Land Surveyor Forum

Replies

• 

  Daniel Palmer October 29, 2011 at 3:42pm

  Sorry about the late response, have been working away for some time. I was after the top centreline of the pipe to allow for design/fabrication of custom pipe bends to match existing pipework. With no pipe openings visible it forced me to do reflectorless scans of the pipework when potentially a few top centreline measurements would have sufficed. Diameters of the pipes were up to around 1200mm. Taking measurements with a total station to a mini prism can still give small height errors such that the highest elevation wins. I'm guessing just a plain old spirit level would work with the point in contact being the top providing no fabrication issues in the pipe rolling process. It just seemed a little agricultural and was wondering if anyone has seen any other tools or methods employed. Thanks for the resposnes from everyone, I learnt a lot about finding centroids now too if I ever need to take that path.
• Government Professional

  Kevin Barry October 3, 2011 at 1:05am

  Can the originator clear up if he wants the centroid of the pipe or as I see it the centreline along the top of the pipe?
• 

  Arnel M. Domag October 2, 2011 at 10:01pm

  I measured in the past machined cylinders, a component of huge mechanical engine, so those are practically of perfect circle cross section. I took two section plane samples. I took 3 distinct shots on each plane. I got the centerline after 6 distinct point shots.

  Antonio is right for a pipeline subject to deformations. You need to identify section planes and take shots as many as you wish along those sample planes. Perform Method of Least Squares (Regression Analysis) for those points to define the cross section profile. Perform statistical analysis for those points to determine the most probable location of pipe centerline. Include in your report any standard deviation values to support your opinion.

  It is very important to undertake Method of Least Squares traverse adjustment not only to get the best-fit solution but also to detect possible blunders in your work. This way, the certainty of your output is high.

  Hope this helps.
• Government Professional

  Kevin Barry September 25, 2011 at 4:37am

  Interesting problem. If I read it right you are not looking for the centre of the pipe, but the mid point on top of the pipe. Hence the plate bubble on the tool you mentioned. If I am wrong I apologise, but thats how it reads to me. What diameters are we talking about? What accuracy are you working to? The tool you mentioned seems to have the answer but you are still at the mercy of a plate bubble which is marginally better than a spirit level. However if you mark it once then flip it through 180 and mark it again then you can get a midpoint of your two marks if they are not at the same place. Good luck Kev.
• 

  cujoBoudreaux September 3, 2011 at 10:31am

  i wish you had a pic, but it sounds to me that they are able to determine the 120 degree points for any pipe. that would be the circle divided into three even sections..... it makes sense to do it that way. anyway, again, i hope i have helped in some small way.
• 

  Arnel M. Domag August 31, 2011 at 4:54am

  If the pipe is certain to be of perfect circle cross section then the job only needs three co-planar point shots for each sample plane. The plane that contains the three points must be roughly perpendicular to pipe's axis. This means that if you have two sample planes, then only a total of 6 point shots is necessary in order to determine pipe's axis lines. If the pipe is not certain to be of perfect circle cross section then along two sample planes, take shots as many as you think that defines the profile. Then perform Least Squares or regression analysis. Hope this helps.
• GEO Ambassador

  Skip Farrow August 30, 2011 at 8:38pm

  anyone want to take a whack at this?