Discussion:
nuts about position
(too old to reply)
Mark Sims
2018-05-30 02:19:35 UTC
Permalink
I finally got in the Ublox M8T and have been testing its RAW data measurements (carrier phase / pseudorange / dopper) data by having Lady Heather write a RINEX file and sending it to CSRS-PPP.

Collecting GPS, SBAS, GLONAS, and GALILEO data (CSRS-PPP ignores GALILEO and BEIDOU data) for 2 hours and having it processed with the "ultra rapid" orbits gives a position with around 350 mm lat/lon and 800 mm alt uncertainty ellipse diameter.

Using 24 hours of data and the "rapid" (24 hour delayed) orbits gives around 125 mm lat/lon and 300 mm alt uncertainty... the 24 hour file was around 100 MB long, .ZIPed to 25 MB. Not enabling GALILEO would have reduced the file sizes around 25%.

I need to see if I can find a program that will split out the GPS/SBAS and GLONAS data into separate RINEX files and see how the two systems compare.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Mark Sims
2018-05-30 19:46:48 UTC
Permalink
Ok, I did the experiment. I took an 8 hour M8T RINEX file with GPS,SBAS,GLONASS,GALILEO data in it. I then used to TEQC to extract GPS only and GLONASS only data (CSRS-PPP ignores SBAS and GALILEO data). The 95% confidence error ellipse estimates ("rapid" orbits) were:

GPS+GLONASS: lat 0.245m lon 0.219m alt 0.582m
GPS ONLY: lat 0.269m lon 0.239m alt 0.616m
GLONASS ONLY: lat 0.610m lon 0.755m alt 1.835m

Conclusion... if you are just interested in an accurate antenna position, a $25 LEA-5T should do just fine. GLONASS brings very little benefit (call it an inch) to the party. 24 hours of observations can reduce those numbers by around 50% I used one of those $70 Chinese L1/L2/GLONASS/BEIDOU antennas from Ebay. Note that my antenna location/environment is HORRIBLE...

Things might change if you get into expensive L1+L2(or L5) receivers.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bob kb8tq
2018-05-30 21:02:39 UTC
Permalink
Hi

It’s not real clear what the magic L1 / L2 / L5 gizmo does for you right now today. As far as I can tell,
the free processing services all only do two at a time. You can do L1 / L2 or L1 /L5. Possibly at some
point in the future that will change.

Even with “only two” a fancy device will get you down to numbers that simply are not believable. Any
device that thinks it’s getting into 0.000x meter precision needs to back up and look at some of the
reference points being used.

Getting back to timing, if you guess that 0.3 m is about a nanosecond, the M8T is off by about a nanosecond
in lat and lon. It’s off by about 2 ns in elevation. The time solution will be some sort of RMS of this. Unless
the one and only sat you can see passes directly over you, the full 2 ns isn’t going to come in.

Location will always be one of a list of errors in getting time data. The bigger question will be how well the
other stuff on the list is controlled / known. Coming up with an RF delay number that is good to < 1 ns on
any of these gizmos may be a bit tough…..

If frequency rather than time is the goal, the position error will indeed add to the “swing” over a day. Your
antenna location / sky view will be a big part of the “how much” part of the question. You might see a
nanosecond of swing on an hour to hour basis …. you probably would see less. Simply put, it’s in the
3x10^-12 or less range frequency wise. Most GPSDO’s struggle below a few parts in 10^-11 for a variety
of reasons ….

Bob
Post by Mark Sims
GPS+GLONASS: lat 0.245m lon 0.219m alt 0.582m
GPS ONLY: lat 0.269m lon 0.239m alt 0.616m
GLONASS ONLY: lat 0.610m lon 0.755m alt 1.835m
Conclusion... if you are just interested in an accurate antenna position, a $25 LEA-5T should do just fine. GLONASS brings very little benefit (call it an inch) to the party. 24 hours of observations can reduce those numbers by around 50% I used one of those $70 Chinese L1/L2/GLONASS/BEIDOU antennas from Ebay. Note that my antenna location/environment is HORRIBLE...
Things might change if you get into expensive L1+L2(or L5) receivers.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Mark Sims
2018-06-01 03:15:41 UTC
Permalink
I did another test to see if the M8T offered any positioning advantage over the old (and cheap) LEA-4T and LEA-5T devices. I drove the M8T and LEA-4T with the same antenna, collected data for 12 hours, beamed the RINEX files to Canada.

