Notifications
Clear all

8 mile level loop with 2 different rods

22 Posts
6 Users
10 Reactions
442 Views
(@ryancj31)
Posts: 74
Member
Topic starter
 

We signed up for an 8 mile level loop (4 miles each way). We are already running a bit behind schedule. I was thinking of running a 3 man crew. One instrument man on the Trimble DiNi and 2 rodmen, one rear and one forward. Each rodman will have their own rod (same exact model).

This should obviously speed up the operation but my concern is if the 2 rods read, lets just say 0.005 different that could accumulate a significant error if we will have 150+ setups. (150 x .005 = 0.75')

Thoughts? Concerns?

 
Posted : December 12, 2024 12:08 pm
Norman_Oklahoma
(@norman-oklahoma)
Posts: 7728
Member
 

Get 2 screwdrivers, one with a red handle and one with a blue. At each turning point stick them both in the ground, adjacent but not exactly the same height (that shouldn't be hard to do). Shoot to and read the red, then read the blue. Keep the left hand page with the red notes, the right hand page with the blue notes.     

 
Posted : December 12, 2024 1:23 pm
(@jon-payne)
Posts: 1611
Supporter
 

Mark the rods so that you are absolutely sure you are running one for foresights and one for backsights.  Do not mix them up.  Make sure you actually physically turn through the point that you start the loop back.  Any difference in rods should have accumulated on the forward run and then be removed in the run back to the beginning.

 

Is it not possible to get a good static GNSS elevation at each end so that you only run one way?

 
Posted : December 12, 2024 2:10 pm
1
(@ryancj31)
Posts: 74
Member
Topic starter
 

@jon-payne as per contract we have the option of not doing a loop. I guess I am of the opinion that when someone requests high accuracy elevation data, a level loop is the only definitive way to be certain of your data. Would the world stop turning if we did not close it out with a loop? Unlikely. I guess I am just trying to perpetuate good survey practices.

Maybe my thoughts on that are getting a bit outdated. I guess I feel a bit iffy on doing high-precision level work and then in the end leave it to the satellites and GPS to tell you how good of a job you did. We will see if my opinion changes after 4 miles.

 

And thanks for input. My head was hid of spinning after trying to work out the procedure and accumulated errors in my head.

 
Posted : December 12, 2024 2:38 pm
(@jon-payne)
Posts: 1611
Supporter
 

@ryancj31 You got to do what you are comfortable accepting responsibility for.  I can certainly appreciate taking extra care.  The last time I was involved in a long corridor level loop was over 20 years ago.  We three-wired control between static GPS derived points and matched so well that we just adjusted between the GPS control points.  That was not with a digital level, so it might be so quick to get readings that any time savings is a moot point anyway.

 
Posted : December 12, 2024 3:15 pm

OleManRiver
(@olemanriver)
Posts: 2542
Member
 

You need to flag one I always did red. It’s called red rod. You neeed to make sure they alternate equally. So if red rod is first BS it ends as closing FS. By alternating them any differences in the rods cancel out. This is all written in old geodetic leveling techniques. You do not want to keep the same one on all BS. You should end no matter what with an even number of turns. If I am remembering everything correctly.  If your guys are use to running levels they should knock that 8 miles out in 2 days tops. A good crew can get about 5 miles a day average. That dini is great. 

 
Posted : December 12, 2024 4:40 pm
2
(@bstrand)
Posts: 2359
Member
 

Posted by: @ryancj31

I was thinking of running a 3 man crew. One instrument man on the Trimble DiNi and 2 rodmen, one rear and one forward.

This is the only way I've done long level loops like that and it worked great.  We used deadblow hammers with leveling pins that had protective caps and carried magnets that we could stick to the bottom of the invar rods so we could sit flush on brass or aluminum cap style monuments.

 
Posted : December 12, 2024 5:41 pm
1
OleManRiver
(@olemanriver)
Posts: 2542
Member
 

Get you a pair of turning turtles. No wasted time driving stuff in the ground. Maybe ste something about every mile more permanent. These things are great for along hard surfaces roads airstrips etc. just make sure the Rodman don’t pull it to quickly. If they are not good pacers to keep the bs FS balanced have them do a distance to FS with Dini before taking the bs helps keep things balanced and moving. 

