The large company I work for bought out a smaller pure survey firm a couple years ago and inherited some of their projects.?ÿ I've gone out on a few of these projects lately and run into various problems when I try to get going on the fieldwork.?ÿ For example:
Headache #1:?ÿ Curb staking in a new subdivision.?ÿ The small company had done everything on this project with GPS (calibrated to project control) but since I was grading hubs I thought it was best to use a robot.?ÿ I pulled the latest .jxl file off the server and headed out the door thinking I would be able to create a job with it and get working.?ÿ I set up the instrument and started checking various control-- everything is off horizontally 1-3 feet.
I am new-ish to Trimble gear so I assumed I was doing something wrong or had a bad setting in the equipment, but at the same time I was thinking with a robot and ground coordinates-- this should be as simple as it gets??ÿ After my own troubleshooting failed I called the equipment guru in the company.?ÿ He said the small company guy (no longer an employee) who set up the project had a habit of choosing state plane as the coordinate system and then calibrating on top of it.?ÿ I've always known this is bad but I've never known the finer details.?ÿ Anyway, I eventually created a new job at scale factor 1, imported points, checked in to existing control and got working.
?ÿ
Headache #2:?ÿ Topo.?ÿ Another project started by the former employee and another site calibration-- (I would come to find out later) on 2 points. ???????? The task was to shoot powerline sag so I set a couple pins, shot them with the GPS, set up the robot and when I backsighted between them the horizontal was good but there's a 3 foot vertical bust.?ÿ What the hell??ÿ There was no cell service in this area, so no outside help.?ÿ Once again I made a new project at scale factor 1, traversed across the site to shoot the powerline and then came back through and tied the traverse points with the GPS at the end of the day.?ÿ When we got the data into TBC the next day we were able to merge the points and see the calibration.?ÿ Horizontally everything jives but the vertical is obviously in question now...
?ÿ
Headache #3:?ÿ Setting pins in a new subdivision (different than headache #1).?ÿ Another site calibration by the former employee.?ÿ I went to the site with GPS this time and... I couldn't even start the base because the coordinates of my points were out of tolerance with the autonomous position the receiver was seeing.?ÿ I ended up doing a whole new site calibration because we had no idea what the former guy had done (field notes were not helpful).
?ÿ
Basically, the pattern here seems to be calibrating on top of state plane.?ÿ Like I say, I know this is bad, but would someone be willing to give a short explanation of the chaos goes on when this is done?
Also, if someone knows Trimble gear-- the .jxl file should contain calibration information, correct??ÿ If I have a calibrated job in data collector #1 and I want to make a copy of it on data collector #2 I should be able to export a .jxl from DC #1 and use it to create a job on DC #2 and everything should work the same, yes?
The reason I ask is because I have tried to do this on all 3 of these projects and it has not worked yet, so either the .jxl files don't work this way, or I'm doing something else wrong, or these frankenstein calibrations on top of SPC are breaking things.?ÿ
Any insight would be greatly appreciated.
Also, if someone knows Trimble gear-- the .jxl file should contain calibration information, correct??ÿ If I have a calibrated job in data collector #1 and I want to make a copy of it on data collector #2 I should be able to export a .jxl from DC #1 and use it to create a job on DC #2 and everything should work the same, yes?
Not sure about JXL as we dont use them.
But if you are going to copy gps points between two Access jobs, you do need to recreate any site calibration in the second job.
Also I find that the 'copy calibration and control' option does not work as expected - seems to expect points saved as 'calibration' and ignores 'observed control' even if they are used in a calibration, which is the way we usually work.
JXL files are just plain ascii text - you could open one with notepad or wordpad and have a look see. I'd be happy to take a look if you want to email one
I used to do calibrations with trimble gear but we were running scs900 software which is a whole other beast than access. It had a .dc file with the calibration parameters in it.
Not sure how access handles it, if it's in the jxl or not, since I stopped calibrating before I started using access.
