So.....
This is my biggest job ever.
The original control was shot in with VRS (ARGH!!!)
No Bench Circuit was ever run.
I'm just curious as to how any of you that have done a similar size job
establish a good set of Elevations for the entire site.
Horiz is no problem. Scale factor is right at 1 (Sweet)
I've already run a bench circuit 1/2 way with my robot (Direct and Reverse Readings)
using a BS/FS length of no more than 400' setting 30" x3/4" iron pins with caps and a tack in the cap.
Using one bench as a reference and Geoid12a I'm seeing anywhere from 0.015-0.05'
In the grand scheme of things I'm not real worried about it but it bugs me because
I'd like to have it tighter.
Suggestions?
TIA
If the design is based on Geoid12A, I'd say keep hittin' that pipe. A protected base station or VRS is your most stable and reliable H and V reference. On site TBMs as you describe are not as stable as one might expect, especially when hundreds of thousands of yards of material is removed/replaced/compacted.
If you really want on-site TBMs for check shots or to help the grade checker (both very good reasons), set 'em with static GPS.
Swim4Life, post: 333432, member: 6035 wrote: So.....
This is my biggest job ever.
The original control was shot in with VRS (ARGH!!!)
No Bench Circuit was ever run.
I'm just curious as to how any of you that have done a similar size job
establish a good set of Elevations for the entire site.
Horiz is no problem. Scale factor is right at 1 (Sweet)I've already run a bench circuit 1/2 way with my robot (Direct and Reverse Readings)
using a BS/FS length of no more than 400' setting 30" x3/4" iron pins with caps and a tack in the cap.
Using one bench as a reference and Geoid12a I'm seeing anywhere from 0.015-0.05'
In the grand scheme of things I'm not real worried about it but it bugs me because
I'd like to have it tighter.Suggestions?
TIA
I'm not surprised you are seeing 0.05', that can be the "error" in the model, they aren't perfect. I would run a bench loop, we always do, it doesn't take long, and it's worth it as far as I'm concerned.
A little confused by how you are applying Geoid12a, are you using the Robot to do that?
With all the tech we have, I can still rely on my 40 year old Leitz B2a. When in doubt, break out the level.
I'd run a digital level through the site setting good control for your robot every 900', run 450' each way so you still get good vertical accuracy when staking.
I think you have to take a more overall approach check into what vertical control you can as run the loop, then check your matches. No matter how good your vertical is it doesn't much matter if you are missing designed elevations by a half foot. It's great to have a nice tight loop (we always run digital loops) but if you are missing the datum the project was designed off of adjustments will have to be made in either design or your benchmarks to the design elevations. I would say set your control see how it matches then sit down with the project staff and make sure everyone is on the same page and it comfortable with what the project will be built off of then distribute your new control to anyone who needs it especially the guys calibrating GPS units for dirt work because even if your right it will be a headache later if they are on a different datum.
Swim4Life, post: 333432, member: 6035 wrote: So.....
This is my biggest job ever.
The original control was shot in with VRS (ARGH!!!)
No Bench Circuit was ever run.
I'm just curious as to how any of you that have done a similar size job
establish a good set of Elevations for the entire site.
Horiz is no problem. Scale factor is right at 1 (Sweet)I've already run a bench circuit 1/2 way with my robot (Direct and Reverse Readings)
using a BS/FS length of no more than 400' setting 30" x3/4" iron pins with caps and a tack in the cap.
Using one bench as a reference and Geoid12a I'm seeing anywhere from 0.015-0.05'
In the grand scheme of things I'm not real worried about it but it bugs me because
I'd like to have it tighter.Suggestions?
TIA
I was always taught to design it to a hundredth, stake it to a tenth and build it with a D9 Cat. What kind of final road surfacing? You're going to get a chance to stake it at least three times. Any grade problems should come up before final grade. They can't set the asphalt screeds to 0.015 ft. I've never done any concrete roads, but from what I've seen they can't set their screeds any closer either. We did an airport a few years ago that was asphalt surfacing and there were several bird baths that were obviously the machine's fault. You are already overkill in my opinion.
Skeeter1996, post: 333459, member: 9224 wrote: I was always taught to design it to a hundredth, stake it to a tenth and build it with a D9 Cat. What kind of final road surfacing? You're going to get a chance to stake it at least three times. Any grade problems should come up before final grade. They can't set the asphalt screeds to 0.015 ft. I've never done any concrete roads, but from what I've seen they can't set their screeds any closer either. We did an airport a few years ago that was asphalt surfacing and there were several bird baths that were obviously the machine's fault. You are already overkill in my opinion.
