hexagon logo
Register Sign in
  • Replies 17 replies
  • Subscribers 32 subscribers
  • Views 276 views
  • Users 0 members are here
Options
Related

Nominals are changing once in a while when I rerun the program

timgramley
We recentely purchased a new version of PC-Dmis 2012 mr1. I've written a few part inspection programs and verified / ran them on different parts. I am seeing the nominals change every once in a while when I just run a program or more so when modify a program by adding a new dimension to the program. It doesn't seem to affect the new dimension, just random dimensions on the report which causes me to have change the nominal back and then refresh the report.

I experience this to a greater occurance when I run programs that I had created using 2010 MR1.

The 2012 MR1 program was installed on a brand new laptop.

I was told that 2010 MR1 had some memory leaks and that was the cause of the nominals changing. Does 2012 have the same issues too? ( we upgraded to get rid of the nuissance of having to shut down and restart every 10 -15 minutes of use on the old system )
  • ...also, make sure you read through the links posted earlier in the thread. They contain some good info regarding settings etc.
  • When you run the "dcc alignment" are you using auto features? Also are you using guided points or just taking the points randomly in the areas? When I was usings arms I found that the changing nominals stopped for the most part when using auto features and guided points on the final alignment. This keeps consistency between runs and hopefully operators. Show some code, also someone else chime in, isn't there a setting about updating nominals? Never had to mess with it but I have heard of people having issues with this.

    The big thing to keep in mind is that the arm is a great piece of equipment that allows us to pretty much measure anywhere, but with that being said it opens up the door for additional issues because there is no fixed location/coordinate system on a table allowing for some error in measurement/setup. This means that the programmer has to take additional sets making sure that the points/programs are as repeatable as possible. This means additional alignments, guided points, pictures of setups, and lots of comments/notes. If things are done right arms will be extremely successful within your company. If these things are disregarded the arm will be put in a closet and will collect dust because everyone will say that the arm is a piece of crap (which at the end of the day is just company/employee lack of understanding).

    Hopefully this helps, and if you have more questions just let us know.

    Wicked
  • there is a setting to toggle in settings editor that fixes the problem with not using a fully constrained alignment.. have to dig a little to find the post but I remember it and it worked.

    make sure that 'allow fine tuning of alignments' in NOT checked in your f5 page. Only use this when you want to update your theoreticals but not your measured values.

    be sure you don't have any warnings defaulted such as 'carry nominal back to feature'; i always reset these to a select few with every program.

    play with the update dependent features option and fine tune alignments in a test program to get a very good understanding of what these will do in the different possible combinations. Those pesky little questions can mess up a lot of things with a quickness if not careful. Good luck!
  • there is a setting to toggle in settings editor that fixes the problem with not using a fully constrained alignment.. have to dig a little to find the post but I remember it and it worked.

    make sure that 'allow fine tuning of alignments' in NOT checked in your f5 page. Only use this when you want to update your theoreticals but not your measured values.

    be sure you don't have any warnings defaulted such as 'carry nominal back to feature'; i always reset these to a select few with every program.

    play with the update dependent features option and fine tune alignments in a test program to get a very good understanding of what these will do in the different possible combinations. Those pesky little questions can mess up a lot of things with a quickness if not careful. Good luck!
    +1 yes the warnings are critical if your clicking yes or no without knowing the repercusions. It can be a disater. Use to run into this problem all the time in my first year of programming
  • there is a setting to toggle in settings editor that fixes the problem with not using a fully constrained alignment.. have to dig a little to find the post but I remember it and it worked.

    I have not heard of this I'll have to check it out. Not going to lie I've check a part say just a diameter check without a full alignment without issues. I believe there is something in maybe the comon core manual or another stating that pcdmis wants full constraint. I could be wrong though. Thanks for the post hemirunner
  • I have unchecked "Ignore CAD<->Part" under the F5-settings and set the registry flag "UseTHEOsForCADToPartLevelAlignment" to "1" (true). This setting is versionspecific so your mileage may vary.
    This has done wonders for my wandering nominals. But as DaSalo says, this is surely mentioned in other posts handling the topic.

    I made a program like this today and after measuring a plane I did a level and origin and then measured a circle and a slot (the THEO's taken from points clicked on the model).
    I ran it and noticed a small deviation in the theo's. I polished these up and moved my part slightly.
    Ran it again and noticed larger deviations in the theo's. Moved the part slightly more.
    Rinse, repeat - still getting theo's that deviated. Polished them again.

    Closed PC-DMIS, fired up the settings editor and looked up the UseTheosForCADToPartLevelAlignment. I changed the setting to TRUE (1), saved and closed the settings editor.
    Ran the program again and still got theo's that deviated. At this time, I almost blew a gasket, thinking that it didn't work.
    Anyway, I polished the theo's once more, saved and closed the program.
    Opened it up again and ran it. No deviating theo's!
    Moved the part slightly, ran it - no deviating theo's!
    I repeated this at least three more times, all resulted in the theo's staying the same.

    Heureka! Finally! (time for a happy dance!)

    So, all in all, it seems that the registry setting did the trick! Too early to tell if the setting affects something else though... (knock on wood)

    From the links posted earlier.
  • UseTHEOsForCADToPartLevelAlignment I must have missed that going through the posts

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