Verifying Control with GPS
Our company was recently asked to verify control points and write up a surveyors report stating their vertical accuracies for a new bridge project. It’s a smaller project < 2 Miles long and has 28 existing control points the engineering company listed on the construction plans with station offsets and elevations and the plans also had centerline coordinates in state plane.I was able to create a centerline file with the coordinates given and calculate the points based off their station offsets. The crew is now needing to find, flag, and verify all the control points to the specified accuracies of 0.1′ Hor and 0.1′ Vert.Since this is a long bridge job and we are dealing with islands, we figure gps would work best for localizing and locating all the control points.We do not know what the original survey company used to establish their state plane coordinates, so we are thinking of setting up our base, localizing on 3 of 4 of their established control points, then locating the rest of the control points and also locating known NGS monuments in the area for a horizontal and vertical check. We are then going to break down the gps, set a different base are re-localize on the same control points and re-locate all the control points as a redundant check.The crew was wanting to localize on as many of the points as they could, rather than just a few. What is the difference and or affects of doing that? Has anyone came across a job like this before? What are the methods you used and how did you ensure that the points are accurate?Any feedback would be greatly appreciated! :{:
- 28 Replies
Well, I wouldn’t localize without first verifying the points.
I like your idea of using known NGS points to establish the coordinate base and then hitting the points to verify coordinates.
Then you should have an ‘apples/apples’ comparison..Yes, I have advice. Most State licensing laws say don’t accept jobs you are not qualified for.
Why would you localize????
You mentioned state plane coordinates. That means NOT LOCALIZATION needed. If you localize, it will create best fit rotation, vertical axis and scale factor for the site. You will NOT be checking the work. You will be force fitting into the other surveyors control.
Just setup your base on one of the points, then using proper projection in the DC, run static at base, then use RTK and rapid static on each point control point with the rover (also include the NGS monument). If you have the budget, I would move the base to a point on opposite end of project. Then locate each control again with same procedure as above. Then send static data for base point into OPUS. Bring all the data into post processing software, process the vectors, and loop closures for each point. Compare coordinates with OPUS solution vs. the NGS monument, then compare with the original survey. This method is much better accuracies than you spec’d.
But never localize when using SPC. I hate in when people localize for no reason. It just creates error, when it should not be used at all.
LeeGreen.com
I’d hate to be so harsh. But Richard is correct.
I’ve seen this many times. With a localizing you may falsely create error in a project where none exist.
Well…there’s a helpful response.
It sounds like you are using RTK for a STATIC job.
So, I LIKE using RTK to FIND the stations.
Then, use POST PRECESSED STATIC, with about 20 mins+ on EACH control station. It is situations like this, that 2 people, and 3 GPS units are handy. Cuts the time. I’d set ONE gps system stationary all day, and move the 2 rovers, getting static for 20 mins, and move ONE rover, and occupy for 20 mins, and leap frog. You will have 20 mins between rovers, and 40 mins to base. This will connect it all together.
THEN, in the safety and comfort of your office, you can actually VIEW the data quality, and you will know what you have.Just me musing. Pretty close to how I’d do it. After doing that, then RE-Observe the one ones with high residuals, or poor pdops.
Nwow that is harsh
i like the rapid static from two points, be sure your timing is different by about 4 hours if you can.
is there any basis for the SPCS from the previous surveyor? it would be a shame to be on the actual SPCS when they were translated by a foot. superior standards and computations will never match inferior work…> Yes, I have advice. Most State licensing laws say don’t accept jobs you are not qualified for.
Rough Crowd:-)
First never use control that has not been provided by and shown on the approved set of construction drawings. This is purely a liability issue protecting the contractor. If a second set of control is provided by the managing (owner), get this control to be an accepted part of the approved plan, protect you contractor.
Now to the question of is the control useable for project at hand. Ask the question is this the control values used to design the project? If so the control has been made local by this fact. If not part of the design run up the red flags. Verification regardless of method is not a one time event, must continue as long as the project does. Set out secondary well protected references always.
A comment on bridge structures; the road will fit the bridge both horizontal and vertical. After initial verification select the ‘hold’ control for the structure(s), if there are several segments connected by roadway each structure should have hold control and stick with them for individual structures.
Lastly understand the effect of the atmospheric on control and layout for the project, use better instrumentation than the auditor’s, you have been hired to protect the contractor.
bridger
I would suggest following licensing laws is helpful.
What, nobody ever used a robot before? Or even one of those old things in the back of a truck called a level? Two miles of levels the old fashioned way, or even the newfangled robotic way, is nothing but a few hours of well spent time. If GPS is the only tool in that box, I’d be looking for another box.
$0.02
BTW, I once got a big chunk bit out from calibrating along an alignment. I think the technical term is inclined plane. Use one point, or a minimum of 4. Certainly not 3.
Thanks for this excellent post.
Regardless of some less than helpful responses.
Beer Leg dot com Rocks!
I’m with the guys who say don’t localize. It just does a “stretch and fit” across the project and can hide problems with the control. That is the opposite of what you are being asked to do.
By the way 2 miles and 28 control points? That’s a lot of control for 2 miles.
You could use each one for a turning point when you run levels.Static, levels, fix to at least two existing control points and process the static to check the horizontal. The levels will check the vertical. If it was all done correctly to begin with you should be good to go by just applying a state plane projection and importing the control. To get GPS to talk well to the elevations may be the most difficult issue you face. I’ve been having very good results with levels and Geoid09 but mileage will vary.
If the control doesn’t check then go to step two and get with the client and go over what they want to do.
I do a lot of these projects and you can usually work through problems with the control.
I wouldn’t worry about how the control checks to OPUS or HARN points. I would look at that but it sounds like this control was used to design the bridge project and needs to be good internally irrespective of it’s actual accuracy to state plane positions.
I’ve read thru the comments and find that foggy has the best idea. It sounds as though this project may have been dormant for awhile and the contractor or client wants to make sure the control points (aka what the project will be staked from) are what are shown on the plans and haven’t changed (bumped trampled on and so forth). The KISS method is best. Yes GPS would be a good tool to do so. The idea, from what I gather is to make sure all 28 points are in relationship to one another. It matters not what the coordinate system is unless you are tying into points outside of the project at the beginning and then at the end. The vertical is most important and while your checking the control points I would also check, based on the final outcome of you verification, that where you start and where you end tie in nicely. As foggy said…check apples to apples.
isn’t this a bridge project? if it’s a new bridge you may find it difficult to level.
Yes, I didn’t get into that; but level what you can and if you can level across the span great. For bridge projects I’ve always used a total station to locate the control at each bridge end and really spend time there, multiple readings and measuring both directions. I would also trig level across the span if I couldn’t level it (both directions at the same time if possible). It is just another check. I would do it at night. T-2’s work great for that.
Static, Static, Static. do not use RTK and absolutly do not localize.
:good:
I’m w/Wayne on this one.
I personally never have never used GPS on a bridge project.
I’d be leaving it in the office.
A good calibrated instrument and level,a chain and plumb bob is all you really need,amongst the obvious other tools 🙂
That much Control in the area,why would you even contimplate GPS on the job?Sounds like the best advice I have seen so far Moe.
jud
Log in to reply.