Activity Feed › Discussion Forums › Strictly Surveying › Running Levels with a Bit of GNSS
-
Running Levels with a Bit of GNSS
Posted by field-dog on April 28, 2024 at 6:14 pmWe had great success with speed and efficiency by running levels and using GNSS to mark out setup and turning points. Our party chief marked the points with pin flags and stored the points for potential future use. He used a Stake to a Line feature and added the Distance to Station Ahead/Station Back readout to the map screen to get good splits. We located the record beginning and ending BMs as well as the new route BMs before beginning. We closed the almost 3 mile run with a 0.003′ error and a maximum allowable error of 0.059′. Distances balanced to +/- 6′. It was a great way to end the work week.
OleManRiver replied 4 months ago 10 Members · 36 Replies -
36 Replies
-
I’m pleased that your levelling worked out so well but merely pacing off the instrument and turning points would do the trick.
-
@ Norman_Oklahoma
What you said about merely pacing off the distances is true, but the topography wasn’t the friendliest for that. Also, we were told to incorporate three additional published BMs into the run for adjustment purposes. In other words the office wanted multiple elevations on three BMs.
MH -
Run a “trig level” in Carlson with a fixed rod. I only say that because the routine will balance your run and it is very slick for when there is a lot of change in elevation. If it is flat, not really worth it.
What was the advantage to having your legs balanced so closely? Is the advantage worth the time and effort in you estimation?
-All thoughts my own, except my typos and when I am wrong.-
This reply was modified 4 months, 1 week ago by
dmyhill.
-
This reply was modified 4 months, 1 week ago by
-
“…we were told to incorporate three additional published BMs into the run for adjustment purposes.”
Glad you had a great end to your work week, so I am not complaining, but am interested in how people do things. Shouldn’t you run your loop, adjust it if you must or need to, holding a single benchmark?
-All thoughts my own, except my typos and when I am wrong.-
This reply was modified 4 months, 1 week ago by
dmyhill.
-
This reply was modified 4 months, 1 week ago by
-
“Shouldn’t you run your loop, adjust it if you must or need to, holding a single benchmark?”
Not necessarily. If your purpose is to densify an existing benchmark network then holding some or all existing BMs in an area is appropriate.
-
“Shouldn’t you run your loop, adjust it if you must or need to, holding a single benchmark?”
Not necessarily. If your purpose is to densify an existing benchmark network then holding some or all existing BMs in an area is appropriate.
Having never worked for a public entity, I haven’t run into that sort of work. My work has been exclusively to establish project control, at times with a tie to a datum, and my control would be considered Temporary Benchmarks, with direct reference to the published benchmarks.
My work in balancing control has always emphasized keeping the constraints as limited as possible.
Say you you run a loop between 3 benchmarks, establishing marks between each one, and you find two original BM’s match within the precision of the closure of the loop. The third BM does not match.
-Do you warp the connecting legs and their new benchmarks to match the third BM?
-Why wouldn’t you republish the elevation on the hypothetical third BM to match the best available information?
-What if all three mismatched to some degree greater than the expected precision? My first instinct would be to establish a single answer.
Perhaps I misunderstand what you are describing.
-All thoughts my own, except my typos and when I am wrong. -
I’ve always included multiple bench marks in a level run if they are available. Much better than hanging out control based on one mark.
It’s a matter of redundancy, better to have more, less accurate closures than one, more accurate closure.
It’s not even a choice if you’re doing DOT work.
-
Having never worked for a public entity, I haven’t run into that sort of work. My work has been exclusively to establish project control, at times with a tie to a datum, and my control would be considered Temporary Benchmarks, with direct reference to the published benchmarks.
It’s less of a “public sector” thing than it is a “network accuracy” thing.
A “tie to a datum” can mean a lot of things, and when there are multiple physical marks representing that datum, it’s often best practices to tie into as many as possible. At least, if the goal is to densify or supplement that network. For a one-off project, sure, it’s probably not necessary.
A minimally constrained adjustment will verify the efficacy of the observations themselves as well as allow us to determine how well published control values fit each other. If the object is to establish a new mark that will be consistent with the entirety of the network, we’re going to want to hold as many of those existing marks fixed as possible.