The results matched to within 3mm... so, again, the M8T doesn't offer much benefit for getting precise positions.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bob kb8tq
2018-06-01 12:10:21 UTC
Permalink
Hi

I can’t say that’s a big surprise. The silicon on these gizmos made it past the “overkill” point at least
a decade ago. There’s only so much you can do with a noisy signal ….. Yes, there are indeed feature
differences, ( like the Furuno pps ). The core of the device seems to have hit a limit a while ago.

Bob
Post by Mark Sims
I did another test to see if the M8T offered any positioning advantage over the old (and cheap) LEA-4T and LEA-5T devices. I drove the M8T and LEA-4T with the same antenna, collected data for 12 hours, beamed the RINEX files to Canada.
The results matched to within 3mm... so, again, the M8T doesn't offer much benefit for getting precise positions.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Magnus Danielson
2018-06-01 17:00:32 UTC
Permalink
Hi,

Well, when the resolution and stability is sufficient, other aspects
weigh in, as being quick to have reduce biases in position and time.
You can resolve much by averaging etc. but that takes time.

Cheers,
Magnus
Post by Bob kb8tq
Hi
I can’t say that’s a big surprise. The silicon on these gizmos made it past the “overkill” point at least
a decade ago. There’s only so much you can do with a noisy signal ….. Yes, there are indeed feature
differences, ( like the Furuno pps ). The core of the device seems to have hit a limit a while ago.
Bob
Post by Mark Sims
I did another test to see if the M8T offered any positioning advantage over the old (and cheap) LEA-4T and LEA-5T devices. I drove the M8T and LEA-4T with the same antenna, collected data for 12 hours, beamed the RINEX files to Canada.
The results matched to within 3mm... so, again, the M8T doesn't offer much benefit for getting precise positions.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bob kb8tq
2018-06-01 17:36:17 UTC
Permalink
Hi

Indeed, if you look at a 30 minute data file from any of these receivers, it will not be giving you
quite the same performance.

Bob
Post by Magnus Danielson
Hi,
Well, when the resolution and stability is sufficient, other aspects
weigh in, as being quick to have reduce biases in position and time.
You can resolve much by averaging etc. but that takes time.
Cheers,
Magnus
Post by Bob kb8tq
Hi
I can’t say that’s a big surprise. The silicon on these gizmos made it past the “overkill” point at least
a decade ago. There’s only so much you can do with a noisy signal ….. Yes, there are indeed feature
differences, ( like the Furuno pps ). The core of the device seems to have hit a limit a while ago.
Bob
Post by Mark Sims
I did another test to see if the M8T offered any positioning advantage over the old (and cheap) LEA-4T and LEA-5T devices. I drove the M8T and LEA-4T with the same antenna, collected data for 12 hours, beamed the RINEX files to Canada.
The results matched to within 3mm... so, again, the M8T doesn't offer much benefit for getting precise positions.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Magnus Danielson
2018-06-01 17:51:32 UTC
Permalink
Hi,

If you move in closer, it will be even more apparent.

You do benefit from the improved receivers and the reduced errors
translates to better timing error, unless the receivers have other issues.

Cheers,
Magnus
Post by Bob kb8tq
Hi
Indeed, if you look at a 30 minute data file from any of these receivers, it will not be giving you
quite the same performance.
Bob
Post by Magnus Danielson
Hi,
Well, when the resolution and stability is sufficient, other aspects
weigh in, as being quick to have reduce biases in position and time.
You can resolve much by averaging etc. but that takes time.
Cheers,
Magnus
Post by Bob kb8tq
Hi
I can’t say that’s a big surprise. The silicon on these gizmos made it past the “overkill” point at least
a decade ago. There’s only so much you can do with a noisy signal ….. Yes, there are indeed feature
differences, ( like the Furuno pps ). The core of the device seems to have hit a limit a while ago.
Bob
Post by Mark Sims
I did another test to see if the M8T offered any positioning advantage over the old (and cheap) LEA-4T and LEA-5T devices. I drove the M8T and LEA-4T with the same antenna, collected data for 12 hours, beamed the RINEX files to Canada.
The results matched to within 3mm... so, again, the M8T doesn't offer much benefit for getting precise positions.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
ew via time-nuts
2018-06-01 13:53:50 UTC
Permalink
I relay like the LEA 5T  board, better than the MT* see attached
 
In a message dated 5/31/2018 11:16:09 PM Eastern Standard Time, ***@hotmail.com writes:

 
I did another test to see if the M8T offered any positioning advantage over the old (and cheap) LEA-4T and LEA-5T devices. I drove the M8T and LEA-4T with the same antenna, collected data for 12 hours, beamed the RINEX files to Canada.

