Xact measure datum shift error

[samus]

Xact measure should utilize the pcdmis size function when evaluating for datum shift instead of what appears to be l^2.



When measuring a casting that had an as cast datum id diameter that was 4.4300 +/- .020, the acutal measured feature was out of round and measured +.0218 and -.0235 on size, but when using the xact measure it calculated an l^2 (least squares) at 4.4286 allowing for .0186 datum shift. When the datum mating evelope is less than the lowest possible size you do not get datum shift. This error needs to be fixed in order to follow asme y14.5.

Has anyone else witnessed this or know if exact measure does in fact use l^2?

[JacobCheverie]

XactMeasure will use the evaluation algorithm specified in the feature definition. If you tell PC-DMIS to measure the feature as a LSQ (L^2) feature, Xact will use a LSQ fitting algorithm. Change the algorithm to a maximum inscribed/minimum circumscribed to adhere to Y14.5.

If you are using PC-DMIS 2020 R2+, the new geometric tolerance command allows you to change the fitting algorithm in the command rather than going back to the feature definition.

[bfire85]

^This

[samus]

MAKES SENSE NOW THAT YOU MEANTION IT, DIDNT THINK ABOUT THAT. I GUESS I WAS UNDER THE ASSUMPTION THAT "XACT MEASURE" WOULD BE MORE... EXACT.

[bfire85]

MAKES SENSE NOW THAT YOU MEANTION IT, DIDNT THINK ABOUT THAT. I GUESS I WAS UNDER THE ASSUMPTION THAT "XACT MEASURE" WOULD BE MORE... EXACT.

Haha!

[samus]

Ok... So what about if datum shift is applied say in the x+ direction to make one hole conforming but then in the next position it applies datum shift in the x- and now that hole is conforming too. Im pretty sure asme y14.5 doesnt explicitly state that you cant but it seems like common sense. Is there any fix or protection from that occuring?

[JacobCheverie]

Like bfire85 alludes to, Y14.5 does explicitly state that you cannot do that. Simultaneous requirements dictate that all tolerances using the same datum reference frame must be evaluated at the same time (using the same datum feature shift).

Refer to ASME Y14.5-2009 Para. 4.19.

[bfire85]

samus

Are the holes the same size? Then you can select multiple holes of the same size in the same feature control frame.

If the holes are different sizes, dimension them all individually, then go to Insert - Dimension - Simultaneous Evaluation and select all the FCF that pertain to that datum setup.

[samus]

Hmmm.... Thats good information. The engineering dept wont like this news...

[samus]

Do you know what formula the size function uses? Is it max inscribed / min circumsribed for uame and just the worst two-point evaluation?