Composite Feature Control Frame

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

  • Composite Feature Control Frame

    To begin, I am using PC DMIS 2016 SP 7 with LEGACY DIMENSIONING.

    On this part, items 1,2, & 3 (see attached picture) pertain to what we call ISOfix wires. Basically another component latches onto these features and it is supposed to engage simultaneously, hence the composite FCF.

    In item # 1,2 & 3 above, I need some additional opinions on how this FCF should be interpreted. The way I have been instructed by our gage engineer to understand this is that each one of these 'wires' individually (measured each as cylindrical features) is allowed to move within a positional Ø zone of 4.0 WITH respect to |A|B|C|. Then, a continuous feature is to be constructed using all of the points from both features and evaluated to itself, regardless of |A|B|C|. This I find hard to understand since I have always thought of the lower portion of a composite FCF as a refinement.

    My interpretation of the drawing would be that both of these wires, based on the FCF having one position symbol, would be tied together in essence. Once each cylinder is measured, then the individual points on each feature would be extrac ted and combined into a constructed cylinder and evaluated to |A|B|C|, NOT as individual cylinders. As a refinement, the same cylinder would be evaluated again to fulfill the secondary segment of the FCF.

    If I am correct, how am I to evaluate the lower portion of the FCF Ø2.2 on the CMM? In other words, how do I actually accomplish this in the software?


    In the past, I have simply set my origin to one of the wires, and then dimensioned the other wire using the 2.2 positional tolerance. I have been informed that this is incorrect, so I have to come up with a different way that would be understandable to all involved.

    Please offer any suggestions if this is something you might do regularly.

    Image 15.jpg

  • #2
    Create a best fit of both wires and dimension them to the 2.2 true position.

    Comment


    • #3
      Once you have evaluated your features to the ABC datum scheem create a new alingment.
      In this alignment you will want to best fit the 2 features with the recall being to the original ABC. This shows nothing governing rotation or axial shifts, so no check marks required to lock axis.
      Report your 2 features to the lower tier tolerance and you are done.

      It is a refinement, the refinement is that in releation to eachother they don't shift more than the 2.2 diametricaly from one another. That's why the tolerance zone is smaller than the ABC tolerance zone.

      Comment


      • #4
        What say y'all about the upper portion of the FCF. Are each of the cylinders treated separately or as one continuous feature? The drawing does say 2x, so I'm wondering if the upper portion of the FCF will have two dimensions while the lower portion now properly understood will have one dimension? Or would both segments of the FCF need two dimensions?

        Thoughts?

        Comment


        • VinniUSMC
          VinniUSMC commented
          Editing a comment
          The upper and lower portions will both have 2 dimensions.

      • #5
        Originally posted by R2ah1ze1l View Post
        This shows nothing governing rotation or axial shifts, so no check marks required to lock axis.

        So, when I am in the Best fit dialog, what is the method that should be used and how can I be sure? Min/Max, Least Squares, etc...



        Comment


        • #6
          johnnyrocket So I would recommend Type User Defined and Best Fit Method Least squares. Then again, I have only a snippit of information to go off, you need to determine the best practice based on the DWG requirements. HTH

          Comment


          • #7
            I don't see how the features to be measured are cylinders, but going under the assumption that they are cylinders:

            The upper frame is the tolerance zone for both cylinders, constrained in location and rotation by the A|B|C DRF. Not know for sure what A, B, and C are, I am again going to just assume that they lock down all 6 degrees of freedom appropriately. This means that, with regards to A, B, and C, the cylinders must be within the 4.0 mm diametrical tolerance zone, respectively.

            The composite means that the 2.2 diameter tolerance zones are free to float within the 4.0 tolerance zones with no constraint on location or orientation, except of course that they be constrained to each other by the basic dimension (the theoretical relationship between the tolerance zones of the two holes cannot move separately of each other).

            So, as said above, dimension the two cylinders as normal to ABC. Then, best fit the two cylinders (3D, rotate and translate) and dimension them again, to the smaller tolerance.
            "This is my word... and as such is beyond contestation."

            Comment


            • #8
              Hi Vinnie, Thank you for your input. I have a couple more questions...


              Originally posted by VinniUSMC View Post
              I don't see how the features to be measured are cylinders...
              Can you suggest a more efficient way to measure these features? I assumed cylinders since there are engagement zones defined on the print.


              Originally posted by VinniUSMC View Post
              The upper frame is the tolerance zone for both cylinders, constrained in location and rotation by the A|B|C DRF.
              I thought that the composite FCF would essentially tie the features together so that they would have to be treated as a continuous feature that would thereby have to be constrained to a 4.0 Ø zone. Are you saying they are reported individually? Same question applies to the lower portion of the FCF.



              Originally posted by VinniUSMC View Post
              of course that they be constrained to each other by the basic dimension (the theoretical relationship between the tolerance zones of the two holes cannot move separately of each other)........Then, best fit the two cylinders (3D, rotate and translate) and dimension them again, to the smaller tolerance.
              When the two individual features are bestfit, thereby creating a new alignment, I would no longer be in my |A|B|C| alignment correct? So if that is the case, then my nominal relationship between the two features should be 0,0,0. Am I thinking correctly?