The results matched to within 3mm... so, again, the M8T doesn't offer much benefit for getting precise positions.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Mark Sims
2018-06-02 16:03:59 UTC
Permalink
I repeated the two-receiver test, this time comparing 10 hours of 1 Hz measurements from an LEA-4T with NVS-08C measurments. The NVS has higher resolution RAW data measurements. But, again, the differences in the error ellipses were in the 2-4 mm range.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Mark Sims
2018-06-03 18:54:30 UTC
Permalink
I did a test on a 1 second vs a stripped out 10 second rate from the same 24 hour run. The differences were down in the noise. Some people have actually reported slightly better results with 30 second vs 1 second data... but I doubt that... I suspect they used different data sets and the differences were just due to different data.

-------

If you are doing a longer run into one of the data analysis services - it does not seem to matter much what the
spacing on the readings is
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Mark Sims
2018-06-05 19:52:56 UTC
Permalink
The typical receiver default self-surveys of 30 minutes to 2 hours are not ideal. They will not include a lot of satellites or the effects of multi-path over time. A survey of at least 12 hours is needed to include all the sats. 24-48 hours is even better.

The effects of things like antenna position changing a couple of cm due to expansion and humidity effects are swamped by the receiver accuracy and noise.

---------------------
That said, there are certainly a lot of issues with how the self-survey
position should be calculated *and* maintained, because it really needs
to change over time for optimal performance.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Poul-Henning Kamp
2018-06-05 19:58:21 UTC
Permalink
--------
Post by Mark Sims
The effects of things like antenna position changing a couple of cm due to expansion and humidity effects are swamped by the receiver accuracy and noise.
Short term: Yes. Long term: No.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
***@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Magnus Danielson
2018-06-03 20:44:42 UTC
Permalink
Hi,
Well, with a little prodding and help from Magnus, I now have the
Trimble devices outputting RINEX files.
Is this applicable to a Thunderbolt, and would this improved position
accuracy be expected to improve the time accuracy from a Thunderbolt
compared to using the older Lady Heather 24 hour self survey method? Or
is ionospheric noise the limiting factor so determining more accurate
position doesn't really help?
Actual ionospheric and corrections isn't matching up too well, so this
can be a way to achieve a better fixed location.

Exactly how the Thunderbolt uses the data in its steering I don't know,
it would be interesting to figure that out.

Cheers,
Magnus
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bob kb8tq
2018-06-05 18:10:53 UTC
Permalink
Hi

Hmmm …… that’s a very big difference between the RINEX and the “precision survey”.

How do each compare to other receivers on the same antenna ( and super duper stable mount) ?
Even with a tripod on the grass, meter level variations likely are not the tripod’s fault. :)

Bob
I did Lady Heather's 48-hour precision survey on an NVS-08 receiver and collected RINEX data at the same time. The NVS was tracking GPS and SBAS satellites.
The RINEX result had lat/lon/alt error estimates of .175/.153/.396 meters.
lat/lon/alt: -.638m/1.190m/1.526m
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Dan Kemppainen
2018-06-06 12:13:20 UTC
Permalink
I would be quite interested in the results of this test from something
like a Ublox 6T or 8 unit.

Please keep us posted.

Dan


On 6/5/2018 2:35 PM, time-nuts-***@febo.com wrote:
The next thing I want to try is a receiver self-survey vs the post
processed results... I may be able to kludge up a test that does a
self-survey / precision survey / post processed survey and compare the
results of the methods on the same data set.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bob kb8tq
2018-06-03 13:37:49 UTC
Permalink
Hi

If you are doing a longer run into one of the data analysis services - it does not seem to matter much what the
spacing on the readings is. One second data does not seem to produce any better result than 30 second data.
I don’t think that the 3 second rate on the Trimble will have much impact on a 24 hour data set.

Bob
Well, with a little prodding and help from Magnus, I now have the Trimble devices outputting RINEX files. They have pseudorange, doppler, and signal strength observations. A 5 hour 1Hz run was sent to CSRS-PPP and the lat/lon/alt error ellipses were in the 250/250/700 mm range... that should improve with a longer run.
Firmware issues in the original Resolution-T limit those to a 3 second observation rate.
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bob kb8tq
2018-06-03 21:30:14 UTC
Permalink
Hi

Location will always impact things a bit. At some point it does become a minor
contributor. What point that is varies with a lot of things. One of them is indeed
the propagation path to the satellites.

