True Position using bad coordinates?

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

  • True Position using bad coordinates?

    Hi All,
    Fairly new to PCDMIS and CMM's in general.. First post.

    I'm running into an issue where one hole I'm measuring TP for is not appearing to use correct coordinates in the calculation.
    Part is a flat piece of .015 6061 stock, I am measuring this all with vision.
    My alignment is pretty simple:
    Constructed center line between top/bottom edge [A]
    Right edge [B]
    Top plane [C]

    I have a hole that is spec'd dia 0630-.0635 coordinates X: -.625 Y: .420 Position tol: |.001S|A|B|C
    My report shows the X/Y as -.0632, .4201 (X dim being way out!)
    However, while in the same alignment if I do a linear dimension from the right edge [B] to the hole along the X axis, it reads .6246!

    What could lead to such a variance in coordinates? I've measured the part with a height gage and I'm trusting that the .6246 dim is correct
    There are more holes on the part, and when in this alignment, the position calculates with a different X than the linear dimension lists.

    I've re-done the alignment multiple times; at this point I'm scrapping the program and starting from scratch. Any ideas on what went wrong? I'm hoping round 2 gets different results, but even if the next routine works,I would like to identify what I did wrong in teh first one to avoid this in the future.

  • #2
    Are you using exactmeasure?
    If so try changing from FCF to use current alignment in the exactmeasure dialog box.

    Comment


    • #3
      Are you rotating the alignment and setting the X origin to [B]? If so, are you using the full length of [B] to rotate the alignment?
      PC-DMIS 2016.0 SP8

      Jeff

      Comment


      • #4
        Are your Datums correct?

        I can’t imagine -A- (primary) not being the top surface of the part, (which you have called -C-).

        Maybe I don’t understand?

        B&S CHAMELEON/PCDMIS CAD++ V2011

        There are no bugs, only "UNDOCUMENTED ENHANCEMENTS!"

        sigpic

        Comment


        • #5
          Assuming that you're using Exact Measure, it sounds like your alignment for the linear measurement and the alignment that Exact Measure is using is not the same. Exact Measure is likely doing something slightly different than you according to its own internal rules.

          Comment


          • #6
            Thanks for the quick replies. I did re-write the program from scratch and it's working, but I'd still like to see where I went wrong.. Since I'm new it coudl be a good learning experience..
            try changing from FCF to use current alignment in the exactmeasure dialog box.
            This was using the "use current alignment" type

            Are you rotating the alignment and setting the X origin to [B]? If so, are you using the full length of [B] to rotate the alignment?
            X origin is [B]. I have built the alignment a few different ways and have always rotated about Z to get my +X.
            I think I'll use a different feature to rotate to +Y and also try to delete and re build [B]

            Are your Datums correct?

            I can’t imagine -A- (primary) not being the top surface of the part, (which you have called -C-).

            Maybe I don’t understand?
            Datums are correct to print. A= centerline of top/bottom, B= right edge C= Top Surface.
            The part is for an RF antenna feed, and will be positioned(sandwiched) centrally in an RF Waveguide. This part has other features that must remain within the center of that channel, and I assume that's why they didn't list the top or bottom face as [A].

            Assuming that you're using Exact Measure, it sounds like your alignment for the linear measurement and the alignment that Exact Measure is using is not the same. Exact Measure is likely doing something slightly different than you according to its own internal rules.
            That's what I was thinking, but I couldn't find anything. I re-wrote the program and i'm getting the correct X,Y locations now.. I'm still going to try to dig around with the old file to see what I did wrong..

            Comment


            • #7
              I've had problems in the past getting TP because I wrote the program to coincide with the print.

              Often, hole A is dimensioned off the datum, but the hole B is dimensioned off hole A. That's the way I wrote the program. That's not what TP wants. Doing the math, and figuring out where Hole B would be in relation to the datum and programming THOSE numbers, instead of what the print says gave me an accurate representation of TP.

              Comment


              • Schlag
                Schlag commented
                Editing a comment
                That not how basic dimensions work. The feature control frame dictates where the features are located from. Basic dimensions are just basic...

            Related Topics

            Collapse

            Working...
            X