Dissemination of unit of mass is one of the key processes in mass metrology and involves a large number of measurements to determine the mass of weights across a wide range (e.g., 1 mg–10 kg in the case of the Czech Metrology Institute, CMI). Evaluation of such measurements can be challenging, and to address this, the European Metrology Programme for Innovation and Research (EMPIR) project 19RPT02 “Improvements of the realisation of the mass scale” developed RealMass software solution (currently available in version 1.1) and a draft calibration procedure. However, standard procedures usually assume either identical densities of the weights or use the volume of the weights for buoyancy correction. In the latter case, if the volume is not known, the usual approach is to estimate it by dividing the nominal mass by the density. If the weights differ in either volume or density, these procedures lead to incorrect results. CMI developed a model and evaluation script to address these issues. The comparison data show that the developed model is consistent with the results obtained by RealMass software and other examples. The examples given in the text show how incorrect assumptions can lead to incorrect results and how they are evaluated by the approach presented in this paper.
Jaroslav Zůda (2026) studied this question.