I will ponder on this some more and if I come up with something I will share.
Also, if someone knows Trimble gear-- the .jxl file should contain calibration information, correct??ÿ If I have a calibrated job in data collector #1 and I want to make a copy of it on data collector #2 I should be able to export a .jxl from DC #1 and use it to create a job on DC #2 and everything should work the same, yes?
Not sure about JXL as we dont use them
But if you are going to copy gps points between two Access jobs, you do need to recreate any site calibration in the second job.
Right, so another reason I ask is because if you go into a job in Access, on the bottom of the screen is a Copy button.?ÿ If you hit that the next window has 3 tabs at the top of the screen, and 1 of them is "Copy from job to job".?ÿ If you click that you have the option of copying the calibration from one job to another.
This is part of my confusion-- if the option to copy the calibration is here then does that mean it's not part of the .jxl?
@bstrand?ÿ
I added a couple of paragraphs to my original answer - there ARE issues with "copy calibration and control"
Trimble documentation also talks about a .cal file specifically for calibrations - I suspect that this carries the seven translation parameters rather than the obs and point data.
Most reliable solution I have found is to copy the whole job file complete.
?ÿ
@jimcox?ÿ
Yeah, a coworker also mentioned that if I were to copy a calibration from inside Access then to choose the option that says calibration only and then link the job with the points separately, so maybe there is some bugginess with the "calibration and control".
I have not seen one of these .cal files yet.?ÿ I've checked the export menu multiples times thinking there should be a way to do this, but nothing there jumps out at me.?ÿ I'll take another look tomorrow and see if there's an option to spit one out.
I think yes and no on copying the whole job.?ÿ Sometimes it's nice to have a job that isn't clogged with unnecessary data from previous days.?ÿ The ability to link jobs is really, really nice imo since you get all of the data but can still export files of just that days work.
Sounds like you're having a hell of a week. Thank god that guy is no longer with the company.
As far as exporting, you can copy the entire system file under Programdata/ Trimble/System/ and just use a thumb drive.
I'll take a look tomorrow, I accidentally created a calibration (we use an LDP) and I'll see if I can walk through what you're doing/experiencing.
Are all the DC on the same current version of ACCESS?
?ÿTHAT could make a difference.
Jesus I need a drink. Mmmm Toki by Suntory.?ÿ Mmmm.
?ÿ
?ÿ
I hate Trimble, but haven't had trouble importing jxl files into Carlson.
That said, why bother with those??ÿ I'd assume they used CAD, C3D, Carlson, etc.?ÿ Why not just go to the dwg and export the control points into your equipment and take that into the field.?ÿ I'd take my laptop too, just in case.
Calibration??ÿ If you're hitting existing control, and you're getting bad checks, then I'd immediately let the uppity ups know about it, and begin the un-F-ing process, if it's even possible at this point.
OK, everyone is making this out to be more complicated than it is.
Basically, the pattern here seems to be calibrating on top of state plane.?ÿ Like I say, I know this is bad, but would someone be willing to give a short explanation of the chaos goes on when this is done?
You can indeed calibrate "on top" of state plane. But you had better be calibrating to values that are close to state plane, or it's going to get all sorts of screwed up.
For instance, if there was a conventional traverse/network done using a state plane value for one or both of the beginning baseline points, it would make sense to start with a state plane system. But if the local ground system is assumed (like 5000/5000 or something), you're going to see some major discrepancies because the calibration is going to warp the hell out of the state plane projected coordinates.
Also, if someone knows Trimble gear-- the .jxl file should contain calibration information, correct??ÿ If I have a calibrated job in data collector #1 and I want to make a copy of it on data collector #2 I should be able to export a .jxl from DC #1 and use it to create a job on DC #2 and everything should work the same, yes?
Yes, JXLs store coordinate system information.
