Roof drainage is a very important issue which can affect the viability of a building. We look at how to get it right and also study changing requirements and new products.
DRAINAGE TO REDUCE LONG-TERM PONDING The concept of the sustainable roof has often focused on the properties of the roof in terms of its energy performance and the effects individual components have on the environment. A third area where designers and contractors alike can contribute to reducing environmental impact is to build and maintain roofs that have a long service life.
Extending the lifespan of a roof is desirable as it postpones the time when new products have to be manufactured, delivered to site and installed, resulting in a better long-term use of raw materials and reducing the volume of bulk waste that will ultimately be dumped in landfill sites.
There are practical measures that can be taken at the design stage of a roofing project to increase the life of a roof. Membrane roofs that have adequate drainage last longer than those with long-term ponding. This view is supported by a survey carried out among roofing professionals in the US where there is a consensus that when a roof ponds, the life of the roof will be shortened by between one and five years.
THE EFFECTS OF LONG-TERM PONDING Many manufacturers of roofing membranes claim that the life of their materials is unaffected by the long-term presence of surface water and there are laboratory test results to support their views. What the manufacturer cannot predict with any certainty is how the roof will be treated and maintained after it has been laid and handed over.
It is a common experience in inspecting aged roofs to find that what were originally intended to be 'temporary' holes, perhaps fitted for new air-conditioning pipework or electrical cables, do not have proper permanent weathering details. On a well-drained roof, rainwater tends to trickle around and away from such open holes. However, if there is local flooding, the situation is much more serious as there is a reservoir of standing water slowly seeping through the penetrations into the underlying insulation and down into the building below.
MINIMUM DESIGN REQUIREMENTS BS 6229: 2003, the British Standard Code of Practice for flat roofs, calls for a minimum finished fall of 1:80 for membrane roofing and 1:60 for aluminium, copper and zinc fully supported metal roofs.
These are absolute minimum falls for any point with no allowance for normal construction tolerances and deflections of the structural deck under imposed loadings.
Thus the design falls should be steeper than the finished falls.
CUT-TO-FALLS INSULATION A popular method in the refurbishment of existing 'flat' roofs is to adopt tapered insulation. Good advice about the practicalities of improving falls on existing roofing is given in the Flat Roofing Alliance Information Sheet No. 27 published in May 1997. It is particularly important to recognise the need to carry out a level survey of the original roof surface at the design stage and before ordering the tailor-made insulation. Without this, the pre-cut tapered boards may not have a properly drained top face and long-term puddles will remain. If the original deck is particularly uneven, then a method for filling depressions should be allowed for as sometimes flood coats of hot oxidised bitumen are not enough to level out significant steps or undulations.
Crickets, laid between low points to introduce a long fall, can further reduce the incidence of ponding, especially in valley gutters and close to drainage outlets. For some existing roofs it may be impractical to use a cut-to-falls system, perhaps because of the long distances between outlets.
In such a situation, increasing the number of outlets may be the most practical means for reducing long term ponding.
FLAT ROOF FALLS - GOOD PRACTICE A totally flat membrane roof relies on a perfect standard of workmanship and regular active maintenance to keep it watertight. In the real world this rarely happens. In the interests of building roofs that last longer, giving economic benefits to the owner and environmental benefits to all, it is good practice to adopt positive roof drainage.
Tenet of sustainable roofing:
'Provide positive drainage to reduce long-term ponding.'
DELUGE RAINFALL August 2004 was one of the wettest months on record.
More than 175mm of rain fell on Bedford, more than three times the monthly average. But what caused the worst damage to property was not the total volume of rainwater but the speed at which it fell. Boscastle in Cornwall had 75mm of rain in just two hours. These heavy downpours remind us of the importance of gutters and downpipes in transferring large volumes of rainwater from the roof down into the underground drains.
RAINFALL INTENSITY The rate at which rainfall lands is known as the 'rainfall intensity'. This is an important measure that forms the basis of the design of a gutter system.
The more intense the storm, the higher the rainfall intensity will be. The units of rainfall intensity are now quoted in litres per second per square metre (l/s/ m 2). This is equivalent to the old mm/hour divided by 3,600.
The British Standard for the design of roof drainage is BS EN 12056: Part 3 that came into effect in 2000. The statistical meteorological data for the UK is given in the figures in National Annex NB.
The frequency and severity of intense short-duration rains are related to geographic location.
Although upland areas of northern and western UK have higher average annual rainfalls, it is in the lowland areas that very intense short-duration rainfall is more frequent.
