How does v4.2 calculate profile of a surface?

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

  • How does v4.2 calculate profile of a surface?

    I have a part similar-looking to a camshaft with 17X profile of a surface (form only) called out all the way down the length of it. It seems that the part is a small fraction of a degree away from the theoretical causing it to read the profile of a surface as being out. It is my understanding that only the shape matters and not the location. As you can see from the arrows in the picture below, the shape remains fairly constant with only the location changing. Does anyone know how PC-DMIS v4.2 calculates profile of a surface when it is form only? IMO, this part is well within tolerance and the shape is good and location is out. Is this correct? Or is PC-DMIS correct?

    Thanks is advance!

    PROFILE.jpg

  • #2
    It should best-fit the profile and it seems that the profile 'dips' at one end (according to the picture).
    There are also differences in the calculation depending on what standard you should use (ISO/ASME).
    PC-DMIS CAD++ 2o19 R1 SP11

    Comment


    • #3
      FORM and FORMONLY are not the same. FORM is to itself, part alignment has nothing to do with it, FORMONLY checks form of a floating plane against your part alignment. In other word, FORMONLY is exactly the same as parallelism.

      Form Only = Highest point minus lowest point (output result is ‘total’ always, –TOL must be .000 always, regardless)
      Form and Location (ASME) = The furthest point from nominal
      Form and Location (ISO) = The furthest point from nominal x 2

      Not sure of v4.2.
      v2009mr1 calculates FORMONLY incorrectly when using a feature set, it must be a plane.

      Comment


      • #4
        Originally posted by DungT View Post
        FORM and FORMONLY are not the same. FORM is to itself, part alignment has nothing to do with it, FORMONLY checks form of a floating plane against your part alignment. In other word, FORMONLY is exactly the same as parallelism.

        Form Only = Highest point minus lowest point (output result is ‘total’ always, –TOL must be .000 always, regardless)
        Form and Location (ASME) = The furthest point from nominal
        Form and Location (ISO) = The furthest point from nominal x 2

        Not sure of v4.2.
        v2009mr1 calculates FORMONLY incorrectly when using a feature set, it must be a plane.
        This statement is incorrect; the feature in the attachment is not even a plane.

        Form and Formonly should report the same thing IF the same points are used in each instance. Form can be applied to a feature, while Formonly requires a Set to be constructed and use of the Profile command.

        vpt.se is correct, Form only will best-fit the selected points (based on which algorithym you select) to try and get them all within the tolerance zone.
        sigpic

        Comment


        • #5
          +1

          B&S CHAMELEON/PCDMIS CAD++ V2011

          There are no bugs, only "UNDOCUMENTED ENHANCEMENTS!"

          sigpic

          Comment


          • #6
            Originally posted by ir a programmer View Post
            This statement is incorrect; the feature in the attachment is not even a plane.

            Form and Formonly should report the same thing IF the same points are used in each instance. Form can be applied to a feature, while Formonly requires a Set to be constructed and use of the Profile command.

            vpt.se is correct, Form only will best-fit the selected points (based on which algorithym you select) to try and get them all within the tolerance zone.
            My bad. I was paying attention to the bottom part of it which is a plane where it shows the points way out. After a second look at the pic and I have to agree with you and vpt.se if the part has no datum structure and it seems to be (not really sure though), I apologize

            Comment


            • #7
              ir a programmer, so I need to construct a set from the points for form only to calulate correctly? Maybe that was the problem, because I just selected all of the points from the surface profile menu. And vpt.se, the part is not bent, it is straight. Idk if you can see it very well, but the deviation is pretty linear down the length of the part.

              Comment


              • #8
                Originally posted by Mike Ruff View Post
                ir a programmer, so I need to construct a set from the points for form only to calulate correctly? Maybe that was the problem, because I just selected all of the points from the surface profile menu. And vpt.se, the part is not bent, it is straight. Idk if you can see it very well, but the deviation is pretty linear down the length of the part.
                Yes -
                1. Construct a Set
                2. Profile the Set with correct parameters/tolerances

                If you so desire, insert a graphic analysis of the Profile to get a nice representation on your report. Everybody likes pictures.

                In the Edit window:
                Insert
                Report Command
                Analysis
                sigpic

                Comment

                Related Topics

                Collapse

                Working...
                X