between the sats clocks and the GPS units clocks may cause
confusion to the GPS receivers.
Anybody know what the truth on this is? I use mine soaring
* * * * * * * * * * * * * * * *
This response came from sci-geo-sattelite-nav:
True, more or less. A matter of firmware design. See posting. A Plessey
spokesman says that it is hard to predict how individual receiver brands
will re-act because each manufacturer uses its own software.. The Plessy
spokesman has it right. The only solution for users is to question their
manufacturer (not dealer) closely.
Joe Gwinn Posting follows:
I have good news and I have bad news. The good news is that the NAVSTAR
Global Positioning System (GPS) will not generally have a "Year 2000"
problem. As part of answering our customers' queries about possible
"Year 2000" problems, I looked into the more general question of
rollover in our time sources. The bad news is that GPS System Time will
roll over at midnight 21-22 August 1999, 132 days before the turn of the
millennium. (To be pedantic, because there was no "year zero", the
millennium actually will turn one year later, at 00:00 UTC 1 January
2001, but never mind.) On 22 August 1999, unless repaired, many or all
GPS receivers will claim that it is 6 January 1980, 23 August will
become 7 January, and so on. Accuracy of navigation may also be
affected, perhaps severely.
Although it appears that GPS broadcasts do contain sufficient data to
ensure that navigation need not be affected by rollover in 1999, it is
not proven that the firmware in all receivers will take the rollovers in
stride; some receivers may claim wildly-wrong locations in addition to
incorrect dates. A few receivers may in addition suffer from a classical
Year-2000 problem. How big a problem is this? For large customers, not
very. They will simply insist that the receiver manufacturers solve the
problem, and they have three years to ensure that the problem is solved.
For small customers? It may cost them a repair. For the manufacturers?
It varies, but they also have three years to solve the problem, and to
spread out and deflect the expense of the solution. Many receiver
warrantees have probably expired by now, so part of the expense will be
deflected to the customers. One would expect that no receiver sold after
1996 will have the problem, so only out-of-warrantee units will be
involved at rollover in 1999. The problem for the manufacturers will be
that customers will demand repair now, while their warrantees are still
valid, and/or complain about "latent design defects" even after the
warrantee has run out. As for the repair cost per receiver, it varies
with the details of construction of the receiver. For the $300
mass-market units: Units returned for repair are probably just replaced,
at customer expense, but at something like the cost of manufacture
(versus the retail sale price). As for the $3,000 to $30,000 units,
these will very likely be repaired. If the PROMs (Programmable Read-Only
Memories) are socketed, then it will be fairly cheap - just mail new
ones to the customer, who installs them. If soldered, it's harder, but
still not that hard. So, GPS rollover is a problem only for those caught
by surprise. As a technical matter, the solution is quite simple. It's
the logistics that will take some years. And, the software using GPS
will need to be imbued with a certain healthy skepticism about the
gospel according to GPS.
The details: Section 3.3.4(b) (page 33) of the ICD-GPS-200 rev B (30
November 1987 issue) states that the "GPS Week count starts at zero at
midnight 5-6 January 1980 UTC, and that the GPS Week field is modulo
1024. This means that the week count will roll over from 1023 back to
zero (1024/52)= 19.69 years from then, or in 1980+19.7= 1999.7 (August
1999), only a few years from now.
* * * * * * * * * * * * * * * *
There then follows a long more trechnical explanation of the matter.
(Anybody who wants it, send me an e-maIL)
So, the question is, will your GPS unit be affected, What about those
expensive you beaut Cambridge units and all the other approved recording
devices, some with built in GPS receivers??
Rudi Treutlein
P.S. By that time replacement cost of small handhelds will probably be
low enough to just chuck your present unit away and buy another.
# # # # # # # # # # # # # # # # # # # # # # # # #
|
