Datum Implications

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

  • Datum Implications



    Can someone elaborate on this one? What if I have a part that is distorted and bowing and can not make perfect perpendicular planes on the datums? How can I also predict where are high points so I can pick them to create a plane? On the surface plate the part sits on high points. Making a plane on the same surface with the CMM you might not pick up the highest points. How do you ensure the simulated datum plane is equal to the functional datum plane?

  • #2
    You use the granite table of the CMM and a precision block or granite square to simulate the datums. CMM's cannot possibly scan every single point on a surface / surface edge. Proper setup on the CMM still requires DATUM Simulators.
    That looks like a great Functional Gauge candidate
    sigpicTAU ALPHA PI INDIANA DELTA CHAPTER
    "Due to the highly confidential nature of my job, I am not allowed to know what I am doing" - author unknown

    Comment


    • #3
      The Datum is a theoretically perfect geometric form that does not exist. A Datum Feature(or the Datum Targets on the Datum Feature) are the places on the part that mate to a Datum Simulator. A Datum Simulator can be a surface plate, a CMM, a Gage Pin, etc. The Simulator does not have perfect form, but it is as close as we can get it and that is very close, close enough for us to use as if it were the Datum. Benedictj1 is spot on with his advice to use plates for Datum Simulators, but it is important to understand the difference between the three concepts even if the part is not dramatically bowed.
      sigpic"Hated by Many, Loved by Few" _ A.B. - Stone brewery

      Comment


      • #4
        These guys nailed it on the head.

        Wish I had something that easy to set up. All my parts are round...

        Comment


        • #5


          Is there a way to prevent what is happening with datum- C-? It seems I can not use the geometry of the part itself to create an accurate datum plane.

          Comment


          • #6
            Originally posted by Andy F. View Post
            [ATTACH=CONFIG]9346[/ATTACH]

            Is there a way to prevent what is happening with datum- C-? It seems I can not use the geometry of the part itself to create an accurate datum plane.
            You can if have a Datum Simulator that is constrained perpendicular to Datums A & B so that it can only slide along the one axis, then it will stop on the single "high point" of Datum Feature C. Keep in mind that Datum C is the tertiary Datum. It is only constraining one degree of translation.
            sigpic"Hated by Many, Loved by Few" _ A.B. - Stone brewery

            Comment


            • #7
              What if I have the same problem on datum -A- ? The geometry prevent me to locate the part on a flat surface and I need to put it up on stands that will not going to hold the part up on the highest spots of the datum - A surface. I actually have a part here that is very narrow and it is twisting and distorting the datum -A surface.

              Comment


              • #8
                Using the simulators eliminates probing on the actual part surface, your going to probe on the datum simulators to establish datum alignment. I would use some refining call outs on the datum surfaces. Flatness and perpendicularity. Material conditions are also important etc etc. What would a functional gauge look like? A two sided bow with pins in it using the material condition and size tolerance to calculate their size.
                sigpicTAU ALPHA PI INDIANA DELTA CHAPTER
                "Due to the highly confidential nature of my job, I am not allowed to know what I am doing" - author unknown

                Comment


                • #9
                  Originally posted by Andy F. View Post
                  What if I have the same problem on datum -A- ? The geometry prevent me to locate the part on a flat surface and I need to put it up on stands that will not going to hold the part up on the highest spots of the datum - A surface. I actually have a part here that is very narrow and it is twisting and distorting the datum -A surface.
                  You need to talk to the engineer. Often thin and flexible parts can be checked in a "constrained" state. Unless noted otherwise a "free state" is the default.

                  Another approach would be to use marker or the like to "color" the surface than rub on a piece of paper on the surface plate to reveal the highest points. With those points identified to be used as Datum Targets for the CMM it does not matter so much if your stands are not supporting the high points, if you are taking touches on the high points to create a plane that you level to as your primary Datum.
                  sigpic"Hated by Many, Loved by Few" _ A.B. - Stone brewery

                  Comment


                  • #10
                    Originally posted by Wes Cisco View Post
                    You need to talk to the engineer. Often thin and flexible parts can be checked in a "constrained" state. Unless noted otherwise a "free state" is the default.

                    Another approach would be to use marker or the like to "color" the surface than rub on a piece of paper on the surface plate to reveal the highest points. With those points identified to be used as Datum Targets for the CMM it does not matter so much if your stands are not supporting the high points, if you are taking touches on the high points to create a plane that you level to as your primary Datum.
                    If you can access that surface with a cylinrical probe, you could come up with datum targets. but I think the cmm without a fixture for this is not the best way to gauge this part.
                    sigpicTAU ALPHA PI INDIANA DELTA CHAPTER
                    "Due to the highly confidential nature of my job, I am not allowed to know what I am doing" - author unknown

                    Comment


                    • #11
                      I always thought the software will help me to eliminate such of problems and reduces the amount physical fixture points need it to hold the part. That it will allow me to pick up the datums from the part itself with less error. I will have to rethink the way I have been inspecting. Thanks for the replies!

                      Comment


                      • #12
                        the exact scenario presented here is covered in a training class
                        I think it is level 3 ---- from what I recall measure line or a plane and extract max x point from that feature to set "C" datum
                        Don't remember all the details---- but set points to an array, extract X components and use max x (min if that is what would give the correct value)---value to datum considered compared to form error; otherwise if in this instance you would use centroid of feature. and as shown the value is off significantly especially when correlation to plate check is considered.

                        Comment

                        Related Topics

                        Collapse

                        Working...
                        X