GPS and Y2K

Questions for any visitor: 

• Is the GPS timing signal susceptible to WNRO and/or Y2K? The concern has surfaced whether failure of these timing signals will affect network synchronization in the national electrical distribution grid, and in the national telecommunication network.
• Will there be any problem because of the lack of synchronization between servers on the internet. I see there is quite a range of dates and times. This will mean a distibution of transition times stretching over several days, rather than just at midnight.

GPS World Home Page

• The GPS End-of-Week Rollover - Richard B. Langley, University of New Brunswick
  • When the rollover occurs, some noncompliant receivers might simply report an incorrect date, or they may incorrectly determine the position of satellites from the ephemeris transmitted in the navigation message, therefore computing and displaying incorrect receiver positions. Alternatively, they might sense that an error has occurred and simply refuse to calculate a position. Such problems may or may not go away if the receiver has its power cycled off and on.
  • If a noncompliant receiver is switched on after the rollover occurs, it could experience longer startup times or fail to lock onto satellites. Yet, it might be prompted to start working normally by performing an auto locate or search the sky operation in order to acquire satellites, collect a fresh almanac, and perform navigation functions.
  • Some units may avoid trouble during the first year or years of the new GPS week cycle but encounter it later when the difference between the programmed reference year and the current year exceeds the length of one GPS week cycle (about 19.6 years).

Sustainable Development Reference Link:  Y2K Bug and GPS EOW Bug

• The Y2K or Millennium bug will affect non-compliant computers, software and devices using microchips on January 1, 2000.  The GPS (Global Positioning System) End of Week (EOW) rollover problem will affect non-compliant GPS receivers or other devices which use GPS satellites to calculate time in August 1999. 
• GPS EOW Rollover Bug
  • The bug is expected to be more of a problem with older equipment (pre-1994) than more recent equipment.  
  • GPS receiver manufacturers
  • two sites which have partial lists of compliant and non-compliant equipment. 
  • GPS time keeping is used to set the clocks which regulate international communications and computer networks.  These GPS clocks are also used in banking for money transfers and bank time locks among other things.  They are also used for time keeping in certain scientific experiments. For a detailed explanation of how GPS time-keeping works, click here to view a web page by the National Research Council (NRC) of Canada.
  • However, by collecting information from a number of sources, it has been possible to identify some of the potential problems:
    • The GPS receiver may not be able to locate the GPS satellites.  In this case the receiver will not work at all.
    • The receiver may take a longer time than usual, possibly up to two hours, to locate the satellites.   Having found the satellites it may or may not display accurate dates, times or positions.
    • The receiver may display an accurate position but the date could be as much as 19 or 20 years in error.
    • The receiver may display a position which is not correct.
  • Manufacturers of GPS timekeeping equipment
    • Magellan Year 2000 and GPS End-of-Week Rollover Information (Ashtech)
      • G12 (One example of compliant with minor issues)

      1. The receiver will continue tracking and generating raw measurement data and position messages over the GPS EOW and Y2K rollover.
      2. Raw measurement data and position computations will continue to be correct and accurate.
      3. At the GPS EOW and Y2K, the receiver will miss one epoch of data.
      4. The receiver will stop computing a position temporarily after Y2K rollover; starting at 00:00:14 January 2, 2000 UTC, and lasting for 48 seconds. The receiver recovers automatically.
      5. The week number will go to 0 in all messages at the GPS EOW. This will not affect the accuracy or the integrity of the receiver’s position computations.

      • Continuous Geodetic Reference Station

      1. The receiver will continue tracking and generating raw measurement data and position messages over the GPS EOW and Y2K.
      2. Raw measurement data and position computations will continue to be correct and accurate.
      3. The week number will go to 0 at the GPS EOW. This will not affect the accuracy or the integrity of the receiver’s position computations.
      4. The date will go from August 21, 1999 to January 6, 1980 at the GPS EOW

    • Year 2000 and End of Week Rollover Information -- CSI

      • GBX-6: GPS Engine: Trimble SVeeSix CM3 v4.06 (NMEA): Compliant with minor issues

        • Position fix may be interrupted and the PPS signal may be unstable for up to 6 minutes after WNRO event.
        • Position fixes and PPS signal will return to normal operation after the interruption.
        • The NMEA RMC and ZDA messages will report incorrect calendar date after GPS EOW.
        • The first warm start or hot start after WNRO may require up to 15 minutes.

      • GBX-12/A: GPS Engine: Ashtech Sensor II: Compliant with minor issues

        • The receiver will continue tracking and generating raw measurement data and position messages over the GPS EOW and Y2K.
        • Raw measurement data and position computations will continue to be correct and accurate.
        • The week number will go to 0 in all messages at the GPS EOW.  This will not affect the accuracy or the integrity of the receiver's position computations.
        • The date will go from August 21, 1999 to January 6, 1980 at the GPS EOW.
        • Navigation is unaffected
      • Week Roll Over Solutions (Leica) (One example)
        • GPS Engine - Version T and older will not navigate through the Week Roll Over. Version V will function normally through the Week Roll Over, but after the Roll Over any power cycle will cause the date to revert to the corresponding date 19.7 years earlier.
      • 1023Y2K (Novatel)

        • 3711 3.351 2 Y

          3751 3.351 2 Y

          3911 3.351 2 Y

          3951 3.351 2 Y

          RT20-PC 3.351 2 Y

          GISROV 3.351 2 Y

          GISRT20 3.351 2 Y

          AG20 3.351 2 Y

          DREDGER 3.351 2 Y

          HYDROSRV 3.351 2 Y

          MILL3151R 4.47 4 Y

          MILLRT20S 4.47 4 Y

          MILLENSTD 4.47 4 Y

          MILLENRT2 4.47 4 Y

          Item of non-compliance:

          1. The week 1023 rollover does not go to 0, but to 1024. This only happens if the card is left on and running through the week roll-over time (00:00 Sunday morning Aug. 22, 1999 GMT, which will be noon Saturday Aug. 21, 1999 to noon Sunday Aug. 22, 1999 depending on local time zones / daylight savings). If the card is not powered during this rollover, the week will report properly when powered after the rollover. If the card is running through this period, the card will need to have power cycled in order to restore it to normal operation.

