Calibration Theory

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

  • Calibration Theory

    **Situation**
    My thought process to be 100% calibrated is to locate the sphere with the master "yes sphere has moved, we cant keep the sphere in the exact place" followed by then selecting "no sphere has not moved" and calibrate all other probes/ angles. It's what I've read on forums and taught from hexagon.

    I've been tasked with trying to shorten the calibration program on each CMM we have 4. each takes a couple of hours to complete. All the probes get used constantly maybe not all the angles but the probes themselves get used quite a bit and i feel its necessary to do a full calibration, We aim for at-least once a week.

    **Theory**
    After talking with other programmers the theory is after you have added and calibrated all the angles you need "a full calibration". Is that you only need to calibrate A0B0 of each probe and the deviations that occur will relate to all the angles of that current probe for future calibrations. Thoughts on this?

    We have HH-AS-T5 5* automatic probe heads
    Attached Files
    Last edited by Jakep379; 04-19-2018, 08:14 AM.

  • #2
    On an indexable probe head (PH10 for example) the calibration of one tip angle only affects that tip angle. Period.

    For a continuously rotatable probe head, I believe you need to calibrate [at least] four angles, and that will cover all angles. I've never seen one of those, so I may be wrong...
    AndersI
    SW support - Hexagon Metrology Nordic AB

    Comment


    • #3
      Originally posted by AndersI View Post
      On an indexable probe head (PH10 for example) the calibration of one tip angle only affects that tip angle. Period.

      For a continuously rotatable probe head, I believe you need to calibrate [at least] four angles, and that will cover all angles. I've never seen one of those, so I may be wrong...
      This would still reduce calibration time significantly. Currently we have HH-AS-T5 *5 automatic rotating heads

      Comment


      • #4
        Originally posted by Jakep379 View Post

        This would still reduce calibration time significantly. Currently we have HH-AS-T5 *5 automatic rotating heads
        The only probe head that I know of that does that is the Renishaw Revo probe head which isn't compatable with pcdmis.

        Hexagon do have a continuous wrist (HH-ACW-43MW) however I don't know anything about it and the website is pretty sparce with the detail. Either way you could easily be looking at a new probe head and possibly a new controller to use it with. On 4 machines that's a lot of $$$

        Comment


        • 2013_junkie
          2013_junkie commented
          Editing a comment
          The PH10M-iQ Plus does it as well.

      • #5
        The only thing i could find on the subject was this "see attached" it's from patch notes from here ftp://ftp.wilcoxassoc.com/docs/old_r...MR1_Readme.txt
        Attached Files

        Comment


        • #6
          I have 12 probe files (2 x 6 slot racks) and I get around constantly calibrating by doing a probe check each week.

          Measure ref sphere with master probe and xyz origin on it.

          Then I measure the sphere with 2 angles (A0B0 and one at A90B(±)90) from each probe and check i'm within ±0.005mm from the origin.

          When probes go out (which they do every so often) then I simply re-calibrate that probe file.

          I actually chart the results as well so I an observe as a probe 'drifts' towards being 'out of tolerance'

          Automettech - Automated Metrology Technology

          Comment


          • Jakep379
            Jakep379 commented
            Editing a comment
            Yeah this is a great idea. I believe we will do something like this if i cant find a definite answer about calibrating all angles

        • #7
          There is a way to highlight which probe angles are used globally. I haven't figured it out yet. I'm sure the more knowledgeable people here could offer more help. I'm trying to work through this as well because the previous programmer added tons of angles that are not used. Presumably to waste time while the machine is "calibrating". Any thought on this would be very much welcomed.

          Comment


          • #8
            Capture.JPG

            Press that button and make your self a pot of Coffee kcassaday
            (In Memory of my Loving wife, "Ronda" who I lost March 7, 2016. I love you baby.)
            They say "Nobody's Perfect." I must be Nobody.

            Comment


            • #9
              Originally posted by KIRBSTER269 View Post
              Capture.JPG

              Press that button and make your self a pot of Coffee kcassaday
              You have to be in the "top" folder (directory) for that to work. For example, if you are in a specific folder for a job, it will only search in that folder and it's sub-folders. You need to be in the top-most folder that has ALL the folders for all programs in it for it to get all the tips.
              sigpic
              Originally posted by AndersI
              I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

              Comment


              • KIRBSTER269
                KIRBSTER269 commented
                Editing a comment
                He or she asked, I pointed, looking to waste time, that's definitely a time waster

            • #10
              AFAIK (I'm an Ex Hexagon Apps Engineer) you need to qualify each tip individually.

              Those release notes are from v3.7 (quite an old version) and I've used every version from then until 2013 and I've never seen that functionality.

              I thought I'd heard it being mentioned but I've never seen it and can't see any option in the SW to achieve it.





