profile of a surface

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

  • profile of a surface

    So I have a customer supplied program.
    Call out for profile of a surface with 4 different surfaces
    They have it originally programmed as
    multiple points like close to 40 along all 4 surfaces and then they are all put into a scn call out and then a profile of surface is called out from the scn. I am not sure that this is right.
    I would think that the scn would need to be made of planes and then profile created from a z+ view. But I could totally be wroung. Please any and all help is appreciated.

  • #2
    I was under the impression that a "scan" was literally a plethora of points (40 seems really low for a scan actually)
    I don't see why this would be bad, especially if it's supplied from the customer.

    If they sent it to you like that, they probably want it like that, is my methodology

    Comment


    • #3
      Normally it's a "feature set" constructed with the multiple vector points to calculate profile error.
      Last edited by Roberto; 04-04-2019, 05:18 PM.
      sigpicIt's corona time!
      737 Xcel Cad++ v2009MR1....SE HABLA ESPAÑOL

      Comment


      • #4
        Yeah so the issue would be... if all dimensions are no more than .0005 +/- from nominal values then how do they exceed a .013 profile tolerance?

        Comment


        • InspectorJester
          InspectorJester commented
          Editing a comment
          Isn't Profile based on T values or something?
          What are you basing this off of; the XYZ nominals?

        • arock
          arock commented
          Editing a comment
          I am not sure what you mean.

        • InspectorJester
          InspectorJester commented
          Editing a comment
          You state the dimensions are no more than .0005 from nominal. In Z? Within the Profile dimension?
          EDIT:
          I believe this is what I was referring to for T value..
          https://www.pcdmisforum.com/forum/pc...-highest-point

      • #5
        So it is a feature control set ... should those be constructed of planes or points? This is a new concept for me. I am not sure how to is calculating the profile and I know CMM are very picky and dimensions can change very easily if not called out properly.

        Comment


        • Roberto
          Roberto commented
          Editing a comment
          Turn on "MaxMin" in the dimension reporting for you to see the high and low. You can also activate the textual analysis to review the individual points & out of spec.

      • #6
        The default "name" for a feature set is "SCN" isn't it ? I always thought that was weird.

        Comment


        • arock
          arock commented
          Editing a comment
          yes i thought it meant somethng different lol but no it is feature set

      • #7
        I think if you use planes / lines etc. in a constructed feature set (scn), it will out put the centroids of those features, rather than give you each individual point within the plane / line, so points are fine. If you put on the graphics analysis of the profile dimension you will see where, and possibly why it is it is exceeding 0.013.

        Comment


        • #8
          if the actual dimensions are with .0005 of nominal basic dimension i dont understnd how it can be out of tol. ....

          Comment


          • InspectorJester
            InspectorJester commented
            Editing a comment
            A location of a plane could be spot on, but the flatness could measure crazy out of spec.
            It all depends on the dimension

        • #9
          Post your code and a sketch or section of the print if possible.

          Code:
          Program code here
          sigpicIt's corona time!
          737 Xcel Cad++ v2009MR1....SE HABLA ESPAÑOL

          Comment


          • #10
            We're missing a lot of information to be able to provide anymore than guesses. Profile dimension of a feature set composed of vector points is common. The issue you're having would depend on how the program is written and the profile FCF. It's possible to see individual point deviation that doesn't match the profile dimension if the program is written in one alignment and the profile is called out to a different alignment.

            Comment


            • #11
              Originally posted by Sora5 View Post
              We're missing a lot of information to be able to provide anymore than guesses. Profile dimension of a feature set composed of vector points is common. The issue you're having would depend on how the program is written and the profile FCF. It's possible to see individual point deviation that doesn't match the profile dimension if the program is written in one alignment and the profile is called out to a different alignment.
              Along the same line, if the points are measured in one alignment and the profile is dimensioned using XactMeasure with datums that are different than the alignment, you would have the same issue.
              PC-DMIS 2016.0 SP8

              Jeff

              Comment


              • #12
                Originally posted by arock View Post
                So I have a customer supplied program.
                Call out for profile of a surface with 4 different surfaces
                They have it originally programmed as
                multiple points like close to 40 along all 4 surfaces and then they are all put into a scn call out and then a profile of surface is called out from the scn. I am not sure that this is right.
                I would think that the scn would need to be made of planes and then profile created from a z+ view. But I could totally be wroung. Please any and all help is appreciated.
                How they did it is exactly right.
                A SCN feature is a "feature set" which is not a geometric feature, it's a collection of geometric features with nominal and actuals.
                Be sure and turn on Max and Min and turn off the Meas value in F10 Dimension Format.

                How it works:
                Each point has a nominal XYZ coordinate laying on the nominal CAD surface.
                Each point could be dimensioned as a T value for deviation as measured perpendicular to surface (along the "normal" IJK vector).
                As a surface profile is an upper limit and a lower limit, both of whom follow the nominal contour but are offset from nominal by half the zone size, it follows that the biggest positive T value on a surface bump should not stick out past the upper limit and the biggest negative T value on a surface dent should not be sticking down below the lower limit.

                How to prove it out:
                Go ahead and make T value dimensions for every point, tolerance + half the profile zone and - half the profile zone, and run the part.
                Write down the Max T reading and the Min T reading.
                Compare these to the Max & Min of the Profile result and you will see that they match (if Legacy or if Xactmeasure & the alignment matches the profile dimension datum scheme).

                Best way to report it:
                In addition to turning on Max & Min, insert an Analysis View or a screen capture with graphic arrows on. These pictures are worth many thousands of report numbers.

                Comment

                Related Topics

                Collapse

                Working...
                X