hexagon logo

After sphere info missing, the calibration did not work any more. Help needed.

{"data-align":"none","data-size":"full","data-tempid":"temp_25362_1697647390655_131"}


{"data-align":"none","data-size":"full","data-tempid":"temp_25363_1697647400292_447"}


Hello all.

I got a bad day. One of our machines, the sphere information was missing, so I copied it from other machine.
Then the calibration did not work any more, It looks like some setting not set properly.
Please help...
Parents
  • When I mentioned manual, I didn't mean just that one manual hit on the ball (needs to be in the direction the probe is pointing, by the way, in case you didn't know).

    When you open the probe qualification dialogue, select angles (like A0B0) and then click measure, another dialogue opens.
    In the upper right area, there are some options with circles next to them you can put a black dot in one of.
    MAN
    DCC
    MAN+DCC
    DCC+DCC

    Put the circle in MAN, and set the number of hits to 5.

    Proceed and take one hit on top of the sphere and four hits around the sphere's equator.

    When back in the original dialogue, look at the results (a button for results).

    It will tell you what it expected and what it measured along with a standard deviation value.

    When you do that manually, how are those numbers?

    Picture looks like a 2mm stylus, so it should say the stylus measured at about 2mm and the std. dev. should be below .1mm (I think, I don't do metric, MURICA!).

    If those numbers are good, go back in the probe qualification, select A0B0, put the black dot in DCC+DCC, set the number of hits to 17, there should be a black dot in the "User Defined" calibration method found just below the MAN, DCC, MAN+DCC, DCC+DCC thing, put in four levels from 0 degrees to 90 degrees (for the stylus in the picture).

    Does it run that now?

    Does it give good results when you open the results and look at them?

    Probe files get corrupted sometimes, and wiping them (or restoring a back up if you have one) becomes necessary.

    Also, some builds that are especially custom might make a situation where you HAVE to do a MAN qualification at all the angles you want first THEN do the DCC+DCC.

    That one hit you are taking is the MAN+DCC option, where you take one hit, in the vector of the probe, and then the machine takes three more before doing the qualifition. It will do that at every head rotation on MAN+DCC.

    If you say the sphere has moved and you want to manually locate it, you'll have to take that hit one time only, first head position, regardless of how many angles you qualify. You should NOT be doing this unless you actually moved the sphere OR restarted the CMM Controller OR crashed the probe into something. You should do this for one stylus and position (having a dedicated master is best), and then never say Yes again until one of the above conditions is met.
Reply
  • When I mentioned manual, I didn't mean just that one manual hit on the ball (needs to be in the direction the probe is pointing, by the way, in case you didn't know).

    When you open the probe qualification dialogue, select angles (like A0B0) and then click measure, another dialogue opens.
    In the upper right area, there are some options with circles next to them you can put a black dot in one of.
    MAN
    DCC
    MAN+DCC
    DCC+DCC

    Put the circle in MAN, and set the number of hits to 5.

    Proceed and take one hit on top of the sphere and four hits around the sphere's equator.

    When back in the original dialogue, look at the results (a button for results).

    It will tell you what it expected and what it measured along with a standard deviation value.

    When you do that manually, how are those numbers?

    Picture looks like a 2mm stylus, so it should say the stylus measured at about 2mm and the std. dev. should be below .1mm (I think, I don't do metric, MURICA!).

    If those numbers are good, go back in the probe qualification, select A0B0, put the black dot in DCC+DCC, set the number of hits to 17, there should be a black dot in the "User Defined" calibration method found just below the MAN, DCC, MAN+DCC, DCC+DCC thing, put in four levels from 0 degrees to 90 degrees (for the stylus in the picture).

    Does it run that now?

    Does it give good results when you open the results and look at them?

    Probe files get corrupted sometimes, and wiping them (or restoring a back up if you have one) becomes necessary.

    Also, some builds that are especially custom might make a situation where you HAVE to do a MAN qualification at all the angles you want first THEN do the DCC+DCC.

    That one hit you are taking is the MAN+DCC option, where you take one hit, in the vector of the probe, and then the machine takes three more before doing the qualifition. It will do that at every head rotation on MAN+DCC.

    If you say the sphere has moved and you want to manually locate it, you'll have to take that hit one time only, first head position, regardless of how many angles you qualify. You should NOT be doing this unless you actually moved the sphere OR restarted the CMM Controller OR crashed the probe into something. You should do this for one stylus and position (having a dedicated master is best), and then never say Yes again until one of the above conditions is met.
Children
No Data