              Automettech - Automated Metrology Technology

              Comment


              • AndersI
                AndersI commented
                Editing a comment
                It might have been tried, and removed because indexing heads have very tight tolerance on *repeating* each angle, but not that tight tolerance on what the angle actually *is*. Any try to re-calculate the offsets from a few angles is doomed in that case... Then add a slightly bent tip to that equation...

            • #11
              Originally posted by kcassaday View Post
              There is a way to highlight which probe angles are used globally. I haven't figured it out yet. I'm sure the more knowledgeable people here could offer more help. I'm trying to work through this as well because the previous programmer added tons of angles that are not used. Presumably to waste time while the machine is "calibrating". Any thought on this would be very much welcomed.
              Ha! I'm not the only one out there...

              I used a script (REFERENCED ON POST #15 BELOW) to figure out what probes and tips are being used in every program. Just snag a copy of all your .prg files and copy them into one folder (no .cad files to speed things up). Once you know whats being used start 'cleaning' up your autocal program. Mine went from +2hrs to just over 1hr and my probes have been standardized. I ONLY use 9 probes for ALL of my CMM needs...so far.

              Glad to hear you're spending time around 'probes and calibration' most 'programmers' overlook that area.
              Last edited by Kp61dude!; 04-20-2018, 11:06 AM.
              PcDmis 2015.1 SP10 CAD++
              Global 7-10-7 DC800S

              Comment


              • #12
                Originally posted by Kp61dude! View Post

                Ha! I'm not the only one out there...

                I used a script to figure out what probes and tips are being used in every program. Just snag a copy of all your .prg files and copy them into one folder (no .cad files to speed things up). Once you know whats being used start 'cleaning' up your autocal program. Mine went from +2hrs to just over 1hr and my probes have been standardized. I ONLY use 9 probes for ALL of my CMM needs...so far.

                Glad to hear you're spending time around 'probes and calibration' most 'programmers' overlook that area.
                Why so many? I have 5, 1x30, 2x30, 3x30, 3x40, 4x30. All I have, all I need.
                sigpic
                Originally posted by AndersI
                I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                Comment


                • NinjaBadger
                  NinjaBadger commented
                  Editing a comment
                  But you mainly measure sheet right?

                  I have prismatic (mainly cylindrical) parts ranging from very small to 20" diameter and 24" long. They (for the most part) have to be held with the axis running along the X of the CMM. I have deep bores (so have a couple of the carbon fibre extension TP20's and a couple of disc probes, plus a star probe and a T-probe. (lost of internal grooves etc).

              • #13
                Originally posted by Matthew D. Hoedeman View Post

                Why so many? I have 5, 1x30, 2x30, 3x30, 3x40, 4x30. All I have, all I need.
                5x20 (master), 3x50, 3x85, 2x50, 1x80, 6x80 (disc), 0.5x38.5, 1x50 (cylindrical), 1.5x20. I use em' all ¯\_(ツ)_/¯
                Last edited by Kp61dude!; 04-20-2018, 10:58 AM.
                PcDmis 2015.1 SP10 CAD++
                Global 7-10-7 DC800S

                Comment


                • #14
                  Originally posted by Kp61dude! View Post

                  1x20 (master), 3x50, 3x85, 2x50, 1x80, 6x80 (disc), 0.5x38.5, 1x50 (cylindrical), 1.5x20. I use em' all ¯\_(ツ)_/¯
                  Ninja, yes, MOSTLY sheet metal, yes, but not only sheet metal.

                  Kp, why such a small master? I thought the prevailing school of thought was for a short BIG ball....

                  I do have "others", but I have no rack, so it's all manual changes here. The 'others' are for the SM25 setup, not "normal" probes. I did just use one of them today. Probably won't take them out of the box again for a month, or 2, or 3, or maybe Monday. They are 'there', just not a normal part of my world.
                  sigpic
                  Originally posted by AndersI
                  I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                  Comment


                  • Kp61dude!
                    Kp61dude! commented
                    Editing a comment
                    See my edit on my master...whoops typo.

                    I run 3 racks: lsp with 6 garages, R rack, extender on R rack has 3 more lsp garages. I also run a laser, that's what the r rack as mainly intended for.
                    Last edited by Kp61dude!; 04-20-2018, 11:02 AM.

                • #15
                  There's an excel macro that will parse a specific folder and produce a "Probes used" report. It's flipping awesome (thanks AndersI i!).
                  https://www.pcdmisforum.com/forum/pc...n-all-programs

                  You will have to take your library of all your routines, copy the library someplace else, then flatten the entire directory of folders and subfolders. There's a MS DOS prompt to get it done rather easily if you google "DOS flatten folders" the line is "for /r %f in (*) do @copy "%f" after you get to the directory.

                  I was able to use it and it reduced our 'all tips' calibration run time from over 4 hours to like 70 minutes. and from over 2 hours on another machine to less than 15 minutes.

                  Comment


                  • Kp61dude!
                    Kp61dude! commented
                    Editing a comment
                    That's the script I use!! It is very awesome. Kuddos to Andersl
                Loading...
                Working...
                X