Welcome
Tuesday, June 6th, 2023
This is a rehash of https://surveyorconnect.com/community/surveying-geomatics/gps-and-total-station-used-together/paged/1/ (March 25, 2011 4:47 pm) by yyamahayzf. My procedure using RTK for a boundary survey would be:
- Set 2 points at each end of the traverse
- Observe each point twice for 3 minutes, a minimum of 4 hours apart
- Traverse using doubled angles, scale factor (SF) = 1 in data collector
- Adjust traverse
- Compute SF
- Enter SF in data collector
- Begin collecting data
Does my procedure make sense?
1.Establish your projection before going to the field, whatever it is State Plane, LDP ect.
2.Load it into the DC.
3.Survey the control points with RTK using the DC.
4.Then survey conventionally using the same projection and the same DC program.
This has been the process since total stations could be hooked up to a data collector.
Be sure the total station isn't double adjusting your distances.
Not sure what you mean by SF in this context. It sounds like you are computing a measured vs measured SF.
What type of projection are you using? SF=1 would imply no projection/no datum. If you’re using a local system, that will work for a small area but fall apart quickly if you expand it out.
Does my procedure make sense?
Not to me. I collect all the data first, then do a comprehensive adjustment. Depending on the complexity of the job I may do one or more interim adjustments and load the results into the data collector for search purposes, but I like to do the final adjustment with all the data, not just some of it.
I collect all the data first, then do a comprehensive adjustment. Depending on the complexity of the job I may do one or more interim adjustments and load the results into the data collector for search purposes, but I like to do the final adjustment with all the data, not just some of it.
This is my workflow as well.
I will tie as many control points as practical with GNSS, not just some at the end. Spreading RTK (or static) observations throughout the network will tighten it up significantly during adjustment. I apply a reasonable combined scale factor and chop off/truncate as well. If the total station is coming out, and it's for design, it's a ground coordinate system.
If you just observe the endpoints of your traverse, depending on its length and the precisions of the GNSS positions, those error ellipses can really blow up in the middle.
Adjusting a traverse by compass rule will give you a result that "closes", but will tell you nothing about how good those final positions really are.
- Set 2 points at each end of the traverse
- Observe each point twice for 3 minutes, a minimum of 4 hours apart
- Traverse using doubled angles, scale factor (SF) = 1 in data collector
- Adjust traverse
- Compute SF
- Enter SF in data collector
- Begin collecting data
Does my procedure make sense?
- Determine appropriate grid zone and set dc to it.
- Set 2 points at each end of the traverse
- Observe each point twice for 3 minutes, a minimum of 4 hours apart
- Traverse using doubled angles
- Adjust traverse
- Begin collecting data
Set your data collector to whatever grid system is appropriate to the area. That takes care of the SF issues for the duration. At no time during this whole procedure should you need to concern yourself with just what the specific CSF is. Except maybe to check what it is, that it is reasonable,
If all the points are GPS-able hit them twice, or more, and then level through the string to tighten up the vertical and dispense with the traversing.
Feeding all your data into a least squares adjustment software does the adjusting work for you. I use StarNet and swear by it. If you have Carlson Survey you have SurvNet, which is a capable alternative. There are others.
For the record, in 2015 I found myself in a meeting with 3 PLS's, including the Survey Manager, who did not understand this simple dc setting and needed a lot of convincing.
I found myself in a meeting with 3 PLS's, including the Survey Manager, who did not understand this simple dc setting and needed a lot of convincing.
I would say that's a weekly occurrence around here, but then again a big part of my job is to deal with problem datasets and try to spin gold out of sh.....straw.
The percentage of individuals with a license and/or a management title who do not understand the fundamentals of data collection and processing, and still want to dictate procedures, is astonishing. Or at least it used to be. Now it's just depressing.
It's hard enough trying to convince field crews to employ best practices, but when the PM hasn't learned anything new since 1992 and is trying to micromanage their every move in the field, it's damn near impossible.
Set your data collector to whatever grid system is appropriate to the area.
We use Florida SPC East. I'm confused about what geoid model is appropriate. Wouldn't you use the latest model?
If all the points are GPS-able hit them twice, or more, and then level through the string to tighten up the vertical and dispense with the traversing.
Level through the string even though the project doesn't require elevations?
I will tie as many control points as practical with GNSS, not just some at the end.
I was just considering starting on 2 GPS-derived points, traversing, and closing on 2 GPS-derived points in order to get a closure. You say all points should be GPS-derived in order to get a more accurate adjustment on the points?
I apply a reasonable combined scale factor and chop off/truncate as well.
Okay, so the CSF is used to correct ground to grid, and you need to level through all the points to get the elevation factor whether or not you are going to use elevations for the project. Does a typical data collector create a CSF? Why would you truncate the CSF?
I think he means truncating your surface values if you intend to combine your terrestrial measurements to the network.
Greatly depends on your data collector, office survey software (and your GNSS knowledge). From the way you seem to approach this, all you are wanting to get from GNSS for your traverse is a basis of bearings.
What type of projection are you using? SF=1 would imply no projection/no datum.
We use Florida SPC East. I was wondering why every time we created a new job in MAGNET Field, a SF other than 1 was already set; it's because of the selected projection in the project parameters. I'm really embarrassed here. Let me ask you something. After the whole traverse is adjusted and a CSF is calculated, would I create a new project, import the adjusted points, and manually enter the CSF as the SF?
Not sure what you mean by SF in this context
At the time, I didn't know what the heck I was talking about! I have a better idea now thanks to everyone!
Be sure the total station isn't double adjusting your distances.
You mean as in temperature and pressure settings? Earth's curvature?
If the project does not require elevations then you don't need to level either.
The geoid model is dependent on the datum (ie/ the flavor of NAD83 in use), and not the grid projection. Presuming that you are using NAD83(2011) then Geoid18 is the datum to use. Of course, if elevations are not required the geoid question is moot.
