Activity Feed › Discussion Forums › GNSS & Geodesy › Odd thing happened, points staked about 4.5′ off
Odd thing happened, points staked about 4.5′ off
Posted by mark-o on October 23, 2018 at 3:58 amFirst time occurence and it scared the …. out of me. I staked about half dozen points using RTK and only learned today (a month later) they were all off by about 4.5′ to the north-northwest. Fortunately this wasn’t a costly mistake. I went back out with both RTK receivers I have and same points staked in correct locations. I’m guessing maybe a default setting may have changed? Maybe “Convert WGS84 to NAD83” was checked, or somehow it loaded an older geoid model?
I initially thought maybe it could be a US/International Survey foot issue, but that doesn’t seem to be the case.
If anyone has ever had this experience and knows what may have happened, I’m all ears.
lee-d replied 5 years, 6 months ago 14 Members · 24 Replies- 24 Replies
It’s not the geoid. I’m betting on wrong datum designation. And yes we have seen it before. In our software it’s the difference between (NAD83 No_Trans) and NAD83. It may be called something else in your software.
Always check into a known point before, after and during RTK sessions.
Sounds like ITRF versus NAD83(2011). Trimble had a poorly named transformation that confused people, but sometime ago they changed the name to be more clear. You don’t say what DC you are using.
Do you use C3D to calc your points? Check to make sure your base file and xrefs are all set to US Survey feet – and don’t forget to look for units. In 2018 you have feet and US Survey feet…If units are not set to US Survey feet your referenced blocks will be off about 4.5 feet.
4.5′ sounds more like feet / international feet to me; ITRF – NAD83 isn’t usually that much in the horizontal (assuming the error was Hz). We see this from time to time with Civil 3D, you have to be very careful that everything is in US feet.
It could be any of the things already suggested but I’m thinking that you had an autonomous position at the base.
My procedure after setting up is to check into a known point before proceeding. And to check into another known point at the end. That has saved my bacon many times.
I multiplied my state plane coordinates by the international/US survey point ratio and got a much larger discrepancy (roughly about twice ~9′), although the ratio of northing/easting was close, so that could explain it,
I’m going to go back out to the site and try the NAD8 Nno-trans vs NAD 83 toggle and see if I get the same discrepancy and post my results.
Lesson learned, like others have said, best to make a habit of checking into a known point or two before setting out.
Thanks for all the help!
ITRF – NAD83 isn’t usually that much in the horizontal
Around here (NorCal) it’s about 4.8 feet.
Wow that’s quite a bit more than here; around here it’s around 3′ – 3.5′.
It won’t be a unit error like US feet vs. International feet. If you think about that, it would cause the point to be off either Northeast or Southwest depending on the foot getting smaller or larger.
The difference between FT & USFT is about 2ppm. The difference depends on the coordinate system your using. In the Portland area the difference is around 14 Ft using Oregon State Plane, North Zone but less than a foot when using the Oregon Coordinate Reference System, Portland Zone. It all has to do with your location in reference to the grid origin of your projection.
Specifically on a point at our office the NAD83 – ITRF difference is 2.675′ Hz (1.942N, 1.84E) and 4.582′ V
US feet – Int feet and feet – meters are always easy to identify here in LA South because our eastings are much larger than our northings.
I ran into this in WA state a few years ago. Using a Topcon GR5 with Magnet on a network. You have to be using NAD83 No Trans with it, or it’s double correcting the baselines. That would have been handy info to have when picking it up from Topcon! Now I always check into a nail in my driveway that I’ve run static on to compare to when fussing with geoids and settings.
As John stated above, when Trimble first put the transformation in they called it NAD83(2011) or something like that. Our VRS was already broadcasting NAD83 so if you chose that one you were double correcting. They subsequently changed the name to ITRF – NAD83.
I would venture to say that very few people, working in the USA, on this list should be using a transformation from WGS84 to NAD83 in their collector. For decades the manufacturers have been telling us to use enter our control as WGS84, the trouble is that with the exception of the military, our control is provided in some flavor of NAD83. The NSRS is currently NAD83[2011 epoch2010] and I do not know of any real time networks that are broadcasting in WGS84 / ITRF. If your WGS84 and local geodetic values are different then you should definitely check your settings.
Even with all of the settings being correct I have had 2 instances where the initial “fix” was resolved incorrectly resulting in a similar error while staking. It is rare but does happen and can’t be detected without a check to a known point. Forcing the RTK engine to reprocess its “fix” was all that was needed to correct the situation.
- Posted by: Lurker
Even with all of the settings being correct I have had 2 instances where the initial “fix” was resolved incorrectly resulting in a similar error while staking. It is rare but does happen and can’t be detected without a check to a known point. Forcing the RTK engine to reprocess its “fix” was all that was needed to correct the situation.
My experience as well, particularly with the R8-3, but also a couple of times with the R10. Seems to be some combination of MP , perfect alignment of satellites and an overly optimistic solution algorithm. SOP for us to do a check at beginning and end as well as re-initializing when in doubt.
Willy Did you start the job with autonomous, opus it, and shift the coords by your 4-5 ft amount, (in the office) then load design coords back into data collector, and USING THE AUTONOMOUS base coord, proceed to stake opus based design points?
If you did, then you must remember, “what you do to your job, you must do to your WHOLE job”.
Or, maybe it got corrections from another surveyor, on that radio frequency.
Were you using your own base?
Did you set base in SAME location both times?
What DC are you using?
What brand gps?
N
I have on a few occasions had a similar issue. I loaded design coordinates into the controller, and then remembered to apply the combined factor. The controller handled it just fine, but the pc software made some sort of wild transformation when I downloaded the as staked coordinates, showing something like the amount you are seeing. Loaded a new file, with combined factor applied in the pc before loading, went out & checked it, all stakes were already in the correct position, and the as staked coordinates came back into the pc just fine. Don’t know what causes it, just know if working with ground coordinates in a pseudo state plane, I need to apply the combined factor before loading the file.
- Posted by: John Putnam
I would venture to say that very few people, working in the USA, on this list should be using a transformation from WGS84 to NAD83 in their collector. For decades the manufacturers have been telling us to use enter our control as WGS84, the trouble is that with the exception of the military, our control is provided in some flavor of NAD83. The NSRS is currently NAD83[2011 epoch2010] and I do not know of any real time networks that are broadcasting in WGS84 / ITRF. If your WGS84 and local geodetic values are different then you should definitely check your settings.
Unless you are using Trimble RTX, then you should be using that transformation
The Louisiana C4GNet has mount points in both NAD83 and ITRF. Slightly dangerous, I try to make sure my crews know not to use the ITRF.
We use the transformation for points collected with WAAS, which we use fairly frequently in the marsh and for boat work.
Log in to reply.