hexagon logo
Register Sign in
  • Replies 7 replies
  • Subscribers 223 subscribers
  • Views 84 views
  • Users 0 members are here
Options
Related

Machine lost orgin of all parts after probe change

pcdmisstudent
Hi guys, this is kinda a continuation of my other thread 
https://www.pcdmisforum.com/forum/pc-dmis-enterprise-metrology-software/pc-dmis-for-cmms/535953-hexagon-wont-help-me-so-now-i-ask-you-for-help"

Where our tips wouldnt relate to eachother and have big deviation in XY.

I fixed this by deleting all our probe files and rebuilding them.
And when i rebuilt them i rebuilt them "correctly" because ver time we've made some changes to our probe system, but never actually updated the probe file, cos the changes were minimal. Ex:

1. Probehead from Tesastar-m to HH-AS8-T5 (Which Tech guy said is identical to the tesa)
2. Tp 1 in our rack (also master) used to be a ⌀2 ruby with 20mm/1⌀ stem + 10mm/3⌀ extension.
Now its a ⌀2ruby with 30mm/1.5⌀ stem.
3. Tip 2 changed from ⌀4 ruby with 20mm/2⌀ stem + 20mm/3⌀ extension
Now its a 4⌀ruby with a 40mm/⌀2mm extension.

So none of these changed affected our measurements/calibration.
Well anyway, so i rebuilt the probe files and calibrated the tips & got good results from calibration and also from our check program (which orgins on sphere with master and then measures the sphere with the rest of the tips and report XYZ-dev)

But when i ran our other check program (which measures a guage ring on the fixtureplate), it could find the center of the ring, but it crashed in Z as is would think the ring is much further down.
The same problem happend in all our part routines, where the machine would think the part is much lower in Z than it is.

Is this really right? shouldnt the probe find the part since we've defined our probes and calibrated them.
Or do we have to rerun the manual alignment on all our programs?

  • Yes, you need to re-run your manual alignments. Since the probe definition was wrong, any alignments previously established using the old (incorrect) probe build will have accumulated errors.
  • Yes, you need to re-run your manual alignments. Since the probe definition was wrong, any alignments previously established using the old (incorrect) probe build will have accumulated errors.

    Ok thanks, I thought calibrating the probes and establishing the new location of the sphere wouldnt affect our programs.
    When we replaced the the old tesahead it didnt affect any programs. (is that because we didnt change definition of the probe file?)

    When running a program the difference in XY is about the same, but in Z its 50mm. I didnt think it would ve made THAT much difference. (especially since the components should have the same dimensions (Besides diameter on stem/extension) )

    Is there a smarter way to bypass this? Since we're talking about a lot of programs.
    Maybe i should just restore the old incorrect probe files again?

    I post a picture of the old mastertip and the new one. Maybe ive done something wrong. (The GLOBAL_TO_TESASTAR_FLANGE missning (?)
    Attached Files
  • I think you need to call the Hexagon Service department for your region. It sounds like you have a problem with the machine mount point or mechanical offset. The 50mm difference you are seeing is due to those two head configurations being significantly different. This will not only affect all of your existing programs, it could also mean that the machine is/was using the wrong part of the error map.
  • I think you need to call the Hexagon Service department for your region. It sounds like you have a problem with the machine mount point or mechanical offset. The 50mm difference you are seeing is due to those two head configurations being significantly different. This will not only affect all of your existing programs, it could also mean that the machine is/was using the wrong part of the error map.

    Ok, but it weird, because it misses the parts by 40mm in Z, but it changes tips in our rack perfectly. (And arent both of them based on machine coords)?
    EDIT: I found something interesting. I looked in the result file for my old tip and the new one and these were the coordinates for the cal.sphere.


    Mäthuvudsfil=D2L20 F1 Datum=2021-11-03 Tid=12:53:41 < OLD TIP

    Kollisionskontroll inaktiverad

    12292 CENT X 216.316 Y 559.346 Z -397.081 D 14.938
    T1A0B0 TEO X 0.000 Y 12.000 Z 230.050 D 2.000
    T1A0B0 MÄTT X -0.000 Y 12.000 Z 230.050 D 1.999 StdAvv 0.002
    ​

    Mäthuvudsfil=D2L20 F1 Datum=2023-09-28 Tid=14:19:15 < NEW TIP

    12292 CENT X 216.577 Y 551.748 Z -361.226 D 14.938
    T1A0B0 TEO X 0.000 Y 12.000 Z 194.400 D 2.000
    T1A0B0 MÄTT X 0.000 Y 12.000 Z 194.400 D 1.995 StdAvv 0.000

    as you can see X is almost identical, Y is almost identical but Z has a big difference. And 397-361 = 36
    and that is kinda the deviation im seeing in the programs
  • neil.challinor
    Hexagon support finally responded.
    they say it's not unlikely there's a 40mm difference between the heads . and its nothing weird about probe being 40mm off when finding part since I redefined my master.
  • so..can I just remove my new probe files and restore the old ones so we don't have to re align all programs?

    only reason I made the new correct probe files was so the collision simulator would work correctly.
  • No matter what you do, it's imperative that you omit spaces from your probe naming convention. CAPITAL_LETTERS_NUMBERS_AND_UNDERSCORES are the only acceptable characters in probe names.
    --and it's also imperative that you do not create multiple probe files/configurations with the same name.
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement