Reasons behind a "bad calibration"

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

  • Reasons behind a "bad calibration"

    Hello all
    This might be a broad question, and not nesscecarily about pc-dmis.

    But im curious why you get bad results (STD) in calibration for certain angles.

    For example, we start our calibration routine at night (with about 100 angles / tip)
    A lot of times when we arrive in the morning, the results are mostly good. But then there are some angles that the std are very high >0.006mm.
    So for example t1a0b0, t1a90b90, t1a90b180 are all under <0.003mm, but then for the same tip at t1A60B90 its too high.

    What are some causes behind this?

    My reasoning is that dirt is the biggest cause for a bad calibration. But when you are calibrating 100 angles and only 5 angles are "out of tolerance", I think its weird.
    Shouldnt the other 95 angles within tolerance detect the dirt on the tip/sphere ? (especailly if you're using a high amount of measuring points).

    Are there some other causes?
    Please enlighten me.

  • #2
    edit: what i think is even weirder is that: In the morning when we see the results. and lets say tip 1 A60B0 has a STD of 0.006mm. We just recalibrate it without touching/cleaning neither the tip nor sphere. And then all of the sudden the results are good. From doing nothing but going into probe utility> Selecting the tip > calibrate.

    Comment


    • #3
      Shooting from the hip here:

      Temperature controlled environment or not?

      ...or the CMM got lonely in the night and what you are seeing as an error is actually CMM tears.
      PC-DMIS CAD++ 2o19 R1 SP10

      Comment


      • pcdmisstudent
        pcdmisstudent commented
        Editing a comment
        Yeah we have controlled enviorment.
        I think you might be right on the second point

    • #4
      Are you resetting all tip back to theo before running that calibration?

      IF SO, you could have a head either out-of-square or slightly not-perfect when seating at certain angles. If IMO, only the MASTER A0B0 should be reset, and if you always say the tool has moved for it, then it should always be perfect anyway so no need to reset it to theo. IF you do not reset to theo, it uses the previous calibration result as the target location for the new calibration, which eliminates any out-of-square or non-perfect head issues. Since you are just re-calibrating the 'bad' probes one at a time, I would guess you are NOT resetting them to theo for that and thus the better results.
      sigpic
      Originally posted by AndersI
      I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

      Comment


      • pcdmisstudent
        pcdmisstudent commented
        Editing a comment
        Ah this explains it!
        Since we only reset the mastertip to theo , but not the rest. So thats why we get better results after recalibrating. Thank you

    • #5
      Well, another thing to consider, level of uncertainty, consider the repeatability of the machine, the head positioning and the TTP itself....then, throw in a degree or 2 temp diff and the elusive few microns you are worried about becomes a moot point.....this cares me a bit also...why are you calibrating 100's of probe angles ? ? ? I find it hard to believe you use all those angles on a regular basis....how often do you calibrate ?
      I could go on forever about how many shops I have seen doing this and wasting TONS of time.....
      Jim Jewell

      Comment


      • #6
        Originally posted by jjewell View Post
        Well, another thing to consider, level of uncertainty, consider the repeatability of the machine, the head positioning and the TTP itself....then, throw in a degree or 2 temp diff and the elusive few microns you are worried about becomes a moot point.....this cares me a bit also...why are you calibrating 100's of probe angles ? ? ? I find it hard to believe you use all those angles on a regular basis....how often do you calibrate ?
        I could go on forever about how many shops I have seen doing this and wasting TONS of time.....
        No we dont use all of them regurarly. But we its just easier to calibrate all rather than planning on what angles we will use tomorrow, the day after that and so fourth.
        The factory closes at midnight, so I dont think we're wasting our time by calibrating all angles during the night when no ones here.

        Comment


        • jjewell
          jjewell commented
          Editing a comment
          I see, but how often are you doing this ? Because if you are doing it often....
          1/ you are doing it often for nothing.....
          2/ putting excessive wear and tear on the head and probe for nothing...
          3/ You can calibrate only as needed, it only takes a couple minutes to calibrate probes you need.....

          Sorry, but this all sounds like a bit of guidance or thought process.....just my 2 cents

        • Peter Fuller
          Peter Fuller commented
          Editing a comment
          I used to work at a shop where the head programmer convinced the quality manager that the probe calibration process was a 3 hour routine that had to be run every day. He would start the calibration process, get his coffee, and read his paper for the first few hours. He wasn't very happy when I came to work there.

        • pcdmisstudent
          pcdmisstudent commented
          Editing a comment
          We run the whole routine maybe once a month or whenever we have a crash- and we run a daily check program that checks the most frequent tips daily

      Related Topics

      Collapse

      Working...
      X