 
Posted : December 12, 2024 6:35 pm
1
(@ryancj31)
Posts: 74
Member
Topic starter
 

Thanks for input and tips. Got it done in 2 days. 4.7 miles one day and 4.7 miles returning the next. The 3 man crew worked pretty good with the rodmen leapfrogging to alternate the rods on each BS/FS. 

We ended up doing a loop and closed under 0.02' feet (way to go DiNi!). BS/FS was balanced under 50'. A couple remain questions:

1. @olemanriver "You should end no matter what with an even number of turns" - I think I know the reason for this but could you elaborate.

2. if you take a shot on a point once on the way up and once on the return, how do you report the adjusted elevation? Don't you end up with two "adjusted elevations" for the same point? In this case adjusting out 0.02' in 160 setups is negligible but for future loops I'm curious.

3. @olemanriver I did find this NOAA publication online https://www.ngs.noaa.gov/PUBS_LIB/GeodeticLeveling_Manual_NOS_NGS_3.pdf I did not read through all of it yet but was wondering if you were thinking of others that might be helpful. 

4. Who has seen a cooler survey vehicle than this one in the attached image?

image
This post was modified 2 months ago by ryancj31
 
Posted : December 23, 2024 7:12 am
1
OleManRiver
(@olemanriver)
Posts: 2542
Member
 

@ryancj31 by ending on even number you are balancing out the rods and turns. 

On adjustments. Of .02 per 160 turns.  You are looking at this wrong.  All you have done so far is meet a closure and have different in heights. You need to apply the corrections of everything to get true NAVD 88 heights and then adjust. By running through the same points forward and back your turning points you could end up with two different elevations unless you perform a least squares and let the software know they are the same point. Running levels is a funny thing gravity also. Depending on a direction of a a run can dictate slight differences in heights along any route. So let’s say you ran around a site. Then run due north across the site without applying the corrections you can have slightly different results. Now this is in the tenths of a millimeter. So are most of the corrections to make the changes. I have no idea on your project requirements but if you do want true navd 88 heights and you came off of and ran to a BM that was of higher order or equivalent to what you are required to do then following the guidelines for all aspects and applying all corrections and adjusting per guidelines you should be good to go. If all you are doing a for a typical survey and all of this is good enough then getting into all the other stuff is not as important I guess. It has been a minute since I have done all this and laid it out. I had guessed if you were running 8 miles with a DINI and two one piece rods and all was good you would have closed around.5 to .7 mm .  But I think your guys did great for 8 miles up and back so 16 miles. And did it at a good pace for not doing that type of work often. You should buy them lunch.  Levels in the geodetic realm long level lines can make a brain short circuit. It becomes monotonous so they are pretty disciplined and that’s a good thing. We’re your rods two pieces or one piece?  Which dini do you have. If all the survey gods are alligned so equipment is all adjusted rods are good. All balanced legs.  The biggest error comes from the Rodman. After running many miles with both reading three wires and also the dini I learned one thing the Rodman was the most critical position on that crew.  And it becomes easy for them to not focus and not stay attentive. Your team did a good job for sure.  

 
Posted : December 23, 2024 7:56 am
1

OleManRiver
(@olemanriver)
Posts: 2542
Member
 

@ryancj31 by ending on even number you are balancing out the rods and turns. 