For the design of roofs we work on a storm of two minutes duration. This is the time taken for rainwater to run off pitched roofs and fill the gutters. For the design of underground urban drainage covering larger areas and taking longer to fill, longer durations are adopted with lower rainfall intensities.
High rainfall intensities can result in local flooding in the gutters, leading to overtopping and water entry into the rooms below. For buildings with valley or boundary-wall gutters, the design rainfall intensity is a critical design issue.
DESIGN CATEGORIES The method of choosing the rainfall intensity is given in the national annex to BS 12056.
The design rate is dependent upon two key factors:
? location in the UK; and - the acceptable risk of water overtopping gutters and draining into the building.
Four different categories of risk are given, summarised as follows:
Category 1. For eaves gutters and flat roofs, where the rainfall intensity could be experienced once every year. Use fi gure NB1 in the British Standard.
Category 2. For valley and parapet gutters, and where the building requires an additional measure of protection. The return period chosen is 1.5 times the anticipated life of the building. Interpolate between figures NB2 and NB3.
Category 3. For buildings requiring a higher degree of security. The return period chosen is 4.5 times the life of the building. Interpolate between figures NB3 and NB4.
Category 4. For the highest possible security. For example, a nuclear facility. Use fi gure NB5.
The procedure for selecting the design rainfall intensity is not straightforward and is open to interpretation and confusion.
This may explain why it is not always done. It would be advisable in the design of a new building for the project designer to specify the rainfall intensity. Alternatively, they should choose a category together with the anticipated life of the building. This then forms the basis for the design of the roof drainage system.
The table gives the rainfall intensities for five different cities, based on the figures in BS 12056 and converted to the more familiar units of mm/hour. For example:
consider a factory in London with valley gutters, requiring a standard measure of protection for an anticipated life of 33 years. The return period would be 1.5 times the lifespan, giving a 50-year period. From the table the design rainfall intensity should be 200 mm/hour. (Refer to Table 1) LATEST MET OFFICE DATA The original rainfall intensity figures from BS 6367 were used in drafting the latest British Standard. These were based on data gathered during the 1970s.
The Met Office set up special recording equipment at selected sites to measure the rainfall over short intervals. The correlations between short-interval and hourly records were then used on the hourly data available throughout the country to prepare the rainfall maps.
Consequently BS 12056 is based on data recorded almost 30 years ago, with no allowances for climate change.
The Met Office measures rainfall in several ways. These include the traditional cylinder rain gauge, the 'tipping bucket' rain gauge and, during the past three years or so, radar data within a 1km grid. The Met Office has produced data from tipping-bucket rain gauges to identify rainfall events with a minimum depth of 1mm.
The apparatus works every time 0.2mm of rain falls, when the bucket tips and the time is recorded. The Deluge software and data tool give the results from 98 different stations, recorded over the period 1987 to 2001. More recent data from 37 stations between 2002 and today has still to be collated and awaits further funding.
Studies are needed at national level to take this basic Met Office data for different locations in the UK and determine the peak measured rainfall intensities for storms with a two-minute duration, to estimate their frequencies and to compare the results with the figures given in BS 12056.
Those who may have a responsibility for the cost of fixing flood damage inside a building should be concerned about this current lack of information. At present we do not have the documentary evidence that says that rainfall intensity rates should change.
DELUGE RAINFALL - GOOD PRACTICE It is worthwhile adopting a robust approach to the design of rainwater goods. If in doubt, increase the design rainfall intensity. This should be defined by the project design team, so that at the tender stage there is fair competition by those offering design-and-build packages. However, it is important to be aware of the imprecise nature of the basic data we use, which was gathered almost 30 years ago.
NEW ROOF DRAINAGE PRODUCTS A new roofing product recently released by Corus Colorsteel is Aquatite TPO, which consists of a waterproof membrane factory-bonded to a galvanised steel substrate. The product has been developed especially for the manufacture of gutter systems, such as internal valley gutters, flashings and parapet copings.
Aquatite TPO is cut and formed using sheet-metal working equipment and then lap joints are sealed using single-ply hot-air welding techniques to form a watertight construction, reducing the need for mechanical fixings and sealants. Specifiers should be aware that TPO membranes are generally more difficult to heat-weld successfully than PVC membranes, due to the narrow welding window of the material. However, skilled tradesmen working in clean, dry conditions can heat-weld TPO membranes to create watertight seals. Alternatively, Aquatite PVC is also available from Corus Colorsteel.
Keith Roberts is a chartered civil and structural engineer based in Oxfordshire who specialises in roofing and cladding. He can be contacted through his website at www. robertsconsulting. co. uk.