The interest in mounting solar-energy systems on the top of exposed roofs has raised questions about the risk of them blowing off.
Experience tells us that lightweight components, such as television aerials and satellite dishes, are often the first to be damaged or come loose during a severe wind storm.
When placing lightweight photovoltaic (PV) panels on a roof, it is important to ensure that they will stay up there securely for the lifetime of the building. The Building Research Establishment (BRE) has recently published Digest 489, giving recommendations for calculating wind loading on PV panels.
The digest was published in August 2004 following work undertaken as part of a Partners in Innovation (PII) project funded by the Department of Trade and Industry. The PII project team included manufacturers, PV specialists, trade associations and the BRE.
To determine whether or not a solar panel will be blown off a roof requires an estimate of the applied wind pressure, which can then be used to check that the applied loads on the fixings and roof structure below are not excessive. The starting point is to determine the wind pressure for the site. The digest helpfully gives a simplified method, based on a minimum amount of information known about the site. For example, the dynamic wind pressure for a 10m-tall building on a level site in Nottingham would be 1.2kN/m 2. The digest gives recommendations for the coefficients of pressure for PV modules mounted in four different ways. The panels are usually aligned at a slope of around 35 to the horizontal and facing south. The digest gives more detailed information about the selection of the coefficients and the assumptions made. Arrays of PV tiles are considered to be 'air-permeable' where the individual PV units are no more than six times the area of the surrounding individual roof tiles. For a solar panel mounted on a stand near the roof edge, the peak coefficient of pressure would be -1.8, acting in suction.
Knowing the dynamic pressure for the site, the coefficient of pressure and the area of each PV module allows you to calculate the design wind pressure. For the worked example, the wind suction-loading acting on a single solar panel - each approximately 2 x 0.6m, or 1.2m 2 in area, mounted near the edge of the roof located in Nottingham - would be 1.2 x 1.8 x 1.2 = 2.6kN. This is equivalent to 265kg of weight acting in suction. One way to picture this is to imagine roughly five 50 kilo bags of cement acting upwards, trying to lift the solar panel up off the roof. The wind can exert large forces on roof cladding and components.
PV modules mounted on roofs are often in exposed locations and will be subjected to buffeting by gusts of wind throughout their lifetimes. It's important to stop them working loose and falling into public areas, so assessing the performance of PV installations under wind loads is important. BRE's Digest 489 is welcomed in giving the basic information for designers and roofers to assess these risks and is essential reading for those involved in this specialist market.
Changing the code Within the next few years, the 'British Standard Code of Practice for Wind Loading on Buildings' (BS 6399 Part 2) is to be replaced by the new 'Eurocode 1: Actions on Structures - Part 1.4 General Actions - Wind Actions' (BS EN 1991-1-4). This standard has now been approved formally by the CEN (European Committee for Standardisation) and it is expected that the full document, including the UK National Annex, will be published by the end of 2004.
This will be followed by a transition period, with the two standards running in parallel. As with the previous changeover from CP3 to BS 6399, time and resources will have to be allocated for training in understanding the new Eurocode and putting it into practice.
Specifications for roof systems that refer to the current standard and give site parameters such as basic wind speed, altitude and terrain will need to be amended, of course.
Let it snow Apparent changes in weather patterns in recent years have reduced the number of times we have experienced heavy falls of snow during the winter months. However, the need remains to assess the design snow load in accordance with BS 6399 Part 3 and ensure a roof structure can withstand this.
The UK is considered to have a 'maritime' climate. This means that most parts of the country will experience a maximum snow-load condition resulting from a single snowfall rather than as a result of snow accumulation over several months. The British Standard is based on this principle of a single snow event.
Essentially, there are two basic types of snow loading: the uniform snow load dependent on location, altitude and roof pitch; and an asymmetric snow load resulting from the redistribution of snow by the wind, causing drifting into valleys and against parapets and obstructions. The weight density of snow in a local drift is assumed to be 2kN/m 3. A method for estimating the snow-load shape coefficients for asymmetric snow loading is given in the British Standard, which was last reissued in September 1996 with further explanation in BRE's Digest 439.