Any gage (tool, measuring instrument) that is used to determine acceptance of a quality characteristic of the product must be calibrated. The calibration must be traceable to national measurement standards, or to another accepted standard.
A test I use is this (adapted from P. G. Stein) --
Does it matter if the measurement is wrong?
If yes, then calibrate the instrument.
If no, then why are you making the measurement?
Axiom: it is always lower cost to determine acceptance of a characteristic at the earliest possible stage of a process. Final inspection is always the second most expensive place to find a problem. (The
most expensive place to find a problem is at the customer's place after delivery.)
If you only have calibrated tools at the final inspection point you are, in my opinion, setting yourself up for a lot of rework.
In some companies and industries it is historical practice to use tools that are marked "For Reference Only". (I work in one of those companies.) I disagree with this for several reasons.
- Putting calibrated and "reference only" tools in the same work area creates a monster control problem. There must be absolutely zero possibility of a "reference only" tool ever being used where a calibrated tool is required. Getting caught once is at least a minor nonconformance. Having it happen several times is a major nonconformance because the system does not work. The consequences could be worse if people's lives are put at risk.
- A rationale used for this practice is that the "reference only" tool is "only used for" troubleshooting, or maintenance, or alignment of the system, or only by Facilities and not by Production, or (and so on). Sorry, but I don't agree. I have done all of those types of work and in every one of them I have wanted some assurance that the meter is correct. This can also cause a lot of rework. For instance, if you align a system with a "reference only" meter but it does not pass the system checkout with calibrated instruments, then you have to align it and check it again.
- There are cases where a calibrated tool may not be "required" but is an important safety factor. An electrician, for example, probably wants to be very sure there really is no voltage on a wire before she grabs it!
Notice that the key factor is
how the tool is used, not who owns it. There is no intrinsic problem with employee-owned calibratable tools, provided they are in the calibration system.
Yes, calibration can be expensive. It's a business decision. But as part of the decision you should factor in the potential cost of not doing it, especially if a failure of your product or service can put people's lives at risk.
[This message has been edited by Graeme (edited 23 August 2001).]
