The present investigation can be divided into two parts: (a) experiments made to examine the mechanism of dropwise condensation of steam with particular reference to the stability of drop promoting surfaces as affected by the material of cooled surface, the drop promoter, the surface finish, the rate of heat transmission, and the presence of non-condensable gas. and (b) a theoretical analysis of the beat transmission through individual droplets, the transient heat transfer through exposed areas, the statistical study of drop size distribution, and the estimation of steam side coefficient. An apparatus was developed to examine qualitatively,the behaviour of drop promoting surfaces on a small scale. It is considered that sufficient evidence was found to show that steam in contact with a cooled surtace condenses as a thin liquid film which later breaks into droplets. surfaces treated to give dropwise condensation deteriorate into mixed condensation in due time, and the duration tor which a treated surface maintains dropwise condensation varies between a few hours to several days, depending on many factors among which· the presence of non-condensable gas must not be overlooked. An approximation to the heat transmission through individual droplets has been worked out with assumed heat flow lines. The result, checked by the relaxation method. is correct within . + 10%. An analysis Of the transient heat transfer through exposed areas was made neglecting the increasing resistance of any accumulating liquid. The drop size distribution was analyzed tor one drop promoting surface at three different heat transmission rates. Based on this drop size distribution, the heat transmission through the drops was estimated by assuming they were held at rest on a cooled surface conducting heat under a steady state. . The estimated coefficient comes within the range or experimental results of many investigators.

Dropwise condensation (DWC) has the potential to enhance heat transfer compared to filmwise condensation (FWC). The heat transfer rates achieved by DWC depend on the drop size distribution, which is influenced by nucleation processes of newly formed drops. In DWC modeling, the nucleation site density Ns is used as an input parameter to obtain the drop size distribution of small drops. However, due to the small scale of the condensate nuclei, direct observation is difficult, and experimental data on the nucleation site density are scarce. In the literature, values in the range of 109 m−2 to 1015 m−2 can be found for Ns. In this paper, we report DWC experiments on SiO2 and 1H,1H,2H,2H-perfluorodecyltriethoxysilane (PFDTES) thin hydrophobic coatings that show significantly different nucleation site densities. Nucleation site densities are estimated from high-speed imaging of small drops during initial condensation and from model calibration using established DWC theory. We have found the values for Ns to be in the range from 1.1×1010 m−2 to 5.1×1011 m−2 for the SiO2 coating and 1011 m−2 to 1013 m−2 for the PFDTES coating. Our results show that there can be large differences in the nucleation site density under similar conditions depending on the surface properties. This underlines the importance of investigating nucleation site density specifically for each surface and under consideration of the specific process conditions used for DWC.