Effective Training Inc. Message Board

I am requesting some opinions on analyzing CMM inspection results.

We are using a very high end CMM software tool that takes into account modifiers on datums of size (MMC, LMC) when reporting results (ie datum shift). There are not many CMM software products on the market that currently do this. When reporting positional or profile results of feature control frames that reference datums that fail their size requirements, the results will always read FAIL regardless of the numerical results.

There is concern in our organization that inspection results labeled “FAIL” solely because the datums fail their size requirement (that would otherwise pass when the datums are at RFS) is misleading and politically dangerous – especially when sharing the reports with customers.

I am told by the software supplier that this has been this way since the beginning and they are reluctant to change.

ASME Y14.5M-1994 does not address this issue - Any opinions?

Hi JRA,

Your question addresses a topic (datum shift) that is not fully understood by many people. First I must preface my response by saying that without seeing the drawing, and your inspection report, and without knowing what CMM software you are using and how it calculates datum shift, I cannot address your situation fully and accurately. I can only speak in general terms that apply to most situations.

You are correct in stating that many CMM's do not have the capability to calculate datum shift correctly. Therefore, it is often ignored by most CMM operators. They treat position or profile tolerances as if the datums are referenced RFS instead of MMC and do not include the datum shift when measuring the part. Since datum features referenced at RFS are more restrictive than MMC, you are allowed to measure a part using RFS to guarantee that you will not pass a bad part. You will obviously have a greater chance of rejecting some marginally good parts that might otherwise pass if the datum shift was applied. There is no rule that says you must use all of the tolerance. You are permitted to use less tolerance if you wish. Therefore, your customer should not object if you wish to treat all MMC datum references as if they are RFS to get valid measurement results to verify that the toleranced features are good and that only the datum feature size is out of tolerance.

A word of caution when applying datum shift. Datum shift is complex, it is not as simple as taking the difference between the actual measured size (actual mating envelope) of the datum and its MMC size and adding the difference to the position or profile tolerance. (AME - MMC = datum shift). Depending on the geometric relationship of the toleranced feature to the datum feature the datum shift may not apply to the toleranced feature at all. Also, if several feature must be gaged simultaneously (simultaneous requirement rule) there will not be any datum shift between them. For example a pattern of holes or multiple features or patterns of features with position or profile tolerances that have the same datum reference frame (same datums, same sequence, same modifiers). The datum shift occurs between the hole pattern and the datum feature not between the individual holes within the pattern. In other words all hols must move in the same direction. That is why datum shift is very difficult to calculate and why most CMM's do not have the capability. The CMM software must create a virtual gage and manipulate the part geometry on that gage so all features move together on the gage to see if all of them will fit their tolerance zones at the same time. Because of this complexity when datum shift is included in the part measurement it is often miscalculated and misapplied and the inspector ends up passing bad parts. Therefore, unless you do have good CMM software that you know is calculating and applying it correctly, I recommend that you treat all MMC datums as if they are RFS to guarantee that no bad parts pass.

Therefore, to get the full benefit of datum shift it is best and easiest to have a functional gage that you can put the part on. If it fits it passes. The gage can take full advantage of datum shift and most CMM's cannot, this is why there is often a descrepency between CMM results and functional (hard) gaging. This is also why many company inspection policies specify that the functional gage is the final arbiter of part acceptance.

If you want to send me an e-mail, I will be glad to send you a simple picture that helps clarify what I explained above.

This concept is fully explained in our Fundamentals and Advanced Concepts of GD&T classes. If you have not already done so, I encourage you to sign up for one of our classes.
Also, we will be glad to consult with you to review your part drawing and inspection report to verify that all geometric tolerances are interpreted, measured, and reported correctly.

_________________
Best Regards,
Roy Cross
ASME GDTP Senior Level S-0488
Dimensional Engineering Mentor
Effective Training, Inc.
2118 S. Wayne Rd.
Westland, MI. 48186
royc@etinews.com
http://www.etinews.com