Land Surveyor Forum

Replies

• 

  Arie Lise, OLS, OLIP, Dipl.T. May 19, 2015 at 12:45pm

  GPS/GLONASS (GNSS) satellites are rarely positioned directly overhead, but always on your horizon.  Your will get good horizontal readings because the geometry of the angles are far apart.  You will always get poor vertical readings because the satellites are not at a large angle of incidence to your location.

  Taking a vertical reading with a GNSS is like trying to calculate a location 50 meters away by doing a distance-distance calculation from 2 points that are only 2 meters apart.  Your error ellipse is huge because the angles intersect only a few degrees apart and you have no redundancy.  It is a miracle that you can actually get something within a few centimeters period.

  GNSS should never be used to get anything but a topographic elevation.
  • Land Surveyor

    Mike Davis > Arie Lise, OLS, OLIP, Dipl.T. May 20, 2015 at 8:05am

    Above 50 degrees of Latitude you probably encounter the situations you listed above.

    Here in Texas I've taken Leica SmartNet measurements on points with at least 2 hours of Static occupation with results with in 0.10' (3 cm) x,y & z. Setting up on these GNSS static points and measuring between them with my Leica Robot normally yields +/-0.02' x & y and +/- 0.05 in z (elev). 

    NGS has been using GNSS for the last decade for Height Modernization projects down here in the lower 48. 

    http://www.ngs.noaa.gov/heightmod/

    Natural Resources Canada (NRCan) has released the Canadian Geodetic Vertical Datum of 2013 (CGVD2013), which is now the new reference standard for heights across Canada. This new height reference system is replacing the Canadian Geodetic Vertical Datum of 1928 (CGVD28), which was adopted officially by an Order in Council in 1935.

    https://www.nrcan.gc.ca/earth-sciences/geomatics/geodetic-reference...
    • 

      Arie Lise, OLS, OLIP, Dipl.T. > Mike Davis May 20, 2015 at 1:15pm

      The 3cm you quote above is about what you can expect - not much better, which is what seems to be the issue in this discussion.
• Land Surveyor

  Bill Schindler May 17, 2015 at 7:16pm

  A company I worked with about two years ago rented a Trimple GPS unit to aid us in a "rush" project in Rahway, NJ. Two Leica units were also being used on the same project. To make a long story short the Trimble units RTK elevations varied consistently by 0.15' to 0.20' when compared to the Leica numbers and NGS published monumentation. The Leica used the New Jersey-Delaware SmartNET and Leica DC software and the Trimble used a Trimble networks do Trimble DC software. We did not investigate it too much further and applied a correction factor after we proved the difference between the two. Naturally the field crew took the initial blame, but I'm pretty sure they were not the problem.
  
  I don't know if this is the same shift you are experiencing and I'm not sure why we encountered it, but it was real. We also had a smaller difference in the horizontal and it too was consistent.
  
  Good luck.
  Bill
  • 

    Paul Quagge > Bill Schindler May 17, 2015 at 7:28pm

    Thanks for your input Bill...this discussion is gaining ground, I appreciate your input and will keep you posted to what I and others may contribute or find...Thanks Again...Paul
• Land Surveyor

  J. Anthony Cavell May 15, 2015 at 7:43pm

  I'm not sure I understand your situation. If you have 55 mount points, I doubt you are getting a network solution (implied by RTN). More likely you are getting corrections from one CORS within a network. If that is correct, getting different results is not very surprising. Every CORS occupies its own location with the unique environmental conditions of that place resulting in differences unique to that CORS.

  Networking software by using the whole network simultaneously uses data from all the stations to determine and mitigate the idiosyncrasies. Mount points then are usually many fewer, identified by preferred formats rather than individual stations.

  I'm still not sure I completely understood your request, but I hope this helps some.

  JAC
  • 

    Tom Homan > J. Anthony Cavell May 18, 2015 at 9:34am

    You are correct in that this is not a network solutions (RTN) It is a collection of single base solutions.

    Tom
    • 

      Paul Quagge > Tom Homan May 19, 2015 at 12:18am

      Hello Tom, thanks...do you use the AZHMP? If so what have you found? Do you use AZGPS?

      Thanks, Paul
  • 

    Paul Quagge > J. Anthony Cavell May 15, 2015 at 8:30pm

    Sorry to all...let me clarify....I have 55 mount points or continuous operating stations sending out corrections that I can choose from within the network. I connect to one mountpoint at a time, yes each mountpoint has a data sheet...I inventoried each mountpoint and found varying degrees of reception from my current location....I understand that the longer the base line isn't always an indication of poor reception, there are other factors...Thank you for explaining the differences J.Anthony, it does help.

    My question is rooted in GPS legacy which is why I'm asking. At one time it was understood that collected data shifted globally along a vertical angle or plane...if I had a better handle on it I would be able to explain it better, I apologize....Thank You, Paul
    • Land Surveyor

      J. Anthony Cavell > Paul Quagge May 15, 2015 at 8:39pm

      After about 4-5 hours of data collection, the geometry of the baseline becomes irrelevant. If performing single base- baselines, after about the same time the length of the baseline becomes almost irrelevant (not counting the ppm of course). 

      Most single baseline processing will give ppm of about 1:1,000,000. Most network based processing will give ppm of about 1:10,000,000. This is primarily because the network can mitigate local errors of any of the individual stations.

      If actually doing RTN (RTK) then measurements should be repeated about an hour or more later for redundancy and conformation of reasult.

      JAC