GPS and Time?
Leith Cassone
old Trimble I used to enter in the offset i.e. I was GMT -8 etc. then
the GPS would calculate my time. On the GM100 I enter both the hour
and the minute.
So:
The times are different on each unit, by about 3 seconds - Which has
the correct time?
Thanks,
Leith
Dave Martindale
>I have two GPS's an Older Trimble unit and the Lowrance GM100. On the
>old Trimble I used to enter in the offset i.e. I was GMT -8 etc. then
>the GPS would calculate my time. On the GM100 I enter both the hour
>and the minute.
>The times are different on each unit, by about 3 seconds - Which has
>the correct time?
Maybe the Trimble, and more likely neither of them. Generally,
updating the display is a low-priority task on handheld GPSes, and the
displayed time is somewhat late. The NMEA output also contains the time,
but it probably isn't synchronized to the top of each second either.
If you want high accuracy timing, you should be using a board-level GPS.
These generally have a 1 PPS output that goes high in synchronism with
the start of each second (plus or minus a microsecond or two), and
NMEA data output that is also synchronized to each second.
The Trimble might be an exception to this - the Trimble Scoutmaster
does have a 1 PPS output option, so even other Trimbles might pay more
attention to timekeeping. Or they might not.
The only way to be sure is to compare to a known good source of
time.
Also, make sure that both receivers have been operating for 15 minutes
or so when you compare them. The offset between UTC and GPS time is
needed to display the correct UTC time, and it is transmitted only
once every 12.5 minutes.
Dave
Bob Gross
earlier time was wrong and the later time was right... right or
wrong being defined by Universal Coordinated Time, not by GPS
time. One receiver (possibly the older one of the two) may have a
fixed number of seconds offsetting Universal and GPS times. That
offset has now been superseded by the newer one, which may
update.
---Bob Gross---
bird...@my-dejanews.com
LCa...@usa.net (Leith Cassone) wrote:
> I have two GPS's an Older Trimble unit and the Lowrance GM100. On the
> old Trimble I used to enter in the offset i.e. I was GMT -8 etc. then
> the GPS would calculate my time. On the GM100 I enter both the hour
> and the minute.
>
>
> The times are different on each unit, by about 3 seconds - Which has
> the correct time?
>
One possibility is one of the receivers is correctly updating the leap second
offset between GPS time and UTC and the other receiver has the offset
hard-wired. GPS time leads UTC due to the leap seconds. A leap second is
being added (i.e. UTC is being retarded) Midnight 31 Dec 98 (1 Jan 99). If
you have a receiver which has all the necessary bells and whistles it will
pick up the leap second as it happens. Others may take a few minutes.
Wait til 1 Jan 99 and make sure the receivers gets a full data download,
Check the time difference again. If they're off buy 4 seconds, the one in the
lead is probably wrong.
JTP
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Scott Pogorelc
>In article <367d6a74....@eastcorp.supernews.com>,
> LCa...@usa.net (Leith Cassone) wrote:
>> I have two GPS's an Older Trimble unit and the Lowrance GM100. On the
>> old Trimble I used to enter in the offset i.e. I was GMT -8 etc. then
>> the GPS would calculate my time. On the GM100 I enter both the hour
>> and the minute.
>>
>> So:
>>
>> The times are different on each unit, by about 3 seconds - Which has
>> the correct time?
>>
>
>One possibility is one of the receivers is correctly updating the leap second
>offset between GPS time and UTC and the other receiver has the offset
>hard-wired. GPS time leads UTC due to the leap seconds. A leap second is
>being added (i.e. UTC is being retarded) Midnight 31 Dec 98 (1 Jan 99). If
>you have a receiver which has all the necessary bells and whistles it will
>pick up the leap second as it happens. Others may take a few minutes.
>
>Wait til 1 Jan 99 and make sure the receivers gets a full data download,
>Check the time difference again. If they're off buy 4 seconds, the one in the
>lead is probably wrong.
Um, I dont think so. The current difference between UTC and GPS is 12
seconds (soon to become 13 on 1-Jan-1999). The non-integral difference
between UTC and GPS is probably < 10 nanoseconds. Neither of these
account for a 3 second difference. It is most likely what one poster
mentioned eariler: updating the display on the hand units is probably not
a high priority item.
Regards,
Scott
--
P.S. Please remove ".ns" suffix when replying via email
Dave Martindale
>Um, I dont think so. The current difference between UTC and GPS is 12
>seconds (soon to become 13 on 1-Jan-1999). The non-integral difference
>between UTC and GPS is probably < 10 nanoseconds. Neither of these
>account for a 3 second difference.
