Thoughts on circular runout

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

  • Thoughts on circular runout

    Good morning all,

    We have a customer that is requiring a circular runout of .0002 with respect to datum A. There are several size IDs and ODs where this requirement is stated. Having outsourced these parts for specialty grinding, due the the tight dimensional requirements, we received some of the parts back and we found some with a circular runout of .0003 - .0004. We returned those to the vendor with a request for verification. They replied that we were interpreting runout incorrectly and that and I quote: "that print calls for runout of .0002, that is 2 concentric circles that are .0002 apart. That is not T.I.R. of .0002, the indicator can move .0004 and those parts are good."

    Thoughts?

  • #2
    .0002 is the tolerance not distance apart. I have never herd anything like that before but made for good laugh this morning.

    Comment


    • #3
      Capture runout.JPG

      Comment


      • #4
        What is Datum -A-?

        Comment


        • Jim Poehler
          Jim Poehler commented
          Editing a comment
          Datum A is one of the ODs

      • #5
        Originally posted by Jim Poehler View Post
        They replied that we were interpreting runout incorrectly and that and I quote: "that print calls for runout of .0002, that is 2 concentric circles that are .0002 apart. That is not T.I.R. of .0002, the indicator can move .0004 and those parts are good."
        A runout of 0.0002 means that the total indicator variation may not exceed 0.0002. Period. They are misinterpreting the description of the tolerance zone.

        AndersI
        SW support - Hexagon Metrology Nordic AB

        Comment


        • #6
          Yeah What Anders said...

          Comment


          • #7
            Even with a Leitz CMM I wouldn't measure it on a CMM!
            I'd use my JENOPTIK form analyzer.
            But I also agree with the above.

            Comment


            • #8
              First of all Jim, Laying out a part manually requires 2 Indicators for a Runout. That picture that Tested shows above, that is the proper way to check concentricity, NOT runout. Watch both indicators and see the difference between the 2. while rotating the part. Technically your indicators can read .010 out of whack, as long as both indicators read that together. As long as they RUNOUT together.

              Untitled.jpg
              Last edited by KIRBSTER269; 11-13-2019, 10:05 AM.
              (In Memory of my Loving wife, "Ronda" who I lost March 7, 2016. I love you baby.)
              They say "Nobody's Perfect." I must be Nobody.

              Comment


              • KIRBSTER269
                KIRBSTER269 commented
                Editing a comment
                This will be easier to do, If you have an old Lathe that you can put this on, easier to hold and easier to rotate.

            • #9
              KIRBSTER269 - are you absolutely sure on that one? I thought the datum would be the center line of the established datum cylinder (min circumscribed cylinder), i.e. a constant datum - your setup incorporates the *form* of the datum? What's really needed is a chuck with an infinite number of grips (?word), at least as long as the datum cylinder...

              But I'm ISO, maybe ISO and ASME do it differently?
              AndersI
              SW support - Hexagon Metrology Nordic AB

              Comment


              • AndersI
                AndersI commented
                Editing a comment
                But once you have established the datum, its center line is the actual datum, no form on the datum involved in the actual evaluation of the runout.

              • KIRBSTER269
                KIRBSTER269 commented
                Editing a comment
                Anders, I have put thousands of parts on lathes and indicated them as perfect as possible, never in my life have I had a perfect part that didn't move the indicator a tad, I'm not arguing the method, with such a tight tolerance, that tad movement on my datum -.0001 <<<maybe. The feature might be +.0002 off in the opposite direction, from centerline, giving it a result of .0003 (bad) with 2 indicators. One indicator isn't going to show you that

              • AndersI
                AndersI commented
                Editing a comment
                I don't argue that the method is bad for verifying the *function* of the part, I'm just arguing that this method is not what's described by the standards as "runout".

            • #10
              AnderSL, you are right. I'm looking at the ASME book right now.

