Activity Feed › Discussion Forums › Strictly Surveying › Trying to Understand GPS Antenna Offests
Trying to Understand GPS Antenna Offests
Posted by The Pseudo Ranger on January 31, 2015 at 1:19 amFor the first time, I’m trying to mix two receivers (Ashtech Locus and Topcon Hiper Pro), so I wanted to do some testing and make sure I know what I’m doing.
I did some testing today, by placing the two next to each other on a flat surface, let them collect data for 2 hours, and processed to see if I could get the elevation to agree.
When I process in Topcon Tools, everything is fine, because Tools has both antenna types programmed in it. The elevations come up less than 1 hundredth apart.
I tried it again in Ashtech Solutions, and it didn’t work so great. Solutions doesn’t have Topcon offsets built in, so I had enter them myself. This is where it gets weird. It appears that both published “up” offsets are significantly wrong (or I don’t understand what they mean).
Ashtech published the “UP” offset as 125mm, and topcon has their L1 “UP” offset as 105.9mm. (0.347 feet). Now that topcon offset immediately looks funny to me, because the Hiper itself is less than that height.
When I used “none” for antenna, and used those heights, I’m off about 0.10′.
However, I look up the NGS numbers for the two GPS, and it has Locus as 139.85mm, and Topcon Hiper as 87.85mm.
So, basically, it’s 125mm published vs. 139.85mm NGS, for the Locus. (about -0.05′)
And, 105.9mm vs. 87.85mm published for the Hiper. (about +0.06′)So there is my tenth.
When I used those numbers, its again within 0.01′ …
So, why the big differences? Am I misunderstanding what the “published” numbers refer to?
paul-in-pa replied 9 years, 1 month ago 5 Members · 12 Replies- 12 Replies
Unless you open the antenna files and compare all the offsets you will never know.
Be aware that some older software still uses an antenna file with the old relative offset differences not the newer NS/IGS absolute offsets.
Also Ashtech has the offsets in the file opposite the NGS published 5° increments.
Some TOPCON antennas have NGS offsets that produced different OPUS elevations from that same data used in Solutions or the Topcon software.
Paul in PA
Just to be clear, I’m using the older “Ashtech Solutions”, not the newer “GNSS Solutions”. “Ashtech Solutions” doesn’t have a lot of antennas included, so I have to enter them manually.
If I understand you correctly, you are asking why the values printed on your antenna differ from those in the NGS antenna information files.
The reason is that there is a physical distance between the antenna reference point (in most cases the bottom of the antenna) and the point at which the signal is captured by the antenna. You will note that the NGS and IGS antenna calibration files include not only a physical offset in terms of N,E and U but also azimuth and elevation dependent values. I am sure that the folks performing the calibrations do not open the antenna to make th physical measurement between the antenna base and each strip antenna. These values are solved from signal tracking data.
Recollect that when you send a file to OPUS it asks for the height to the Antenna Reference Point and the exact antenna model. If you carefully examine the OPUS extended output you will see the offset values used in processing. In addition to the UP offset, there are horizontal offsets though usually not enough to worry surveyors.
Bad news was that Ashtech’s original locus antenna file was based on physical antenna measurements and not on a NGS calibration. Unfortunately NGS did’t do a calibration on it until near the EOL for the units.
Good news was that under the Sokkia brand NGS did do a calibration and it can be found on the NGS site under Sokkia Locus.
Bad news is that most people then using locus with opus could tell that it was not right for some reason and based on messages on the then still active ashtech forum users came to other numbers based on actual experiences. Ashtech forum messages may still be available. I think I have some emails on the topic archived.
So caution is advised. Also as I recall it was revealed the AS didn’t use any of the non vertical data in the file. i.e. only straight up offsets.
SOK_LOCUS NONE Sokkia Locus (L1, display to South ) NGS ( 1) 04/08/24
2.5 0.9 139.9
0.0 -0.1 -0.5 -1.0 -1.7 -2.4 -3.0 -3.5 -3.8 -3.7
-3.6 -3.3 -3.1 -3.0 -3.0 -3.2 -3.7 0.0 0.0
0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
– jlw
Jerry nailed it. The Locus values were physical, not the more precise electronic phase center offset. That 125mm offset (from way back memory) would produce 0.06′ error when mixed with other receivers. As a system, ie Locus to Locus, the errors canceled and great elevations could be determined. Mixed with other receivers, though, users had to determine a custom offset.
Also, as Jerry mentioned, I think most consumer post processing software uses the vertical offsets and not the individual elevation offsets. I could be wrong about that.
Thanks to everyone so far for the help. Some addition notes.
