Repeatability estimation of torque and thrust results in open water tests as a function of Reynolds number.

According to current European regulations for calculating measurement uncertainty, besides computation of sensibility coefficients by, for example, OFAT tests, we have to evaluate the own repeatability of measurement device and technique, in order to estimate the influence of not controlled parameters. The aim of the present work is to show that repeatability of open water test results depends on the performed Reynolds number. A large set of tests carried out on four screws models, with different shapes, diameters and blades numbers, has shown that, the coefficients of variation of torque and thrust measurements, as well as standard deviation of hydrodynamic coefficients KQ and KT, decrease if Reynolds number increases, with a trend well fitted by a rational function like kx-?. Results show that, if Reynolds number is about 750.000, standard deviations of torque and thrust are about 0.5% of the obtained measures, as well as those of KQ and KT that are about 0.001, that is, in the worst case, the 0.5% of the mean value. In addition to what already shown by ITTC (International Towing Tank Conference) Procedures, this conclusion provides useful information about the minimum Reynolds number to perform open water tests limiting the uncertainty intervals.