Coordinate System Placement: Where (or how)

[[Ra...]]

 

Hello everyone,

I am writing to request your expert assistance regarding a complex metrology challenge.

My experience has primarily involved programming inspection routines for standard rectangular or square components. However, I am now tasked with a part that presents a higher level of alignment complexity due to its geometry.

Our current customer requirement is to accurately inspect specific inner radii and several hole locations. I am seeking confirmation on the best approach for establishing the coordinate system alignment for this specific component geometry.

I have included an images of the part showing the customer-specified positioning.

If you were tasked with creating the measurement program, how would you approach the alignment? Specifically, I would greatly appreciate guidance on the following points:

1. Coordinate System Placement: Where (or how) is the most effective place to establish the Coordinate System Origin ($X, Y, Z = 0, 0, 0$) to facilitate the inspection of the required features?

2. Alignment Features: Which specific geometric features on the part (e.g., planes, cylinders, points) would you use to define the primary, secondary, and tertiary axes of the alignment?

Any feedback, suggestions, or ideas regarding the alignment methodology would be highly beneficial to me.

Thank you very much for your time and expertise.

Sincerely

[[DW...]]

Please sign in to view this username.

 If you don't have a 3d file available, you are more than likely going to need to have one made from a drawing. The next step would be to use the CAD file to create a reference point system (RPS) alignment. This is a common alignment system used in the auto industry for things like body panels that lack standard geometries. There are threads on here about RPS alignments in Calypso. I haven't used RPS alignments all that much in Calypso, but I do remember that looping the RPS alignment at least 3X is important. Hopefully someone with more RPS alignment experience chimes in. Best of luck sir.

[[Ra...]]

Thank you, I do have a 3 cad file. 

[[Ma...]]

First of all - i would do some basic determination how much are holes variable in position.

For alignment - i would use start alignment for manual location.

For base alignment - i would use sperical probe, but touching by shaft ( like using cylindrical ) just around and then touch in middle.

Then using 9 points for cylinder ( 3 circular paths ).
If thickness of material with combination of differences in radius can matter to correct scan, then scan area around to obtain correct origin of circle/cylinder ( depending on scan path )

[[No...]]

If the part and hole pattern is symmetrical and the accuracy needed allows it, you could measure the four bores in the outer corners as circles, together with 4 circular planes very close to the edge of the hole. Then make a perpendicular between each hole and plane and intersect them with the planes. That would give you four stable points to be recalled to a big plane, which could serve as the spatial alignment.

 

[[Ch...]]

Based on Nobert's idea, I would create a rough outline of the launch system.

Is the plate a cylindrical piece? If so, this would be a rotation in space, any outer edge as a planar rotation...

[[Ri...]]

To help me understand my points, I make them symmetrical over the entire surface used.
If you randomly select points, its more difficult to determine what you have compared to symmetric points.

I can't recall the post, but I uploaded a program demonstrating this technique with a part that has zero normal surfaces.