Best practices would be to generate a JXL from TBC in the office with control-class points embedded, and send it to the project as a template through the cloud, or just upload it to the controller and then use the "create template from job" command. Then anyone may begin a job from that template, with the calibration and the control already there, and link additional files as needed.
But yes, you can manually import a JXL from one controller to another.
The reason I ask is because I have tried to do this on all 3 of these projects and it has not worked yet, so either the .jxl files don't work this way, or I'm doing something else wrong, or these frankenstein calibrations on top of SPC are breaking things.?ÿ
You're likely not doing something wrong, it's more likely they royally screwed up the calibrations before you ever came along. If they supposedly already have a good calibration, why aren't they sticking with that and setting up a template in the office in TBC? Far easier.
The software is fine, the hardware is fine, they hardly ever screw things up without human intervention. Knowing how it operates is a prerequisite for performing quality work. This is a good place to start:
https://help.trimblegeospatial.com/TrimbleAccess/latest/en/Site-calibration.htm
Also, QC/QA is a prerequisite for quality survey work, so I am more than a little surprised that the firm did not know about severe coordinate system problems across three (!) projects.
Paraphrasing Will Rogers, I only know what I read online, but this may help:
RTX-Site-calibration-Final.pdf (neigps.com)
?ÿ
I'd start a new job and check the project control, the property corners and as built any infrasture constructed.?ÿ Use software and equipment you know.
Edit: Verification should have already been done in anticipation of a request for construction staking.
?ÿ
?ÿ
That's pretty much what I'd do.?ÿ But if all the design's been done from wonky data, there'll be some major issues
Just had a wee play with this.
Yes - the jobxml file caries both the coordinate system and calibration
Yes - when I create a new job by importing the JXL file, the calibration is carried across.
This can be seen in both TBC and on the TSC3
And
Can confirm that 'copy between jobs' does not 'copy calibration and control' if the calibration is done using 'observed control points' - no points copied
@mathteacher Notice the screenshots in that pdf are from a TSC3.?ÿ If you are using newer, up to date gear along with TBC, you no longer have to go through that process.?ÿ Access and TBC now work together to automatically transform the ITRF coordinates over to your datum/SPCS of choice.?ÿ?ÿ
If the original calibration was done using Access and those .job files were stored on the server, it might be a simple enough matter to load them back up into Access and go to the ??Review Job?? function to see exactly what was done. You mentioned the field notes weren??t much help but the details of what was done should be preserved in the original .job file. Might provide some insights. You can only hope that if procedural mistakes were made, at least they were consistent about doing it.
Just because I'm paranoid, doesn't mean they aren't out to get me.
Calibration??ÿ If you're hitting existing control, and you're getting bad checks, then I'd immediately let the uppity ups know about it, and begin the un-F-ing process, if it's even possible at this point.
Yeah, it's existing control.?ÿ And yeah, this appears to be the case-- a combination of poor calibrations and calibrating on assumed coordinates + state plane.?ÿ These are all legacy projects that are short and/or winding down so I think the goal will be to ride them out while keeping the damage to a minimum.
Run static GPS. Put it on SPC,?ÿ Never look back.
I see too many surveyors calibrate to SPC when it never br done. I follow many surveys who did for for site work and screwed the datum. Got one now on Grand Island that off by 9.2ft. They even moved the Georeferenced Imagery. Some of this error stems from Acad 2017 and beyond finally recognizing there is a difference between Feet (International) and US Survey Feet. But many CAD monkeys still do not know this.
When done correctly we should be able tie into every project with VRS, and double check with OPUS. Especially true when Engineers are adding GIS layers like wetlands, soils and lidar.
@jimcox?ÿ
See, it makes sense that it should work this way.?ÿ I'm really curious to try it on a brand new project so I can see it working the way it's supposed to.
Can confirm that 'copy between jobs' does not 'copy calibration and control' if the calibration is done using 'observed control points' - no points copied
Is there some reason you're shooting your calibration points as observed control points instead of calibration points?