Water density formulations and their effect on gravimetric water meter calibration and measurement uncertainties

In the gravimetric calibration method of water meters, the volume of water that has passed through the equipment under test (EUT) is generally collected into a tank and the quantity (mass) determined by weighing. The mass of water collected is then converted into a volume. This conversion of mass into volume requires knowledge of the water density, which can be estimated, measured directly or determined by other means. The error of measurement of the EUT is determined by comparing the volume recorded by the EUT and the volume collected in the tank. The density of water is, therefore, one of the major causes of measurement uncertainty in laboratory calibration of water meters using the gravimetric method. Water meter calibration facilities commonly use density formulations proposed by the International Standards Organisation (ISO) and the Organisation for International Legal Metrology (OIML). In Australia, additional guidance in water density determination is provided by the National Measurement Institute (NMI). In this study, testing was undertaken using ten positive displacement water meters arranged in series in the test rig to evaluate some of the common water density formulations used in Australia. The effect of these different formulations on the water meter error measurement was determined, as well as the effect on the measurement uncertainties. The results shows that the use of these different density formulations evaluated do not significantly affect the water meter error of measurement or the uncertainty of measurement. There was no apparent correlation between the water meter error and the meter position in the test rig. It was also determined that if the water density was adjusted only for temperature effects, a maximum of 0.05 and 0.15% drift in meter error and measurement uncertainty respectively, can be expected.