              Comment


              • #9
                Originally posted by johnnyrocket View Post
                Hi Vinnie, Thank you for your input. I have a couple more questions...

                Can you suggest a more efficient way to measure these features? I assumed cylinders since there are engagement zones defined on the print.

                Nevermind, I see what they are now. They are wire clips. I thought it was some sort of section view the first time looking at it. I also see now why you are asking about continuous features.

                I thought that the composite FCF would essentially tie the features together so that they would have to be treated as a continuous feature that would thereby have to be constrained to a 4.0 Ø zone. Are you saying they are reported individually? Same question applies to the lower portion of the FCF.

                The "composite" part of the position callout is the lower frame. The top frame works exactly how any other TP would work. 2x means it's 2 features, not a continuous feature. Each feature is separate, but simultaneous requirement applies. Same for the lower frame.

                When the two individual features are bestfit, thereby creating a new alignment, I would no longer be in my |A|B|C| alignment correct? So if that is the case, then my nominal relationship between the two features should be 0,0,0. Am I thinking correctly?

                I also see now exactly why composite with no datums is appropriate. They are, as drawn, coaxial. You are correct that the bottom frame (the composite part) is no longer in ABC (but it is constrained within the 4.0 to ABC zone). The nominal relationship is 0,0, but there is a distance between them. Within the engagement zones, they must be coaxial to within 2.2, and located within 4.0.
                See blue responses above.
                "This is my word... and as such is beyond contestation."

                Comment


                • #10
                  Think of a bolt pattern. The holes are individual features, but they all have to meet the tolerance simultaneously. Same applies here. The top locates them individually to the datums and the bottom locates them as a single feature to themselves for mating purposes.
                  PC-DMIS 2015.1, SP10, CAD++
                  Global 7/10/7, 5/5/5
                  Renishaw PH10MQ, PH10M, TP200

                  Comment


                  • VinniUSMC
                    VinniUSMC commented
                    Editing a comment
                    "as a pattern" not "as a single feature". But, I get what you're saying.

                • #11
                  Originally posted by VinniUSMC View Post
                  I also see now exactly why composite with no datums is appropriate. They are, as drawn, coaxial. You are correct that the bottom frame (the composite part) is no longer in ABC (but it is constrained within the 4.0 to ABC zone). The nominal relationship is 0,0, but there is a distance between them. Within the engagement zones, they must be coaxial to within 2.2, and located within 4.0..
                  So, considering everything you have clarified, would this change your CMM method as you described above?

                  Comment


                  • VinniUSMC
                    VinniUSMC commented
                    Editing a comment
                    No, that would still be my recommendation.

                • #12
                  Think car wheel and hub.

                  The studs in the hub can't move (press fit in place). They are in a certain pattern.
                  The holes in the wheel can't move. They are in a certain pattern

                  The pattern of studs (hub) determine the pattern of holes (wheel)

                  If the 2 patterns aren't the same, the wheel won't go on the hub. Try putting the wheel from a Suburban on a Chevette.....

                  The actual location around the wheel of the holes makes no difference, as long as the pattern is correct, it will go on the hub. There is a certain distance each individual hole must be within from the center of the wheel, but then all of them, when looked at as a whole, must maintain a specific pattern or the wheel is useless.

                  5x127 (or 5x5) is the pattern for the Suburban. 5 holes, 5 inches from the center of the tire. Each hole has a TP of 2.0mm to the wheel alignment (very generous IMO) Alignment is center of wheel XY, no rotation, level is to the 'through' axis of the wheel. Each hole individually can be TP 2.0, but, you can't have one hole that is 1mm too close and one hole that is 1mm too far from center, nor can you have one that is 1mm too close to the hole 'clockwise' and 1mm too far from the hole 'counter-clockwise', so after they are 'good' to the 2.0TP, now you have to look and make sure that they are all within 0.5 of each other AS the pattern. Else you'll have a car rolling down the road like a clown car.
                  sigpic
                  Originally posted by AndersI
                  I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                  Comment


                  • johnnyrocket
                    johnnyrocket commented
                    Editing a comment
                    Thank you for the input Matt. So based on the original print that I've attached with only two features rather than a bolt hole pattern, would you agree with VinniUSMC regarding the CMM inspection method?

                • #13
                  Vinnie's method would work. I don't do much with cylinders (once in 9 or 10 blue moons) but for circles (just circles mind you!) another way to an answer for the relationship of JUST 2 circles is a very simple 2-D distance, of +/-1.1 (for your 2.2 refinement). The nominal spread is hard to read (280?) so the allowed spread is 280 +/-1.1 is the same thing as TP 2.2 for TWO features. This won't work for 3 or more.
                  sigpic
                  Originally posted by AndersI
                  I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                  Comment


                  • VinniUSMC
                    VinniUSMC commented
                    Editing a comment
                    Not spread, the features are 2 coaxial cylinders.

                Related Topics

                Collapse

                Working...
                X