Reproducibility error between instruments for automatic measuring equipment

J

jackson

Hello,
My customer requested us to evaluate the reproducibility error between instruments for automatic measuring equipment. We have many automatic measuring equipment. I don't know how to pick out 3 pieces of measuring equipment to evaluate the reproducibility error.
My question is:
What are your rules to select three of all measuring equipment to do reporducibilty analysis?

Best regards
 
J

jrubio

Hi, Still alive.

Reproducibility. Page 7 MSA third Edition.

Variation in the Avarage of the measurements made by different apprasiers using the same gage when measuring a charasteristic of one part.

Just make a GRR (REpeteability & REproducibility) study to each equipent.

There are some GRR files in the cove. Use Search.

Regards.
 
J

jrubio

Other important consequence of R%R is that allows us to detect if the operator needs training therefore I think that is a good thing to make a R&R to the new operators and let us know if they are well trained.


:notme:
 
J

jackson

Thanks for your response so quickly.
According to the Page 53 MSA third Edition "This is often true for manual instruments influenced by the skill of the operator. It is not true, however, for measurement process (i.e., automated systems) where the operator is not a major source of variation. For this reason, reproducibility is referred to as the average variation between-systems or between-conditions of measurement"
This statement implies the reproducibility between-appraisers can be neglected for the automated system. That is why our customer requests us to evaluate the reproducibility error between instruments rather than between the appraisers.
According to the page 54 MSA third Edition:
“Between-instrument: average difference using instrument A,B,C etc., for the same parts, operators and environment.”
My question is how to select A,B,C from many same type instruments. Do you just select any 3 pieces of instruments at random or based on some rules (for example, the two instruments with the biggest accuracy)?
 
J

jrubio

Hi.

You were right. :thanx:

But if you have the GRR for each one (You must have it according to the norm) thus you can compare them, isn´t it?

If you want to freeze all the conditions using the same 10 parts, the same operator, Temperature, et...

I would take those Gage with Minimum.maximum GRR and I would do other GRR

(I do not know if exists an official method)


Best Regards.
 
J

jrubio

Hi.

I woud create a matrix like this.

..Gage.........PN Charasterist.......GRR............Conclusion.
Gage 1........Interna Diameter...............6 ........... OK

Gage 2........Internal Diameter................7.............OK

Gage 3........Internal Diameter...............10............OK
 
J

jrubio

..Gage.........PN Charasterist.......GRR............Time.......Conclusion.

Gage 1........Interna Diameter...............6 ......January ..... OK

Gage 2........Internal Diameter................7......January.......OK

Gage 3........Internal Diameter...............10.......February.....OK

Gage 1........Interna Diameter...............7 ......February ..... OK,increasing

Gage 2........Internal Diameter................22......February.......nO OK. RECALIBRATE

Gage 3........Internal Diameter...............12.......February.....OK
 
B

briggs_joe

jackson said:
According to the Page 53 MSA third Edition "This is often true for manual instruments influenced by the skill of the operator. It is not true, however, for measurement process (i.e., automated systems) where the operator is not a major source of variation. For this reason, reproducibility is referred to as the average variation between-systems or between-conditions of measurement"
This statement implies the reproducibility between-appraisers can be neglected for the automated system. That is why our customer requests us to evaluate the reproducibility error between instruments rather than between the appraisers.
According to the page 54 MSA third Edition:
“Between-instrument: average difference using instrument A,B,C etc., for the same parts, operators and environment.”
My question is how to select A,B,C from many same type instruments. Do you just select any 3 pieces of instruments at random or based on some rules (for example, the two instruments with the biggest accuracy)?

Correct me if I'm off here, Jackson. I don't want to mislead you.

I'm assuming you're asking about limiting the number of test systems you will use in your reproducibility study and still get a representative sample of your overall test measurement process.

First, you need to use your knowledge of the target process that the ATE is used in, and understand all the variables that can result in varying measurements. This could range from environmental effects across the working area to the age/revision of the instrumentation in your ATE.

Then, you need to understand how much the chosen number of ATEs for the experiment will statistically represent your overall complement of ATEs. I don't personally know of any hard and fast rules on this; you have to use good judgement, and base it on risk associated with quality vs schedule vs cost. With this result you might choose 10% of your ATEs, 20%, 50% or even all of them (this obviously would be extremely expensive and take a long time if you have a lot of ATEs).

Unfortunately, there are no set rules for this that I'm aware of. This is going to be one of those "good judgement" calls. Like any statistical analysis, the more data points across all variables, the more representative the final result will be of all of the equipment.
 
J

jackson

Hi jrubio and briggs:)
I greatly appreciate your suggestions. Now, I have two
solutions which I can take them up with my boss.
1.To select the Gages with Minimum and Maximum GRR.
2.To select the Gages by statistical methodology.

Best regards
 
Top Bottom