Runout - Axial vs Radial

  • Filter
  • Time
  • Show
Clear All
new posts

  • Runout - Axial vs Radial

    I have a part that is constantly out of tolerance on radial runout. My datum is a cylinder and my runout feature is a circle. When I change runout from from radial to axial, it’s now within tolerance.

    I looked up radial runout and axial runout and the difference is radial is how far off a feature is off the datum axis but still parallel and axial is how much of a tilt the feature has to the datum axis. Which one should I use and why and I out of tolerance with only one and not both?
    Last edited by acgarcia; 11-29-2022, 11:28 AM.

  • #2
    Can you show the print or GD&T callout?


    • #3
      a circle does not have an axis so there is no tilt to measure in comparison to the cylinder.


      • acgarcia
        acgarcia commented
        Editing a comment
        That makes sense. I though about checking my runout features as cylinders but there isn't exactly a lot of meat on them, my CMM head will hit other surfaces so I chose to stick with circles.

    • #4
      runout image.png
      So here's an pic of what i'm checking. FYI, this program has been running on the CMM for 2 years and the parts were 95% good. Part was moved to a different CNC and now we are having out of tolerance problems on a daily basis. Not saying the CMM is perfect or even correct. Just pointing out the only change that has occurred.

      The datum cylinder and the runout feature to the right are machined at the same time with a tail-stock installed. Then the part is taken out of the chuck and flipped around to chuck on the newly machined runout feature and then it machines the other runout feature, that feature is the one that almost always comes out of tolerance.

      GD&T callout is Single Runout of 0.006" to datum A.


      • Matthew D. Hoedeman
        Matthew D. Hoedeman commented
        Editing a comment
        maybe they should be chucking on the datum cylinder. Sounds like a CNC problem, not a CMM problem.

    • #5
      What does the dial indicator say?
      Sheffield Endeavor3 9.20.8, Tesastar-SM, Leitz LSP-X1s & LSP-X1M, PCDMIS 2011 MR1


      • acgarcia
        acgarcia commented
        Editing a comment
        I suppose that's the next step...

    • #6
      Put it a "V" Block and check with an Indicator before CMM check. Validate results. it's the process.............


      • #7


        • #8
          What exactly does the print say?

          Are we talking circular runout vs total runout here? They are very similar to measure them both the exact same way & collect the same data.. the way the final result is calculated is different depending on which one the print requires


          • #9
            Symbol on the print is a single arrow.

            We actually indicated a part this morning still in the machine. The feature that reports out of tolerance runout indicated to about 0.0005" on the indicator. I asked the machinist to move the indicator to the large cylinder since that is the datum and it measured about 0.004" on the indicator on the end to the large cylinder. I asked him to move the indicator to the other end of the large cylinder is it measured about 0.003" on the indicator.

            The machine has the tail stock engaged and I think its moving the part when it engages.


            • #10
              Put the part in a good "V" Block and check concentricity of the two diameters cut during first operation. They should be as good as machine can make them!
              This will help determine where in process the problem occurs.​


              • #11
                acgarcia this is how to measure circular runout
                circular runout.jpg


                • #12
                  Thanks for the info DAN_M as this always confuses me. Due to the size of my Datum A which is like 8 inches in diameter by 8 inches long, and my runout feature being about 3 inches in diameter but only like 1 inch long, it would only make a cylinder about .0.100" long because the the extension on my touch trigger head would crash into the part. I thinking about checking a part on 1 of my LSP CMMs. I might be able to make a longer probe to try and get a nice spread out cylinder. I haven't tried but I don't think that a short cylinder would make a difference compared to a circle.

                  I went ahead and reported true position for reference and you can clearly see the axial deviation. You can rotate the part 180 degrees on the CMM and my axial deviation move to the opposite directions so that pretty assuring. A machinist manager checked things out and found runout in the CNC. The drill that makes the hole and chamfer for the tail stock is slightly off center causing the tail stock to **** the part over. I haven't been asked to check anything differently or change anything like number or points or circles to cylinder yet.


                  • #13
                    DAN_M How would cross section 2 give you a value of 0.0008? Asking honestly.


                    • Beck's
                      Beck's commented
                      Editing a comment
                      min+max = 0.0008

                  • #14
                    mathematically, you should have the entire datum surface measured and then as much of the evaluation surface as you can. Only measuring .100 of a 3 inch long diameter will not yield correct or repeatable results

                    I would definitely look into using larger hardware if you need to keep this on the CMM or just do it on the surface plate

                    The example of shared with you, the result you would report is .0008 because its the distance between the lowest & the highest point of the “worst” analyzed cross section


                    • #15
                      DAN_M Thanks for the help. I got some work to do.


                      Related Topics