Hey guys,
I'm working for a transportation construction company that does their "in-house' site calibrations using Trimble Business Center. For example, if they have a control point list from the state in Excel format they just insert these points on the software and then they get the necessary Lat/Long for each of them until they tie State Plane coordinates to Lat/Long coordinates, without even going to the field.
When we go out to the field to stakeout these points we said calibration they are within 0.05 ft in horizontal but in vertical they might be off by 3 inches or 0.25 ft.
Is this method of site calibration better than the traditional site calibration method of going to each point in the field and taking static measurements? If not, how can the in-house method be done better? (Assuming the vertical information we have is "correct".)
Somebody will give you a more technical answer, but a site calibration is meant to get you a best fit from actual observations on the points compared to what the plans say. Not sure what you accomplishing with your method. Assuming the vertical information is correct is a bad assumption.
Surv3251, post: 451287, member: 12451 wrote: Is this method of site calibration better than the traditional site calibration method of going to each point in the field and taking static measurements? If not, how can the in-house method be done better? (Assuming the vertical information we have is "correct".)
No. And... If you mean "in the office" when you say "in-house" e.g. no field component - It can't be made better.
If I'm understanding this correctly - it sounds like a good way to end up with a variety of problems - like a drainage feature that doesn't drain. This could work out OK 9-1/2 times out of 10.
This is a business decision so if the company understands the risk and still wants to do it - good luck to them - save a few bucks but risk a layout that doesn't match the plan location (Hz and/or Vert).
Option 1 - input the project coordinate system from defined parameters - put your base on control points identified in the planset (or use a VRS) and CHECK into other control listed on the planset.
Option 2 - calibration/localization using control points and control values identified in the planset - control should bracket the work area - Check your residuals.
I'm glad to see that you are doing checks to validate the input data - maybe being off 0.25' vertically is within the error budget for the kind of work being done.
P.S. Calibrations/localizations can make you think things will drain when they won't - it all depends on the original control. If the planset control has vertical problems, so will you and you might not realize it until the concrete is curing. If there are vertically critical elements, break out the level (or 1" TS) to verify the vertical. Generally: The flatter the site - the more critical the vertical.
Surv3251, post: 451287, member: 12451 wrote: Hey guys,
I'm working for a transportation construction company that does their "in-house' site calibrations using Trimble Business Center. For example, if they have a control point list from the state in Excel format they just insert these points on the software and then they get the necessary Lat/Long for each of them until they tie State Plane coordinates to Lat/Long coordinates, without even going to the field.
When we go out to the field to stakeout these points we said calibration they are within 0.05 ft in horizontal but in vertical they might be off by 3 inches or 0.25 ft.
Is this method of site calibration better than the traditional site calibration method of going to each point in the field and taking static measurements? If not, how can the in-house method be done better? (Assuming the vertical information we have is "correct".)
Why are you doing site calibration on a state coordinate project?
When projection is on state plane coordinate, you use a predefined state plane projection. Why do so many Trimble users not know how to use a projection? Seems odd.
What geoid was used to provide the elevations. Were they leveled in?
If their control points are correct in relation to each other, then the only way they could be off when you shoot them would be if you were using a vrs or rtn system. If you had a base on one of the control points, then the others should check.
It's important to get the terminology correct. From what you describe the office guys are not calibrating. As Lee says why would you calibrate to state plane GPS control.
What they are doing is inserting the given coordinates into a state plane file set up as a projection in TBC. And yes it is the correct way to do it. You were given a gift of existing GPS control and it's up to the surveyor to know how to use it.
How can you be "off" .25' vertically to the control that's the CONTROL.
The proper thing to do is to set up on it, accept that it's correct until you check it and find that for some reason it has an issue.
For that check DO NOT use some external control system!!!!!!!!!
If you aren't geared up enough to use a base rover system, then your VRS or whatever needs to adjust to the control, TBC has routines to do just that without a calibration. Do not do a calibration.