We had a 19 mile road re-construction project where we did just that. Surveyed in the control, brushed the r/w and measured cross sections every 100'.
That got us quantities for an estimate. We let the profile float and held the contractor to the super elevation profile. They pulverized the roadway, we checked slopes after they graded it.
The pay units were set up so that everything was easily measured and accounted for (with the exception of surplus spoils - they got us on that one).
Turned out good.
Get your level out and loop the job. You'll sleep better at night.....
What kind of job is it? What's your DOT requirements? 5 miles of two lane asphalt? Or 5 miles of 8 lanes of concrete with four asphalt shoulders. I know the DOT here has tight tollerances. 0.02 +/- for concrete. And yes, the contractor CAN build it that close. But he needs good lay out" to achieve that.
As always - It Depends. How much vertical relief is there? Is there curb and gutter? Are you matching into a lot of crossing streets or driveways? Are there Storm Drain manholes? Do you know how accurate your geoid is in this area?
You should almost always stick with the control that is the basis of design - just like FrozenNorth and mvanhank222 advised.
However - If it's flat and drainage is critical then it's time to consider the good advice that T-Ray and mvanhank222 are giving. If there are drainage problems after it's built and you are trying to explain how geoids work to the contractor, hearing administrator, attorney then you are in a bad spot. Run closed loop differential levels (preferably digital) in these situations. Design based on geoids in drainage critical low-grade areas are often problematic and will require you to assess the quality of the geoid early in the process. The checks mvanhank222 advised are in order and depending on the results - the sit down chat he described may be necessary - earlier is better.
If there is a lot of vertical relief and there are no low-grade areas of drainage - run with the geoid and VRS. Don't spend your (or your client's) money trying to resolve issues that won't affect the function of the roadway or drainage structures. The error in the geoid will be low frequency (so it won't show up in short distances) and if you have the relief, they typically won't negatively impact the final product (but you've got to consider what the impact of differences between geoid based elevations and actual elevations could be).
Is the .015 to .05 your seeing vs plan control or your own VRS shots. if that is all you are seeing keep leveling. .05' differences in a VRS system vs leveling is not at all uncommon in my experience. Also if you are going to share the level data done with a total station I would put it in a bs,hi,fs level book format. I find just giving elevations with out math insettles some engineers and contractors.
MightyMoe, post: 333439, member: 700 wrote: I'm not surprised you are seeing 0.05', that can be the "error" in the model, they aren't perfect. I would run a bench loop, we always do, it doesn't take long, and it's worth it as far as I'm concerned.
A little confused by how you are applying Geoid12a, are you using the Robot to do that?
The geoid12a is with my rtk setup. Base and rover. VRS can be a little flakey in our area 🙂
So here is what I did....
RTK Base and Rover, Automonus Position, Geoid12a
Located all site control, moved to a site control point, adjust to a bm,
checked how the rest fit.
Saw some differences in elevations in the middle of the job
so we decided to run a bench with the robot.
Going to experiment with a different geoid
If that don't work, I'll localize the the robot benches.
Site is pretty flat.
Contractor will be using a topcon stringless curb machine, Yikes.
It's using GPS for horiz and a lazer for elev.
Swim4Life, post: 333514, member: 6035 wrote: So here is what I did....
RTK Base and Rover, Automonus Position, Geoid12a
Located all site control, moved to a site control point, adjust to a bm,
checked how the rest fit.
Saw some differences in elevations in the middle of the job
so we decided to run a bench with the robot.Going to experiment with a different geoid
If that don't work, I'll localize the the robot benches.
Site is pretty flat.Contractor will be using a topcon stringless curb machine, Yikes.
It's using GPS for horiz and a lazer for elev.
Sounds like the control is set for you. Work between what they gave you and report anything that is out of the specs. My 2 cents, Jp
Got it!!!
Ran bench with robot through my control points.
Horiz over the entire site difference between grid and ground .04' so i left it on grid.
Edit the points elevation and localize to the new points.
Everything matches. Yeah!!
Swim4Life, post: 334709, member: 6035 wrote: Got it!!!
Ran bench with robot through my control points.
Horiz over the entire site difference between grid and ground .04' so i left it on grid.
Edit the points elevation and localize to the new points.
Everything matches. Yeah!!
I'm assuming by localize you mean setting on the elevated point and just using it without changing projections (some SPC grid).
It's best to just let the Geoid from the elevated point slope in your rover elevations, and I usually don't travel very far from a control point/base with the rover.
The good thing about highway control is that are lots of them to check to,,,,,,,,,,,,,
Check in, check out, check often,,,,,,,,,,,,,,