Reference Standard for Dimensional Control

Post Reply
Koutinho
Frequent User
Frequent User
Posts: 33
Joined: Wed Mar 18, 2015 5:11 am

Reference Standard for Dimensional Control

Post by Koutinho » Tue May 26, 2020 5:26 pm

Good Evening for all,

I hope that everything is fine with all of you and your family and that we can overcome the current situation as best as possible.

I'm trying to develop an internal procedure to perform normal measurements like distances, diameters and features like this.

Anyone know a reference standard or any document that have some guidelines about this?

My questions are pretty simple but difficult to answer.

For instance, how many points should i use to evaluate a circle? 3? 10? 50?

Best Regards,

Koutinho

User avatar
CrashN8
Site Admin
Site Admin
Posts: 546
Joined: Thu Feb 09, 2017 4:07 pm

Re: Reference Standard for Dimensional Control

Post by CrashN8 » Wed May 27, 2020 9:54 am

I would look into artifacts used during ISO 17025 CMM Certification. There are a few known artifacts out there that have assortment of measured feature types - i.e. Sphere, Cone, Planes, Cylinders, etc. As far as measured points, I would measure 10-12 points for circles, no more than 15-16 on 3D feature types. Position and size won't vary much by adding so many points while the form will slowly grow towards the CMM's accuracy limit as you add more and more points to each feature measurement.

Koutinho
Frequent User
Frequent User
Posts: 33
Joined: Wed Mar 18, 2015 5:11 am

Re: Reference Standard for Dimensional Control

Post by Koutinho » Mon Jun 01, 2020 5:35 pm

Good Evening!

Thanks for your feedback.

What do you pretend to mean by "artifacts used during Cmm Certification"?

User avatar
CrashN8
Site Admin
Site Admin
Posts: 546
Joined: Thu Feb 09, 2017 4:07 pm

Re: Reference Standard for Dimensional Control

Post by CrashN8 » Tue Jun 02, 2020 9:57 am

Could look like this... several companies make one, they all look a little different. Main point is you measure it once as a reference then repeat measurements at later time to ensure stability of CMM performance over time.

17025 artifact.jpg
You do not have the required permissions to view the files attached to this post.

User avatar
CMM Guru
Moderator
Moderator
Posts: 98
Joined: Thu Feb 24, 2011 7:49 pm

Re: Reference Standard for Dimensional Control

Post by CMM Guru » Tue Jun 02, 2020 11:55 am

Simple artifacts such as - ball bar, step gauge, gauge blocks, and ring gauges.

Nobody has created an "official" standard for how many points or placement of points. It depends on your use case and needs.

The technical answer?

Based on your requirements for tolerance and the various tolerances of the machine, probing, and artifacts used, you need to use sufficient points to ensure that your measurements meet the requirements of your specification and that they are sufficient to be repeatable and meet the statistical stability requirements.

So -- this gets really difficult. We have test blocks that are meant as functional testing only. When I had them made, I said "Make them however the machine makes them, and we will then inspect and make a list of actual sizes after we get the artifacts". They are typically within 20 microns. For demo purposes, 3 points on a circle (6 on a cylinder) are more than sufficient. If I have a part with a cylinder with a very tight tolerance and, in particular, a very tight position/perpendicularity/cylindricity tolerance, I would not depend on 2 "circles" of 3 points. I might do 5 "circles" of maybe 7 or more points. Maybe more. Maybe less.

Sorry the answers are so vague.

*At a minimum*, you should use at least the minimum number of points to mathematically define the feature +1 so you can get form (although incredibly unreliable with just one extra point).

Post Reply