Rain & Wind Driven Rain
Rain penetration associated with PV or solar thermal panels tends to come from two factors: the fixings on stand-off (traditional) and flat roof designs, and around the edges of modules on integrated designs. Existing European testing standards, by comparison, only "assess the extent to which glazed collectors are substantially resistant to rain penetration" (EN12975), ie. how much rain enters the thermal collector itself, and are not intended to assess how watertight the solar systems are when installed onto roofs or whether the solar system causes the existing roof to leak, through the fixing penetrations for stand-off systems or through the seals and flashings for integrated systems. They also tend to focus on simple rainfall without wind-driven effects.
The EurActive Roofer project undertook extensive laboratory testing of fixing systems, with rain coming at different angles and intensities in order to create a methodology that could be used for a future European standard (prEN16501). The approach was based on varying the conditions:
A: Low wind speed with severe rainfall rate;
B: High wind speed with high rainfall rate;
C: Severe wind speed with low rainfall rate;
D: Maximum rainfall rate with no wind (deluge).
The actual values should be selected to represent a one in 50 years occurrence; analysis of UK data shows that there is actually a very low correlation between wind speed and rainfall so these tests provide an adequate range of conditions.
Rain Penetration round Fixings
The most common type of PV in the UK is where they are fixed above a traditional tile or slate roof (sometimes referred to a stand-off arrangement). The actual fixing is then normally by way of hooks under the tiles screwed into the roof battens; a fixing rail then usually connects the hooks and the modules are slotted into the rail (or the rail is attached to fittings on the rear of the panel) and secured by locking bolts. This enables the load to be spread, but does allow for the possibility of water ingress around the hooks.
Testing was carried out with the roof angled at 12.5°, 22.5° and 45°. On the graph above, which shows penetration of rain at 45° roof angle with a 75mm headlap, the lines represent rain penetration expressed in grams per square metre over a 5 minute period (y axis) plotted against pressure factor (in effect wind speed). The different coloured lines represent different gap sizes from 0mm (no gap) in 5mm increments to 15mm. Although it is not possible to show all the graphs here, at a 12.5° roof pitch there is little difference between the performance of the roofs with 5mm, 10mm or 15mm gaps. At the 10g/m2/5mins recommended reference rate, 5mm is marginally better, but as 12.5° would not be recommended for roofs in any case and is too low for a tilt for optimum PV generation in Europe, this is not too important. Conversely at the relatively steep pitch of 45°, although it leads to a significant loss of weathertightness, the effect of a 5mm gap is much better than 10mm or 15mm (the turquoise line compared to the pink or dark red ones).
The EurActive Roofer project recommendation was actually for a maximum 6mm gap, as there are a number of existing 6mm thick fixings on the market, and this should not perform too much worse than 5mm. Insistence on a smaller gap would lead to a widespread chamfering of tiles or slates to accommodate the fixing, which might have unwanted side effects of reducing the structural integrity of the tiles themselves, weakening both the panel fixing and the roof itself.
In contrast, fixing panels with a membrane or bitumen coated roof, as above on a Polish roof, is likely to give fewer problems, providing manufacturers' guidance on penetrating the roof covering is followed, especially on metal standing seam or mineral roofs. Note that the positioning of the screw into a roof timber is important, as leakage would be more likely if it went through into the void below the roof.
The EurActive Roofer project also looked at integrated panels, and undertook a number of tests on Biohaus systems in Germany. Essentially infiltration can occur in the flashings around the eaves (above, below or down the sides) or in the crossing point where four modules abut. Evidence collected showed that leakage was not a major problem when fitted in line with manufacturers' recommendations, with water penetration similar to a normal tiled roof, but the area at greatest risk was the crossing point in the centre of the four panels.
NEF would like to thank BRE and the other partners in the EurActive Roofer project for their assistance with this page. The information given is for guidance only and should not be used in place of proper engineering calculations in accordance with the relevant British Standards.