When you translate points in Access in the middle of job, say just a rotate, does your next shot get the translation applied? I've assumed No but have been afraid to try it as I can tell that undoing a rotate of more than just a few points may be very difficult. Would this be devastating on a GNSS project?
If it's a no go, is there a best practice for accomplishing this in the field apart from a calibration, which seems to bring in it's own issues (like scaling the entire project)? What might be nice is if I could temporarily translate the underlying CAD...
YGATB (you guys are the best) 🙂
Personally, I would not rotate anything mid job. There are many who won't agree with me, but I would not rotate anything in the field.
Like Chris I am not a fan of field calcs - they are risky and difficult to check.
But sometimes you gotta do what you gotta do
I save the newly calculated points separately with new names.
Dont overwrite or change the originals
Then you can delete and recalc if you need to
Field calcs have their place but rotations in the field and collecting data post rotation can lead to a major train wreck, I have seen it in the past. It's better not to have to perform a postmortem when the patient could have lived on their own.
"What are you trying to do?"
POLs and staking proposed site features (construction staking). Depending on level of acceptable tolerance, sometimes I'll put lathe or nails on line for setbacks, boundary or planned irregular features (ie curving walls or driveways) so the owner can visualize or measure on the ground themselves, Sometimes site conditions make added POLs very helpful. To do that accurately I shoot all or most found corners and would like to adjust the CAD (or project orientation) in the field to perpetuate relative locations as best possible even if just a slight translation.
I may be missing some procedure but it seems it can spare me a 2nd trip out and to give the client a better feel.
Personally I'd feel more comfortable tweaking the CAD, but I don't know that that's possible on the fly.
Another vote for "don't do it."
I've majorly screwed up data orientation at least twice by doing field "tweaks"...cause I didn't learn the first time.
We do it often. We never translate and rotate the points we have collected though. We leave those in their "raw" state. I think you are going out with pre-calced points. Establishing a coordinate system on site. Then finding you need to translate and rotate your pre-calced points to match your site coords. Yes you can do this and the points you choose for the translation will remain translated. They do not revert without you going through the whole process again. If you want to be cautious, save your pre-calced points somewhere else so that you can always access them again with their original calced coords.
If I were your supervisor my first question for you is how far off are your found corners that you feel a need to adjust the CAD points? If the found points are off enough to cause an adjustment you might be trying to fit a round peg in a square hole. Conversely, the found corners may just be off enough to indicate normal random field measurement errors. And as for the POL's keep it simple. Occupy one end, sight on the other end using field measured coordinates, set a point on line using each end of the line as control. It seems to me the found corners would need to be off quite a bit to cause concern for staking out most of the construction features mentioned.
Agreed. Keep the raw data raw. Screwing around with it will always cause a problem your confidence and ego will be beat down by eventually. And it will be painful. Don't ask me why I know this....😉
When I started in this business c.1990 we were expected to do calcs of this sort in the field on a daily basis, but I haven't done so -much- for over 20 years. It can be done, but you have to check, check , check. It gets easier with practice but never approaches foolproof.
As far as what happens to your orientation when you do such calcs - that is easily test-able. Let us know what you find out.
What others have stated. Translate or rotate in the middle of a project can get one in trouble. I always try and take my computed points and rotate those to my actual surveyed raw data. If and when I choose to do a rotation in field. This allows me to make a copy of the original computed data. And rename or renumber them in Xcell. I have spent hours I mean hours fixing raw data where numerous rotations and such were done through a job and everything got all messed up. I actually hardly ever do a translate rotate on simple boundaries where a deed or play is computed to gain confidence it closes and have search points. I usually simply comp a new search point by different methods bearing and distance or turned angle and distance from the difference in the bearings so internal angles etc. that is one of the set things about Trimble access anywhere you have numerical values in the program like rod heights bearing or azimuth input distance etc you also have a calculator function that can be used on the fly and then accept that value to populate the field for any cogo function. Now I am not opposed to translate and rotate but it is not my first go to solution.
I had a client once his crews did a here position for base. Then made a couple control points. Then randomly shot in with an assumed set of coords with total station in same job on state plane. Then rotated and translated both differently and as they traversed and did more here position s along the route with here positions and gps and more assumed stuff yeah they never linked all together with common points. Like they thought they did what a mess.
It's the 'undo' bit that will make you see the error of your ways. I do it regularly but only with grid points in specific ranges once I have the GNSS points I want to hold which I keep in different point ranges. One of the best ways to gob things up in the field if you aren't paying close attention to what your doing.
Personally I’d feel more comfortable tweaking the CAD, but I don’t know that that’s possible on the fly.
By cad you mean your calc points? If so then absolutely. Bring a .csv of the calcs with you so you can reload things if something goes to hell on ya.
What I would actually suggest is exporting your calc linework as a .dxf and then using the georeference routine in Access to transform it onto your boundary.
Unfortunately Access2017 does not support the Geo-Referencing.
Presumably he's using that version because its a TSC3 (2017 is the last/only version that runs on that controller).
To get the Geo-referencing you would have to upgrade the controller to a TSC5 or better.
