Calibration - Out of Tolerance - Trace Back System

[D_Wood]

Out of Tolerance - Trace Back System

Can anybody help me out with a procedure for tracing back a piece of test equipment to see where it was used and what tests were performed?

Irregardless of a calibration interval, sometimes a piece of equipment is found to be out of tolerance while being calibrated. You don't know at what point it went out of tolerance, so you may have to review every test performed back to the last calibration date.

My problem at the moment is that not enough information is being recorded in our database to permit isolated searches. Too much information is still done by hand.

I am looking for a generic process to start out with for now. One suggestion is to start attaching a "log" with each piece of equipment which records the product tested, test request number, type test performed, etc., along with the normal stuff like date, technician, blah blah. This may work until we can get a better database established, and more electronic tracking in place.

Thanks,
D_wood

 

[tomvehoski]

You need to be able to track the serial number of the gage against the inspection performed. If you are using a database, add a field for gage #. The inspector will have to enter this for each inspection. In the event a gage is found out of calibration, you can run a query to check all inspections performed with it since the last good calibration.

If you are still paper based, add an area to enter the gage number on the inspection report. Searching becomes much more difficult, but it is still possible.

You may not always need to actually revisit the inspections if you can justify the lack of a need for recall. For example:

1. Micrometer is 0.001 outisde of calibration tolerance.
2. Used only on parts with a tolerance of +/- .1
3. Parts are not safety critical
4. Funcitional testing was performed after mechanical inspection, so problems would have been found there.

Now if the part being checked is a sensor for an air bag which may not deploy if out of spec, you need to look more at recall.

Document the above on a corrective action (or whatever form your system requires) and look at changing the calibration frequency of your gage, or possibly your measurement system if it is that critical.

Tom