hexagon logo
Register Sign in
  • Replies 10 replies
  • Subscribers 224 subscribers
  • Views 638 views
  • Users 0 members are here
Options
Related

Question about calibration & master probe.

nighthawk9008
Morning, All.
I'm sure this question has been answered a million times on here, but I couldn't find specifics of my situation during a search. At out shop, we have 2 Global 5.5.5 CMMs that use the Hexagon/Tesa touch trigger probe system. With that said, I'll tell you how we calibrate the probes here. Both CMMs have a sub-plate. We screw the calibration sphere in the same location every time we calibrate. We will write a simple calibration program for all part numbers, so if me or the QE isn't here, they can screw in the calibration sphere, and open the program to calibrate. We use parameter sets for each part number, so we don't have to calibrate all tips. For every tip we use, we always say "No" when PCD asks if the calibration sphere has moved. Can we loose accuracy by doing this? My QE says that since we never move the actual location on the subplate, it shouldn't matter.

Also, we use one of our commonly used tips (1mm X 27.5mm A0B0) for the master probe, but we never use it to find the sphere (as its always placed in the same place on the subplate.

Here is the code for our typical calibration program: 

STARTUP    =ALIGNMENT/START,RECALL:USE_PART_SETUP,LIST=YES
            ALIGNMENT/END
            MODE/MANUAL
            FORMAT/TEXT,OPTIONS,ID,HEADINGS,SYMBOLS, ;NOM,TOL,MEAS,MAXMIN,DEV,DEVANG,OUTTOL
            MOVESPEED/ 300
            COMMENT/OPER,NO,FULL SCREEN=YES,AUTO-CONTINUE=NO,
            MOVE THE PROBE UP ABOVE AND OUT OF THE WAY
            IN A SAFE POSITION.
            COMMENT/OPER,NO,FULL SCREEN=YES,AUTO-CONTINUE=NO,
            REMOVE ALL FIXTURES AND PARTS FROM THE TABLE.
            PLACE THE CALIBRATION SPHERE IN LOCATION I9.
            FORMAT/TEXT,OPTIONS, ,HEADINGS,SYMBOLS, ;NOM,TOL,MEAS,DEV,OUTTOL, ,
            LOADPROBE/EPS 1X27_5 NO EXT
            TIP/T1A0B0, SHANKIJK=0, 0, 1, ANGLE=0
            AUTOCALIBRATE/PROBE, PARAMETER_SET=EPS C504-002 1X27_5 NO EXT, QUALTOOL_MOVED=NO,
                          CHECK COLLISION=NO, SHOW_SUMMARY=NO, OVERWRITE_RESULTSFILE=NO
            COMMENT/OPER,NO,FULL SCREEN=YES,AUTO-CONTINUE=NO,
            MOVE PROBE UP AND IN A SAFE POSITION.
            COMMENT/OPER,NO,FULL SCREEN=YES,AUTO-CONTINUE=NO,
            REMOVE THE CALIBRATION SPHERE AND RETURN IT
            TO THE HOME LOCATION.  ​


Thanks!!!
Attached Files
  • Whether or not the probe is located in "the same place" every single time why not say it has moved? This way whatever variance there is between screwing and unscrewing the sphere is not important as it would eliminate itself. Are you wanting to save time by not picking up the master, locate the sphere, and then calibrate probes and tips used for measuring your part?

    That little bit of error doing it this way introduces isn't what I would stress first, though it is something I would investigate or flat out trust the 'pros' at Hex without my own research, what I would stress first is process. There are ways this can bite you somewhere down the line if you aren't keeping your master 'updated' on where the rest of the probes are getting their reference frame from. I invested a good amount of time into automated calibration when I worked on a CMM that was part of a 'lights-out' automation cell, things really stuck out like a sore thumb when you made 'little' mistakes like potentially this one (notice I said potentially? It could very much not be an issue at all).
  • If the probe is located in "the same place" every single time why not say it has moved? This way whatever variance there is between screwing and unscrewing the sphere is not important. Are you wanting to save time by not picking up the master and locating the sphere?

    Pretty much the logic here. My concern would be that I may screw it snug, but the next operator may make it a 32nd of a turn tighter... I have the mode set as "DCC + DCC" so that the tip does 4 hits near the center, then starts the 10 hits around the sphere to help compensate for any variance like that.
  • "DCC + DCC" is for when the theoretical hits around the sphere are struggling to happen. I personally only used it when I brought a new probe into my setup to 'help' it find the sphere better because the theos were sometimes really off (especially a fancy non linear tip angle) but once calibrated I would turn "DCC + DCC" off to save time. I basically had a parameter set to introduce a new probe to the system and a different parameter for production. I also had one program to program all tips used on all parts thanks to whoever wrote that script to find all tips in all routine programs in a folder ;-)
  • There are ways this can bite you somewhere down the line if you aren't keeping your master 'updated' on where the rest of the probes are getting their reference frame from.

    ↑↑↑ THIS ↑↑↑

    Especially if the program utilizes multiple probes and one gets recalibrated but not the others. Then callouts like position or runout where features are picked up with different probes in which the relationship has been broken, your numbers will be wonky!
    Technically, the sphere moves every time the machine is homed.

  • Just to add: If you have slight variations where the cal sphere mounts each time you can answer Yes and DCC to find it each time, then repeat and answer No to refine the location again before you start calibrating your probes. Sorry. Meant yes each time. Basically run Yes and DCC twice.
  • in reply to Zen Cat
    Right, your master will 'find' the cal sphere in DCC mode (after you've found it manually once) rather than prompting an operator to drive with the joy stick.

  • I always thought the master probe is the probe used to locate the calibration sphere. Ideal master probe is a short stylus with a large diameter ruby like a 5mm x 10mm.

  • in reply to TonyYaka

    That is the general consensus, but I have seen people use different builds for master probe as well.

  • in reply to TonyYaka

    You are supposed to only use the master probe for location of sphere and it is good practice to use this probe to calibrate prior to calibrating any other probe.

    The size of the master probe should be what hexagon or manufacture suggest. As far as I know that is the 5x10mm probe for all CMM's. You want a nice size ruby and a very stout length of the shaft that is ideal. I have heard this from multiple trainers at the Hexagon facility.

  • If you are already in the habit of using parameter sets, make one for your master probe (tip A0B0).

    Perform an Autocalibrate on this with the master parameter set and set the qualtool moved to yes(dcc). All other probes/tip angles have qual tool moved=no.

    Even though the sphere is placed approximately in the same place, it is not micron perfect positioned the same each time it is assembled to your fixture plate. 

© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement