PT Journal
AU Thompson, A
   Southon, N
   Fern, F
   Stupfler, G
   Leach, R
TI Efficient empirical determination of maximum permissible error in coordinate metrology
SO Measurement science and technology: devoted to the theory, practice and application of measurement in physics, chemistry, engineering and the environmental and life sciences from inception to commercial exploitation
PD July
PY 2021
VL 32, 2021
IS 10
PU IOP Publ.
DI 10.1088/1361-6501/ac0c49
WP https://www.db-thueringen.de/receive/dbt_mods_00051594
LA en
DE maximum permissible error; metrology; fringe projection; co-ordinate measurement
SN 1361-6501
AB Maximum permissible errors (MPEs) are an important measurement system specification and form the basis of periodic verification of a measurement system's performance. However, there is no standard methodology for determining MPEs, so when they are not provided, or not suitable for the measurement procedure performed, it is unclear how to generate an appropriate value with which to verify the system. Whilst a simple approach might be to take many measurements of a calibrated artefact and then use the maximum observed error as the MPE, this method requires a large number of repeat measurements for high confidence in the calculated MPE. Here, we present a statistical method of MPE determination, capable of providing MPEs with high confidence and minimum data collection. The method is presented with 1000 synthetic experiments and is shown to determine an overestimated MPE within 10% of an analytically true value in 99.2% of experiments, while underestimating the MPE with respect to the analytically true value in 0.8% of experiments (overestimating the value, on average, by 1.24%). The method is then applied to a real test case (probing form error for a commercial fringe projection system), where the efficiently determined MPE is overestimated by 0.3% with respect to an MPE determined using an arbitrarily chosen large number of measurements.
PI Bristol
ER