How much the ionosphere and troposphere mess things up is very much a “that depends”
sort of thing. There are corrections applied to the data as part of normal GPS L! operation.
The degree to which these corrections work depends on how close things are to the “normal
model”. That in turn depends to some degree on how active the sunspot cycle is at the time.
Right now we are in a period of relatively low activity. That equates to the models mostly fitting
better most of the time.

If solar activity was somewhat higher, then things get more dynamic. The magic models and
the broadcast data can’t keep up as well. That translates to more noise on the estimates and
worse timing ( as well as an impact on location).

Bob
Well, with a little prodding and help from Magnus, I now have the
Trimble devices outputting RINEX files.
Is this applicable to a Thunderbolt, and would this improved position
accuracy be expected to improve the time accuracy from a Thunderbolt
compared to using the older Lady Heather 24 hour self survey method? Or
is ionospheric noise the limiting factor so determining more accurate
position doesn't really help?
--
Chris Caudle
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Bob kb8tq
2018-06-04 00:45:52 UTC
Permalink
Hi
As far as I'm concerned anything that you can do to improve the position accuracy, environmental changes, noise environment, etc is a good thing. Minimizing errors and disturbances can't hurt and may even improve things. How much any improvement provides ... ??? But time nuts tend to be a bit nutty about minimizing our therbligs ;-)
Most receiver self-surveys seem to get your lat/lon to the 2-3 meter range. Heather's median survey is in the 1-2 meter range. PPP data is in the < 0.25 meter range... seems like something worthwhile. (altitude errors are usually around twice the lat/lon error).
There is always the possibility that some receiver model's computation of lat/lon/alt could have some intrinsic bias in it. If so, a position calculated by an external source could possibly degrade performance...
If you go back to the NIST papers where they were testing timing modules, they indeed did find “gotcha’s” with putting in survey based coordinates.
I don’t think they ever did PPP on the modules they published data on.

Bob
----------------
Is this applicable to a Thunderbolt, and would this improved position
accuracy be expected to improve the time accuracy from a Thunderbolt
compared to using the older Lady Heather 24 hour self survey method?
_______________________________________________
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Mark Sims
2018-06-05 18:35:42 UTC
Permalink
The tripod is a survey grade tripod on a limestone terrace (in a horrible location for an antenna).

Heather's precision survey uses the receiver's reported position data. It does not take advantage of carrier phase/pseduorange/doppler data and post-processing. Unlike the simple averaging self-surveys that most receivers do it processes the average of medians over 1 minute/1 hour/24 hour intervals. It typically produces results 1-2m better than the simple self-surveys. The next thing I want to try is a receiver self-survey vs the post processed results... I may be able to kludge up a test that does a self-survey / precision survey / post processed survey and compare the results of the methods on the same data set.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Poul-Henning Kamp
2018-06-05 19:33:00 UTC
Permalink
--------
Post by Mark Sims
The next thing I want to try is a receiver self-survey vs the post processed results...
I fell into that trap many years ago as well :-)

The post processed result eliminates and compensates for a lot of
effects which are invisible to the receiver, that is the entire
point of post processing.

But if your goal is to get good 1PPS out of your receiver, forcing
it to hold a "perfect" position, means asking it to deal with effects
it cannot measure, and as a result you get *worse* 1PPS performance
than if you use the "wrong" self-survey position.

That said, there are certainly a lot of issues with how the self-survey
position should be calculated *and* maintained, because it really needs
to change over time for optimal performance.

Almost any suboptimal antenna-position means that you should vary
the position-hold coordinates over the orbital period (but see
below), but over longer periods also environmental factors come
into play.

For instance if you put the antenna on a wood-construction, be it a
pole or a house, it coordinates will vary at the cm level over the
seasons as the wood expands and contracts.

The good news is that the receiver gives you the input data to work
and model these biases, the bad news you don't get a lot of data.

I did some experiments when ten Oncore M12+T's passed through my lab
many years ago: I plotted the reported "residuals" vs. satellite
position, and I tried to update the held position on a daily basis
using the trend, and the results were measurable at the several ns
level:

http://phk.freebsd.dk/raga/sneak/

I didn't persue this further, because my latitude is particular
unfavourable to the simple (N/S+E/W) approach (no birds north of
me) and with only a single receiver the amount of input data is too
low for such a crude algorithm.

A more complex geometrical model could probably get more mileage
out of the sparse data, as would tracking more birds than the
12 the M12 could track.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
***@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
_______________________________________________
time-nuts mailing list -- time-***@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Loading...