manual probing very inaccurate

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

  • manual probing very inaccurate

    Our Excel 9-12-9 with version 2009 MR1 is very inaccurate when checking parts in manual mode. When reading a fixed gage pin, for instance, you can set the origin on the pin, come back and read it again and it will report a 0.005" different position. I would expect it to repeat within 0.0005" or close to that. It does not have that issue in DCC mode.

    We are a very low run shop doing a lot of prototype work where we don't want to create a program for each part or fixture we check. This CMM is not trusted by anyone any more, so we check parts manually on an older machine or on the surface plate.

    Any suggestions? Some parameter we may not have set properly? Or, as we have been led to believe, this machine will just not work accurately in manual mode?

  • #2
    Manual probing is inherently inaccurate.
    A huge point of DCC is to eliminate the variation in touch and vectoring.

    Even using manual, you need to have a good solid alignment. Are you doing that?
    Perhaps that is part of the variation you are seeing.
    When in doubt, post code. A second set of eyes might see something you missed.
    sigpic

    Comment


    • #3
      You can do a compromise of manual/DCC.

      Get into DCC mode.
      Start an autopoint. Check the "Measure" box on the right side.
      Manually probe the point, then hit "Create".
      DCC will take over and automatically measure the point.

      This works with any auto-feature.
      Lately, it occurs to me
      What a long, strange trip it's been.

      2017 R1 (Offline programming)

      Comment


      • #4
        Originally posted by George Holz View Post
        Our Excel 9-12-9 with version 2009 MR1 is very inaccurate when checking parts in manual mode. When reading a fixed gage pin, for instance, you can set the origin on the pin, come back and read it again and it will report a 0.005" different position. I would expect it to repeat within 0.0005" or close to that. It does not have that issue in DCC mode.

        We are a very low run shop doing a lot of prototype work where we don't want to create a program for each part or fixture we check. This CMM is not trusted by anyone any more, so we check parts manually on an older machine or on the surface plate.

        Any suggestions? Some parameter we may not have set properly? Or, as we have been led to believe, this machine will just not work accurately in manual mode?

        Holy crap, seriously? Have you done a Gage R&R on this? I never would have guess that my manual driving was less accurate than a Computer Numeric Control... I am just as pissed as you to think this is even possible...

        Comment


        • #5
          prototype work. high accuracy in dcc.

          Write a program to measure in dcc. It doesn't take very long. perhaps some training would help. you 're using high tech equipment. Would your toolmakers rather use a manual mill or CNC? why? repeatability and accuracy.

          B&S CHAMELEON/PCDMIS CAD++ V2011

          There are no bugs, only "UNDOCUMENTED ENHANCEMENTS!"

          sigpic

          Comment


          • #6
            Originally posted by dph51 View Post
            prototype work. high accuracy in dcc.

            Write a program to measure in dcc. It doesn't take very long. perhaps some training would help. you 're using high tech equipment. Would your toolmakers rather use a manual mill or CNC? why? repeatability and accuracy.

            those old manual mills were made for "manual" milling. A DCC or CNC CMM is NOT made for manual use. probe comp just doesn't work in manual mode and we all ( well, almost all ) know that.

            Comment


            • #7
              maybe its PCDMIS?

              Comment


              • #8
                Originally posted by Ironhoe View Post
                maybe its PCDMIS?
                I think Slug would agree with you.
                Always blame the software, never the programmer.
                When in doubt, post code. A second set of eyes might see something you missed.
                sigpic

                Comment


                • #9
                  Originally posted by John Kingston View Post
                  I think Slug would agree with you.
                  Always blame the software, never the programmer.
                  +1 !

                  Comment


                  • #10
                    +1 to what everyone else has written. Also your approach and retract directions are critical in manual mode, you must approach and retract as perpendicular to the feature as possible. That is how CMM software works.
                    If I have offended anyone with this post, I'd like to take this opportunity to say BOLLOCKS
                    Dry your eyes Princess and man up.

                    Comment


                    • #11
                      For things like a circle being that bad, I would BET MONEY that these 'errors' you are seeing are from BEFORE to AFTER an alignment. If your part isn't 100% square and perpendicular to the machine axis, then you ARE GOING TO SHOW a "ton" of error from working in a machine workplane to working in an ALIGNMENT workplane.
                      sigpic
                      Originally posted by AndersI
                      I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                      Comment


                      • #12
                        If you insist on measuring manually... lock your x & y axis when measuring in z. For diameters lock your z. You should see a marked improvement.
                        sigpic
                        I am fluent in three languages:
                        English, Sarcasm & Profanity.

                        Comment


                        • #13
                          Just one point to consider. What controller do you have? If you are hooked up to a Leitz controller, when probing manually you must have the jogbox set to SLOW (light illuminated) or you will not get good readings. "FAST" is only for moving the machine around not measuring points.

                          Comment


                          • #14
                            Thanks for the replies. I am an old time CMM guy, but no recent experience (until now) with PCDMIS. I do know for sure, that the PCDMIS Excel that I ran 12 years ago WAS repeatable to 0.0005" by manually jockying the joystick. They must have changed someting in the software to screw it up! Now it's only good to 0.005"?????????? Not very usefull.

                            Comment


                            • #15
                              Originally posted by George Holz View Post
                              Thanks for the replies. I am an old time CMM guy, but no recent experience (until now) with PCDMIS. I do know for sure, that the PCDMIS Excel that I ran 12 years ago WAS repeatable to 0.0005" by manually jockying the joystick. They must have changed someting in the software to screw it up! Now it's only good to 0.005"?????????? Not very usefull.
                              Im pretty sure you lost something in translation here... its not the software here its the operator and the operators methods. but wtf why not blame the software, everyone else does.

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X