NGS also has an Ashtech Locus calibration, which looks identical:
ASH_LOCUS NONE Integrated L1 only Display to South NGS ( 1) 04/08/24
2.5 0.9 139.9
0.0 -0.1 -0.5 -1.0 -1.7 -2.4 -3.0 -3.5 -3.8 -3.7
-3.6 -3.3 -3.1 -3.0 -3.0 -3.2 -3.7 0.0 0.0
0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 END OF FREQUENCY
END OF ANTENNAI looked at GNSS Solutions, and they have the “UP” as 159mm.
So, we have 0.125 published in the original manual, 0.1399 by the NGS, and 0.159 by GNSS Solutions…. Weird that there would be so much variation. Did Ashtech really blow it by 0.11′ on their first try? I’m going to download some CORS data, and run the pair in Ashtech Solutions, GNSS Solutions, and Topcon Tools, and see what I come up with. I’ll post the results.
Yeah, that’s always how I have used them in the past … Locus to Locus, and whatever height error was present, it cancelled out. Now, I have a project that I could use few extra GPS on, and still have 3 working locus laying around, so I’m going to give it a shot.
So, here are the results for the two “side by side” points when I process against a CORS about 3 miles away:
Tools – Using the built in antenna models … also should note, the vector from the CORS to the Hiper was L1/L2, were as all the other solutions were L1 only.
Locus
N 42.443
E 09.852
Z 19.020Hiper
N 41.443
E 10.412
Z 19.042GNSS Solutions – Using built in antenna models (Note:, had to use Topcon Hiper Plus G/G because the Hiper Pro Model is blank). L1 Only for both points.
Locus
N 42.421
E 09.826
Z 18.996Hiper
N 41.427
E 10.387
Z 19.016Ashtech Solutions, using the NGS heights manually added to the 2 meter poles, and the antenna type set to “none”. L1 Only for both points.
Locus
N 42.444
E 09.870
Z 18.996Hiper
N 41.449
E 10.419
Z 19.021So, between all 3, I have a range of about 0.025′ vertically, between the topcon and Ashtech results. Could be something having to do with Tools using L1/L2 for a vector, where as I don’t have that activated for the Ashtech processing. The two Ashtech products agreeing exactly for the LOCUS is interesting. The key thing I was looking for was to make sure I could integrate the two units without a systematic error, and I think that’s about as close as I’m going to get to weeding out the odd vertical offset component.
Solutions Locus Antenna Offsets
From Ashtech Solutions antenna.inf file.
[800372]
Description=Locus
Radius=100.0
SHMPOffset=125.0
L1Offset=0.0 0.0 125.0
L1Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
L2Offset=0.0 0.0 0.0
L2Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0From GNSS Solutions antenna.ini file.
[800372]
Description=Locus
Radius=100.0
SHMPOffset=125.0
L1Offset=0.0 0.0 125.0
L1Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
L2Offset=0.0 0.0 0.0
L2Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0Paul in PA
Solutions Locus Antenna Offsets
If I go to the GNSS Solution: Tool/GNSS Antennas/Double Click Ash_Locus
I get this:
Am I misreading that? Why a difference between the antenna.inf and what it shows internally?
Solutions Locus Antenna Offsets
There are actually two entries in the .inf file for the Locus:
[ASH_LOCUS NONE]
Description=Integrated L1 only Display to South
Radius=0.0
SHMPOffset=0.0
L1Offset=1.9 1.3 158.0
L1Phase=0.0 0.0 -8.2 -3.9 -0.7 1.6 3.1 4.1 4.5 4.5 4.1 3.6 3.0 2.2 1.5 0.9 0.4 0.1 0.0
L2Offset=0.0 0.0 0.0
L2Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0[800372]
Description=Locus
Radius=100.0
SHMPOffset=125.0
L1Offset=0.0 0.0 125.0
L1Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
L2Offset=0.0 0.0 0.0
L2Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0I beleive the [800372] is the left over one from Ashtech Solutions.
GNSS Solutions Locus Absolute Antenna Offsets
[ASH_LOCUS]
Description=Integrated L1 only Display to South
Radius=0.0
SHMPOffset=0.0
L1Offset=2.5 0.9 139.9
L1Phase=0.0 0.0 -3.7 -3.2 -3.0 -3.0 -3.1 -3.3 -3.6 -3.7 -3.8 -3.5 -3.0 -2.4 -1.7 -1.0 -0.5 -0.1 0.0
L2Offset=0.0 0.0 0.0
L2Phase=0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0The absolute antenna offsets differ from the older relative offsets. You can use either but you must not mix and match. International antennas probably will not get an old style NGS relative test, so if you use any of those antennas then you have to use the absolute for all your future work.
Notice radius is zero, so if you use slant height measurements in the field you may have to do the math yourself. When I query the antenna in GNSS Solutions I get an h value which is needed because the slant height is to a point on the measurement hat which needs to be accounted for.
Paul in PA
Log in to reply.