Similarity rules of cavitation tests : the case of the Francis turbine

On the basis of experimental studies on the influence of the test head and the water nuclei content on the performance of a Francis turbine, we have shown that outlet cavitation is influenced by the active nuclei content of the test loop water. Numerical simulations of the cavitation nuclei dynamics in a Francis runner allowed us to associate the leading role played by the water nuclei content in the development of outlet cavitation, with the particular pressure distribution over the blades. It was found that the saturation of the nuclei content influence was driven by a limitation of the volume available to the bubble growth in the region where the flow pressure was lower than the vapor pressure. If this fundamental result is applied to the problem of transposing model cavitation tests to the prototype case, it can be shown that, if the saturation effect is reached during the model test, this effect will also be present in the prototype. In order to guarantee the same cavitation development and characteristics using the model or the prototype, it is necessary to perform the cavitation tests with Froude similarity. The saturation characteristics for the model will then be the characteristic cavitation behavior of the prototype.