Do you generally prefer to use RTNs from start to finish on jobs, or do you just use them to set control for RTK? I understand there are many factors at play here such as the size/scope/overall goal of your project, proximity to a base station, cell signal, etc.--so let's just say it's a standard boundary job, and the limits of the work you expect to do would be confined to a single PLSS section. Thoughts on topographic surveying are also welcome!
The firm that I work for uses the ARDoT and SmartNet RTNs. We've primarily used them to set control to take off of with RTK at the start of each job. We do seem to get pretty repeatable results using the RTN, and control points generally seem to check in as they should. SmartNet seems to work better in most places around us than ARDoT, presumably due to a denser network of base stations in our area. ARDoT is usually fine when you're close to a base, but it can be hit-or-miss in some places.
We have run the RTN from start to finish on some jobs, and it's worked pretty well for the most part. Granted, these jobs were close to a base and had good cell signal and relatively open sky. However, I have heard in the past that RTNs can be more susceptible to bad fixes in certain situations. It seems to me that switching to RTK would generally be a safer option when you know you have good control to start from.
I'd love to hear everyone's thoughts on this, as well as any personal experiences with ARDoT, SmartNet, or any other local RTNs people might use in other states!
I use all the tools I can to get the project done. I believe the relative precision is usually better for base and rover. Not all NRTK are equal. So I will set a point with NRTK. The. Set my base on that and log data and tie in property corners and additional control. I will even after observing a base rover point switch to NRTK at times on some of my points. Also use robot and might throw up my rover on a point to log static while using a robot. It takes only a 1/2 a minute to do so and let that collect while I might be traversing through some thick areas that GNSS doesn’t perform well enough in. Remember base is logging as well. We use smartnet here as well. What I see is it performs well in most of my work areas. Not as good as base and rover in a canopy situation. I just use all the tools based on proven results over time and incorporate all the data into my survey after throwing out blunders and or bad observations.
I'm seeing more and more plats stating that the survey was performed with GPS. I'm in Northeast Georgia where canopy is the rule. We typically use a Carlson Brx7 on the eGPS network to set control in areas of less or no canopy and robot the rest of it. I recently found both Brx7s not being used by the crews and decided to try base and rover in a mostly open area. It was a small topo in the open mostly so we did the topo with base and rover AND with the robot off of the same control. I must say that I was rather amazed at the comparison because it was almost perfect. There was one bad check shot on control with the GPS. About 0.20' in the vertical. When we ran it with the robot we verified the base and rover elevation on the point in question. It turned out the bad shot was with the GPS on the network. An old friend. Phil Stevenson (God rest his soul), used to say "the proof is in the dirt".
My $0.02 worth.
Me. "What's the difference?"
T.C. Carroll "It's the difference between right and wrong!"
@stacy-carroll almost all the latest receivers Carlson Brx7 Trimble r12I Leica etc etc. Javad. When you can set the base up in a good environment the rover can do amazing things in some nasty canopy areas. I went in to verify a Topo along a creak. All originally ran with conventional survey means. I ran my Topo observations up to more epochs and had great results. I do a lot of boundary and control for boundary in canopy. What I personally have learned is this. If I have at-least 21 satellites and a couple have L5 band on the GPS constellation and I can get a 180 epochs without loosing lock the rms values raised up to 95% confidence level truly work. Now I do come back and hit those same monuments again and sometimes it takes a third time or an independent location from good control from robot can meet all our requirements. Now I say all this but I would never shoot or locate one shot on any property corner even in the wide open with GPS period for a property corner with no manufacturer equipment. This comes from my time monitoring the satellites themselves. A lot can happen and some things can happen that no manufacturer can solve for. Topo I do one observation but randomly I set a temp shot or something I can come back to and hit again as a check when using gps. Might be a paint dot on a rock or a nail stuck the ground etc. or like you did set up and spot check some shots with conventional. I do believe we still are not far from a time when only a rover will go out no base no NRTK just raw satellite data and achieve great results. How we go about proving those results are still up for debate but it’s coming and coming fast. One will simply grab his rover if we even call it that at that time go out and get results that will be as good as what we go through today with base n rover. The URE is getting so good that it will become unnecessary to do RTK or even static in the future. True PPP will be all that’s needed unless something happens. Oh I do hope everyone is still watching the solar activity as that is still a factor.
A surveyor I knew a few years ago simply despised GPS. He said the "Surveyors would eventually only visit the corners and not walk the lines. You aren't truly surveying if you don't walk the lines." He retired with those same sentiments. And we've all probably seen the high accuracy survey that got the corners tight but missed the encroachments. Can happen with most any tool in the tool box in the hands of some...
I started using static GPS in 1999 or so. As with any new technology, I pushed it to its limits so I could learn it's capabilities. There were so many different factors that could kill the accuracy and also so many times when the black box lied and said you had a good solution when you didn't. The biggest improvement, in my opinion, is a receiver that doesn't lie to you. I've gotten good results in heavy canopy, and I've also been unable to get a fix at all. I know there are places and times that GPS won't work the way I want it to. I can always change my method as long as I know I need to. I guess those Sunday school classes were correct " The truth will set you free"
Me. "What's the difference?"
