Hello,
Back to this business of antenna calibration, I have a question. On the Trimble 5800, the manufacturer specifies the distance from base of antenna mount (ARP) to the L1 antenna phase center to be 96.8 mm (64.9mm + 31.9mm). The data from government calibration tables shows 93.0mm. See below for details.
If I understand correctly, processing GPS5800 survey data with TBC, the software automatically treats the ARP-APC height to be 96.8mm (if the ARP is bottom of antenna mount). The software uses the height from the antenna reference point (64.9mm) and then calls the file t5800.pct which adds the extra 31.9.mm.
Processing GPS5800 data using a generic antenna with the values published by the government, the ARP-APC height would be 93mm.
My question is: Does one of the solutions introduce a systematic error? The heights resulting would be different by 4mm, wouldn't they?
When I verified with field observations the ARP-APC height, I obtained a result of 90mm. Of course my method is not as accurate as a true calibration. I just wanted to verify what kind of numbers I would get.
For simplicity, I only use the L1 values in my tests.
So, 97mm or 93mm, which one is it?
*
*
*
*
*
2.10 OBSERVATION DATA G (GPS) RINEX VERSION / TYPE
DAT2RINW 3.10 001 gc 22SEP11 16:41:41 PGM / RUN BY / DATE
gc gc OBSERVER / AGENCY
4534155914 TRIMBLE 5800 Nav 2.31 Sig 2.31 REC # / TYPE / VERS
TRM5800 ANT # / TYPE
----------------------------------------------------------- COMMENT
Offset from BOTTOM OF ANTENNA to PHASE CENTER is 64.9 mm COMMENT
----------------------------------------------------------- COMMENT
59142650 MARKER NAME
5914 MARKER NUMBER
-1220278.9227 -2695431.5811 5631560.9405 APPROX POSITION XYZ
-0.0649 0.0000 0.0000 ANTENNA: DELTA H/E/N
*** Above antenna height is from mark to BOTTOM OF ANTENNA. COMMENT
----------------------------------------------------------- COMMENT
Note: The above offsets are CORRECTED. COMMENT
Raw Offsets: H= 0.0000 E= 0.0000 N= 0.0000 COMMENT
----------------------------------------------------------- COMMENT
1 1 0 WAVELENGTH FACT L1/2
5 L1 C1 L2 P2 D1 # / TYPES OF OBSERV
15.000 INTERVAL
2011 9 22 18 3 30.0000000 TIME OF FIRST OBS
2011 9 22 20 3 0.0000000 TIME OF LAST OBS
0 RCV CLOCK OFFS APPL
13 # OF SATELLITES
4 227 227 227 227 227 PRN / # OF OBS
*
*
*
*
*
;Processor name : MRGPCT 1.0
;Merge time : Wed Aug 07 01:45:00 2002
;Calibrated antenna : 5800 Internal
;Mean phase center (mm) North East Up
L1NominalOffset = -0.4 0.4 31.9
L2NominalOffset = 0.9 0.4 18.7
;Elevation range (deg) Start Stop Step
ElevationRange = 5 90 5
;Azimuth step size (deg)
AzimuthStep = 0
;Azimuth/elevation corrections (mm)
AZ=0
;L1
-9.9 -5.5 -1.8 1.1 3.6 5.4 6.8 7.5 7.6 7.3
6.6 5.6 4.5 3.4 2.3 1.4 0.6 0.0
;L2
-7.7 -2.9 -0.6 0.7 1.7 2.5 3.2 3.8 4.2 4.2
4.0 3.4 2.8 2.0 1.2 0.6 0.2 0.0
*
*
*
*
*
[R8/5800/SPS78x]
Name=R8/5800/SPS78x Internal
DCName=R8/5800/SPS78x
Manufacturer=Trimble
Class=Survey
PartNumber=
CharCode=K4
Type=97
MeasMethod0=0.00000,0.06490,0.00000,"Bottom of antenna mount"
MeasMethod1=0.09100,0.00970,0.00000,"Center of bumper"
RINEXMethod=0
RINEXName=TRM5800
RINEXName=TRMR8
RINEXName=TRMSPS780
RINEXName=TRMSPS781
PhaseCorrTable=t5800.pct
NGSCorrTable=t5800.ngs
Freq=2
LowVoltage=1
Internal=1
GraphicsFile=5800.jpg
IGSAntenna=TRM5800
IGSDome=NONE
AddDate=
*
*
*
*
*
TRM5800 NONE Antenna inside TRIMBLE 5800 and TRIMBLE NGS ( 3) 02/12/05
0.2 1.5 93.0
0.0 1.6 3.4 5.0 6.5 7.9 9.1 10.0 10.6 10.9
10.8 10.3 9.4 8.2 6.5 4.5 2.1 0.0 0.0
0.2 -1.6 84.9
0.0 -0.9 -0.6 0.4 1.8 3.4 4.8 6.1 7.0 7.3
7.2 6.6 5.4 3.8 2.0 0.0 -1.9 0.0 0.0