If the one receiver has a UTC/GPS time offset of 9 seconds hard-coded
into its firmware, while the other receiver uses the broadcast 12 second
offset (which is correct), this would explain the 3 second difference.
Apparently some early GPS receivers did use a fixed offset.
Now, a new leap second is coming at the end of December. So if the
difference between the two receivers increases to 4 seconds, this is
pretty good evidence that one is tracking changes in the number of
leap seconds and the other isn't.
And that's what the previous writer meant.
Dave
Scott Pogorelc
>I didn't say that one receiver "ignored" the leap second, I said one probably
>hard-wired the difference and the other one is updating the difference. At
>the time of manufacture of the hard-wired unit, the integer difference
>between GPS time and UTC may have been 9 seconds. Today, it's 12 seconds.
>Hence the "difference" between the hard-wired unit and the unit which updates
>correctly would be 3 seconds. Note, this is only a suggested possible cause.
> Let's wait until the New Year and we should know.
[snip]
I apologize, I misunderstood what was written in the post. I get a duh here.
bird...@my-dejanews.com
sco...@tidalwave.net (Scott Pogorelc) wrote:
>
>
> Um, I dont think so. The current difference between UTC and GPS is 12
> seconds (soon to become 13 on 1-Jan-1999). The non-integral difference
> between UTC and GPS is probably < 10 nanoseconds. Neither of these
> account for a 3 second difference.
I didn't say that one receiver "ignored" the leap second, I said one probably
hard-wired the difference and the other one is updating the difference. At
the time of manufacture of the hard-wired unit, the integer difference
between GPS time and UTC may have been 9 seconds. Today, it's 12 seconds.
Hence the "difference" between the hard-wired unit and the unit which updates
correctly would be 3 seconds. Note, this is only a suggested possible cause.
Let's wait until the New Year and we should know.
By the way, there are three basic ways a receiver manufacture can accomodate
the difference: Hard-wire, Basic Data Message Offset Update, and Real-Time
Update. Hard wired just puts whatever offset currently exists at the time of
manufacture (or the clever one's even subtract a few seconds). This should
keep the time within a few seconds for several years. Some might even say
for the life of the receiver. If time isn't the main feature, this is the
easiest and cheapest approach. The Basic Message Offset Update is the
intermediate approach (which I recommend for basic receivers). These
receivers just apply GPS to UTC offset as noted in the data message. These
receivers need to actually read the data message offset before applying the
correction, so the receiver will be off one second after a leap second until
the data message is read. The Real-Time Update is the most accurate
approach. The data message not only provides the GPS to UTC offset, but it
also tells when the next leap second will occur. Thus, if programmed to
respond to the next leap second warning, a receiver can apply the offset when
it actually occurs, even if the receiver isn't even tracking the signal!
If anyone is interested, they might be able to check out which method is
employed in their receiver by watching the response of their receiver as we
cross over to the new year. A receiver who's time hiccups at midnight uses
the Real-Time Update. A receiver which take a few minutes (or possibly
several minutes or even hours under some circumstances) to correct the second
offset uses the second method (conceivably, some of these receivers might
even need to be turned off and back on again to catch the change). And those
receivers which don't correct at all are hard-wired.
As for me, I don't have a GPS receiver (unless there's one under the tree),
so I won't be checking this year. However, I'm having great concerns as to
when to pull my party popper. Do I pull it when the count reaches zero, or
do I wait one more second?
Ooops... I just remember the leap second is added with respect to GMT, so
everyone needs to check their receiver according to the GMT crossover
(sometime in the afternoon of 31 December for the US folks.) Now, I feel
much better about that party popper. Of course, you folks in the UK are
going to have problems.
Cheers,
Sam Wormley
>I have two GPS's an Older Trimble unit and the Lowrance GM100. On the
>old Trimble I used to enter in the offset i.e. I was GMT -8 etc. then
>the GPS would calculate my time. On the GM100 I enter both the hour
>and the minute.
>
>So:
>
>The times are different on each unit, by about 3 seconds - Which has
>the correct time?
>
>Leith
To find out which is closer to the correct time call 1-303-499-7111 which
is the National Institute of Standards and Technology (NIST) time-of-day
message. While listening and observing, you should be able to determine
what the "average" delay for each display is.
___________________________________________________
Sam Wormley - http://www.cnde.iastate.edu/time.html