There’s a point at which “adjusting” does become “warping” (I guess one could argue that all adjustment involves warping) but that’s variable depending on the work being done. As the observed values drift away from the published values, the adjusted error ellipses will go out of tolerance and that’s when the decision needs to be made about what to hold fixed, or how much to let a particular control value (or values) float.
“…people will come to love their oppression, to adore the technologies that undo their capacities to think.” -Neil Postman -
How long did it take you to run through that line (s). Nice job. I use to do similar with gps rtk. We would be surveying airports military. So we would have a window at certain times to run levels from one end to the other. Not a lot of time for any error. What we would do is paint the instrument point and where the turn points needed to go and I would use a old 4000ssi rtk as pacing I could do but when the tower gets you on and off the cl of runway so much during touch n goes and training it made up time to be able to do a lot at night and throughout the day as we set packs n sacs and other things. This made for the time for running levels a breeze everyone knew exactly where to go and skip right through quickly. We used Tuttle’s for turning points. I would also rtk cl of runway back then in various spots . We had devised a device that hung a plumb on off the hood of the survey vehicle and had gps antennas on it so we would start at end of runway point do ppk to other end turn around and come back usually at minimum we had 3 static receivers set up on pacs sacs all logging 24/7 for the whole two weeks roughly we did all the surveying. Even back then we got some pretty good results with some stringent field procedures. I am sure now days they are not doing all that as much as technology as become so much better. Of course we were surveying on the ellipsoid not orthometric heights and all data was reduced to the ellipsoid as that was what was needed. Fun times.
Now because you were so strict with balancing bs fs distances between each turn along with the total of bs distance and fs distance that’s a great thing it works it makes for great results no matter if you are reading three wires or using a digital level. The old standards and procedures manual calls that out which most look at just the closure that counts to in geodetic leveling . All I know is when we followed those principles we had great results almost every time no matter where we were or how far we traveled or how the conditions were .
The many benchmarks in a level network is nice when you adjust and reduce for correct data c/r orthometric corrections you can weight those BM’s accordingly to their uncertainty if NGS published now you are tied to the datum and have a relative uncertainty and also network uncertainty great job . Always great seeing good work done
-
“Say you you run a loop between 3 benchmarks, establishing marks between each one, and you find two original BM’s match within the precision of the closure of the loop. The third BM does not match….. …”
There is a lot of “it depends” in any answer that can be given. But if one doesn’t match the other two (and you are quite sure that it isn’t a blunder in the current work) then you probably need to update the data on that rogue benchmark. No, you do not warp to fit something that is proven wrong.
If the benchmarks are all within tolerances and have existed for a while, and been used a lot, you probably don’t want to fidget with the published values. Just what “within tolerances” means is open to interpretation.
-
This reply was modified 4 months, 1 week ago by
Norman_Oklahoma.
-
This reply was modified 4 months, 1 week ago by
-
I have worked on a few military runways and local airports that receive commercial air traffic and have never come across one that did not have 1st order centerline control monuments established and proven on both ends of the runway. You either go from A to B and check in within tolerances allowed, or you drop back to see where you went wrong in your procedures or calibration of your equipment.
-
I think that everybody is going off topic and not considering the GNNS aspect of the conversation and the achieved “balance” in leveling stations. It seems to me that, based on the method mandated, the mandate for the procedure was dictated by office staff.
Was it really necessary to send somebody ahead with GPS/GNSS equipment to stake to a line that was predetermined by office calcs, or, could that have been easily been accomplished by either pacing or slap chaining distances between turning points? Are a few feet of distance between back and fore readings going to change the results?
Things got done accurately before the advent of GPS that had a very limited constellation available and required more than the OPUS hours of static observation times. The younger Crowd seems to focus more on technology and less on the methods to do things terrestrially to achieve the desired results, based on proper procedures.
-
@chris-bouffard i think that the OP mentioned gps as pacing in that environment would not have yielded the balance probably as quickly. I think they used gps as a tool and it proved in this situation to be beneficial. We would pace for miles running geodetic levels. That fore site rod hit the ground and the first thing read was top n bottom wire. For distance. Bs distance was known then I could say come or go or hand signals and how far. To keep the balance. Remember the bs fs distance must be within a certain range as well as the total of bs fs distance for entire line or run.
