Composite Tolerance

I am having trouble understanding composite tolerance, especially when it comes to a bolt hole pattern like this. I am wondering how you would dimension these true positions. Here's my idea using legacy dimensioning:

Level and translate to A, rotate to B, and report the position of 1 hole (any hole?...lets say the 12'O'Clock hole) and give it a .001 tolerance. Then, would I stay leveled to A&B, translate to the hole that I just reported on (12'O'Clock hole), and report the positions of the rest of the holes with a .0005 tolerance?

Thanks so much...I prefer to use legacy...

Neil Challinor over 2 years ago

Lukas No, that isn't the correct approach. It's pretty difficult to validate composite position callouts using legacy because you need to perform various best-fit alignments - and even then, it's not always possible to fully achieve the design intent.

This previous post should explain things: https://www.pcdmisforum.com/forum/pc-dmis-enterprise-metrology-software/pc-dmis-for-cmms/517387-bolt-hole-with-composite-tp?p=517437#post517437
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
Lukas over 2 years ago

Hm...I'm looking at Xact measure and it seems really straight forward - it does all the work for you. Anything I need to know with that method? Do you think my interpretation of the ASME standard is way off (referring to what I laid out above)?
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
Lukas over 2 years ago

Is my interpretation of the standard wrong because the FRTZF circular tolerance zone must lie within the PLTZF circular tolerance zone (not overlapping the border of the PLTZF zone)? Please see attachment...sorry I'm really just asking about GD&T here

Attached Files
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
Schlag over 2 years ago

You need a best fit alignment for the 2nd layer. You would never choose any " 1 " hole and use it as a 0.0 point. You are looking for the pattern and not related to any 1 " random " hole you call 0.0.
Look at pages 224/225/232 in the 2018 Y14.5 as it shows similar examples of your call out.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
Krisj over 2 years ago

From my understanding the lower segments of composite only constrain rotational degrees of freedom. To check the lower segment create a feature set out of the holes, create a best fit alignment using the feature set allowing translation in the X and Y (assuming Z is looking at the holes) and set the best fit method to min/max. After the alignment redo the position of the holes. Technically you can also allow translation in the Z but you can't do a min/max best fit in 3D for some reason and least squares evaluation may not give you the best result possible
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
NinjaBadger over 2 years ago

For top segment, level and origin to A, rotate to B.
report position of all 6 holes.

For lower segment, use the above alignment, then just best fit, 2d translation only to the 6 holes, then report the position of the 6 holes
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
Mike Rivard over 2 years ago

What version of PC-DMIS are you using?? If its after 2020r2 the software does it for you?? someone please correct me if I'm wrong.

Attached Files
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
NinjaBadger over 2 years ago

No you're correct, but he said he prefers to use legacy
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
kingsld1 over 2 years ago

Also, the lower frame should only have 1 datum. As written, the FCF is basically just repeating itself, with a tighter tolerance on the lower frame.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
Henniger123 over 2 years ago in reply to kingsld1

Thank you, I was wondering why there are two tolerances at all when a and b are the same
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel