An approach for estimating the uncertainty of measurements is outlined based upon the "error budget" concept using the measurement tolerances specified in the testing and calibration standards. It should be noted that it is not possible to calculate a single value for the measurement uncertainty for all materials since different materials exhibit different response characteristics to some of the specified control parameters, e.g. straining rate or stressing rate [3]. The error budget presented here could be regarded as an upper limit to the measurement uncertainty for a laboratory undertaking testing in compliance with this International Standard (class 1 machine and extensometer).
It should be noted that when evaluating the total scatter in experimental results the uncertainty in measurement should be considered in addition to the inherent scatter due to material inhomogeneity. The statistical approach to the analysis of intercomparison exercises (Round Robin experiments) given in appendix K does not separate out the two contributing causes of the scatter. Another useful approach for estimating interlaboratory scatter is to employ a Certified Reference Material (CRM) which has certified material properties. The selection of candidate materials for use as a room temperature tensile CRM has been discussed elsewhere [3] and a 1 tonne batch of a material (Nimonic 75) in the form of 14 mm diameter bar is in the process of being certified in a project under the supervision of the Community Bureau of Reference (BCR).