MSA - CMM measurements are not used in a production environment

[s-bell]

First post to the Cove

We deal with Plastic Injection Moulded components and measure direct to CAD data using a CMM. However the CMM measurements are not used in a production environment but more so for ISIR and Annual recertification. Our Control Plans relate to MSA on in process checks but not to the ISIR process.

Considering we have approximately 50 measurement points on components, we do not wish to perform an MSA on each measurement point. At present we get around this by picking a point in X, Y and Z axis on the part and performing the study on these three points.

I think this covers the MSA requirements for R&R in that we are proving the gauge, person and part. However I am not sure whether this is the correct way to go. My gut feeling is that we should perform the study on each measurement point but this would be a lengthy exercise and I don't want to waste my operators time.

Does this seem OK to everyone, or does anyone have similar experiences they have learnt from.

 

[ben sortin]

Two suggestions:

Conduct the study on the characteristic with the tightest tolerance.

Conduct the study on the characteristic most affected by "injection profile."

Conduct the study on the characteristic most affected by "end of fill."

If all three criteria are met with one part then try for some tolerance relief (molding nightmare).

 

[Rob Nix]

Welcome to the Cove, s-bell!

I'm not sure I understand. You are trying to determine how to perform MSA on your CMM, yet you do not use the CMM to check parts in production!?! Ensuring the accuracy and repeatabilty of the CMM is important, especially if used to verify your product's capability in your PPAP, so verifying in all three axis seems OK.

HOWEVER, what are you using to check parts in production? Are all 50 dimensions critical? It seems the gaging you use for critical dimensions during production are the ones to do MSA on.

Then again, I haven't had my 2nd coffee and may have missed something.

 

[s-bell]

Our in process checks are based on vernier / dial indicator measurements on a separate gauge in most cases incorporating gap & profile checks. We perform MSA on each of these measurements as specified in our control plans.

The CMM measurement is performed on a specific gauge separate from production to facilitate A & B surface measurements for ISIR / Concern investigation. Therefore we are looking to get a R&R result on this gauge. My fault for not clarifying earlier.

I am trying to determine best practice for an internal procedure, do you think the following would be good for R&R in my case given that a 50 part MSA seems unrealistic.

1) Determine where the XYZ part location points are on the gauge.
2) Locate point furthers away from the X datum as measurement point 1
3) Same as above for Y and Z.
4) Perform the R&R on these points

My thoughts are that the points furthers away from gauge locators / datums are going to be most subject to gauge variance. Whilst not achieving the 50 points on the gauge we would cover the worst possible area's on the gauge.

It's late in the day for me now so if I'm confusing people appologies.

 

[Sam]

Our in process checks are based on vernier / dial indicator measurements on a separate gauge in most cases incorporating gap & profile checks. We perform MSA on each of these measurements as specified in our control plans.

The CMM measurement is performed on a specific gauge separate from production to facilitate A & B surface measurements for ISIR / Concern investigation. Therefore we are looking to get a R&R result on this gauge. My fault for not clarifying earlier.

I am trying to determine best practice for an internal procedure, do you think the following would be good for R&R in my case given that a 50 part MSA seems unrealistic.

1) Determine where the XYZ part location points are on the gauge.
2) Locate point furthers away from the X datum as measurement point 1
3) Same as above for Y and Z.
4) Perform the R&R on these points

My thoughts are that the points furthers away from gauge locators / datums are going to be most subject to gauge variance. Whilst not achieving the 50 points on the gauge we would cover the worst possible area's on the gauge.

It's late in the day for me now so if I'm confusing people appologies.

Your PPAP (ISIR) should come from a production run. Therefore your MSA should be done on those gages used in production.

 

[Howard Atkins]

I am trying to determine best practice for an internal procedure, do you think the following would be good for R&R in my case given that a 50 part MSA seems unrealistic.

I think that this thread is confusing the 2 seperate points.

1. The requirements for ISIR/PPAP- the control plan referenced gauges etc. Here again the gauge needs to be checked and iy you use the same caliber for 5 measurements then you do not need to check it on each measurement , just approve the gauge and operator once.
2. Internal procedure- it is very good to do the same on non PPAP required equipment but again one dimension or type of dimension for each gauge.

 

[JohnR]

First post to the Cove

We deal with Plastic Injection Moulded components and measure direct to CAD data using a CMM. However the CMM measurements are not used in a production environment but more so for ISIR and Annual recertification. Our Control Plans relate to MSA on in process checks but not to the ISIR process.

Considering we have approximately 50 measurement points on components, we do not wish to perform an MSA on each measurement point. At present we get around this by picking a point in X, Y and Z axis on the part and performing the study on these three points.

I think this covers the MSA requirements for R&R in that we are proving the gauge, person and part. However I am not sure whether this is the correct way to go. My gut feeling is that we should perform the study on each measurement point but this would be a lengthy exercise and I don't want to waste my operators time.

Does this seem OK to everyone, or does anyone have similar experiences they have learnt from.

We have parts with a similar situation. ON our drawings we specify about 75 measurement points on a door beam. These coordinates are what we call "LEVEL 1" Points. From those points we have determined based on fit, function and critical characteristics anywhere from 5 to 9 points that we call "LEVEL 3" points.

The Level 1 points (75 XYZ points) we measure on the CMM for a full blown PPAP and we measure 6 pieces of those.
The Level 3 points we measure on the CMM during prototype & measure 20-30 pieces of using those points.

In the meantime we have a gage designed that can picked up the coordinates of those 5-9 points and that way we can monitor the SPC and capability of those "key characteristics" during production. Once the gage is completed & Certified we use this gage for the 100 pcs capability study required for PPAP and continue to moniter just those Level 3 points during production. The only time we go back to check the Level 1 points is during a yearly part layout or if we re-cut a tool and re-PPAP it.

Does that make sense?

John

 

[s-bell]

Makes sense, very similar to what we do, our Level 1 measurements are the ISIR CMM points, with our level 3 points the in process check.
From what I am gathering from the posts, I am OK in performing a gauge R&R on three points on the part, XY&Z. With additional R&R checks on in process work. This means that we verify the CMM and in process gauges and therefore meet all requirements for acceptable gauging. Thanks