Another calibration question..

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

  • Another calibration question..

    I know there are tons of threads like this, and ive searched around and found some good sources. (uploading a tech sheet here)

    According to this document, IF you calibrate ur full program with all angles, tips etc..then you move your cal.sphere to another position. You'll only need to recalibrate the master tip and the new angle/damaged probe, and then all the tips in the rack will relate to eachother. (yes it has moved on master tip, no it hasnt moved on the other tip(s)).

    So if i see that for example my tip 4 A0B0 is measuring weird, I can just put the cal.sphere wherever on the table, and say "Yes it has moved" to master tip, then "no" to the A0B0 on tip 4?

    I tried this, measuring only the master tip with (yes it has moved man+cnc), and then the tip 4 (no it hasnt moved)
    Then i measured a guage ring (first with the master tip a0b0, aligned it to the ring once manually and then twice in DCC). Then i went to measure the diameter of the ring with the other tips: the result was that the diameter of the ring was identical on all tips. But only the XY values were within tol. for tip and tip4.
    So is this document valid or am i doin something wrong?

    Is it also wrong to asume that aslong you say "no the sphere hasnt moved" you can calibrate any tip and get relation to the other tips?
    So that you for ex. can measure the tip with "man+dcc" but answer: "no it hasnt moved" to the 2nd question?
    Attached Files

  • #2
    Yes you can do as you did and it should work. It sounds like somewhere along the way the calibration chain has been broken however.

    If you click 'Setup' in the probe utilities dialog there is the option to append calibration results, I do this so I can see if anything has gone wrong.

    This might be a pain to do, but if you do the following there are a few things you can do to validate the results:

    1) Reset all tips (this wipes all calibration data) but gives you a clean slate to work from
    2) For each probe switch on 'Append' calibration results

    Now calibrate master (Yes sphere had moved)
    Calibrate rest of your probes (No sphere hasn't moved) *(see below)

    From this position you can see if anything has gone wrong by doing the following.

    Locate the .RESULTS files for the probes (if you right click on them in Windows you can select Open With and use Notepad to view them)

    1) If you look at the probe results for the Master probe-tip you will see the THEO and MEAS are identical. If there is ever a difference then at some point someone has said 'No - sphere has not moved' for the Master probe

    2) If you search in any other results files for CENT then this probe at some point has been used to locate the reference sphere (i.e. someone has said Yes sphere has moved)


    * It is only the 'Yes - sphere has moved' or 'No - sphere has not moved' question that has an impact on the probe relationship.

    You can use the Man+DCC option for probes which won't calibrate straight in DCC (this is usually when there is a long stylus, the probe head isn't mounted very square, or a tip is bent)
    Applications Engineer
    Hexagon UK

    Comment


    • #3
      Originally posted by NinjaBadger View Post
      Yes you can do as you did and it should work. It sounds like somewhere along the way the calibration chain has been broken however.

      If you click 'Setup' in the probe utilities dialog there is the option to append calibration results, I do this so I can see if anything has gone wrong.

      This might be a pain to do, but if you do the following there are a few things you can do to validate the results:

      1) Reset all tips (this wipes all calibration data) but gives you a clean slate to work from
      2) For each probe switch on 'Append' calibration results

      Now calibrate master (Yes sphere had moved)
      Calibrate rest of your probes (No sphere hasn't moved) *(see below)

      1) If you look at the probe results for the Master probe-tip you will see the THEO and MEAS are identical. If there is ever a difference then at some point someone has said 'No - sphere has not moved' for the Master probe

      2) If you search in any other results files for CENT then this probe at some point has been used to locate the reference sphere (i.e. someone has said Yes sphere has moved)

      I think i found the problem for my results: Im the new CMM-guy here (and the only one), and the previous autocal program they had, the master tip was being calibrated in A0B0 and all (40-50) angles for that tip as "yes it has moved". (which means ill have to calibrate the a0b0 and all the 40-50 angles for the other tips to relate)
      I changed that now so the only tip/angle whose the answer is "yes it has moved" to is a0b0, and the rest of the angles of that tip "no, hasnt moved". So now a0b0 theo = meas, (except for STD (around 0.001) and the diameter which is about 4-5 microns under nominal diameter. (which i assume it should be?)


      Originally posted by NinjaBadger View Post
      * It is only the 'Yes - sphere has moved' or 'No - sphere has not moved' question that has an impact on the probe relationship.

      You can use the Man+DCC option for probes which won't calibrate straight in DCC (this is usually when there is a long stylus, the probe head isn't mounted very square, or a tip is bent)
      And regarding this, why can(should) i use this option only för probes that wont calibrate straight in dcc?
      The advantage i see to "No - sphere not moved" and "man+dcc" is that you dont need to recalibrate the master tip, thus reducing the calibration time by a little.






      Comment


      • Douglas
        Douglas commented
        Editing a comment
        1) yes exactly, do not recalibrate the master tip after the ball has moved, what you are saying you found causes a level of error. Take the master tip out of the full tips calibration. XYZ will match THEO and MEAS but the diameter will still compensate and there might be a very slight difference from THEO only on tip diameter

        2) what he means is that a sketchy probe build might not be very close to exact THEO build and may collide, doing the man+DCC just gives it the ball location... you would still say the ball did not move. You can also try just increase pre-hits for this, honestly if resetting to theo and increasing pre-hits to .150" doesn't do it the probe is pretty fkd up

      • pcdmisstudent
        pcdmisstudent commented
        Editing a comment
        Douglas

        1. With ball i assume you mean the calibration sphere? We move the cal.sphere all the time (because lack of machine volume).
        We only have one full-tip-all-angles calibration program which we run once a day.
        So how would we remove the calibration of master tip in the full calibrationprograms and how would we then locate the sphere? With another tip and "No - sphere hasnt moved" and "man+dcc" ?
        Also for master tip A0B0 i always reset back to theo during calibration.

        2. Ah okay, that made it clearer

    Related Topics

    Collapse

    Working...
    X