PETpla.net Insider 01+02 / 2016

INSPECTION 19 PET planet insider Vol. 17 No. 01+02/16 www.petpla.net Figure 1 - Semi-Auto Preform Inspection Gauge with telecentric lens and colli- mated light source Precision is not accuracy Precision is the expected spread of results when a specific measure- ment is repeated under fixed condi- tions. Use of the word precision is dis- couraged by measurement institutes and standards authorities worldwide who instead entreat the use of defined terms such as capability, repeatability and reproducibility. The difference between accuracy, precision and resolution are illustrated in Figure 2. Figure 2 - Illustrative difference between accuracy, precision and resolution System repeatability Repeatability & reproducibility, often referred to as R&R, is a quanti- tative value representing the expected scatter of measurement results when the same test is performed multiple times. It is generically expressed by the equation where GRR is the calculated gauge R&R, EV is equipment variance (repeatability) and AV is appraiser variance (reproducibility). Repeatability , or equipment variance, is the contribution to overall R&R that comes from the instrument. Specifically, the range of obtained measurement results when a particu- lar measurement is repeated with no external influence. Reproducibility , or appraiser variance, is the expected measure- ment change that can be attributed to operator influence. Contributing factors include operator training, load- ing factors, positional variation and misuse. Torus Measurement Systems’ R&R study is historically derived from the AIAG (Automotive Industry Action Group) MSA (Measurement System Analysis) and performed on all gauges prior to shipping. The study evaluates measurement results of 10 parts each measured three times by three individual operators, using a 5.15 sigma value to represent 99% of the area beneath the normal distribu- tion curve. Subsequently, equipment and appraiser variation are evaluated by identifying trends present in control charts, see Figure 3, Figure 4 and Figure 5, for examples. Figure 3 - Points beyond either of the control limits # Figure 4 - Points consistently one side of the centreline Figure 5 - Points consecutively rising and falling Traceable standard Measurement quality is ascer- tained by comparison to a reference instrument of higher accuracy or to a master calibration. Either method must be certified, directly or indirectly, to a primary standard. The series of documented links between the highest level calibration and the final gauge accuracy is known as measurement traceability. The primary, or national, standard is defined by a country’s National Measurement Institute (NMI), see Table 1. They are responsible for providing guidelines, working practices and calibration services. By collaborat- ing with other NMIs they ensure stand- ards are internationally recognised and adhered. This means a traceable gauge accuracy is valid in many coun- tries.