Reference part calibration or verification

[Marvin16]

Hello,
I would like to ask you about one issue. In our production we use test station for checking printed circuit board
for corect capacity of condensator. On the begining and on the end of the shift we release this station with reference parts.
We have 3referene part 1-OK, 2-capacity is under the limit, 3-capacity is above limit and station must evaluated correctly.
We check these reference parts 1x/year with measuring capacity 3x, then calculate avarage and uncertainty and check if this is in spcification.
I would like to ask you if this process of checkikg reference parts (measuring capacity 3x, then calculate avarage and uncertainty)
is calibration or verification.
And how we can do MSA for these reference parts.
Thank you for your support.

Martin

 

[Jen Kirley]

Welcome Marvin!

I want to better understand how you confirm these reference parts are adequate as verification devices. I am not clear on what industry you supply for, or what standard you are subject to. Can you tell us?

I am also wondering what is your process for checking these reference parts:

• What is your process tolerance?
• What are your limits?
• Can you explain for us what you mean when you say "We check these reference parts 1x/year with measuring capacity 3x, then calculate avarage and uncertainty and check if this is in specification."

Thank you for your help!

 

[Marvin16]

Hello Jen,
thank you for your response. We are involve in consume industry according ISO 9001.
We assembled printed circuit boadrs.
At one board we check if capacity of assembled condesator is in the specification. This is done by test station, which measured capactiy.
Nominal value is10nanoFarad, limits are 9 - 11nF.
We have 3 reference parts:
1 - OK part (capacity is cca 10 +/-0,5nF),
2 - NOK part (capacity is less than 9nF, cca 8,0 +/-0,5nf)
3- NOK part (capacity is higher that 11nF, cca 12+/+0,5nF).
With these reference part we do release of the test station. OK part must be evaluated as OK, NOK part must be evaluated as NOK parts.

• Can you explain for us what you mean when you say "We check these reference parts 1x/year with measuring capacity 3x, then calculate avarage and uncertainty and check if this is in specification."

It means that we have to check if capacity on the reference parts is still according specification.
Therefore we measure capactiy of refence parts 1/year to known if capacity on the reference part is still in specification. This check is done by calibrated RLC meter.
And would like to know if process of checking refenrece part by multimeter is called calibration or verification.
Thank you.

 

[Sidney Vianna]

We check these reference parts 1x/year with measuring capacity 3x, then calculate avarage and uncertainty and check if this is in spcification.
I would like to ask you if this process of checkikg reference parts (measuring capacity 3x, then calculate avarage and uncertainty)
is calibration or verification.
And how we can do MSA for these reference parts.

From an ISO 9001 perspective, what you propose is perfectly acceptable. The "golden" boards OK and Not OK are being verified, so is the test station. But why an MSA? MSA's are not an ISO 9001 requirement.

 

[dwperron]

From what I am reading the reference parts are calibrated when you make traceable measurements of the values of the capacitors on them.

You then are using these calibrated reference boards to perform a verification on the test station.

 

[John Predmore]

Measuring your reference parts is verification. If you made adjustments to your test station based on measurement of the reference parts, that would be closer to the idea of calibration.

One measurement system analysis you can do is to measure the reference parts more often than once/year, maybe weekly, and compare separate Statistical Process Control charts of the results for each reference part. I suggest you purposely measure your reference parts at different times of the day, and if you have more than one test station, keep traceability so you know which readings come from which station. When you average your readings now, you lose potentially valuable information about the influence of warm-up time or between different stations. By charting measurements over time, you will see an early warning if the reference parts drift, maybe with air temperature, or there is some unwanted interaction between one reference part and a particular test station.

 

[Marvin16]

From an ISO 9001 perspective, what you propose is perfectly acceptable. The "golden" boards OK and Not OK are being verified, so is the test station. But why an MSA? MSA's are not an ISO 9001 requirement.

Thank you. MSA is required by customer. But I do not know how I should do MSA for reference "golden" part.

 

[Guest]

Thank you. MSA is required by customer. But I do not know how I should do MSA for reference "golden" part.

You don't have to because it's not calibration. It's verification. Explain to the customer you're doing verification and MSA doesn't apply. If they insist, charge them for doing an MSA study, by increasing the price.

 

[Marvin16]

You don't have to because it's not calibration. It's verification. Explain to the customer you're doing verification and MSA doesn't apply. If they insist, charge them for doing an MSA study, by increasing the price.

Thank you. I will try it. But if MSA will be nessesary, could I ask you how the MSA for reference part should be done?

 

[Guest]

How familiar are you with MSA?