There are two choices, either the control is internally consistent or it isn't. If it is it's good to go, you have to use it. If you set up on it and for some reason there are issues with it, then you need to figure out a solution for that situation.
I've found all the GPS control I've ever been given by the DOT to work just fine (some disturbed points, some removed and one odd situation where two different crossing surveys didn't quite match), no it won't "fit" to OPUS, no the elevations will be "off" to some external solution from CORS. That is to be expected.
As stated by others, the site calibration or localization can be done in the field or office. What does matter is that the initial data is good. I'm going to assume that the plans are in a local system and that some of the points have geodetic values (or SPC that can be converted to geodetic). In the office I would do compare is the relationship of the vertical deltas between points in the local system and the in the geodetic system. They should be close in an area small enough to use a site calibration. If not then something is wrong with the initial data and it should be verified. I would also review the residuals reported in the calibration to see if they look good (I would hope someone would have already done that). The first thing I would do in the field is to check some of the geodetic control to see if you match it, using the original SPC. If you are consistently off in your checks, within the tolerances or RTK, then I would start looking at things like antenna types, antenna heights, geoid models and/or base control values. If the error you are describing is consistent then this could be the cause of you problems. If the difference is random then I would say you have a problem with the geodetic component. If the error you are describing is inconsistent then this could be the cause of you problems. You can do the same thing looking at the changes in height between the localized values.
I would never bid a construction job without including sufficient time to review and verify provided control.
just some FYI, I've been told by someone that sets up jobs for machine control, that they can't work in State Plane Coordinates, they have to do a calibration. I don't know if thats correct or not. If thats not true, I know for a fact they teach doing a calibraion and hammer it home.
David Livingstone, post: 451405, member: 431 wrote: just some FYI, I've been told by someone that sets up jobs for machine control, that they can't work in State Plane Coordinates, they have to do a calibration. I don't know if thats correct or not. If thats not true, I know for a fact they teach doing a calibraion and hammer it home.
I've been told the same thing.
I sat in a meeting where I tried to get them to use the correct projection with a new Geoid model, they said they can only calibrate.
In a very messy geoid area they tried to build a road using machine control 7-8 miles south beyond the south limit of the calibration, needless to say it didn't work.
David Livingstone, post: 451405, member: 431 wrote: just some FYI, I've been told by someone that sets up jobs for machine control, that they can't work in State Plane Coordinates, they have to do a calibration. I don't know if thats correct or not. If thats not true, I know for a fact they teach doing a calibraion and hammer it home.
I've heard that too.
I NEVER work around Machine Control (or construction of any sort), so I have no idea how "true" that may be.
However, it make absolutely no sense to me whatsoever. If true, then the software sucks, it untrue, then the training sucks.
Personally, I think that "calibrations/localizations" are spatial STDs, and should avoided. Granted, there IS a time and a place for everything (I guess), but I have yet to see a "good" calibration (although I'm [pretty] sure that they exist).
Loyal
Ha, calibrations are a virus, infecting the files of the surveying community.
Imagine how deadly they could be if they get loose.
David Livingstone, post: 451405, member: 431 wrote: just some FYI, I've been told by someone that sets up jobs for machine control, that they can't work in State Plane Coordinates, they have to do a calibration. I don't know if thats correct or not. If thats not true, I know for a fact they teach doing a calibraion and hammer it home.
I here that a lot from Trimble user's. It's not true. It is Just the only way they know how to setup up a job, as directed by their Trimble rep or dealer.
leegreen, post: 451426, member: 2332 wrote: I here that a lot from Trimble user's. It's not true. It is Just the only way they know how to setup up a job, as directed by their Trimble rep or dealer.
