Comparing True Position of Two Holes on Manual CMM

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

  • Comparing True Position of Two Holes on Manual CMM

    We have an older Brown & Sharp Gage 2000 which is a manual CMM. This part is machined on a 5 axis mill so we are trying to measure the position of the holes compared to each other (concentricity). Right now I am setting surface A as the reference plane and cylinder 1 as the origin. I then go to the other side (exact mirror of pictured side) and measure the cylinder over there. It gives me a X and Z axis position that is way off of what was recorded on a new off-site Zeiss CMM. Also, I tried the same exact measurement method three times in a row to check repeatability. Without moving the part, the X and Z offsets change as much as .003in. I have also tried several different measuring methods and have still come up with a no-good repeatability.

    Does anyone have any suggestions or can point out what I am doing wrong?
    Attached Files

  • #2
    Does the GAGE 2000 run on PC-DMIS? Or does it have the box that came with it? I would construct a cylinder from both cylinders and align to that before checking concentricity. It sounds like you are projecting the axis of one cylinder over the length of the part, which will make it very difficult to get consistent results, especially on a manual machine

    Comment


    • Kryptor
      Kryptor commented
      Editing a comment
      It only has the box unfortunately. My boss is telling me that I'm not supposed to project a line between the two cylinders and align to that. As I understand it, the reference plane sets the alignment normal to it. I am then just setting the origin to the cylinder closest to the plane.

    • Kryptor
      Kryptor commented
      Editing a comment
      I have tried aligning the axis between the two cylinders and still came up with a variance when I repeated the measurement.

  • #3
    I would measure a ring gage vertically positionned, and look at the capability like this.
    Then, you should check the calibration results, and calibrate with much more hits, so that the measurement defects would be partially included in the calibration (?).
    Then, I don't know about this cmm, but there are sometimes a fine mode to take hits (block the axes and turn a calf screw).

    Comment

    Related Topics

    Collapse

    Working...
    X