Alignment with Datum Targets and Geometry

[[Cl...]]

Please sign in to view this quote.

Similar to the best-fit iterative alignment used in Virtual DMIS for non-prismatic surfaces.

[[Cl...]]

Please sign in to view this quote.

No picture 4-50?

[[To...]]

Please sign in to view this quote.

There is a Figure 4-50 on page 88.

[[Cl...]]

Thank you.

[[Ja...]]

Please sign in to view this quote.

I am thinking the same as Clarke but offset the alignment in +Z by .4808. That would give you the theoretical axis based off those 4 points.

[[To...]]

Please sign in to view this quote.

Unfortunately, this does not capture maximum condition of the 2 geometry surfaces, but might be an acceptable solution.

[[To...]]

Please sign in to view this quote.

As a Primary datum, there should be no limit to rotation or translation. That being said, I would guess that rotation and translation on the X axis would not take place or be minimal. I agree, testing is necessary. Too many tests on my plate right now...lol

[[Ri...]]

Please sign in to view this quote.

I agree with your statement. My statement just comes from some of the "wonkiness" I've seen with Curves when creating alignments with it allowed to translate/rotate in all 6 axis.

[[Pe...]]

With v-blocks the part will always sit on the four highest points, regardless of variation in diameter. One of the troubles with using four single points to simulate v-blocks is you never know if you're really hitting the high spots. I've had some success in the past scanning four 2D-lines set to outer tangential with constrained normal vectors. Then use theoretical features with formulas referring to the 2D-lines to build up a 3D-line feature that represents the -A- axis.

[[An...]]

_

[[Ja...]]

Question: If you have your base alignment setup using datum B and datum C. Would it not suffice to just use the GBF you created without needing to make secondary alignment "Alignment1"?
To my understanding, any DOF not constrained from GBF features will fall back onto the alignment selected (base alignment in this case)
The problem with having these four points is that you cannot constrain Z translation DOF unless you consider the nominal diameter as basic dimension for alignment purpose or you use a cylinder made from two circles at the cross section of those 4 points. Meaning that your Z origin will differ between the methods depending on difference between actual diameter and nominal.
If using two circles, datum B and datum C, you can use RPS321 alignment with looping but I am not sure if that is the right way to go about it. I used this method for a similar part but I had single points for datum B and C. I did try using 4 points instead of circles for looping but it just kept getting worse with each loop.