Recall Feature Points from a Single Circle Path of a Cylinder, High Volume (Loop or Pattern)

[[Ra...]]

It is an absolute shame. I still can't get it to work to be honest. I've tried right clicking everywhere.

 

[[He...]]

Please sign in to view this quote.

Very strange. I'm doing the same exact thing and it works just fine. 

I have tried it with 6.8.2002 and 6.8.24.

What service pack are you using?

Edited Tuesday at 02:00 PM

[[Ra...]]

6.8.08. Maybe I need to update. I mean I might as well. I'll report back if IT has time today.

 

[[Iv...]]

I'm running 7.8.12, and I can access the formula option in the Recall Feature Points. @Henrik Lindahl, I think your idea is great and I'm verifying it now. Unfortunately, I encountered an issue where my cylinder list starts at "106" instead of "1". I'm not too familiar with the PCM coding, so does anyone have any recommendations on how to have the loop start at 106? I tried variable functions/combinations within the list formula, but I could not get the loop to start at "106" properly. Any tips? Should be fairly easy.

Also, what coding language does PCM most closely follow? C++, Java, Python?

[[Ma...]]

Please sign in to view this quote.

I would say none of them. But backstack is closer to Java.

[[Iv...]]

Unfortunate 🙄 I have a light background in C++, so that's what I was hoping for. Thanks nonetheless 

[[Ma...]]

Please sign in to view this quote.

Python is not using brackets like here, but also is using indentation to separate levels ( like for loops )
C++ and Java are using brackets, but you have to write full definitions if, then, else.

PCM looks for me more like a VBScript from Microsoft.

 

[[He...]]

Please sign in to view this quote.

Do you have any gaps in your names or does it run uninterrupted from 106 to 2000+?

If you have an uninterrupted series it should work if you change "Cylinder" +LOOP1 to "Cylinder" +(LOOP1+105)

The brackets at important in 6.8 and I think it is the same in later versions. I'm not at my computer right now so I can't verify this but I think it will work. 

[[Iv...]]

Thanks, Henrik! That worked perfectly. I originally tried adding that "105" index outside the loop, so it makes sense why my combinations did not work 🤦🏼‍♂️. But I have my cylinder interrupted in quarters—cylinder 106 to 696, 3065 to 6025, etc. While 2000+ already seems like a lot of cylinders, the actual part has about 9000 total cylinders. We are only completing a 25% sample coverage 😆 super fun stuff! 

Due to these interruptions and my limited knowledge in PCM. I planned to have 4 different "CylinderX's" so that I could capture each segment. But happy to hear if you have any recommendations to account for the interrupted series in the loop, I'd be interested. 

I'm currently implementing your suggestion into my program. I will update you guys shortly! Great thinking on the formulas in each nominal for the CylinderX feature. I mistakenly thought these were highlighted in yellow since the recall feature points were a formula. Nice touch! 

Thanks again for the help so far!

[[He...]]

Please sign in to view this quote.

Do you have PCM?

Is there any other cylinders in the program that you do not want to recall?

What the hell is this part containing almost 10k holes? How long does it take to measure? I thought my parts could be crazy with 1000-2000 holes but this is insane. 

Edited 20 hours ago

[[Iv...]]

I have PCM, but I don't have the skillsets to use it effectively 😭

Aside from the at-interest cylinders, there is one other cylinder that I do not want included in the recall. 

Yes, around 9600 holes total. To measure the 25% sample, it takes about 26 hours. In this particular case, we are dealing with through holes that are about .030" in diameter but 0.500" long!!! So I use two 0.5mm styli, inspect 50% length with the +X stylus and the other 50% with the -X stylus. Not ideal, as far as timing goes, but because of the size and length, I could not get a working stylus that could inspect the full length in one shot. This two-styli process approximately doubles the time since for each hole it needs to go up and over the part. Tangent, but a PCM script that could make the measurement plan measure all +X cylinder portions first and -X cylinder portions last would be killer. Albeit, it would add a lot of concern to processing time on the backend. The first hole would not be fully complete until all the other 2000+ holes are measured with the +X stylus. 

BUT!!! I was able to validate your theory, and it does work!! As mentioned earlier, through my SMA, I did get a working PCM script that was able to recall circle paths. But even the ZEISS application engineer was not thrilled with the script that was needed, since it was a little clunky. Your process is exactly what I envisioned when I first started this conversation. Thank you so much for the help!!

[[Iv...]]

Seperate issue, but when generating the PiWeb results, I've been hit with this System Error that halts the PiWeb results from saving/opening. Just in the last couple of days, I've seen it across multiple workstations and programs. I searched the forum and couldn't find any other tips to resolve this. 

And in case you wonder, how I was able to verify the process without results. This error only blocks the PiWeb report; the Tables Files are still exporting out. Because these parts are so data-dense, we use a combination of the Table Files and PiWeb reports to analyze the reporting. 

[[He...]]

Please sign in to view this quote.

It is not possible to separate measurements within single features. The feature would have to be separated in two different cylinders and then recalled into one. But that is a lot of work for a part like this. 

The PCM to create a list with gaps is quite simple. I will try to do it tomorrow or the day after that. 

[[Ch...]]

Ivan,

 

 Heck of a job you have there - I would personally split up the cylinders to measure all one side first then the other, would likely produce slightly better repeatability. Anyway, my hat is off to you on this part. Like others said you can recall into a cylinder later and usually makes analysis cleaner - for me anyway at the end.