Backsight check using Total Station doesn't match GPS

Your willingness to explore options is half the battle...
In our early days with Carlson we discovered that 'process GPS' didn't really 'process'. It simply applied the projection to the stored geographic position. Errors at the base had to be cleaned up with manual translations.
Happy to say if you check the right box it stores the standard G lines with delta x,y,z and quality data. If you didn't store the data there are options. A closed loop through the base and a few control points should tell you a lot. I would definitely go that route before shipping off the gun and losing productivity. The only thing worse than spending money is doing it while not making any.
Good luck, Tom

Yes, I am familiar with this adjustment. I'd have to look it up but "CCD" comes to mind - the "robot eye", as you put it. I have performed this adjustment in the not so distant past. Doing it again might tweak it up a bit, but I don't think it would amount to more than 0.01' at this point.

Is the setting that you are referring to located in File>Job Settings>Options>"Store GPS Vectors in Raw File"?

Height Errors Shoud Balance Out With Same Equipment

Just looked up that red line in the RINEX Format Document

So, Height was 2.000 meters, and there are no eccentricities to the East and North for the Hiper Pro....at least, that's the way I read that line. And, it's consistent with what I believe to be the case in the field.

|ANTENNA: DELTA H/E/N| - Antenna height: Height of bottom | 3F14.4 |
| | surface of antenna above marker | |
| | - Eccentricities of antenna center | |
| | relative to marker to the east | |
| | and north (all units in meters) | |

The horizontal distance between base and rover was about 214 feet. I just realized that in the field, using SurvCE, I didn't use a geoid. But, when I sent in my base data file to OPUS, NGS applied a geoid to the observations and provided me with an orthometric height, which I used to translate all the field points with. Could this be a source of error?

That's it. It will add the data you need for post processing or adjustment. Doing so is enlightening. Its our best tool for learning how the equipment works (or doesn't) in your location and environment.

UPDATE-Backsight check using Total Station doesn't match GPS

Well, I've done some testing, and I've been able to duplicate the problem.

Here is a link to a folder containing my test data, including CRD and RW5 files

I ran 4 tests on 3 nails in my driveway:

Test 1 - Topcon Hiper Pro receivers at both Base and Rover, no geoid applied
(Base at 1 autonomous, log static data for OPUS, RTK points 2 to the north and 3 to the south)

Test 4 - Topcon Hiper Pro Base and Rover, no geoid
(Base set up on 2 autonomous, RTK 1 and 3)

Here's what I'm finding - as suggested by others, whether or not a geoid is applied to the work, the differential in elevation between the points is the same.

Also, I've been very careful measuring HI and HR and checking my setup with the total station, as in Test 3, and I believe that the differences in elevation between the 3 points as determined with total station are correct. Assuming this hypothesis is correct, the true elevations of the points are as follows:

Point 1 (from OPUS) 786.757
Point 2 (calculated) 786.615
Point 3 (calculated) 783.813

In Tests 1 and 2, occupying Point 1 with the base, I calculate the average height of Point 2 with the rover to be 786.246, or about 0.37' lower than measured by total station. That's about the same difference as I've experienced with previous jobs (backsights measured 0.37 feet higher and 0.41 feet higher than calculated using GPS coordinates).

In Test 4, I swapped Base and Rover, putting Base on Point 2 and Rover on Point 1. In this case, assuming an elevation of 786.615 for Point 2, I measured an elevation at Point 1 of 786.490, again lower than the assumed true value (786.757) but by 0.27 feet instead of 0.37. On measuring to Point 3 in Test 4, I measure about 0.22 low.

So, it seems that for some reason, I'm measuring low, but that the magnitude of the difference might vary. I'm stumped for now....

UPDATE-Backsight check using Total Station doesn't match GPS

Try levelling between the points as well - an independent check on their relative heights?

UPDATE-Backsight check using Total Station doesn't match GPS

Great suggestion, thanks.

So, as another check on this matter, I ran a static session on point 1 using the Base, and a separate static session on the same point using the Rover.

Submitted each to OPUS.

For the base I got 239.804 meters (RMS = 0.021)
For the rover I got 239.830 meters (RMS = 0.028)

The difference between these two sessions (0.026 meters = 0.085 feet) was a little more than I was expecting, but it seems reasonable to me to conclude that the two receivers are operating similarly. That is, the 0.4 foot difference on backsight check does not appear to be a result of hardware problems.

I was looking at the raw data,

--BP,PNBP001,LA42.092425084133,LN-73.260161693810,EL196.5442, AG2.0000,PA0.1060,ATAPC,SRBASE,

BP,PN_0000,LA42.092425084200,LN-73.260161693900, EL196.4382,AG2.0000,PA0.1060,ATARP,SRROVER,--

On the R8's there is an option to measure to the ARP or the rubber bumper (which is handy when the base is setup on a tripod)

It looks like the rover is measured to the ARP but the Base is measured to ATAPC ??

Neil

> It looks like the rover is measured to the ARP but the Base is measured to ATAPC ??

I'd say you've solved the mystery. Per the Topcon HiPer Pro manual, the L1 PC is 105.9mm (= 0.35 ft.) above the ARP.

http://www.manualslib.com/manual/807838/Topcon-Hiper-Pro.html?page=73

Thank you Neil !!!!!!

Now at least I know what I'm looking for. I'll do some research and see if I can figure out where that setting is made.

EDIT - I looked on the data collector and in the SurvCE manual, but couldn't find the setting. I've sent an email to Carlson, and maybe they can help me track down how to make this setting.

Cheers!

Big Al