We used quad and Topo maps and a scale to plan the route and to have an estimate of total distance and apps number of turns on long runs before even setting foot on the job . As we ran we ever so often checked to see how we matched up as we also had to have an even number of turns. So BS rod and FS rod landed correctly. This would if done correctly cancel the error of any rod discrepancies which were also known before hand as they were calibrated. When we first started using and testing the digital levels there were no guidelines written for that yet. As a matter of fact I was out of the USMC for at least a couple years before they came up with a guidance on using them here in the USA. For federal work NGS. I spoke with the geodetic advisor at the time and told him how we had tested ours in the service against our wild and such geodetic levels and developed the procedures to use them. Some of the same conclusions are in the current system now. I agree that technology is somewhat over used but it’s not the technology that’s the problem. That’s fine. I believe it’s when one doesn’t understand the principles and the technology in a way to meet and or exceed what was done before it came about. I mean they surveyed the coast with a 1” transit that was designed by Ferdinand Hessler and it was so heavy it was carted in a buggy. Also so sensitive it could only be leveled at night. That was around 1809 ish I believe he was appointed in like 1807. I hope my dates are right. I often think. Take a well used transit for many years like the wild T-2 T-1 T-1A T-3 my favorite. Those things were light could literally be dropped out of a plain attached to a soldier. Could easily be adjusted on site and still be as accurate and precise as we are today. Angles. Now what is better is we have compensators that now can automatically detect and adjust to settlement. So one doesn’t have to be as untuned with that equipment as once was necessary.
As far as military having first order BM at end of runways. That’s neat. We never had that. We had PACS and SACS and often other marks HARN azimuth marks even 1st Oder BM on some bases. Never at end of runways. But I mostly did USMC airstrips a few air force but that was OCONUS . That would be a nice set up though.
I did a commercial airstrip easement last year. Had to tie it to end of runway and saw something you would never see on a military airstrip well not supposed to especially if harriers land was PK MAG nails set at each end. All I can say is the surveyors I was retracing and the NGS published values 2000 feet away when I hit we missed each other by less than 10” in azimuth and less than a tenth in straight delta. That was holding all the corners 2000 feet away for all the several different properties that the easement ran through. The NGS values were all tight as well. I think it was 5 different surveyors imthat did several different bouncier retraced to tie down the easements. Awesome following that type of stuff.
-
I can do GPS spacing for level work with my DC, it’s no bigger than a cell phone and I don’t need a base, pole, or receiver head to use it. Quick, easy, as fast or faster than pacing. Within a meter each time, plenty good for that type of measurement. And if you don’t want that get a golf rangefinder.
-
Yeah. Thats a great way to do it for sure. I had not thought of that. I remember the back pack days no rod needed for certain things lol.
-
@ dmyhill
There’s no advantage to having the legs balanced so closely except to make the office happy because they like it that way. The advantage was worth the time and effort. We saved time, especially when tying into an existing run. The effort was minimal. The purpose of the project was to extend the county network. We started and closed on published BMs from one run and ran through three BMs from another one. Those three BMs now have multiple elevations on them. Do you use least squares to come up with the best elevations for those?
MH -
@ OleManRiver
How long did it take you to run through that line (s). Nice job. I use to do similar with gps rtk.
It took about three days. We used GPS RTK. Thanks for the compliment.
MH -
“Those three BMs now have multiple elevations on them.”
What purpose does adding yet another elevation serve?
-
@ chris-bouffard
It seems to me that, based on the method mandated, the mandate for the procedure was dictated by office staff.
It was not. We thought it might speed things up.
Was it really necessary to send somebody ahead with GPS/GNSS equipment to stake to a line that was predetermined by office calcs, or, could that have been easily been accomplished by either pacing or slap chaining distances between turning points?
It was easier and more efficient than pacing. There were no office calcs.
Are a few feet of distance between back and fore readings going to change the results?
Certainly not. FGCC (1984) third order leveling specifies difference of forward and backward sight lengths never to exceed 10 m per setup and 10 m per section. I think that’s been supersede by standards that address digital leveling.
MH
Log in to reply.