In-house device calibration record - Actual values obtained required?

[Helifreak]

Hi all,

My first post on the Cove...

Are actual values obtained during device calibration required on the in-house calibration record or are reference to standards used in calibration and traceablity to NIST sufficient?

Thanks

Rick

 

[Ninja]

Up to you, really...but it is hard to question recorded actual vs. known good (standard) data.

It would be easy to question "We used this standard and the gage passed".

 

[Helifreak]

Thanks Ninja!

I'm guessing it's safe to assume measurement results from device calibration would be of no use on the device record.

 

[AndyN]

Hi all,

My first post on the Cove...

Are actual values obtained during device calibration required on the in-house calibration record or are reference to standards used in calibration and traceablity to NIST sufficient?

Thanks

Rick

Hi Rick: Welcome!

Firstly, if we understand WHY we calibrate that helps. Calibration is the act of determining how far from a value (spec), the measuring instrument value is reading. Based on that understanding of the comparison, you can make 2 fundamental decisions - based on the value reported: is the product (feature) in spec? How reliable is the measuring equipment?

Without recording that value when calibrating, how will you be able to make valid decisions?

 

[Ninja]

Without recording that value when calibrating, how will you be able to make valid decisions?

100% agreed (with the whole post, not just this snippet)...

While calibrating/validating/assessing, the values have to be known...and typically that means recorded either visibly or within some piece of equipment.
...and they have to be compared to the known standard...
...and a decision must be made based upon the differences seen...

All of that MUST be part of the calibration.
Whether or not all of that needs to be part of the calibration record, though...I don't see the Shall.

@ the OP: All the same, is there a reason why you are trying to avoid keeping the data?
It seems such an easy thing to keep to avoid any future question that it seems odd to be looking for a loophole here. What drives your question as to whether to keep the data used for the decision?

 

[Ninja]

I'm guessing it's safe to assume measurement results from device calibration would be of no use on the device record.

Not sure I'd say that...

I've looked back on cal data for trends and to revisit calibration periods.
Having a record of Pass, Pass, Pass, Pass, Fail does nothing for me in determining why the fail happened, or whether to replace the gage with the same model or to look elsewhere.

I've also looked back at "Passed by how much" to lengthen or shorten calibration intervals (and to defend a gage cal cycle to customers).

While you may not HAVE to keep the data...I would require it in my shop. It would never occur to me to discard data used as the basis of a decision...especially if it was directly related to Acceptance Criteria like calibration.

 

[AndyN]

Another, often overlooked part of an effective calibration scheme is trend analysis. Capturing the "as found" values of the measuring equipment gives information on drift, stability and, hence, adjustments to calibration frequency. If the concern is the maintenance of such records over time, compare that to the cost of scrap/rework/repair/recall associated with NOT keeping records...

 

[Helifreak]

OK I'm convinced. After reading your helpful responses, I've decided to include the calibration data on the record. Now I need a calibration record with provision for recorded data; do you know where I can get a sample form?

Thanks

 

[AndyN]

OK I'm convinced. After reading your helpful responses, I've decided to include the calibration data on the record. Now I need a calibration record with provision for recorded data; do you know where I can get a sample form?

Thanks

TBH,and depending on the population of measuring devices you have, a better solution to management of gauges is a database - you can find free ones on the interwebs, but investment in Gagetrak or similar might be best in the long run...

 

[Ninja]

OK I'm convinced. After reading your helpful responses, I've decided to include the calibration data on the record. Now I need a calibration record with provision for recorded data; do you know where I can get a sample form?

Andy already gave some guidance...I would simply point out that a "Form" is not required.
Most of the time I kept my in house cal data in XL (or in whatever form it originally output...csv, .txt, whatever).
I would print a copy of the data from the XL file for our cal records simply because they were still on paper at the time.

Computer storage is cheap...as long as you can find it when you look for it.

\\Network drive\calibration records\Gage number\raw data 072018.xlsx
Pretty easy to find...
...and easy to copy/paste for trend analysis...