On adjustments. Of .02 per 160 turns.  You are looking at this wrong.  All you have done so far is meet a closure and have different in heights. You need to apply the corrections of everything to get true NAVD 88 heights and then adjust. By running through the same points forward and back your turning points you could end up with two different elevations unless you perform a least squares and let the software know they are the same point. Running levels is a funny thing gravity also. Depending on a direction of a a run can dictate slight differences in heights along any route. So let’s say you ran around a site. Then run due north across the site without applying the corrections you can have slightly different results. Now this is in the tenths of a millimeter. So are most of the corrections to make the changes. I have no idea on your project requirements but if you do want true navd 88 heights and you came off of and ran to a BM that was of higher order or equivalent to what you are required to do then following the guidelines for all aspects and applying all corrections and adjusting per guidelines you should be good to go. If all you are doing a for a typical survey and all of this is good enough then getting into all the other stuff is not as important I guess. It has been a minute since I have done all this and laid it out. I had guessed if you were running 8 miles with a DINI and two one piece rods and all was good you would have closed around.5 to .7 mm .  But I think your guys did great for 8 miles up and back so 16 miles. And did it at a good pace for not doing that type of work often. You should buy them lunch.  Levels in the geodetic realm long level lines can make a brain short circuit. It becomes monotonous so they are pretty disciplined and that’s a good thing. We’re your rods two pieces or one piece?  Which dini do you have. If all the survey gods are alligned so equipment is all adjusted rods are good. All balanced legs.  The biggest error comes from the Rodman. After running many miles with both reading three wires and also the dini I learned one thing the Rodman was the most critical position on that crew.  And it becomes easy for them to not focus and not stay attentive. Your team did a good job for sure.  

 
Posted : December 23, 2024 7:56 am
Norman_Oklahoma
(@norman-oklahoma)
Posts: 7728
Member
 

Posted by: @ryancj31

2. if you take a shot on a point once on the way up and once on the return, how do you report the adjusted elevation? Don't you end up with two "adjusted elevations" for the same point?

Least squares adjustment makes this question moot. 

 
Posted : December 23, 2024 9:18 am
2
OleManRiver
(@olemanriver)
Posts: 2542
Member
 

@norman-oklahoma that is correct which is how an adjustment should be done at least that’s how we were required to do them when I was running geodetic levels. After all other computations had been performed. I think it has been a while for sure.

 
Posted : December 23, 2024 3:49 pm
Norman_Oklahoma
(@norman-oklahoma)
Posts: 7728
Member
 

In Oklahoma, the procedure demanded by the DOT was to make up a table with columns for Run#1, Run# 2, and Average. Supposing you started on BM "A" and ran through B, C, D, E, F....... etc. The field staff would run the levels in that order and end the day a few miles up the road from their truck, which was back at "A", and waste quite a lot of dead time walking back to it. The next day they would start at "A" again and repeat.   You can't believe how much trouble I had convincing them that it would be alright to work outbound for half the day, then run back in reverse order, ending up back at the beginning without a lot of deadhead walking, and how I could reorder their measurements to the DOTs satisfaction.     

 
Posted : December 23, 2024 5:42 pm
Norman_Oklahoma
(@norman-oklahoma)
Posts: 7728
Member
 

In Oklahoma, the procedure demanded by the DOT was to make up a table with columns for Run#1, Run# 2, and Average. Supposing you started on BM "A" and ran through B, C, D, E, F....... etc. The field staff would run the levels in that order and end the day a few miles up the road from their truck, which was back at "A", and waste quite a lot of dead time walking back to it. The next day they would start at "A" again and repeat.   You can't believe how much trouble I had convincing them that it would be alright to work outbound for half the day, then run back in reverse order, ending up back at the beginning without a lot of deadhead walking, and how I could reorder their measurements to the DOTs satisfaction.     

 
Posted : December 23, 2024 5:43 pm

OleManRiver
(@olemanriver)
Posts: 2542
Member
 

@norman-oklahoma good grief. lol. When we were evaluating the Digital levels to see if they would perform as well as our geodetic Wild levels. For most of our work we did test the dini double run at same time was neat to have everything recorded as you went. That was in the early 2000’s. Honestly at that time we could move just as fast reading three wires I am sure they have become better now. But sunlight angle and such caused some down time trying to get readings early on. But having a digital file was a time saver on the back end for sure. We had a seat of TGO and terramodell in which we could import data but it was never authorized to meet our requirements. I had a smart kid right a program to pull all the data out and place it into our forms. Made that better.

