Calibration Frequency for Rarely Used Gages

[Wes Bucey]

In Reply to Parent Post by blemon

I was scanning the board for an answer to a similar question...we have a Fowler Optical Comparator Set which might be used once a year, maybe twice. I was told by an engineer at the OEM that the shift over time in accuracy is negligible. Basically, what I am looking for is substantiation for extending the interval out as far as feasibly possible. If I can justify it on paper, management, auditors, and integrity will be satisfied.

Any help will be greatly appreciated!

Good question, blemon! Welcome to the Cove
Al Rosen has good advice here.

Typically, an OEM will have suggested intervals for calibration and recertification of instruments and gages based on amount of usage over time. A big company like Fowler should have this available as a simple handout (it may have even been included with the set when purchased.)

If the manufacturer or its engineering staff is unavailable or reluctant, you can arbitrarily extend the time between calibration/certification events to any number of days, weeks, etc., maintaining a protocol or instruction to check its reading against a known standard for calibration purposes before measuring a workpiece. (This is also a way of helping reorient the technician to an unfamiliar instrument and helps take the human error out of a measurement.)

[Jim Howe]

I am fascinated by all of the good positive comments on calibration frequency and over the years (Mil-Q, Mil-I and Mil-C environments) have used most of the rules for extending calibration intervals. Let me add just three notes:
1) Make sure your calibration procedure spells out the rules for extending intervals and who will make the decision to extend.(Usually QA Manager)
2) In the past I have had great success using seals. (I have used wire with lead seals, plastic seals, wax seal and tape seals). These seals work quite nicely on crimping tools and electronic test equipment. If the seal is unbroken at time of recalibration then the tool/instrument has not been used and may be extended.
3) Finally, if the "test sets" inputs and outputs are independently verified by calibrated meters such as Fluke, etc. then there is no need to calibrate the "test set". Only the instrument verifying final acceptance needed to be calibrated. The "test sets" were marked "no calibration required". These test sets were used to verify the performance of printed circuit boards and wire harnesses.

I might add that many of the test were witnessed by customers and government reps and they preferred the independent verification methods.

Jim Howe

[airaza08]

is it true equipment for temperature, pressure and electrical must be calibrate once a year? can it be longer/extend?

[Mchurch - 2008]

Is it not possible to just get an onsite service from you calibration provider and get those items verified/calibrated at your location? This would reduce down time to a matter of 15-20 mins (per dial indicator) on the item instead of 5 working days, esentially eliminating the need for redundent items for down time.

Depending on your location, the money you could save from not getting the others calibrated might cover the onsite fee the provider would charge.

Get your back ups calibrated and sealed up in a bag, keep them purely as backups incase something breaks.

Just a thought,
Mark.

[bobdoering]

One tool that is often ignored in determination of calibration frequency is stability. For dimensional gages, it would make sense that your calibration data over time should show the rate of wear. That is one reason why your calibration standards should be at 10:1, to have enough resolution to detect such changes. You may consider using that wear rate from a linear regression of your calibration data, and extrapolate it to the endpoint of your gage tolerance. If you are making adjustments, you will have to track the before and after dimension, and add the change to original dimension to chart the actual wear. Divide that time by 3 or 5 (whatever you feel comfortable based on your usage so as to not overshoot the limit) to determine a sane, justifiable calibration frequency. Not the best explanation - but, in essence, handle the issue that same as tracking tool wear. (Yes, that means do not X-bar R the stability measurement, it should be non-normal. You might have figured that was coming...)

[n866fandp]

It is a judgment call, that you should make based on the severity of your measurements. Depending on the gage software you use, you may be able to generate a graph to show the results of wear, and you should be able to use this info to help you make an informed decision. I use GAGEtrak v.6.5 and it has a graph feature that I use quite frequently.

[bobdoering]

One factor that does not come out of tracking wear is the risk of a 'special cause' such as dropping a gage. If one decided to make the calibration period 2 years, but the gage is dropped in 30 days, you have 1 year and 11 months of suspect product. It has been known to happen with dire consequences. That said, the point is the decisions need to be made with "profound knowledge" of the entire system - how it is used, risk of damage, wear rate, etc. There are a lot of folks looking for the magic bullet of when to calibrate (and how to calibrate as little as possible). Usually, the bullet they get is in their foot...

[Mark Paul]

Maybe I missed the answer to this.
We have 5000 ring, thread gages and thread rings. The company was sold and we are now gaining back customers. 2100 gages are active, the balance are in a climate controlled "inactive" room.
The inactive gages were dipped in wax or have been sprayed for corrosion.

The question is this: can we take a gage, calibrated,verified, and mothballed before the original cal due date, that has been sitting on a shelf for 4 years and activate it for use? It would be "calibration extended" as the last verification and cert are old as well. This first "new" use would then start the original 1 year calibration cycle.
The only gages I am leary of are the ring gages as they have an adjustment screw.

Any ideas? Mothballed, wax dipped, certed gages, 3 - 5 years since last use.
Much appreciated.