Mark Mayer, post: 445146, member: 424 wrote: Commonly I get control from the topographic/boundary survey phase of a project, by others, to use for the construction layout. And just as commonly that control consists of nothing more than Mag Nails in the asphalt and sidewalk cracks reported only as points in a CAD drawing file. Naturally, the control is just where it will be the first thing destroyed by site demolition.
What I do is establish more stable, more permanent points (capped rebar or brass plugs epoxied into solid sidewalks or curbs) across the street for the construction site. I then traverse through these points and side tie the pre-existing control and boundary monuments. Sometimes I'll incorporate RTK ties as well if the site conditions are favorable for it. I adjust the lot in StarNet with the pre-existing control fixed or with a very small standard error (unless the control is found to be to loose for the purposes).
For layout control I resect my instrument positions to these newly established points.
I had 10 pts with opus only positions. And distances 350-700 ft apart. I triangulated the points returning a very rigid network. Then applied 5 mm SE to the opus positions, making best fit. And it worked out very well, the opus fit better than I'd expected.
Existing control, some years old, unknown methods, best fit (assuming enoughnpoints, and reasonable residuals) by least squares, is the right way - imo. It's a case by case by call.
4 pts 500 ft apart, not a hard call to make a network out it, and evaluate it.
Control, mine or anyone else's, that ain't been checked is not control. Verify it.
The dependability of someone else's control is no better than the dependability of someone else's used prophylactic.
That's my take on the issue.
I went on a recon mission today with my dad and only found that only one of the control points was spared by a fencing crew from somewhere around '99. It was not the 3" aluminum cap in concrete it was supposed to be but rather a disturbed ?«" iron rebar.
The building dimensions were not at all accurate. The design was not carefully followed, that's for sure.
Now I'll need to make a report to give the client to make a new game plan about how we'll establish new control. We'll probably end up calling one of the long sides of the compressor building North and then translating everything to the one rebar we found.
Sent from my SAMSUNG-SM-N920A using Tapatalk
I plan on occupying all of my control points and recording direct and reverse measurements to all visible points from each station. Does StarNet have any quirks about how the data is collected? I've never used it. Any tips for making a control network with it?
Sent from my SAMSUNG-SM-N920A using Tapatalk
arctan(x), post: 445277, member: 6795 wrote: Does StarNet have any quirks about how the data is collected?
The very best part of StarNet is that it is extremely un-quirky.
Awesome! So I'm using Magnet Field. When reobserving all of the control points, do I use the same point number and just store it as a check shot? Or add an alpha character on to the the point number? Does it care either way?
Sent from my SAMSUNG-SM-N920A using Tapatalk
arctan(x), post: 445280, member: 6795 wrote: Awesome! So I'm using Magnet Field. When reobserving all of the control points, do I use the same point number and just store it as a check shot? Or add an alpha character on to the the point number? Does it care either way?
Sent from my SAMSUNG-SM-N920A using Tapatalk
Can't speak to Magnet Field, but if a "check shot" is the equivalent of SurvCE's "redundant observation", that's what you want. In Starnet, if you don't enter the same point number, they wont be adjusted as if they ARE the same point (which is what you want).
Every time I've had to go back and find control from years ago on a construction project, there were buildings in the way and we could not see a control directly from another to verify if it was the point or a reference point.
Then the facial addition around the IBeams was no help in being able to confirm or act as control.
In our quests, there was temporary wall in place we could dismantle to get to the IBeams and establish control for the new additions.
On one site we had to make a window thru metal walls to obtain alignment from two coal powered generators to be able to set control for alignment of a third unit.
Building owners and corporate heads and site managers cringe at this kind of news.
arctan(x), post: 444831, member: 6795 wrote: I'm due to survey a site next week that has existing control. The control is about 30 years old. There are 4 points that still exist on site. What would be the preferred methodology for checking that the control is still good?
My plan right now is to set up on one and backsight one, turn to the next one and traverse to it until I work my way around the site and close back in on the first occupied point. Recording direct and reverse measurements on each.I'm sure there are a few ways to skin this [emoji250]
Sent from my SAMSUNG-SM-N920A using Tapatalk
I would absolutely do what you're thinking then cut the sucker in half after you close the first loop.
arctan(x), post: 445280, member: 6795 wrote: When reobserving all of the control points, do I use the same point number
Yes. Use the same number. Unless you want an observation to just be a check shot.
Another way to think of it - If instead of numbering your stations you gave them names, such as FRED, BARNEY, WILMA, BETTY (You could do this in StarNet and in the data collectors I'm familiar with, you can't do it in CAD). When you come around to tie station FRED a second time are you going to call it PEBBLES? No. It's name is FRED. Call it FRED. These numbers are station names. Not point numbers.
All that said the latest release of StarNet includes a proximity detector. StarNet will figure out when you have retied a point with a different name and deal with that somehow. Since I don't have that very latest release I can't advise you just exactly how that works.