 
Posted : December 23, 2024 7:21 pm
(@ryancj31)
Posts: 74
Member
Topic starter
 

@olemanriver and @norman-oklahoma you seem to have a wealth of knowledge on this subject so if you don't mind I will pick your brain one more time.

To clarify any previous questions about our project: The DOT hired us and had already set Feno monuments, mag nails and 3/4" bars about 500' apart along a 4.67 mile highway corridor that will be reconstructed sometime in the future. This section of the highway is pretty much due N-S. They are holding their X & Y coordinates that they measured with GPS. They hired us to get the Z value on all the control points. The DiNi we are using is the 0.3mm/KM model. (does anyone know what DiNi stands for???). The 3 section folding barcode rod was used for this along with a turning turtle whenever appropriate. We took numerous GPS shots an established NGS monument near the project which checked out with the recorded elevation on the NGS Datasheet. This recorded elevation was held as our "gospel" elevation.

 

We are trying to send out our results to the DOT in the next day or two. I hate the Trimble Business Center interface and usually don't touch any TBC software as my co-worker has handled that on the other 2-3 times we used the DiNi.

After watching over the shoulder of how TBC seems to be adjusting this data I have come to the conclusion that the software is adjusting out the misclosure error via a correction value applied to every setup that is weighted by accounting for the BS & FS distances.

It also seems to know any point that has been measured more than once and will give you a little yellow circle on your spreadsheet. I assume this is doing the least squares adjustment on these points?

I guess I am most confused (or hesitant to send the final report) without fully understanding @olemanriver's comment "You need to apply the corrections of everything to get true NAVD 88 heights and then adjust" If you could elaborate on this that would be wonderful.

Happy new year all!

 
Posted : January 2, 2025 11:53 am
OleManRiver
(@olemanriver)
Posts: 2542
Member
 

@ryancj31 TBC can do a least squares on the levels. If you had your DINI set up and C/R and have the project set up in the correct datum and all you should be good for a DOT project. I would follow the DOT standards and such if they document them. The corrections I was speaking of are the orthometric corrections c/r temperature etc. NGS probably has a free online webinar that you could watch for true navd 88 type  work and using there software which you could do with your data. It has been 15 plus years since I went through the geodetic leveling class. It has been double that since I have actually performed a true geodetic leveling run.

A few years ago I did do a static control network along with some traversing and some rtk along with a few miles of leveling to meet a project specs for some monitoring.  I used TBC with all data and performed a least squares. I also took the same data through opus projects for static to compare the horizontal results as an independent check and a make me feel good about my absolute position on datum around the project site.  I took the level data which derived from a 1st order BM through project site control all control and all monitoring objects except some in water. Multiple different level runs. And all that was to a 2nd 1st order BM to make sure we were good. And tied to a 3rd BM to much was 2nd order BM. One 1st order BM had proved to be out from the others. It was where a new bridge replaced an old bridge and the old bridge embuckment had stayed but a new one placed right against it. Upheaval in the new build lifted it best I could tell. We had about 0.2 mm of error in those verification runs and closed back on ourselves. What was bad was the starting BM was the bad one. So I held the other 1st order BM only and hit the 2nd order BM well under the error estimate via NGS data sheet. All gps static came in under .015 ft vertical. The RTK was .025 greatest errors. I didn’t perform the field work just planned and guided the crews and prime company on the hat and how to achieve the results and I did all the adjustments and qa/qc and a few blunders where I had crews re-observe.  TBC is actually a very powerful little software for this type of work. It has some shortcomings for true geodetic computations but for most DOT and everyday surveying it is nice.  

what my experience can say is when it comes to highly precise and accurate work. The most important aspect is the field procedures period. I will say a lot of talk on this site and others about the instruments. The level rods are no less important than that dini. If you were running the 3 section rods that to me explains the .04 ft at first check. I would adjust in TBC via least squares and call it a day if your results meet the requirements and cash the check. Not bad work at all. I saw more than that from guys running levels around a small project site and they were tickled that they closed that tight. Your crews did a great job the results show that. I would say you would have had the invar rods one piece that same run would have been around .3 to .5mm easily I bet. 

for giggles how many bottom wires were less than say a couple feet above ground. How many mean middle wire were within a mm or so of being at the connection of the rods where they came together. Were there any chances that a top wire was above are very close to the top of the rod. These things can introduce error. How does your results compare to the gps orthometric derived observations jive compared to your geoid model in that area. Easy to simplify is to get diffences in heights between main points via both leveling and gps.  

people like @norman-oklahoma  and John Hamilton are way more versed in this than I am. They are more familiar with US CONUS requirements than I am. I did a lot of geodetic work but most was not in the USA so while similar some things are different. 

