Alignment to match Datum Scheme of Drawing

[Inspector_Gadget]

I am going to attempt to write a macro that aligns multiple parts to a single CAD reference file per the drawing data scheme.  I have two strategies that I am considering:

1. Create features on data_1, do an auto extract, plane/line/point alignment, then try to transfer the auto-extract option to data_2 and perform the alignment again.

2. Use an alignment for each datum, using ignore for alignment on non-datum surfaces and the best-fit alignment with specific constraints for each datum.

I am running into some dead ends with both options.  I dont know how to transfer the measured features from one data to the next.  Also I dont know how to automate the plane/line/point alignment very well.  For the second one, I am not sure how I am going to automate the constraints.  Maybe I could calculate the normal vector from the surface or something.

Can someone point me in the right direction if they have an experience with doing something like this.

Thanks in advance.

[prehistory]

Hi,

It totally depends on where the data objects are to begin with.  Are they all in different positions, or are they all in the identical place?  As I understand it, this is how I would approach the problem.

1. get each of the models really close to the CAD using a Best Fit (with automatic initial guess used).  If the data objects are in various locations, you will need to do this iteratively- that is, ignore data_2 while best fitting data_1, then ignore data_1 while best fitting data_2.

2. now you are ready for some sort of align refinement.  If you have b/p callouts for an alignment scheme, you must use that.  Probably it will be a datum alignment like plane/line/point.  If it is not plane line point, n pairs of points, or 3 perpendicular planes, please let me know and I will post specific instructions for such a case.

3. have your alignment datum features defined on your CAD (as nominals). In the feature properties, set the Auto Extract parameters properly.  For instance, a plane might use the Best-Fit technique, with a max angle of 25 degrees, etc.

4.  With your data_1 best-fit to the CAD, you can ignore the other Data object, and auto-extract the measured features from data_1 and do an alignment based on them.  (The trick with any of these alignments is to actually do them, then copy the Command History to the macro).  Make sure to set the plane axis point alignment sequence to reflect the b/p (e.g. datum A, B, C, in that order).

5. Lock data_1

6. With your data_2 best-fit to the CAD, you can ignore the other Data object, and auto-extract the measured features from data_2 and do an alignment based on them. These new measured features will replace the old ones from data_1.

7. If you want to keep both sets of measured features, you'll need to duplicate the alignment features...

Am I on the right track here?  Let me know if you are trying to do something else...