Calibration

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

  • Calibration

    Just go out 1st Global Advantage setup got the racks setup and now time for me to create a qualification program.

    As CMM guru's, what do you think is the right amount of hits, number of rows and speed to calibrate. I know I've been doing CMM quite a while. But MM4 software never really got into this much details as [email protected]@ does.
    sigpic

  • #2
    I'm not a guru but I set mine to 25 points, 4 levels, movespeed 25% and touchspeed 2%. HTH.
    DeWain Hodge

    Comment


    • #3
      No matter what you do, no matter HOW many hits or how many levels, use the SAME move and touch speed that you will use in your program.

      As for touches/levels.... 5/2. Good enough for sheet metal and the equipment I have (TP2). More touches does NOT give a better STDDEV or size with what I have.
      sigpic
      Originally posted by AndersI
      I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

      Comment


      • #4
        As Matt stated, ALWAYS use the same move and touch speeds in your program that you use in your probe qual. This is more important than number of rows and hits. If you qualify at one set of speeds and then measure at another set, you effectively nullified your qualification as it accounts for over shoot and probe flex at only the touch and move speeds in your qual routine..

        Oh yeah I also use 5 hits and 2 rows.

        Bill

        Comment


        • #5


          We use 8 hits, 3 rows (1 at pole, 3 halfway, 4 at equator):
          DCC+DCC
          Number of levels : 3
          Start Angle : 0.0
          End Angle : 90
          Last edited by John Riggins; 02-13-2008, 06:14 PM.
          Lately, it occurs to me
          What a long, strange trip it's been.

          2017 R1 (Offline programming)

          Comment


          • #6
            Thank you my fellow cmm measuring, underpaid & overworked kinda of guys......
            sigpic

            Comment


            • #7
              As Matt stated, ALWAYS use the same move and touch speeds in your program that you use in your probe qual. This is more important than number of rows and hits. If you qualify at one set of speeds and then measure at another set, you effectively nullified your qualification as it accounts for over shoot and probe flex at only the touch and move speeds in your qual routine..
              Interesting. I have been doing things wrong then. I run my programs a LOT faster than the qualification. My machine is an old LK-G80. It does not have move circular capability (or more accurately, it will do move circular, but in a series of right angle steps). The service tech just replaced our x axis gearbox. One of the things he mentioned as creating wear and tear on the gearbox is the canned cycle of the probe qualification which does move circular around the qualification sphere, so I qualify at 20% move speed and 2% touch speed and run my programs at 85% move speed, otherwise they'd take forever. I stay away from move circular as much as possible since it is so jerky.

              So I should speed my qualifications up? We also have an old DEA Epsilon that does pretty much the same thing....would it be in the same situation as well?

              ...oh yeah....13 hits three levels. We used to us the default setup with five levels, but the folks at Hexagon recommended we up the ante. One guy told as 20 hits, 3 levels (but that takes forever), and anther said 20 was too many and recommended 13, and that's what we have stuck with.
              sigpic

              Comment


              • #8
                WMS.

                I run our CMMs DCC+DCC with 5 hits. 70 movespeed and 2 measurespeed. As Matt said, measure at whatever speed you calibrate with.
                Xcel & MicroVal Pfx & Global 37mr4 thru 2012mr1sp3
                Contura Calypso 5.4

                Lord, keep Your arm around my shoulder and Your hand over my mouth. Amen.

                Comment


                • #9
                  Would it be a benefit to calibrate at fast speeds, like 80%, then run slower in the programs, like at 30-40%?

                  Aaron Fenner
                  Quality Engineer
                  PC-Dmis v4.3
                  B & S Global Status 7.7.7
                  6 years xp 32-bit

                  sigpic

                  Comment


                  • #10
                    If you run programs at a different speed than you calibrate you WILL have error due to this.
                    It may be small, but there will be error as a direct result of the speed difference.
                    Lately, it occurs to me
                    What a long, strange trip it's been.

                    2017 R1 (Offline programming)

                    Comment


                    • #11
                      13 and 3 as per Hexagon Tech.

                      It's touch speed that matters during qualification, not move speed.
                      sigpic GDTPS - 0584

                      Comment


                      • #12
                        I believe that move speed affects touch speed.
                        Lately, it occurs to me
                        What a long, strange trip it's been.

                        2017 R1 (Offline programming)

                        Comment


                        • #13
                          Originally posted by inspector212 View Post
                          13 and 3 as per Hexagon Tech.

                          It's touch speed that matters during qualification, not move speed.
                          Depending on your controller, your TOUCH speed is a percentage of your MOVE speed. I have proved this out on the Sharpe32 controller. SO, to keep the TOUCH speed the same fopr calibrating as well as measuring, the MOVE speed MUST also be the same.

                          Don't assume that your touch speed is anything OTHER than a percentage of your move speed without proving it.

                          Program a simple point with a simple program, BUT, I would suggest that the touch not be perpendicular to the shank, but make it a side hit.

                          Set the MOVE speed to 100 and the touch speed to 2 and watch how fast it moves for the touch.

                          Then set the MOVE speed to 10 and the touch speed to 20 (which is teh maximum it will allow) and run it again. If it moves at the same speed for the touch that it did the first time, then it is a percentage of teh MOVE speed. If it SLAMLS into the thouch, then you are at a set touch speed based on WHO KNOW WHAT.
                          sigpic
                          Originally posted by AndersI
                          I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                          Comment


                          • #14
                            Originally posted by inspector212 View Post
                            13 and 3 as per Hexagon Tech.

                            It's touch speed that matters during qualification, not move speed.
                            Agreed. Move speed should not matter.
                            Links to my utilities for PCDMIS

                            Comment


                            • #15
                              Originally posted by cmmguy View Post
                              Agreed. Move speed should not matter.
                              As long as your prehit/retract distance is great enough for the cmm to 'settle' to the touch speed, then yes, move speed should not matter.

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X