hexagon logo
Register Sign in
  • Replies 7 replies
  • Subscribers 222 subscribers
  • Views 108 views
  • Users 0 members are here
Options
Related

Program Accuracy

dhoughton
I've been programming CMMs for about 8 years now programming in PC-DMIS for almost 2 years. I've been interviewing for a programming position's lately because I'm moving and I keep getting asked how do I know the program data is correct or accurate. I haven't really been asked this before and I've been saying that I check the data of my features used for alignment and calibration results but I'm not sure if this is the right way of checking.

Does anyone know if this is correct or if there is a better way to check the accuracy of the data generated from a program?

Thanks
  • Experience tells me I'm doing things the right way, especially with GD&T. Understanding how that stuff works and proving it out is a good indication you understand the print. There are only a few ways to measure lengths, diameters, angles, etc. Kinda hard to jack that basic stuff up (if you're seasoned). However, if I can double check with another calibrated Gage to validate the CMM, I will do that as well. Especially in cases where I'm unsure. I will also perform a Gage R&R to make sure I have my bases covered.

    I've never had to answer that question either. But good luck in your future endeavors!
  • Check some parts (sampling of linear dimensions in all three axes and sampling of diameters) with tools other than the CMM and compare the results (enter the data in an MSA form).

    Run the same part 3 times and enter the data in an MSA form.

    You can do it without the MSA form (and should) for all your programs.

    You are checking the machine to still be ok since calibration and checking your programming methodology.

    If you only use dimensions checked on the floor by operators and use their data against your CMM data, if there is a difference, you can fix all sorts of measurement problems on the floor and in inspection.
  • They want to know that you can back up your CMM inspection with traditional inspection methods.

    If you write a program, you or SOMEONE should be able to prove out each and every dimension repeatably via plate layout or other means.

    Believe it or not...I've ran into some programmers that can't even use a pair of calipers
  • in reply to DAN_M
    I truly believe, if you can't layout a part manually, you shouldn't be allowed to touch a CMM.
  • When I program a part, I can verify that the program is correct by re-measuring dimensions on a surface plate set-up, calipers, optical comparator, etc.
  • They want to know that you can back up your CMM inspection with traditional inspection methods.

    If you write a program, you or SOMEONE should be able to prove out each and every dimension repeatably via plate layout or other means.

    Believe it or not...I've ran into some programmers that can't even use a pair of calipers

    +1

  • MSA / comparison studies confirm your results are "accurate", and gauge R&R will show you can get those results time after time.
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement