The 1. jan 2004 around 18.30 UTC, the clock in one of the
GPS-satellites had a malfunction. This resulted in a distance-fault
from this satellite of up to 285 km (normally under 10 meter).

A fault like that is not serious if it is spotted immediately, and the
satellite is put out of play, so it isn't used in the position
calculation.
In this case the fault wasn't spotted by the GPS-earth-station until 3
hours later. Then the satellite were labelled "unhealthy" so it would
be left out of the calculations.

For 3 hours GPS receivers calculated faulty positions up to several nm
out of place.

Published 30. jun 2004: http://www.frv.dk/ifm/navigation/gps.htm
This link is better: http://www.frv.dk/nyheder/nyheder.htm#54

This illustrates the limitations of this in many ways wonderful and
cheap device.


Peter S/Y Anicula