hexagon logo
Register Sign in
  • Replies 2 replies
  • Subscribers 221 subscribers
  • Views 27 views
  • Users 0 members are here
Options
Related

D6-51991 Requirements

mrollings53
We are updating our CMM manual to be compliant with D6-51991. Inspectors were here last week to watch how we ran the CMM and how we handle "Product Acceptance Software" verification. While they were here we identified areas that need attention. Once we get everything hammered out we will create work instructions and templates so that all programs are compliant moving forward.


Datum Alignment Verification:

Reporting position of a datum feature is simple enough. I've always programmed using the datum structure as my part alignment. Datum feature locations should be zero for the axis they origin. All well and good for legacy dimensions. I use, and would like to continue using, the Geometric Tolerance command. However, PC-DMIS doesn't always solve the datum structure the same way as the alignment I used, especially when a pattern of features is a datum. I am not smart enough, nor skilled enough to program the equations required manually to solve these datum structures just so that I can report the location of the datum as required by D6-51991. Is there some way to report the alignment that the demon calculates when using the GeoTol command? Will I be limited to using only the current alignment? I think this would be hurting ourselves when it came to datum shifts from MMB.


Datum Construction:

I know it's been asked a bunch before about constructing a datum plane from the three highest points. Can someone point me to a good example of the code needed to filter through a set of points and construct a plane with just the three highest points? Alternatively, Primary Datum Plane is a thing. I worry that it is going to construct a plane inside my part. How does this construction actually work? Does it use the workplane to construct the plane, the point vectors? I don't understand it well enough to trust it. Once I do I can explain why it's used instead of the looping command needed to filter through a bunch of points.


NIST test and the NIST Cert:

What does the NIST Test and Cert available from Hexagon actually tell me? Is the NIST Cert good for all builds of the version it is written for? The inspector would have accepted the cert for PAS if it had matched our installed build exactly.


Thanks for your help.
Parents
  • Hi , as you have found, the way in which the geometric tolerance command fits datums differs from how traditional alignments function. In many cases it is not possible to construct a traditional alignment that exactly matches the DRF - particularly when there are datum patterns involved or when MMB/LMB are invoked. We provided the constructed primary datum plane, secondary datum line and tertiary datum point as a way of creating a traditional alignment that functions the same way as a basic DRF but it only really simulates the most basic cases - combinations of planes lines and points, no orientation constrained hole fitting. See here for a detailed explanation of how PC-DMIS fits datums: https://docs.hexagonmi.com/pcdmis/20...ves_Datums.htm
    Having said that, you may still be able to satisfy the datum validation requirements by reporting the position of the datums to themselves since this would be ​the same as aligning to the datums and then reporting their location. For example, If you had a pattern of holes (B) relative to a primary datum plane (A), you could report the position of the pattern of holes to A|B.

    In terms of finding the high point on a datum plane, the constructed primary datum plane should do exactly what you want. It simulates how a part would sit on surface plate and will always be outside the part (external to material). If you are working to ASME, the appropriate fit would be constrained L2. It constructs a least squares best fit plane through all of the hits and then shifts it to the high point along the planes resultant vector. There is a detailed explanation in the help file here: PC-DMIS Help Center - 2020.2 (hexagonmi.com)

    Both NIST and PTB only certify the least squares fitting algorithms used by a metrology software, they do not currently offer validation of GD&T, although we are actively pursuing this with several vendors - NIST included.
Reply
  • Hi , as you have found, the way in which the geometric tolerance command fits datums differs from how traditional alignments function. In many cases it is not possible to construct a traditional alignment that exactly matches the DRF - particularly when there are datum patterns involved or when MMB/LMB are invoked. We provided the constructed primary datum plane, secondary datum line and tertiary datum point as a way of creating a traditional alignment that functions the same way as a basic DRF but it only really simulates the most basic cases - combinations of planes lines and points, no orientation constrained hole fitting. See here for a detailed explanation of how PC-DMIS fits datums: https://docs.hexagonmi.com/pcdmis/20...ves_Datums.htm
    Having said that, you may still be able to satisfy the datum validation requirements by reporting the position of the datums to themselves since this would be ​the same as aligning to the datums and then reporting their location. For example, If you had a pattern of holes (B) relative to a primary datum plane (A), you could report the position of the pattern of holes to A|B.

    In terms of finding the high point on a datum plane, the constructed primary datum plane should do exactly what you want. It simulates how a part would sit on surface plate and will always be outside the part (external to material). If you are working to ASME, the appropriate fit would be constrained L2. It constructs a least squares best fit plane through all of the hits and then shifts it to the high point along the planes resultant vector. There is a detailed explanation in the help file here: PC-DMIS Help Center - 2020.2 (hexagonmi.com)

    Both NIST and PTB only certify the least squares fitting algorithms used by a metrology software, they do not currently offer validation of GD&T, although we are actively pursuing this with several vendors - NIST included.
Children
No Data
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement