Generic Measurement Best Practice
The general principles of best practice in dimensional metrology are very clearly defined and well documented by:-
- NPL's Fundamental Good Practice in Dimensional Metrology - Measurement Good Practice Guide No. 80
- NPL's dimensional metrology courses
The importance of considering accuracy when deciding whether a measurement proves that something is within specification is set out in:-
- BS EN ISO 14253-1:1999 (Decision rules)
Estimation of Measurement Uncertainty
The following link gives a simple explanation of what measurement uncertainty is and why it is important.
One-Dimensional Measurements
The best practices for uncertainty estimation of 1D measurements are documented by:-
- UKAS document M3003, an excellent introduction to measurement uncertainty.
- The NPL Best Practice Guide No. 6 - Uncertainty Evaluation, a more technical guide
- The definitive standard regarding the expression of uncertainty in measurement is given in the following standards:-
Coordinate Measurements
Uncertainty estimation, calibration and verification for coordinate measurements does not currently have such clearly defined best practice.
The following standards cover verification tests for specific instruments. Although they separate some of the major sub-systems they do not provide full estimations of uncertainty parameters:-
- For CMM's: ISO 10360-2
- For Machine Tools: ISO 230 (many parts)
- For Laser Trackers: ASME B89.4.19
- For Photogrammetry: VDI/VDE 2634 Parts 1 and 2
State of the art methods are being developed by NIST/NPL/PTB which will provide full rigorous uncertainty evaluation for coordinate measurements. These include:-
- Laser tracker uncertainty methods being developed by NIST and others.
- For CMM's and Machine tools the Etalon system combined with methods being developed by LIMA and other organisations.
- These techniques also have some similarities with the bundle adjustment techniques which are well established for photogrammetry.
The generic approach is to create a model of the uncertainty parameters for the instrument and then best fit measurements to reference coordinates while optimizing the uncertainty parameters. The fundamental mathematics and application of this approach are documented in publications by Forbes et al.