Best Fit Alignment - Iterations?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Best Fit Alignment - Iterations?

    I have a long 8mm pipe with 14 bends in it. The print only has XYZ intersect coordinates of the bends and the 2 end points. I do a quick manual alignment then go to DCC and probe all of the pipe sections with learned hits. Then I'm doing a best fit alignment using the XYZ theoreticals of all 16 points, and then reporting the actuals. Nearly every point is out except for the 2 end points, which are most critical.

    My question is, should I go back and remeasure everything after the best fit alignment? I ask because in the best fit window I see where it says "Iterate and re-pierce CAD". Not quite sure what that means and I'm not using CAD model. Any help is appreciated.
    sigpic
    Global Advantage 12-22-10
    TESASTAR M SP25 4.3mr2

  • #2
    Does the intersection points have ± tolerances or what kind of callout is it that you are checking?

    Pipe drawings sometimes seem to 'lock' the start point and only allows for rotation around that point (bestfit 2D).
    PC-DMIS CAD++ 2o19 R1 SP11

    Comment


    • #3
      Change you best fit alignment to "Specify Constraints" I don't trust 2D, 3D since V4.3. You have to type in the nominal values in order to use best fit alignments. You cannot use learned points without typing in the nominals.
      sigpic GDTPS - 0584

      Comment


      • #4
        I typed in nominals when constructing the intersect points. I'm using 3D best fit with these. Seems to work but I just need to know if I should go back and re-probe each pipe section and do another best fit, i.e, another iteration. What I mean by learned points are the actual probe touches for each cylinder. I'm creating them manually with the jog box then running them DCC.
        sigpic
        Global Advantage 12-22-10
        TESASTAR M SP25 4.3mr2

        Comment


        • #5
          With all your XYZ and vector theoreticals intact (correct vectors a must) I would do 2 iterations rough then tight to float to best situation. If the need be from that point I would use best fit with weights added to my most critical features to change there influence.
          sigpicThanks a Millionth

          Comment


          • #6
            You should not have to re-probe each section.
            sigpic GDTPS - 0584

            Comment

            Related Topics

            Collapse

            Working...
            X