After working on a few jobs in which the contractor is 'trying to use machine control' that is also my understanding. The sales reps are seriously doing the contractors a severe disservice by requiring them to calibrate on every job. In my experience the contractor's GPS guy typically knows just enough to be dangerous and on occasion knows nothing at all except 'give me your model and site calibration information so we can get to work' scenario. I put a stop to that really quick on projects that are not setup or capable of handling their request
I seem to remember one point calibration translates, two points rotates and three points warps the whole system by some secret method that even Trimble people don't understand. I use two points calibration s quite frequently, but have had some weird results with more than two points.
How do you do calibrations without a GPS position on the points you're configuring to?
Go to a job site.
Setup State Plane projection and GEOID pertinent to your job location.
Setup your base on a known SPC, start logging static data and output RTK corrections.
Connect your Rover.
Stakeout to each control point from dot control report, with RTK
Occupy found control point. Measure for 180 epochs.
Do this for all points.
Manually compare each points, coordinate, elevation, inverse from others. Verify if there are any disturbed or damage control points. Do not occupy points that are not suitable for GPS, due to canopy or man made obstructions.
Send static data to OPUS for verification of SPC at base.
This is the fundamentals of surveying, a system of checks and balances. Not until like a boundary survey. Surveyor must measure each point with confidence and proven procedures. Then make an educated decision on each point to verify the location.
Yet most people want the majic box to make the decision for us. If you are using all points under all conditions, you using a recipe for disaster.
leegreen, post: 451435, member: 2332 wrote: Go to a job site.
Setup State Plane projection and GEOID pertinent to your job location.
Setup your base on a known SPC, start logging static data and output RTK corrections.
Connect your Rover.
Stakeout to each control point from dot control report, with RTK
Occupy found control point. Measure for 180 epochs.
Do this for all points.
Manually compare each points, coordinate, elevation, inverse from others. Verify if there are any disturbed or damage control points. Do not occupy points that are not suitable for GPS, due to canopy or man made obstructions.Send static data to OPUS for verification of SPC at base.
This is the fundamentals of surveying, a system of checks and balances. Not until like a boundary survey. Surveyor must measure each point with confidence and proven procedures. Then make an educated decision on each point to verify the location.
Yet most people want the majic box to make the decision for us. If you are using all points under all conditions, you using a recipe for disaster.
That's not a Trimble Calibration that I'm familiar with. I wouldn't say that was an Office Calibration routine.
Skeeter1996, post: 451438, member: 9224 wrote: That's not a Trimble Calibration that I'm familiar with. I wouldn't say that was an Office Calibration routine.
I don't mean to speak for Lee, but I would agree that he isn't explaining a calibration, he is just explaining how to survey. No calibration, you are working in the control using it's native format. It's very important to see the difference, particularly with respect to the vertical. Calibrations have always been a poor way to get on vertical control and new geoids are far more useful.
I would only have a very gentle disagreement with Lee. It really depends on the age of the control if OPUS offers much use. The existing control supercedes everything else.
Surv3251, post: 451287, member: 12451 wrote: Hey guys,
Is this method of site calibration better than the traditional site calibration method of going to each point in the field and taking static measurements?
This subject always ends up like herding cat.
Is it better? It depends on if the coordinates that are being used came from a good control survey to start with. If they did the project parameters should be known and neither an office or field calibration should not be needed. Handling vertical is a separate issue that an office or field calibration will just cloud. If the quality of coordinates are unknown a control re-survey is the place to start.
MightyMoe, post: 451439, member: 700 wrote: I don't mean to speak for Lee, but I would agree that he isn't explaining a calibration, he is just explaining how to survey. No calibration, you are working in the control using it's native format. It's very important to see the difference, particularly with respect to the vertical. Calibrations have always been a poor way to get on vertical control and new geoids are far more useful.
I would only have a very gentle disagreement with Lee. It really depends on the age of the control if OPUS offers much use. The existing control supercedes everything else.
Moe, I agree. Original control always supercedes. But I use the OPUS as additional verification for future ties to SPC datum.
You are correct, that was not a procedure for calibration, only to verify.
I prefer to use 5 to 7 static receivors when verification of NYSDOT control.
