INTERNET MALL (October 01, 2000)
Whether you inject, extrude, blow, rotomold, or simply work with polymers, chances are you'll have to measure part dimensions. What can be simpler than taking a measurement on a part, you ask? The ans...
Whether you inject, extrude, blow, rotomold, or simply work with polymers, chances are you’ll have to measure part dimensions. What can be simpler than taking a measurement on a part, you ask? The answer to that question really depends on how deeply we look at the measurement process. And that process could include anything from a ruler to ultrasonic or laser non-contact scanning equipment.
Chances are your operation falls closer to the “steel rule” end of the spectrum, but there are many ways to boost accuracy and repeatability that don’t cost a dime.
One is to know the difference between accuracy and repeatability.
Unfortunately, “accuracy” is the universal word (I use it too) for both concepts, although they’re not the same thing. Accuracy is the degree to which a measured quantity agrees with a desired value or specification. Repeatability is the ability of your instrument to deliver the same reading measurement after measurement. To put it into perspective, remember that it’s possible to produce out-of-spec parts with terrific repeatability.
Can’t hold tolerances on your night shift? Consider a comparison test between measurements from each inspector on a part of known dimensions. Digital readouts on precision measuring instruments have taken away the need to read vernier scales, but they don’t remove the need to handle micrometers and vernier calipers correctly.
It’s common, for example, for inspectors to reef down on micrometers instead of using the friction thimble for consistent results. A common method of standardizing tools such as “mikes” and calipers is to measure standards or gauge blocks. The “jo-blocks” will certainly provide a consistent standard but they’re also hard, damn hard, so the clumsiest operator will produce consistent numbers. Resin parts are softer, and more prone to distort under measurement.
Solutions? One is to encourage a light touch, and to educate your inspection personnel that there’s no such thing as one “correct” measurement. A little variation is normal.
For small measurement accuracy and repeatability, improvements can be had through many low-cost methods.
Checking fixtures to hold parts steady during measurement can be simple and cheap, and combined with height gauges or stand-mounted mikes, make testing much more consistent.
Randomizing test instruments can help, instead of assigning an instrument to a machine or inspector permanently.
Another method is test instrument maintenance. Sure you check out the measuring tools as required by quality procedures, but have you looked at the instrument calibration logs lately? Entries such as “calibrated October 23” followed by a signature don’t tell you much. How far off the instrument was and what it took to bring it back to repeatability says a lot more, and if one tool is consistently off by a mile, get it serviced or replaced.
And don’t forget to encourage your team to “fess up” when a tool is accidentally damaged or abused. I once worked in a shop where every miscue was met by a stream of hostile invective. When the junior man dropped his micrometer on the concrete floor, reporting the incident depended on whether the instrument seemed to survive the fall. Naturally, it should have been recalibrated, and it was eventually. And the novice inspector survived to write technical columns in Canadian Plastics.