XVII IMEKO World Congress Metrology in the 3rd Millennium June 22 − 27, 2003, Dubrovnik, Croatia PROPOSAL FOR A PRACTICAL PROCEDURE FOR THE EXPRESSION OF UNCERTAINTY IN HARDNESS MEASUREMENT Giulio Barbato*, Alessandro Germak°, Samuel Low^ *Dipartimento di Sistemi di Produzione ed Economia dell’Azienda - Politecnico di Torino, Torino, Italy ° CNR, Istituto di metrologia “G. Colonnetti” – Torino- Italy ^ National Institute of Standards and Technology - NIST, Gaithersburg, USA Abstract − Hardness measurement is widely used in industrial applications for quality control and acceptance testing of products because it is fast, inexpensive and relatively non-destructive. Uncertainty evaluation is complicated because calibration procedures require the use of direct and indirect verification tests, but the effects on the measurand itself due to test parameters variations are difficult to predict, such as the force-time pattern, inelastic performances of Rockwell indenters and the numerical aperture for Brinell and Vickers indentations measurement. This paper starts by accepting as a matter of fact that standard specifications have demonstrated acceptable performance, indicating that their application is correct for most of the materials generally used. Accepting this premise, the next step was to try to translate the practice, confirmed by many years of experience, into the new language of uncertainty, strongly required by quality documents. Keywords: Hardness, Uncertainty, Calibration. 1. INTRODUCTION Hardness measurements have to guarantee to produce results within a reasonable difference between the different parties (producer, customer and, if present, an independent third party). Moreover, this condition should be verified not only for materials of common use such as steel, but also for different metallic materials frequently used today. The problem to face, therefore, can be considered as split in two: • guarantee the measurement compatibility when common metallic materials, similar to that of the reference blocks, are used; • guarantee the measurement compatibility when metallic materials different from that of the reference blocks are used. The usual practice of indirect verification of hardness testers is sufficient for the first circumstance: when two different hardness testers give compatible results on hardness reference blocks, it is likely that the results obtained on test pieces of similar materials should be compatible. This is confirmed by common practice. However, even if the indirect verification has the merit to take into account the effects of the whole measurement procedure, having compatible results on hardness reference blocks is not sufficient to assure that the two testers will give compatible results on materials different from that of the test blocks used. In fact, it certainly could be possible to have a tester that uses lower forces and a different indentation measurement scale which are matched in such a way to obtain results well in line with the values of specific reference blocks. This is a common occurrence, for instance, with hand operated testers. It is certainly evident that measurement compatibility when testing the material of reference blocks, will not automatically apply to any different material. This gives reason to also require that a direct verification be performed, aimed to assure that the measurement procedure of the hardness tester represents, within given tolerances, the principal features of the hardness scale used. The solution, therefore, is not so simple, but it should not be overcomplicated, as industrial practice requires an uncertainty evaluation that can be applied by any normal workshop that uses hardness measurements for quality assessments. To address the two sides of the problem from a practical point of view, the common practice used up to now of using both indirect verification (performed by reference hardness blocks) and direct verification (performed by direct measurement of force, length and time) has been shown to be acceptable both for the practical application and the general guarantee of acceptable compatibility of results between the three parties involved. Therefore, even if we know that this is an approximate solution of the problem, we shall accept the traditional method that requires a direct verification, to be sure that the hardness tester examined operates in line with the standard procedure, and an indirect verification, to take into account little, non-measurable causes of variations, as the load-time pattern, indenter micro-geometry and hysteresis, etc. Being that the direct measurement is only an evaluation of conformity of the procedure, uncertainty could be assessed on the basis of indirect measurement only, at least when a well defined starting point is available. The problem of the starting point is the international or national hardness standard references, for which national and international metrology organizations are working. Proceedings, XVII IMEKO World Congress, June 22 – 27, 2003, Dubrovnik, Croatia TC1 Proceedings, XVII IMEKO World Congress, June 22 – 27, 2003, Dubrovnik, Croatia TC5 986