              Comment


              • #11
                Originally posted by KIRBSTER269 View Post
                First of all Jim, Laying out a part manually requires 2 Indicators for a Runout. That picture that Tested shows above, that is the proper way to check concentricity, NOT runout.
                Hola Kirby,
                I have to disagree, Concentricity is the derived "median points" of the feature with respect to the axis established by -A-.
                sigpicIt's corona time!
                737 Xcel Cad++ v2009MR1....SE HABLA ESPAÑOL

                Comment


                • KIRBSTER269
                  KIRBSTER269 commented
                  Editing a comment
                  Yeah you have to establish 3 points, the picture above does not do that. can you create a circle with 2 points Robert? I probably did not go into detail about it, but I was showing the best way to check it with little tolerance

              • #12
                Runout is how much one given reference feature or features vary with respect to another datum when the part is rotated 360° around the datum axis. It is essentially a control of a circular feature, and how much variation it has with the rotational axis. Runout can be called out on any feature that is rotated about an axis. It is essentially how much “wobble” occurs in the one part feature when referenced to another.

                Runout is all about "Wobble" I covered this before

                image_16239[2].jpg
                (In Memory of my Loving wife, "Ronda" who I lost March 7, 2016. I love you baby.)
                They say "Nobody's Perfect." I must be Nobody.

                Comment


                • AndersI
                  AndersI commented
                  Editing a comment
                  Sorry Kirbster, but that description of runout doesn't run in ISO land, at least. The tolerance zone is always circular, not depending on the *form* of the datum. Once you have established your datum, it is just a *line* which you are rotating about.

                  And concentricity I *know* is different in ISO (distance of center points/lines) and ASME (non-existing nowadays, really different definition earlier).

              • #13
                Right from the standard...
                Attached Files

                Comment


                • #14
                  Glad you posted that MikeD6 Indicator Fixed in a position normal to the "True Geometric Shape" If a part isn't perfect, which they never are, then you are just getting the results close.

                  Capture.JPG

                  Jim Poehler use the 2 indicators
                  (In Memory of my Loving wife, "Ronda" who I lost March 7, 2016. I love you baby.)
                  They say "Nobody's Perfect." I must be Nobody.

                  Comment


                  • Mike Ruff
                    Mike Ruff commented
                    Editing a comment
                    What if your part was tri-lobed from being too tightly clamped in the lathe chuck, say it is out of round by .001"? If your datum and feature with the runout callout were turned at the same time and both exhibit the tri-lobe condition, and both were out of round within .001" but the indicators on the datum and feature both move together, you would report a runout of zero?

                    In 9.4.1 of ASME Y14.5, it says "Where applied to surfaces constructed around a datum axis, circular runout may be used to control the cumulative variations of circularity and coaxiality." I believe your method with 2 indicators would negate the circularity portion if the form of the datum and feature coincide.

                  • Mike Ruff
                    Mike Ruff commented
                    Editing a comment
                    However, to support your argument there is this:

                    9.3.2 Rotation About an Axis

                    When the part is rotated about the datum axis, full
                    indicator movement (FIM) for each considered feature
                    must be within its runout tolerance. This may also
                    include the datum features as a part of the runout tolerance
                    control where so designated.

                    The key being "where so designated"

                  • AndersI
                    AndersI commented
                    Editing a comment
                    I read that as "the datum surface may not vary in form more than the runout tolerance". The *datum* is still the established *axis*, no wobble there.

                • #15
                  Datum A in this case would be the theoretically perfect cylinder that contains all of the elements of the cylinder that is being defined as datum A. It is the axis of this perfect cylinder that the part is being rotated around. Its like if the Datum A portion was press fitted into a sleeve that rotates. The runout of the datum A cylinder would be irrelevant. The axis of the sleeve would the axis of rotation and the runout would be checked with one indicator as the callout specifies.

                  Comment

                  Related Topics

                  Collapse

                  Working...
                  X