Any coating used on the external surface of a window needs to be a hard coating that can resist the wear this surface is exposed to. For this reason hard oxides or nitrides should be used, and for the low emissivity coating fluorine doped tin oxide was used; a coating readily available on the market. The hydrophilic coating tested with respect to external condensation was a titanium oxide coating which is also already available on the market [9]. It is marketed for its “self cleaning” properties, however, and so far not for its effect on external condensation. Two test windows were prepared which consisted of triple glazed IG units with silver based soft low-e coatings on surfaces 3 and 5. For one of the two windows the outer pane was ordinary uncoated float glass and for the other one the outer pane was a hydrophilic coating on the external surface. This coating does not change the surface emissivity of the external surface and the U-value was identical for both windows and with argon filling the centre of glass U-value was 0.65 W/m2K. Both windows had warm edge spacers, although this had no effect on the formation of condensation in the middle of the window.

3.2. Experimental details

The test was conducted during the period September to December 2007, and the condensation was monitored in terms of its occurrence rather than number of hours. No detection equipment was used as was the case for the tests in the test box. A reason for this was the cost together with the fact that the windows were used in an ordinary family house. The condensation rate was thus recorded as “number of times condensation occurred” rather than as “number of hours”. It turned out that the formed condensation could look quite different from one time to another and a subjective division was made between “full” condensation and “moderate” condensation. Full condensation means that the amount was sufficient to completely obstruct the view across the whole window, while moderate condensation means that the condensation was visible, but more like haze or just at the lower part of the window, not covering the whole surface. Although most of the occasions with condensation were detected early morning, there were also some days when condensation was noted in the afternoon or evening, defined as “daytime” condensation. On several occasions during the second half of the tested period it was also noted that the surface temperature was low enough for the condensed water to freeze on the window surfaces. Since the visual appearance of the condensation layers is the important parameter, the occurrence was recorded by pictures taken with an ordinary camera.