Gage Measuring 15 Parts at a Time

[Ranjit B]

Gage Measuring 15 Parts at a Time, so should we use Standard AIAG form with 10 Parts measurement to validate Gage?
Should we scale up our measurement samples to 15 pcs?

 

[Miner]

I recommend performing an expanded R&R study, which would allow you to include nest to nest variation as an additional factor. I use nests to designate the 15 different measurement positions. This would separate the nest to nest variation from the part to part variation allowing you to see whether action should be taken to reduce nest to nest differences.

 

[Ranjit B]

That is really helpful, Miner
Do you have a template to perform expanded study?
You got my point right as there are 15 nest to measure parts.

 

[Ron Rompen]

There are a couple of things about this gauge that concern me.....and I'm not really sure how to address them.

1) Should an R&R be performed on each nest? I don't think it is required, in the same way that you don't do an R&R study on each caliper in your shop - they are considered a 'family' of gauges.
2) Should parts be moved from nest to nest during the measurements? This will affect both repeatability and reproducability. However, I'm a little concerned that this movement of the parts from nest to nest may end up MASKING some issues. I would definitely defer to someone (like Miner) who has more extensive experience with something like this.

 

[Miner]

That is really helpful, Miner
Do you have a template to perform expanded study?
You got my point right as there are 15 nest to measure parts.

I do not. Minitab supports the expanded R&R, but any software that can perform an ANOVA can be used to duplicate it. This is a link to the Minitab method.

 

[Miner]

There are a couple of things about this gauge that concern me.....and I'm not really sure how to address them.

1) Should an R&R be performed on each nest? I don't think it is required, in the same way that you don't do an R&R study on each caliper in your shop - they are considered a 'family' of gauges.
2) Should parts be moved from nest to nest during the measurements? This will affect both repeatability and reproducibility. However, I'm a little concerned that this movement of the parts from nest to nest may end up MASKING some issues. I would definitely defer to someone (like Miner) who has more extensive experience with something like this.

1) I would start with a random sample. If the effect is insignificant, leave it at that. If significant, a separate study of all nests and analyzed by ANOM would be recommended. In my experience, the multiple nests of an ATE (Automated Test Equipment) are not built to the same exacting standards of calipers or micrometers.
2) There would be some movement of parts from nest to nest. If there is an issue, it will show up as a part x nest interaction.

 

[Bev D]

…and like any gauge there should be some level of ‘calibration’ for each nest. Just as every caliper is calibrated.

 

[Ranjit B]

There are a couple of things about this gauge that concern me.....and I'm not really sure how to address them.

1) Should an R&R be performed on each nest? I don't think it is required, in the same way that you don't do an R&R study on each caliper in your shop - they are considered a 'family' of gauges.
2) Should parts be moved from nest to nest during the measurements? This will affect both repeatability and reproducability. However, I'm a little concerned that this movement of the parts from nest to nest may end up MASKING some issues. I would definitely defer to someone (like Miner) who has more extensive experience with something like this.

Your points are valid, because just using standard MSA will not help me understand variations in play.
Any thought Miner

 

[Miner]

Your points are valid, because just using standard MSA will not help me understand variations in play.
Any thought Miner

See my post #6 above.

 

[Miner]

…and like any gauge there should be some level of ‘calibration’ for each nest. Just as every caliper is calibrated.

I could go on for a long time about what should be done on ATEs, but is usually not done. Calibration of all nests is just the beginning. Almost never are they calibrated as a "measurement system." They will usually isolate the measurement device from the system and calibrate the device in isolation, which is wrong.