ON PREDICTING THE PERFORMANCE OF SHIPS FROM MODELS

A method of estimating the performance of ships from model tests is described. It is based on propeller coefficients and wake fractions which apply in operating conditions with the screws at power in the non-uniform flow astern. A rotative factor for thrust is introduced relating the propeller coefficients in working conditions to those in uniform inflows: the assumption of thrust identity cannot be true in general, and is avoided. The model ship is tested in such a way that the flows which give rise to thrust are dynamically similar at model and at full size. In this condition and not otherwise, a valid law exists for scaling up the thrusts measured on the model. Examples show how the method is applied and how the proper propulsion point is determined. The rotative factor for thrust must be measured during the propulsion tests. Estimates of wake fraction, rpm and shaft power at full scale differ significantly from predictions based on thrust identity, and on propulsion data from tests in which dynamical similitude for thrusts is not ensured.