Curve Filtering Knowledge Dump

[[Je...]]

Based on the attached image, tell me what comes to mind regarding filter settings and other settings for closed curves.

What are some important things to keep in mind when evaluating the profile of a small, tightly toleranced closed curve?

Requirement:  You are not allowed to say the word "Cookbook" in your response.

 

 

 

Filtering and Wavelength.pdf

[[Ma...]]

I think 0.25 is ideal, but it all falls to what you need to get out of it.

But that jump from line to radius is strange. Is it closed loop or just scanned L shape? If closed, did you clicked on "close loop" on curve feature?
You can also try to use Nom -> Act if your actuals are more dense than nominal points.

[[To...]]

How to determine the proper wavelength is the real challenge.  In the early days, Calypso defaulted to some value that I don't recall, but now it seems to automatically adjust based on the number of point (but this is only a guess).  I tend to look at the wavelength setting and make sure I am collecting at least 7 points per wavelength, because if you don't have enough points, the filter won't be applied.  There are some really smart people out there that believe 21 points per wavelength is the best scenario. 

An old forum friend, whose name escapes me but I seem to recall his initials were HK or KH, created a series of plots in Planner to show the effects of various filter settings.  One has 0.0 dispersion and the other has 0.05 dispersion. 

 

Curves with 0deviations.pdf Curves with simulated deviations - 0.05.pdf

[[Je...]]

Please sign in to view this quote.

Thanks, Tom!  That is useful feedback.  I appreciate sharing the simulated plots.

[[Je...]]

🤔     Moderators:

I see you moved this post into the curve subtopic.

This thread is as good as dead now.  It's like putting a fast-food drive thru on a dead-end gravel road in the wilderness.

Very little traffic.

I understand that this was done for the sake of organization.  However, a more effective modern solution is to attach tags onto forum posts rather than include many subtopics.  If the quantity of unique visitors to the Zeiss forums was approximately 10-15x its current number, subtopics would make sense.

However, in the current situation they are wholly ineffective.

Edited Thursday at 09:02 PM