Calibration grace period after due date

[GreatManBehinD]

Looking for some of my concerns for calibration grace periods. We have established our calibration frequencies and what my question is there any documented evidence of grace periods after due date. Example 1 month freqency has a grace period of 1 day, 6 month has 3 day, 1 year has 5 days. Please provide me a link to some kind of document which contains these grace periods if possible and if not what is the right approach, Should we have a grace period or not.

[Tom W]

Originally Posted by GreatManBehinD

Looking for some of my concerns for calibration grace periods. We have established our calibration frequencies and what my question is there any documented evidence of grace periods after due date. Example 1 month freqency has a grace period of 1 day, 6 month has 3 day, 1 year has 5 days. Please provide me a link to some kind of document which contains these grace periods if possible and if not what is the right approach, Should we have a grace period or not.

Not sure what you are calibrating but if it has to do with Pyrometry then AMS 2750 D has a table 10 that states allowable grace periods based on test frequencies.

[GreatManBehinD]

Originally Posted by Tom W

Not sure what you are calibrating but if it has to do with Pyrometry then AMS 2750 D has a table 10 that states allowable grace periods based on test frequencies.

These are temperature / pressure / humidity chart recorders, pressure guages, thermometers, transducers, sensors, etc.

[Tom W]

I would start with AMS 2750 Revision D Table 10 and see if that helps - we use it extensively in the Heat Treating industry for our calibrations on the temperature controllers, uniformity surveys for the furnace hot zone and temperature probe checks or system accuracy tests. It's got a tight tolerance on the grace period to be honset with you but it might be able to be applied to ther typs as well if you are willing to accept the time frames.

[BradM]

Tom's point is a good one. Do you have any industry requirements that establishes the interval? If you're aerospace, then do what Tom suggested. If you have no industry requirements, then that's another discussion in itself.

[Jerry Eldred]

What you are calling a "grace period", I have commonly either called a "temporary interval extension" or "late." I'll discuss both briefly below and some common industry practices.

TEMPORARY INTERVAL EXTENSION- If an instrument has (for example) a 12 month interval, that was selected for a variety of reasons; the most common of which is the manufacturer's recommended interval. The basis (in good OEM quality) is the 2 Sigma (or so) statistical probability that that is how long it can be expected (on average) to remain in tolerance (to a percent confidence level). To exceed this manufacturer's prescribed interval, it is common practice to have some historical data to support that there is not significant risk to do so. Typical data would be multiple intervals wherein the instrument has continuously remained in tolerance without adjustment. For the example, if a 12 month interval instrument must be extended for operational purposes, and it has remained in tolerance for three consecutive 12 month intervals without adjustment, you could support that (for example) a one or two month extension would not incur significant risk increase of out of tolerance readings.

LATE- As is often referred to be "grace period", how many days late an instrument can be turned in and be considered acceptable..... I'll have to give you "the company line" (so to speak) on this. I don't believe in grace periods. I don't want to sound hardnosed on this, but I feel obliged to try and provide my best advice. If I were writing a calibration quality system policy (and I have written a few), I would not include a grace period as such in them, as this defeats the statistical purpose of the prescribed interval. If I had to somehow write some such detail, I might write a "reverse grace period." That is, perhaps create an early recall date, and a true due date. I question whether a normal grace period would with stand some audits. But with a "reverse grace period", you could have the recall date lets say 2 weeks prior to the due date (including the due date on the cal label). It must then be written into your quality manual that the recall due date is two weeks prior to the interval, and the instrument must be turned in within 14 days from the due date. I don't like even that idea either, but it might hold up better in some audits.

Back to some of the prior responses. The only time I would allow grace periods would be within specific compliance to some standard such as previously listed.