Facades and Summer Performance of Buildings - Low Energy Cooling for Sustainable Buildings - page 49

Figure 2.25 PV fa¸ade element with series-connected air collector at the Stuttgart test site
temperatures on the double-glazed window ( T s -Window) correspond nearly exactly
to the measured data. This is a good indication that the heat transfer rate within the
gap must have been calculated correctly.
The air temperature increase reaches 15 K at average flow velocities of 0.3 m s − 1
(see Figure 2.30). These high temperature levels in the air gap and the rather poor
thermal separation from the room by uncoated double glazing then lead to inside
surface temperatures up to 37 ◦ C and consequently to high secondary heat flows to the
adjacent room (see Figure 2.29).When comparing the fa¸ade outlet temperatures in
Figure 2.30, different approaches for the calculated Nusselt numbers were used. The
newly developedNusselt function aswell as the correlation formixed convection over a
single vertical plate fromChurchill (1977) provide nearly the same temperature profile.
Because of the very slow air flow rate of a maximum of 0.3m s − 1 , the correlation of
Petukhov (1970) for fully developed forced convection is not useful for reproducing
results obtained from the Matar ´ ofa¸ade. The Petukhov correlation for fully developed
flow in an air channel is proportional to the measured flow velocity and gives very low
heat transfer coefficients below 2Wm − 2 K − 1 . The highest heat transfer coefficients
over 4Wm − 2 K − 1 are obtained from single plate flow (see Figure 2.31).