          2. During the 1023 week rollover, there is a significant jump in position (several meters). This error will stay in the position solution until the power to the card is cycled. If the card is not powered during this rollover, the position will report properly when powered after the rollover. If the card is running through this period, the card will need to have power cycled in order to restore it to normal operation.

          3. After the 1999-2000 rollover, the NMEA GPZDA log (which reports a date like 12/31/1997) reports a date which is not accurate. The card does not report a correct date in this log from Jan. 1, 2000 onwards. However, the GPS week and seconds of week continue to report correctly, subject to item 1 above.

          4. The RBT and SBT logs (Satellite Broadcast Data, raw bits and symbols) do not report a correct GPS week near the 1023 week rollover (for only a few seconds). These two logs are not commonly used logs. They report the unaltered broadcast messages from the satellites. NovAtel intends to fix this shortcoming in a future release.

Clock Server Hardware Products

• GPS introduction
• GPS world
• GPS-GIS information
• GPS links
• GPS Resource

GPS Technical Note (FCC) - Stan Trost, FCC Y2K Task Force "

• GPS Use in Timing Systems - As mentioned above, GPS systems rely on a precise timing source (clock) to perform their calculations. This aspect of GPS receivers allows them to generate both relative and absolute time. Fast digital networks use an external clock to enable synchronous data transmission. For some data transmission systems, GPS based timing systems are used to provide the necessary timing information.
• Conclusion - GPS is a widely used system in both military and commercial application. Satellite systems will be unaffected by any dates in 1999 and 2000, but receivers and ground equipment are subject to two date failures. Ability of this equipment to respond correctly needs to be checked with equipment manufacturers."

• GPS Joint Program Office (Department of Defense)
• a partial list of manufacturers

Estaban:  http://home1.gte.net/esteban/GPSBUG.htm Bug in GPS Time Standard receiver equipment

• I am a SCADA (computer automation) Technician for a large utility in California. So when I heard about another Counter Register with a ROLLOVER problem I discovered after some research not only would most GPS navigational receivers need a software upgrade some will be unrepairable. The BUG is related to the 1024 Week counter on August 22, 1999 all GPS units will roll from 1023 to 0000 this may cause widespread failures on a large scale!
• I started to wonder about all the TIME STANDARD receivers that are tied into most every Telecom site, mainframe computer and Frequency deviation controller for Electrical Grids. So I called a popular manufacturer of these devices and spoke to the service department. Just as I suspected this is a "BIG PROBLEM"

TrueTime: http://www.truetime.com "TrueTime's lengthy experience in producing premier, high-precision time and frequency products uniquely positions us to best understand the intricacies and importance of delivering hardware products that are Year 2000 compliant. To this end we are thoroughly testing our products with the best means available."

• Year 2000-GPS Week Rollover Overview
• TrueTime Rollover Test Methodology
• New GPS-Based NTS-90 Network Time Server from TrueTime Provides Cost-Effective Way to Synchronize Clocks on a Network

Measuring instruments & systems, meters, analyzers, monitors - ElectricNet

• DATARADIO Inc. - Dataradio's Year 2000 Statement
• Magellan Systems Corporation - Magellan Products In 1988, Magellan introduced the first hand-held Global Positioning System (GPS) receiver. Since then, we've guided more than a million people on adventures all over the world.
• Meinberg Funkuhren - Meinberg Product Overview
• Motorola -
• Sokkia Corporation
• Trimble

NavStar GPS Joint Program Office NavStar GPS Homepage

• Year 2000
• GPS Control Segment is certified
• Year 2000 and GPS End of Week Rollover
• GPS for Air and Space #136 (GPS Newsletters List)

GPS User Equipment Home Page - Joint Service System Management Office (JSSMO) located at Robins Air Force Base, Georgia, home to the Global Positioning System Joint Service System Management Office (JSSMO)

US Naval Observatory GPS Operations

• GPS Week Number Rollover
• GPS Time Transfer Operations
• Computer Time Services
  • Time Synchronization Software by Dr. David L. Mills, University of Delaware
  • Network Time Protocol (NTP)
  • Time Synchronization Explanation

US Coast Guard : Week Rollover

• The firmware in all affected (mostly older) receivers will have to be replaced. This will involve replacement of PROMs; some are socketed, some are soldered. As a technical matter, the solution is quite simple. It's the logistics that will take some effort.
• Without a GPS Simulator, there is no way for users to test a GPS receiver for this problem. Users are encouraged to contact their receiver manufacturer to determine if their receiver will be affected, in particular if a failure of navigation could put lives or property at risk."
• GPS Manufacturers List (No compliance information, but links to sites)
  •  
• The GPS Week 1024 Rollover is on Midnight August 21, 1999
• Joint Program Office (JPO) Web site on Y2K

GPS Risk Assessment - This report documents a study performed by the Johns Hopkins University Applied Physics Laboratory (JHU/APL) to assess the risks of relying upon the Global Positioning Satellite (GPS) navigation system and proposed augmentation systems for air navigation. The study was performed under contract to the Air Transport Association (ATA) and was jointly sponsored by the ATA, the Federal Aviation Administration (FAA), and the Aircraft Owners and Pilots Association (AOPA). On a quick read, this report does not address Y2K or WNRO.  It is 125 pages and 2+Megabytes.