Hi,
I have been tasked with determining the deformation in a raked pile via total station survey. There are a few ways to do this, but I would like to see if there is an easier way than manually creating cross sections at certain elevations. Due to constraints on time and safety, it was easier to simply "scan" the effected pile area, than to try and collect neat offsets and elevations.
I have the asbuilt raked pile drawn as 3D object and have used a total station to record a grid of points along the 3D axis of the pile, in the area that is suspected to be damaged. The points have been imported as a TIN surface, and it fairly closely matches the asbuilt pile (which it should).
Due to constraints on time and safety, it was easier to simply "scan" the effected pile area, than to try and collect neat offsets and elevations.
The area is on the underside of the pile rake.
Methods I have tried:
-I have tried to draw all objects, then rotate the UCS so that the X axis is along the centre line of the pile, created an alignment along the centreline and then create section views. First problem was that the alignments kept disappearing off screen, to a random location. I think the problem is that they need to be flat ie. have no change in real Z value. Second problem was the section view would not sample the solid object (pile), so this wouldn't show the deformantion clearly.
-I have tried the above, but using the section tool from 3D Modelling. This created sections along the pile centreline, perpendicular to the centreline, however, the section wouldn't sample the TIN.
This seems like there will be a simple solution, but it is eluding me at the moment. Any advice would be greatly appreciated!
Regards
Ben
You need to get away from the vanilla ACAD stuff and use the Civil 3D tools. Set the screen back to 'World'. C3D doesn't like UCS and certainly doesn't use it. Create an alignment that follows your pile. Then, setup a sample line group that will section the pile as you need it and sample the TINN of the pile. It should also be able to label the grade breaks within the pile. Once you generate the multiple section views, you should be able to extract the data you're looking for in the sampled data for each section line.
"Just exactly is this about?"
We are building a wharf structure, using steel piles, driven into the seabed at 1:4 rake.
The piles are 1500mDIA, ~40m long and are driven with a large hydraulic hammer.
We align the piles using a gate, which is a guide that hangs off a barge or the already built wharf. The gate has a top gate and bottom gate. Top and bottom gates have UHDPE rollers that sit on rams. The top gate has hydrualic rams and the bottom gate has screw rams. We position the pile in 3 dimensions by first positioning the gate and then the rollers, then pitching the pile and adjusting the rollers.
As you can imagine, when we start banging in the pile, there is a huge amount of force. The inside roller on the bottom gate, the roller that the pile leans on, broke, so the pile was leaning on steel, as it was being hammered in.
The steel to pile contact removed a strip of paint from the pile and appears to have deformed the pile.
As an engineering surveyor, my job was to determine if and by how much the pile was deformed. Given that the pile is so large and the paint so thick, about 3mm, the strip of paint that is missing makes the pile look creased, when you look down from the top.
Sorry, I don't mean to lecture you in pile driving, but I have no idea what you do, or do not know about it, and I regularly have to explain this to people who have no industry experience.
jeff_rivers wrote:
So if I understand what you are asking, you have a material pile, probably cone-shaped, and you wish to have horizontal cross-sections of the pile at certain z-elevations.
Sort of like contour lines, yes? And these would show the deformation as anyplace where the contour line wasn't close to a perfect circle (assuming the pile is close to conical everywhere but the deformed area.
Is that close, or did I misunderstand?
The Pile is a steel tube, 1500mDIA.
However, the contours is pretty much what I ended up doing. Because the area of the pile that I was surveying was the underside, there was no vertical surfaces, otherwise this wouldn't have worked.
I created a TIN with major contours every 1m and minor contours every 5mm. Sounds excessive, but we are looking for sudden changes.
When I was satisfied that the damage was consistent, I created an eclipse that match the piles design shape and put that at every major contour, to create a line to visually compare to.
Then I just extracted the major contours and moved them with their respective eclipses into a grid pattern, printed on A3 and handed it to the engineeer.
