Setting plane boundaries when measuring parallelism

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

  • Setting plane boundaries when measuring parallelism

    Team,

    I'm about one year into using PCDMIS and a question came up regarding parallelism. My part is laying flat on the CMM table. The GD&T call out for \\ is between the top side and the side that is laying on the table. The table is datum Z.

    I simply took 6 pts on the top side and "constructed" a BF Plane1. I also, took 6 points on the table to construct the datum BF Plane2. When I run the parallelism measurement, I simply select Plane2 as number 1 (the datum) and Plane1 as number 2 (the feature).

    My question is, what is PCDMIS using for the boundaries to calculate Max-Min. I don't see where PCDMIS lets me set the planer boundaries in which to run the math. I have also searched in "F1" but all I get is lots of stuff on scan boundaries.

    Thanks,
    Charlie

  • #2
    When using Points to construct features, it is best to use BF Recomp, as this correct for any incorrect vectors during the point measurement. Second, PC-DMIS uses the SECOND selected feature for the datum, not the first.

    For the calculation, PC-DMIS makes a perfect 'average' plane with your datum, and gives you the range of the highest point to the lowest point of the feature from the datum.

    Stu

    Comment


    • #3
      I never use the cmm surface for any datums unless it's going to function that way by sitting on it's highest points.
      I used to be high on life but I built up a tolerance.

      Brown & Sharpe Global Advantage
      PCDMIS CAD++ v2011mr2
      PH10MQ/SP600M


      sigpic

      Comment


      • #4
        Gday,
        Wherever possible place part on parallels and actually probe datum Z face rather than cmm table. You may get a false indication of parallelism if there is any flatness issues on DatumZ. Amount of hits taken would vary depending on accuracy desired and size of part ( 6 hits may not be enough ). Cheers Paul.

        Comment


        • #5
          [echo]Select feature first, datum second. Try to set up the part so you can probe both planes rather than the plane and the table.[/echo]

          As far as boundaries go, it is limited to the points you take. It will not iterpolate between points. See drawing below. Blue points are the hits. As you can see, both cases the plane extends past the points but the reported parallelism value isn't accurate because of where the points fall. In order to be more accurate, you would need to move the points very close to the edges, use more points, use a method other than the CMM.

          http://i214.photobucket.com/albums/c...arallelism.jpg

          Comment


          • #6
            Thanks for the responses. I think I understand now !

            I appreciate it,

            CharlieJ

            Comment

            Related Topics

            Collapse

            Working...
            X