How Accurate Is The Gps Clock?

[+jotne]

How accurate is the GPS clock?

 

Is guess it's synchronized with the satelite.

[+º]

Yep the gps clock is synchronized with the satellite's atomic clock. Since time is a essential part of the navigation with satellites, the clock of the GPS even has to be synchronized.

[EagleI]

I have a MeriGreen, my son has a MeriGreen and my son-in-law has a MeriYellow. Their clocks are exactly together, mine is 56 seconds behind theirs. All 3 are running 4.03 firmware.

 

This condition seems to be permanent, since it has been observed over a period of weeks. The lat/long readings, however are all together. How can this be?

[+nincehelser]

It depends on what you consider to be accurate.

 

Time from NIST and the US Naval Observatory (they control the GPS system) often vary from each other by several nanonseconds. These variances are constantly changing.

 

At one time, GPS time was as much as 11 seconds from Universal Coordinated Time, but that interval difference is pretty constant. I'm not sure if that gap is still the same today.

 

Several years ago there was an issue with the way the GPS system keeps time. I don't recall the details, but it boils down to something of counting the number of blips from a certain point of time. The blip counter needed to "roll-over" and there was a lot of concern it would screw up older GPS receivers...it was something similar to the Year 2000 bug.

 

George

[+nincehelser]

I have a MeriGreen, my son has a MeriGreen and my son-in-law has a MeriYellow. Their clocks are exactly together, mine is 56 seconds behind theirs. All 3 are running 4.03 firmware.

 

This condition seems to be permanent, since it has been observed over a period of weeks. The lat/long readings, however are all together. How can this be?

You got me. I've seen similar differences between Garmin units.

 

I'm guessing it has something to do with the way they are going about computing the "clock" time. A lot of clocks that are controlled by central source only synch to the central source once in a while...maybe once a day, maybe once an hour. The rest of the time they are "free running", and can loose some accuracy during these times.

 

Who knows how often a consumer GPS receiver re-synchs its time?

 

It might be interesting to "reset" both GPS units and see if they are then in agreement. If there's still a difference of 56 seconds, that would be really odd.

 

George

[EagleI]

It might be interesting to "reset" both GPS units and see if they are then in agreement. If there's still a difference of 56 seconds, that would be really odd.

 

I might try that this weekend, when I will be with my s-i-l again. I did my firmware upgrade before they did theirs, so that might have something to do with it. I just hate to do a system reset, because it takes so long to get all the settings back to the way I want them.

[+DocW]

There is some trouble observed by people (me too) using Magellan Meridian units. This was discussed in the Yahoo-Group and in some other forums too. The Magellans seem to have some trouble in showing the exact time. My Merigold was behind > 2minutes. Similar time delays were observed by others in the group, some from Europe, some from abroad. Mostly this gets to attention around new year ;-) I observed it with the 4.0x firmware last year, didn't care for it for a year and now observed it again with the 5.12 Firmware again.

 

You can cure your Magellan by initializing and clearing all memory to the factory status. It then gets back to the correct time (But how long?)

you can manually change the clock in the Magellan, if it is not locked to the sats. Also if you only re-initialize it, you can change the time to whatever you like. But if it then lock to the sat, whuuup, it switches back to the wrong time.

I didn't observe this with my Garmin Geko.

 

Just to explain the timebase:

Here in Europe, we have an extreme stable time base: Its called DCF77 and its an atomic clock located near Frankfurt/Main supplying official time signal to the continent (1200km radius) by a longwave (77kHz) radio transmission signal. You can buy cheap clocks which receive that signal. This is my timebase to compete against, I guess the error is somewhere in the sub second area.

 

I guess the clock for the user is not linked to the satellite clock base. It only gets its second signal from that clock.

 

Wulf

[+EraSeek]

A GPS unit's time accuracy and the GPS time 'displayed' can be two different things depending on the unit and its firmware. Generally (at least the better units) syncs with the atomic clock on the sats with each almanac update (about every 12 mins, and so the accuracy of the time 'display' should be less than a second. Other units, well, I own Garmin. I have looked at this before and have the link somewhere. I see if I can post something later. By the way, radio controled "atomic accuracy clocks' are updated at differing rates. Some like twice a day, others more often. I believe most common is updates in evening, night, and early morning (avoiding ionospheric delays of signal).

[Kerry.]

GPS time as against GPS timing, two completely different things. GPS "clock" simply tells/shows the time and is not a critical function of a GPS and hence doesn't have priority as far as CPU processing is concerned.

 

Basically some units can be up to and around 2 seconds behind as far as knowing when lunch time is and yes it is basically controlled through the nav mesagge data.

 

Cheers, Kerry.

Edited January 15, 2004 by Kerry.

[+Oneyedjack]

This threw me for a loop the other day. I had been caching for about 3 hours when I looked at the time display on my Meriplat and it was 12.5 hours off. I don't know how the time was off that far, but I reinitialized it, and it is still OK today. I don't think is was effecting the accuracy of my finding caches because I found 5 before I noticed the time difference. Go tell!

Edited January 15, 2004 by oneyedjack

[+EraSeek]

By the way, radio controled "atomic accuracy clocks' are updated at differing rates. Some like twice a day, others more often. I believe most common is updates in evening, night, and early morning (avoiding ionospheric delays of signal).

Here is some info on the radio atomic clocks from NIST.GOV

 

"Once your radio-controlled clock has synchronized, it won’t decode the signal from WWVB again for a while. Some clocks only decode the signal once per day, others do it more often (like every 4 hours or every 6 hours). Those that decode the signal just once per day usually do it at night, since the signal from WWVB is much stronger once the sun goes down. In between synchronizations, the clocks keep time using their quartz crystal oscillators. A typical quartz crystal found in a radio-controlled clock can probably keep time to within 1 second for a few days or longer. Therefore, you shouldn’t notice any error when you look at your clock display, since it will appear to be on the right second, even though it has probably gained or lost a fraction of a second since the last synchronization."