Disagreement CMM operator - QE 2

[gabimo]

One of our QE claims that in order to get the “correct” datum displacement when parts are measured on the CMM the secondary datum feature (B) is to be measured twice.
Details: Parts similar with 7-24 or 7-26/ 2009
On the initial base alignment, A primary datum feature and B secondary are measured as shown on the drawing (similar with the pictures from the standard). The size of datum feature B is qualified and reported.
Then the size and location of the thru holes are measured. –same base alignment is used .
- On the bonus side we have no disagreement.

- On the datum shift/ datum displacement (from B at MMB) the disagreement is because of how the size of B is measured (UAME versus RAME as explained by the QE, but not understood very well by the CMM programmer and I, as well)

As we were told, B “should” be measured twice -- base alignment (coordinate system on the CMM) shall be changed in such of way (what exactly is the practical method, not understood) that size of B is to be measured perfectly perpendicular to the datum A---

Is the QE correct? And if yes, then why the same feature should be re-measured? Any changes in the CMM algorithm that can be done to avoid redundant feature measurement?

Could you shed some light on our dilemma for the theoretical side and also on the practical side of things?

 

[greenimi]

http://www.cmmquarterly.mbccmm.com/index.php/past-articles/172-apply-datum-shift-in-cmm-software

 

[3DDave]

I like how the article said MMB was a new concept with '1994, rather than just a new name for an existing concept of virtual condition.

Anyway, it seems like it should be measured twice - once to determine if it meets the size requirement, which is independent of orientation, and second, in the orientation of the primary to determine it's condition as a datum.

It's up to the CMM software to determine if the second round of measurements need to be physically done or if the captured coordinates from the first round can be transformed to be used for the second purpose.

 

[pmarc]

gabimot,
You may want to take a look at fig. 1-1 in Y14.5-2009. I know that the geometry that you are dealing with is different but the figure is good in showing that there are different "sizes" of a datum feature.

In your example I assume you have a perpendicularity callout applied to datum feature B wrt A (otherwise there is no way to calculate datum feature B shift properly). So what you actually need is assessment of datum feature B against 3, not 2, different purposes (as 3DDave just mentioned, it does not mean that you have to make multiple rounds of measurements - first round can be utilized for other purposes):

1. Actual local size - this is so called two-point measurement and verifies whether the feature is within its size limits.

2. Size of Unrelated Actual Mating Envelope, UAME - this one is used to verify whether datum feature B conforms to Rule #1 and additionally it could be used to calculate available amount of bonus tolerance for the mentioned perpendicularity callout, if applied at MMC.

3. Size of Related Actual Mating Envelope, RAME - this one is used to figure out available amount of datum feature B shift. The difference between MMB size and actual size of RAME divided by 2 is the available radial amount of datum feature B shift.

 

[AndrewTT]

pmarc, why can't you simply use the UAME to determine if the feature complies with the size limits? If you use the actual local size to check if the feature is within its size limits then you should take multiple measurements. How many measurements are needed to satisfy you that the feature meets the size requirement? The UAME takes the entire feature into account. Why not just use this measurement?

 

[greenimi]

How many measurements are needed to satisfy you that the feature meets the size requirement

AndrewTT,
Let me ask you a simpler question: in how many points you measure a simple cylindrical pin to make sure the size requirements are meet? The theory says that "all points" should be within the size requirements. My asnwer is: control plan/ risk management/ risk mitigation. Your answer:?

The UAME takes the entire feature into account. Why not just use this measurement?

A gage is designed using the VC condition (4.5.2./ 2009)-basic location relative to other datum feature simulators-- and you might want agreement between your functional gage and the CMM measurement. (both inspection techniques to accept/ reject the same part).
So, the gage elements, are perfectly ( perfectly = in the gage manufacturing tolerances) oriented relative to one another, hence RAME should be used.

 

[AndrewTT]

My answer to taking into account "all points" of the feature is to use the UAME.

My question was for measuring the feature to verify if it complies with the size requirement only. I was not talking about gaging/verifying the orientation or location of the feature. I was only interested in how the size of the feature was being measured, that's why I was speaking of using the UAME and made no mention of RAME.

 

[greenimi]

Then I will copy-paste from pmarc's post:
"1. Actual local size - this is so called two-point measurement and verifies whether the feature is within its size limits.

2. Size of Unrelated Actual Mating Envelope, UAME - this one is used to verify whether datum feature B conforms to Rule #1"

Two points measurements must also be taken because UAME will NOT fulfill the print requirements. (lets say a hole or a pin toleranced with a direct toleranced ± dimension)

The question was "how many two points measurements" ---pmarc's #1 assessment---- will you take?

 

[greenimi]

Sorry pmarc that I have jumpped into this conversation.

Could you, please, provide some answers to AndrewTT (if the ones I have tried to offer are not quite correct/ relevant on what he is asking).

Again, sorry and thank you pmarc

 

[powerhound]

why can't you simply use the UAME to determine if the feature complies with the size limits?

Because this may not tell you that the actual local size has violated the LMC limit. Actual local size is part of a size requirement.



John Acosta, GDTP Senior Level
Manufacturing Engineering Tech

 

[AndrewTT]

thank you powerhound, that makes sense.

 

[gabimo]

Thank you pmarc for the clarification.....excellent rundown and summation.

 

[pmarc]

You are welcome, gabimot.

Also, my apologies for no reaction on AndrewTT's question. Fortunately, powerhound hit the nail on the head.