A statistical approach to estimate the experimental uncertainty of viscosity data obtained by the oscillating cup technique

The determination of the dynamic viscosity of metallic alloys at elevated temperatures is a key experiment for materials design and characterisation. To do so, the oscillating cup method is the most suited and commonly applied one. However, critical estimations of the experimental uncertainties are scarce, complicating the comparison of results obtained by different authors as well as with models set up in order to describe the changes of the dynamic viscosity with temperature and the alloy's composition. We present a statistical approach, first to estimate the uncertainties of the actual measured parameters of a torsional oscillation, and second to analyse the propagation of these errors when calculating the dynamic viscosity. Overall relative uncertainties in the range of 5–10% are obtained for selected experiments carried out on pure liquid Al, Sn and Cu. This figure is in good agreement with previous error estimations. However, method is ready to be involved in the usual numeric calculation procedure of the dynamic viscosity in the actual experiment and is thus able to take the individual conditions with a given sample into account.