Hi guys I recently completed a steel column survey. The client is designing another floor to be built on top of some existing columns so needed the exact locations of all of the heads of the columns.?ÿ
I was told that the co-ordinates had to be with 6mm (1/4 inch) of their true location. Levels weren't needed just x and y co-ordinates.
I applied reflective tape to the steel, shot face left and right, shoot from as many stations as possible (3 where possible but as least two), and shot as perpendicular to the faces of the steel as possible (within 25 degrees of perpendicular).
The survey itself was a breeze. But there was?ÿ a lot of prep work. Because I want to shoot the steel from many stations, as as perpendicular to the targets as possible the steel flanges (flat sides) were not always visible so I put timber blocks inside the webbing flush to the edges - I applied to targets to the timber (see photo). This gave me another face of the steel that I could shoot. I also marked vertical cross hairs on the steel 15mm in from each corner of the steel. This meant that on CAD I could locate the corner of of the steel ( as useful check ).
I wondered if there was a quicker way, that was just as accurate. How would you have done it?
By the way the results seemed great. All obs were within 3mm (1/8 inch) of each other.
Cheers, Andrew
?ÿ
?ÿ
I wondered if there was a quicker way, that was just as accurate. How would you have done it?
You could have saved a little time using pre-printed retros - vis
<IMG></IMG>
They are also sticky enough and stiff enough to go direct onto the flange so that you would not need the wooden blocks.
But your results speak for themselves - Well done, that man.
?ÿ
Good job on the accuracy, that takes time, care and experience and willingness to just figure it out sometimes.
The only thing you may have considered would be renting a scanner of some sort and comparing the cross-sectional?ÿcut of the point clouds and overlaid them onto your points. It adds?ÿpeace of mind but also a good visual for the client.
I only see one picture. This client or the site may not support such a service or the need. Other than that you are on point.
It looks like he used some consistent offset from the edge so I think it was probably easier to draw a line than try to match up a pre printed line.
using pre made targets would have been quick to put up but drawing my own cross hairs mean that I could locate the corner from my shot. All cross hairs were drawn 15mm in. This mean that I could shoot one side of the steel from one station and then shoot an adjacent side from another station. In the office if the corners of the steel were in the same place on CAD I could be confident that all was good.
I think it would have been difficult to line up cross hairs of a target to an exact offset on the steel.
I could have made use of an assistant to help me on the day to hold targets up to the steel and maybe have done horizontal angle offsets but I don't know how much I'd have trusted them to do things accurately.
?ÿ
Having a target line offset from the corner is not needed if your data collector has an offset routine, shoot your laser distance and turn to the corner. Typically I will take several distance shots as I turn into the corner, when I see no observable distance change I record the distance and turn to the corner. That way I can quickly compare shots from two different axis. For far side column shots use a rectangular reflective sheet and fold over a protion so that the sticky side can go on the column and the fold over can give you the face.
Paul in PA
That's a good point Paul - thanks.
I curious how wide a target you need to get a reliable reflection. The thickness of the steel flanges were 11mm (0.4 inch). I was shooting from 50m (150 ft) away.?ÿ Maybe I could just have put 10mm (3/8 inch) pieces of tape on the flanges, instead of fitting timber blocks to carry a 25mm wide target.
?ÿ
?ÿ
?ÿ
I would have been comfortable with using reflectorless. I would have marked the spots with chalk as an aid to aiming.
According to Korec my Trimble S7 has a "general purpose" laser, not a HD laser. That's why I used targets - they advised it would be more accurate and precise.?ÿ I've had some imprecise results in the past shooting reflectorless.?ÿ These results were very precise.
It's seem your approach gave good results - nicely done!
Regarding faster, I have a few questions...
How much time did it take for setting up the targets,?ÿ boards, and vertical lines?
Time to set your control/baselines?
Time for measurements?
Time for analysis to produce results for reporting.
?ÿ
Then compare to scanning and include time:
Go to rental shop and get the scanner.
Time learn what buttons to push.
Set control targets if required.
Download data... etc... compute results you're comfortable with...
Return to rental shop...
I suspect for a one time task that your approach would be less total time.
?ÿ
?ÿ
Strange no one speaks about the alternative with the mini prism, less time in prearing and great accuracy! Shoot 2 corners per column and using F2F coding 90% of CAD is done in the field.
i gues if more then 100 columns renting a scanner may be faster.
Chr.
?ÿ
?ÿ
?ÿ
?ÿ
Christ - I though about using a mini prism and I would have preferred to use one. I assume your plan would be to do angled offsets?
I just don't trust a chain guy to get it right for the level of accuracy required for this job. They'd have to line up the plumb point of the prism holder against the corner of the steel and then hold it perpendicular to the instrument.
Heck I don't know if I trust myself to get it right! I think there's less to go wrong using targets for this sort of work.
I would guess 11mm at 50M would work?ÿ
I assume your plan would be to do angled offsets?
If it were me I think I'd make a jig to slip-fit the column face (I'm assuming that they're all the same dimension).?ÿ I'd install a quick-release holder for a Leica miniprism on the jig, aligned with the column face and at a known offset.?ÿ I'd probably code the shots to correspond to the column corner (e.g., NE, SE, SW or NW) to avoid confusion.?ÿ Then I'd shoot the prism direct (no offset shots) and correct the offsets to the column faces in CAD.
Jim -?ÿ A jig is a good idea but all the steels were slightly different sizes . My steel tape measured their sides ranging from 202mm - 210mm.
But you could make a jig that you pushed tight to one corner and them flipped it over to measure the other corner.
Have you made such a jig before? A photo or sketch would be great
?ÿ
?ÿ