@ccoles wrote:You need to get away from the vanilla ACAD stuff and use the Civil 3D tools. Set the screen back to 'World'. C3D doesn't like UCS and certainly doesn't use it. Create an alignment that follows your pile. Then, setup a sample line group that will section the pile as you need it and sample the TINN of the pile. It should also be able to label the grade breaks within the pile. Once you generate the multiple section views, you should be able to extract the data you're looking for in the sampled data for each section line.
The problem I encountered trying this was the sample lines will only cut vertically. Because the pile is raked at 1:4 (14degrees off vertical), the difference between the scanned surface and the ASB pile surface appears huge, even if they are only mm's appart.
To create the ASB pile, I shoot the highest and lowest visible points of the pile and draw a 1500mDIA cylinder between those points. Because the pile isn't actually round (manufacturing defects, weld distortion, bending from hammering, etc), even if there wasn't any major damage, the points that I scanned on the pile surface would never really match the ASB pile.
The other two problems I encountered trying this method was:
1- the sample lines wouldn't sample the cylinder. It was a solid, so perhaps I needed to make it into a mesh? As I understand it, TINs can't have vertical surfaces, so I don't know how to deal with that.
2- I tried rotating all the objects on a horizontal plane, to make it easier to use sample lines. However, the TIN wouldn't rotate like that. I could move it, but not lay it flat.
Ben,
"The piles are 1500mDIA, ~40m long and are driven with a large hydraulic hammer."
Wow! That sounds like one hell of a project. I'd love to see that hammer! And you're slugging those big nails in with a hammer on a boat? I've never worked with piles before, and certainly nothing on water. (I'm in the Mohave Desert.)
Can you post a photo or two? And how about posting your drawing? There's lots of creative brains on this forum that might come up with a solution.
Dave
Dave Stoll
Las Vegas, Nevada
Bill,
I'm sure that he meant 1500mm = 4.92 ft. in diameter and 40m = 131.23 ft. in length. If not, then I'm doubly interested. I'd still like to see the hammer that drives these in.
Dave
Dave Stoll
Las Vegas, Nevada
Indeed I did mean 1500mmDIA, certainly not 1500mDIA. Not sure about the precise Imperial units values as we do not use them at all here in Australia.
I will see if I have any pictures of this project. We have strict rules about taking and posting photos from this project, people have been fired already.
The data is in a local grid, so that should be ok if I post the points file.
Thanks for the interest so far!
Sorry for the delay in getting back to this, I've been pretty busy.
There is a photo attached which shows typical piles on this site. I can't get a photo of the actual pile in question due to the works going on in that area.
I've attached the csv file which contains the relevant information. Column order is Point Number, Easting, Northing, Elevation, Description
Codes are
8ABOT - Bottom centre of pile
8ATOP - Top centre of pile
SCRATCH - Top and Bottom shots on the centre line of the scratch
SCAN - General scan of the pile to accurately determine the shape of the pile surface
SCRATCH SCAN - Scan of the pile, focused on the area effected by the scratch
SETUP - Position of the total station
Pile has a radius of 750mm.
There are some important things to consider.
The shots that have a low angle of incidence are much less accurate, therefore, shots that further around the side of the pile are woefully useless.
The pile itself is not perfectly round, straight or smooth. It is hard to say how much the surface varies from perfect, but it is quite a lot.
The results of the survey showed that the amount of deformation to the pile was minimal, in most areas the measured deformation was less than the thickness of the paint.
Ben,
Cool! I've never seen piles driven in at an angle. You must get some rough seas there. Thanks for posting the pic.
Dave
Dave Stoll
Las Vegas, Nevada
Ben
I've been playing with the CSV points and re-reading your posts, trying to understand what it is you're doing. All to no avail. I can be fairly dense. What deformation are you trying to quantify? The shape of a particular pile?
Dave
Dave Stoll
Las Vegas, Nevada