Ball Bar Calibration

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

  • Ball Bar Calibration

    Just wondering if anyone has a list of the positions to preform during the ball bar check. I have a ball bar and the program just would like to know how to preform it and evaluate the results correctly.

    Thanks!!!

  • #2
    Originally posted by Nada_Inc View Post
    Just wondering if anyone has a list of the positions to preform during the ball bar check. I have a ball bar and the program just would like to know how to preform it and evaluate the results correctly.

    Thanks!!!
    ASME (American Society of Mechanical Engineering ) Publishes the standards. They will gladly sell you all the information you are requesting. Start by looking for B89.4.
    Last edited by Wes Cisco; 11-13-2014, 01:26 PM. Reason: 4 not 7
    sigpic"Hated by Many, Loved by Few" _ A.B. - Stone brewery

    Comment


    • #3
      If you buy a ball bar you can get a program to run it as well?
      Last edited by Tinyfxds; 11-13-2014, 02:35 PM.

      Comment


      • #4
        The ball bar program itself isn't anything special.

        The one I have you take a point on top of ball 1, then a point on top of ball 2. It measures sphere 1, then sphere 2, then re-measures both, and applies two 3D distances. The average of which is your number for that run. The spacing doesn't matter because it changes origin back and forth between tooling balls. There are probably more sophisticated programs, but it isn't really necessary.

        What positions and how to evaluate though, I have no idea.
        "This is my word... and as such is beyond contestation."

        Comment


        • #5
          some links you might find helpful

          http://webstore.ansi.org/RecordDetai...E+B89.4.1-1997

          http://www.precisionballs.com/cmm_cal_aids.php

          http://www.precisionballs.com/all_ball_bar.php#howto

          http://www.precisionballs.com/RECENT...TECHNOLOGY.php
          sigpic"Hated by Many, Loved by Few" _ A.B. - Stone brewery

          Comment


          • #6
            Should be able to get the positions off your last cmm cert.
            Myself, I do MORE positions than is called for (I use 29 positions of a bar that is 31.4" long).
            The exact length, as stated, isn't important. What IS important is the consistance of the measured length.
            My program measures the spread 5 times.
            My Excel form uses the average measured length from ALL positions, ALL 5 'trials' so use as the nominal length.
            It then averages the length of each position and compares each position to all the other positions
            1) "square" in X (3 positions)
            2) "square" in Y (5 positions)
            3) 45* in XY (3 positions)
            4) 45* in YX (3 positions, 90* from XY)
            5) 45* in XZ (3 positions)
            6) 45* in ZX (3 positions, 90* from XZ)
            7) 45* in YZ (3 positions)
            8) 45* in ZY (3 positions, 90* from YZ)
            9) 'square' in Z (3 positions)

            1&2 are used for XY square
            1&9 are used for XZ square
            2&9 are used for YZ square
            3&4 are used for 'diagional' square in XY
            5&6 are used for 'diagional' square in XZ
            7&8 are used for 'diagional' square in YZ

            My values for ALL of those are within OEM specs for the machine. More than is required for calibration/certification. I know my machine, right down to the dust on the way covers.
            sigpic
            Originally posted by AndersI
            I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

            Comment


            • #7
              Ball bar length should be 80% of the length of the shortest axis. I think that's per the standard.

              Comment

              Related Topics

              Collapse

              Working...
              X