Help with making the best GPS Localization / Site Calibration possible.

You can only do a single point localization if you're in a standard projection and you know what that projection is. If it's something else you'll need to bracket the area your working in and look at how well things fit horizontally and vertically in which case you'll end up holding those that have the lowest residuals for rotation and scaling in a multiple point calibration. What bothers me is that apparently nobody knows the basic answer and that's a recipe for disaster. The last thing you want to do is include a control point in your calibration that's off because it will really mess things up. Ideally you want to bracket your work area and don't go outside of it. If after calibrating your missing other control points beyond tolerance, something is hosed up somewhere. Check the calibration report and look for one of the points with the highest residuals and remove it and resolve. Hope that helps some. Good luck.

First off, you can't localize without good control. What makes you think the control you were trying to use was any good? Usually when you localize to several points, before you accept the data, it give indictators to the quality of each point in the vertical and horizontal.

We once worked on a large job, i'm talking 3 or 43 miles long by a mile wide that non one had run a level circuit to or really any good horizontal control run. They suggested we calibrate to this poor control, everyone else did.

Williwaw, post: 445095, member: 7066 wrote: apparently nobody knows the basic answer and that's a recipe for disaster.

I don't know how you priced this project, but I *Suspect* that ALL the original control was run with RTK GPS, and that it's common for Tribrachs, and Prism poles to be out of adjust.
So, BEFORE I'd work on a project like this, I'd ASSUME several days of Localization, and then STUDY of this localization, to KNOW WHAT'S REAL.
Also, I'd ASK for the field book, where the levels were run.
Is this an understatement? I don't think so.
In our modern times, many Construction Projects are done, with the Sup, Assuming, and all others believing. And, you are the ODD one out.
Yup.
You should have a clause in the contract that says you would like to have a copy of the Data Collection File, (you said that you saw a GPS base on it the other day) . And, I'd bet a burger and fries, that they are all Irreverent Catholics, who pray to "Our Lady of ASSUMPTION". Are we ready to rock?
Fact is, many folks BELIEVE their Data Collector. Without knowing.

Case in point, I asked another surveyor, what his Basis of Bearings was. He said he had tied into another survey, and was using that.

So, I asked for the control he had localized on. We went, and tied those with my Javad.

His first Job, was done with a "here", and "True north, Ground scale on first Job.

Second Job was a couple miles west.

And, it was some 3 minutes off of "True" or Geodetic.

I gave him a complete rundown on what he actually had. I got a blank stare. I later gave him coords, in "His system". With Meta Data parameters. I double that he really understands it. Though, he vaguely understands what I was getting at.

Now, that was a Land Surveyor. He has been using GPS ALL of his career. I know his Mentor.

I only went to 6th Grade, and 10th grade. Dropped out and went to work, with 1923 Adolph Lietz transit, and 200' tape.

So, you are going to NEED to get up to speed on this. It's not too hard, 2 days, in the office, learning the theory, and then, a day or 2 of field work.
To develop CONFIDENCE in what you know, will take longer. But, it's some of the BEST investment in knowledge you will ever make. Find a surveyor in your 'hood, that is up to speed, and hire him to show you.

Nate

First, I would put up a base pole at a convenient location, set your base on it, then calibrate to the existing control and see what you got. It's possible the one you set up on is off and the others are not. At least then you can see if any of them fit together.

Step one: Verify the control. Never assume that anyone on the site knows the facts. Never use one control point unless the Client has released you from all liability.
Step Two: Report control discrepancies in writing to your Client. Don't just tell them that the control is bad. Do the points not fit relative to each other or some other Datum? Give your Client a solution to the problem, even if they do not want to hear it. Relative position is often more important in construction than Datum. Differences in elevation than the National Vertical Datum.
Step Three: If the client is not interested in the issue, do not proceed without a written liability wavier that clearly outlines your findings. Most problems can be avoided if the time and energy to solve the issue is invested before and not afterward.

When the steel don't fit, everyone will not suddenly become interested in the control. If you staked the piling wrong, The old "but the Super told me this was fine" will not fly. They will be looking at you. I always assume that all the control is wrong until proven otherwise. I have royally p_____ off Clients with this approach. I have also had Clients who recognized my value and did not select me based on price. Take your time and sleep good at night.

Site calibrations can do more harm than good... it really depends on how many control points you have available and what their geometry is with respect to the site. Verifying the control by other means might be a better first step.

Well, and first I believe the person that told you to figure it out doesn't know what is going on (wrong person to ask) or doesn't want to help you. If you are in a DOT or county job you should get the full set of plans, roadway, and structures plans. Verify what system is being used for example state plane coordinates and what elevation reference for example NGVD 29. If you are working at a bridge with piles you may find that the bridge structure is an independent structure with a unique system of itself. You may need to locate the existing structure and key features like forward face of backwall, pier caps, gutterline, top of beam seats, the centerline of beams and cut line to mention some of the bridge elements. If you compare your data with the plans and you find distances, alignments don't match you should call the engineer of record or the designer and inform your finding. I have done in the past and they usually react with appreciation for your time to share the information. Another thing please request control points and benchmarks to the contractor or inspection team as they should have this information available. If you have to set control points request to have one or two weeks to set control ahead of time. If you want to feel confident use a total station to set control and a level to set elevations, then bring in your GNSS unit to locate and verify your control.

Good luck.