Symetry of a profile ?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Symetry of a profile ?


    I have to measure these 2 FCF on this drawing:


    OK for the 2D profile, as we have an Optiv 321GL, i'll align to the CAD and try to do a scan with the camera to have it.
    But fot that symetry ? how would you do it guys ? does PCDMIS allow to use symetry on a profile feature ? or do i have to do some constructions to obtain the middle line ?
    Attached Files

  • #2
    I believe the symmetry callout is placed wrong. The symmetry should be placed so it applies for the width, not the surface. As it is now, the symmetry callout (in my point of view) is invalid.
    According to the current drawing, you'd need to measure the top surface and the surface the leader line connects to and somehow create a new feature that encompasses the median plane from the angled section with the median plane from the symmetrical section and evaluate that.

    On the other hand, why didn't the engineer use datum A for the profile and ditch the symmetry altogether?
    PC-DMIS CAD++ 2o19 R1 SP11


    • VinniUSMC
      VinniUSMC commented
      Editing a comment
      I was curious about Datum A being a cylinder. If the profile is a revolved surface versus if the profile is opposed surfaces (maybe planar?) then the answer might be different.

  • #3
    I asked the question because yes, it sounded a bit strange to me too.
    I would have construct some median points for each points measured on each side of the profile, then construct something like a LeastSquare line and then report the symetry of that line to A

    Yes A is the cylinder axis, and the section of the surface with profile looks like this:


    So if the engineer use something like Profile/0.05/A without symetry it should do the same job ?
    i'll try it on the CMM today.


    • #4
      For your last question, I would say "no, but..."
      The symetry is about the "mid line" of the whole profile (construct a set of mid points on each side, then all mid points must be in a a tolerance zone centered on A, and measuring 0.05 of width).
      The profile is about the external surface. If you add "A" to the profile, I think that all the mid points will be in the symetry tol zone (so the job is the same), but the feature which is dimensionned is not the same (mid line vs external line).
      Hope it's clear, really not sure of it !


      • #5
        I guess you're right, JEFMAN, i gave it a try today, asking for some different dimensions.
        First time i was trying to use a DXF, but PCDMIS seems to have some trouble to detect the good edges… so i used the 3D model which i have cut in half before, so that it can easily detect 2D edges.

        Form looks good, and logically when adding the location to A it is worse. I also tried to create some mid-points along the profile, to see if i can ask for some symetry, but as you can see the result is weird... (values are exactly double the position of the points along X axis).



        Why is the profile without datum showing an unilateral tolerance and not like the one with datums ?
        Is it really possible to ask a symetry of a point to a line ? why those strange values ?


        • #6
          Are you using ASME or ISO? Profile without the datum is reporting according to ISO and the profile with datum is reporting according to ASME... Could be a bug or maybe something wrong with the PPAP report template.

          Zero is nominal for the symmetry for the points in relation to A. Change to Z-workplane, and evaluate a 2D-distance between a point and datum A, perpendicular to the X-axis. The nominal should be zero and your tolerance for that dimension should be symmetry / 2 = ±0.025. The deviation * 2 is the symmetry measured value.
          PC-DMIS CAD++ 2o19 R1 SP11


          • #7
            I tried both ASME and ISO, with the original report:



            Only the one with datum in ASME is inerpreted diffently. Dunno why the datum seems changing Something

            Concernning the points, I tried to ask for 2D distances, yes it works, however the engineer will be disapointed if he don't see a symmetry into the report , anyway those tolerances seems too tight for this tiny part, i have to discuss with him.


              Editing a comment
              Well, you get disappointed when you see incorrect GD&T on his drawings so... ;P

          • #8
            Originally posted by pab39 View Post

            Why is the profile without datum showing an unilateral tolerance and not like the one with datums ?
            Is it really possible to ask a symetry of a point to a line ? why those strange values ?

            Profile without datum is like a flatness, so the min is zero.
            With a datum, the part can be smaller or bigger than the theoritical. The tolerance is defined as the diameter of a ball which goes along the nominal, and the measure gives a result as a radius (distance between hit to nominal profile), so the result is multiplied by 2 to match to the tol.


            • pab39
              pab39 commented
              Editing a comment
              OK i see what you mean, it is just a bit disconcerting to see a result of, for example, 0.042 showing OK with a -0.025/+0.025 tol on the report. -_-

          Related Topics