 
Posted : January 2, 2025 12:45 pm
OleManRiver
(@olemanriver)
Posts: 2542
Member
 

Posted by: @ryancj31

@olemanriver and @norman-oklahoma you seem to have a wealth of knowledge on this subject so if you don't mind I will pick your brain one more time.

To clarify any previous questions about our project: The DOT hired us and had already set Feno monuments, mag nails and 3/4" bars about 500' apart along a 4.67 mile highway corridor that will be reconstructed sometime in the future. This section of the highway is pretty much due N-S. They are holding their X & Y coordinates that they measured with GPS. They hired us to get the Z value on all the control points. The DiNi we are using is the 0.3mm/KM model. (does anyone know what DiNi stands for???). The 3 section folding barcode rod was used for this along with a turning turtle whenever appropriate. We took numerous GPS shots an established NGS monument near the project which checked out with the recorded elevation on the NGS Datasheet. This recorded elevation was held as our "gospel" elevation.

 

We are trying to send out our results to the DOT in the next day or two. I hate the Trimble Business Center interface and usually don't touch any TBC software as my co-worker has handled that on the other 2-3 times we used the DiNi.

After watching over the shoulder of how TBC seems to be adjusting this data I have come to the conclusion that the software is adjusting out the misclosure error via a correction value applied to every setup that is weighted by accounting for the BS & FS distances.

It also seems to know any point that has been measured more than once and will give you a little yellow circle on your spreadsheet. I assume this is doing the least squares adjustment on these points?

I guess I am most confused (or hesitant to send the final report) without fully understanding @olemanriver's comment "You need to apply the corrections of everything to get true NAVD 88 heights and then adjust" If you could elaborate on this that would be wonderful.

Happy new year all!

https://www.ngs.noaa.gov/locus/equip.xhtml

nos 34 for geopotential comps for geodetic leveling .

 

the DINI was a carl Zies level I believe and like Trimble bought out the old geodometer/spectra precision robot they bought the digital level from Zies if my memory serves me correctly  I I think dini might have maybe been someone’s name but I have no idea I been down with fever last couple days so just can’t focus very well yet  

 

 

 
Posted : January 2, 2025 12:54 pm
Norman_Oklahoma
(@norman-oklahoma)
Posts: 7728
Member
 

I use StarNet. That software has the option of weighting observations either by the number of turning points between marks or by the distance distance between them. Apparently your TBC is set to weight by distance. I'm guessing there may be a setting to weight by number of turning points. Really, their isn't going to be a huge difference. 

I will enter observation data into Excel, creating a field book layout, run the pluses and minuses, and get elevations. I can add proportional misclosure values to the foresights and backsights to adjust to a perfect closure, and calc the elevations that way. Then I compare what the LS adjustment results of Starnet to my Excel field book results. If the misclosure is small, and everything is working right, the differences between the two will be so small that I rest easy knowing nobody is going to argue with me about it. 

Naturally, there is also the residuals report from the LS adjustment to look at. You should be seeing individual adjustments applied that are infinitesimally small.        

As far as OMR's comment ....... I think that he was speaking of manually adjusted loops. When you have common points in your outbound and inbound loops that amounts to having two or more smaller loops chained together to make up the big loop.  So you are best to adjust the smaller loops individually, then adjust the whole. But since you have only held a single point (ie/minimally constrained) you will have no such secondary adjustment to make. 

 

 
Posted : January 2, 2025 1:02 pm

Page 1 / 2