T.C. Carroll "It's the difference between right and wrong!"
A surveyor I knew a few years ago simply despised GPS. He said the "Surveyors would eventually only visit the corners and not walk the lines. You aren't truly surveying if you don't walk the lines." He retired with those same sentiments. And we've all probably seen the high accuracy survey that got the corners tight but missed the encroachments. Can happen with most any tool in the tool box in the hands of some...
I started using static GPS in 1999 or so. As with any new technology, I pushed it to its limits so I could learn it's capabilities. There were so many different factors that could kill the accuracy and also so many times when the black box lied and said you had a good solution when you didn't. The biggest improvement, in my opinion, is a receiver that doesn't lie to you. I've gotten good results in heavy canopy, and I've also been unable to get a fix at all. I know there are places and times that GPS won't work the way I want it to. I can always change my method as long as I know I need to. I guess those Sunday school classes were correct " The truth will set you free"
For sure The whole walking the lines thing has honestly improved in some scenarios vs a robot or total station. Let me explain Crew is given a compile of the boundary hand to be surveyed obviously this is rough Crew finds the first corner some monument on the site could be as joined or clients as they search they find another monument They simply take the dxf or dwg and translate rotate much the same as we do in the office as we start resolving the boundary this gives them a physical line to walk as they locate fences etc I always tell crews look at the plats walk the lines if you can get on the line it will lead you to the corners the same thing can be done with staking a line between two points or translate rotate points But even that can become an aid to install laziness if crews just want to drive around and find corners from search points etc I update the compiles as more information is collected and update the crews on hey can you give me more fence locations in this area or are these sheds still here etc
Back to the original post. I've been using RTK in various forms since before O.J. became a glove model. I have found that a Base/Rover better vertical precision than NRTK solutions, but not enough to warrant setting up a base when a network is available. Over time, that gap has been greatly reduced. It is interesting that most data sheets indicate a smaller ppm for NRTK versus single base while SmartNet automatically changes its corrections to single base when you are within a certain distance of a CORS.
For the most part I limit my use of base/rover RTK to location without NRTK or cellular coverage. With the use of satellite communications, even the lack of cellular coverage is removed from the equation. The only real exception for me is when using our rail geometry trolly were I like to ensure the base is as close to the project area as possible.
We rarely do boundary surveys at the county; however topographic surveys are done frequently. Currently we're doing topographic surveys of parts of a subdivision that are experiencing drainage issues. Water is pooling at the turnouts of street intersections and some street straightaways too. We used base/rover to set control only. Everything else is being done with our total station. We used the Florida Permanent Reference Network (FPRN), the Carlson BRx7 base/rover, and the Carlson Listen Listen correction service.
Back to the original post. I've been using RTK in various forms since before O.J. became a glove model. I have found that a Base/Rover better vertical precision than NRTK solutions, but not enough to warrant setting up a base when a network is available. Over time, that gap has been greatly reduced. It is interesting that most data sheets indicate a smaller ppm for NRTK versus single base while SmartNet automatically changes its corrections to single base when you are within a certain distance of a CORS.
For the most part I limit my use of base/rover RTK to location without NRTK or cellular coverage. With the use of satellite communications, even the lack of cellular coverage is removed from the equation. The only real exception for me is when using our rail geometry trolly were I like to ensure the base is as close to the project area as possible.
You are correct that most manufacturers spec is 1/2 ppm less when using NRTK. The problem arises with that 1/2ppm from the closest physical base. When you are 18 miles from the closest physical base and less than a mile at 1ppm to a typical base rover. The base rover wins all day long. It’s a math game. Now where NRTK shines is most of them are modeling the troposphere so the game changes when your old base rover are in two different tropospheric conditions. I use smartnet here as well and the Trimble VRS. They both have pros and cons. I have used cell to cell for base and rover also. That doesn’t make it 1/2 ppm . Relative elevation differences truly depends on the network spacing and how well that provider monitors that network. When I operated a network we had alarms set up in a sense that alerted us when any base moved within the network by a certain amount from its primary position. Things like hot days and the antenna mast moves the buildings it was mounted on moved between seasons. You might be surprised just how much things are moving way more than you think. One of the owners was a civil engineer as well as structural engineer so it didn’t take long to find the best possible steel and method to eliminate the expansion and contraction of the mast itself. Now we had to monitor how much each building was moving as well. That’s just a small piece of the puzzle. The next is the latency from the base station to the server itself. Most network GNSS monitors this also. Then the latency from the server going to you via cell and your information to that server. These things can systematically cause you to be relative a noon but be off more at 1600. Again not all NRTK are created equal. It only takes one bad network station within your triangle to introduce this error. Oh we had some zepher geodetic antennas that literally separated from the GA heat. The casing. Choke rings are much better. Again antennas with radomes or without all have a different APC even with the exact same model. Now we have not discussed multipath or other signals in the areas of the base. So until one has built in some independent checks in there field procedures and are constantly sampling that in there work area as antennas go bad they get replaced or upgraded one never knows truly how good a network is.