ORIENTATION

[[Je...]]

Please sign in to view this quote.

Zoltan,

You make valid points about asking "why" is a start alignment necessary and how programming for cnc machining is similar but not identical to metrology programming.

A start alignment is not a requirement. It's an added help to decrease the chance of alignment error. It's a poka-yoke (mistake-proofing tool). Start alignments have proved value-added for me, so I've made them a standard part of my workflow.

Perhaps a reason why alignments in metrology software like Calypso behave differently than alignments in CNC machining is because CNC machining can assume a fixed normal for the cutting tools based on spindle orientation. In CMM programming, although the sensor head may be fixed, the probe can vector from a variety of space planes, including underneath the part. Also, CMM programming is focused on evaluation of characteristics. This happens both at the machine and offline. The versatility needed to do this requires that alignments can be fully rotated, even inverted occasionally.

I know this doesn't alleviate your current challenges. I hope I can simply encourage you that it does get better. Over time and with practice you become fluent in Calypso. Muscle memory builds and success rates climb. Sure, there will still be the occasional head-scratching issues, but that's what this forum is for and the talented application specialists at Zeiss. And don't forget about stress eating from the vending machines... it's amazing what a good Reese's can do for dopamine levels 😂

[[Ni...]]

Please sign in to view this quote.

Was this issue fixed?
I believe the reason your alignment is flipped is because you're using -X for the Rotation in Space. According to your images, the plane is on the Nut side which should be +X.

As long as you're not measuring the nut itself, there is no reason to have planar rotation. If we ignore the nut on the right side, the part is symmetrical and round. You can rotate it in any direction inthe X axis and it will not matter. The program will still run.
So the planar rotation shouldn't be the issue.
I believe your issue is that you're using -X instead of +X for your Rotation in space.
Change that and then run it using a manual alignment in simulation to see the results right away. Or do it on the CMM but run it slow.
Let me know if it fixed it.


Also regarding a separate conversation others are having about Circles vs 3D lines for Planar Rotation, I've done circles for planar rotations and then 3D lines to that same circle and it always shows no difference.
The one example someone posted about getting bad results when using circles in a threaded bore, I believe the issue is that threaded bores are random when you use a single circle. Needs a cylinder. But in that case, 3D line would be more stable I believe.

[[Ri...]]

Please sign in to view this quote.

Calypso is easy when it comes to alignments and I`ve ran some old software in my day.
I have said this many times on this forum.
I use the Base Alignment to describe where the part exists in the volumetric area of the CMM, Base Alignment features are never used for measurements of any kind, they are as simple as can be and I prefer a 3-2-1 or similar alignment wherever possible.
Base Alignment features do not have to be any kind of Datum.
I create Base Alignments that are simple to probe on features and provide the minimum of 6 degrees of freedom. Basically, a Start Alignment without having to deal with a start alignment, just skip over it.
Then I create an Alignment of actual Datums that all features adhere to. I coined this the "CNC Alignment".
Example "CNC Alignment A_B_C".
Once the CMM starts in CNC mode, it runs thru the Base Alignment features and then the CNC Alignment.
This method is robust and after 20+ years of using it, it has never failed to properly align a part.
This is how I was taught in my first Zeiss class back in 2001. :shrug