Abstract
The results of the International Wet Steam Modelling Project presented at the Wet Steam Conference in Prague in 2016 again gave rise to a discussion about the suitability of the flow solver ANSYS CFX for the prediction of spontaneous condensation. In this article, the main reason for the discrepancy between results obtained with CFX and measurement data is identified. It could be found that in CFX the temperature of already existing droplets is used for evaluating the fluid properties involved in the nucleation process. This is not in agreement with the isothermal classical nucleation theory, which is based on the assumption that clusters of critical radius are formed at vapour temperature. The most dominant parameter affected by this is the surface tension, the evaluation of which can be altered easily by the user. The influence of the correction is illustrated by means of standard Laval nozzle test cases, and a significant improvement of the results compared to measurement data can be observed.
Get full access to this article
View all access options for this article.