ngs.noaa.gov/ANTCAL/LoadFile?file=ant_info.003
> If I understand correctly, processing GPS5800 survey data with TBC, the software automatically treats the ARP-APC height to be 96.8mm (if the ARP is bottom of antenna mount). The software uses the height from the antenna reference point (64.9mm) and then calls the file t5800.pct which adds the extra 31.9.mm.
>
> Processing GPS5800 data using a generic antenna with the values published by the government, the ARP-APC height would be 93mm.
>
> My question is: Does one of the solutions introduce a systematic error? The heights resulting would be different by 4mm, wouldn't they?
If TBC uses the Trimble antenna phase center correction tables, wouldn't you expect the absolute calibration errors to cancel? I'd think any problem would occur when one mixed NGS antenna models and Trimble models.
Take for example a baseline measured from a CORS monument to a survey marker (occupied with GPS5800).
The baseline is processed with two different software. One of that uses the published values (93mm), one that uses the manufacturer's values (97mm).
Which height value of the survey marker is closer to its true value?
I went to ALL NGS models for this very reason, if you only use one brand, it doesn't really matter, but as soon as you start mixing and matching, then you need to use models all from the same source. Since each brand usually only supports the color of antenna they sell, it is best to just run with NGS models. I even loaded the NGS models for my equipment into the firmware, I can select the NGS model in the field, when I bring the data back into the office it is ready to play nice with CORS data, just makes life easier. And then there is relative vs absolute models.... There will be a screw up or two before the dust settles on this...
SHG
Howdy,
If you review the text message on the antenna diagrams at the NGS antenna calibration site you will see a note indicating that the antenna offsets (with respect to the antenna reference point) and phase center variation values are determined without regard to the manufacturer's nominal values.
In other words, the NGS values are determined by observations and not by opening the antenna and precisely measuring the relative locations of the antenna strips. If I remember correctly, there were some concerns early on with individual antennas of the same model varying more than they should. I also recollect that this issue was resolved (and verified) in subsequent testing.
I would use the NGS values to be consistent. As indicated by other replies, mixing antenna makes and models is where the problem in processing can become significant. Different antennas have difference phase center variations which must be accounted for in processing.
I was rather concerned years ago that some once-a-year observations diverged at the decimeter level from previous years when processing without a phase model for a new antenna with respect to a CORS. When the new model was released we were happy to see good agreement < 2cm over < 20 km baselines.
I would not expect millimeter-level accuracies in the height component any time soon. Researchers are still unable to model the wet component of the troposphere at that level. The approach used by OPUS-RS where the tropo parameters are solved at network sites and the effect interpolated at the unknowns is probably the best approach today.
Working at the millimeter level is pretty demanding. The antenna phase center offset is only one element.
Having gone on too long. I quit.
Cheers,
DMM
