Predicting how long a roof will last is a difficult question to answer with certainty, but it is an important question for building owners and funders of PFI projects. We take a look at relevant standards and at several new books published on the subject. Condensation risk and the latest tiling and slating code of practice are also discussed At the specification stage of a project, when comparing the overall project costs of two different roofing systems, one of the most critical pieces of information is the prediction of the likely service life of each system.
Equally, for environmental audits or assessments for sustainable construction, the predicted life of the building element is an essential piece of information that forms the basis of the ultimate results.
The increasing use of private finance for major public buildings, where the financial promoter takes on the responsibility for long-term maintenance and repair of the building envelope, has created a need for reliable predictions of the life of the roof and the annual maintenance demands.
As a starting point, there is the British Standard Guide to the Durability of Buildings, BS 7543, that was last amended in 1998. In addition, BS ISO 15686 parts 1, 2 and 3 on Service Life Planning were published between 2000 and 2002. They apply primarily to new buildings and their components, rather than to alterations and repairs.
The design life is defined as the 'period of use intended by the building designer'. BS 7543 offers a helpful table giving five different categories of design life and relating them to typical building examples. These vary between 'temporary' of less than 10 years for a non-permanent site hut, to 'long life' of 120 years for a civic building.
The standards suggest that, when stating the design life for a building or its parts, you should allow for a time shorter than the predicted life where the usage of parts of the building cannot be foreseen, when untried product s are to be incorporated, or where exceptionally large assemblies of components are involved. BS 7543 discusses briefly a simple categorisation of failure effects, enabling a ranking to be given for a particular building. For example, the building element would be category A if failure would involve a high risk of injury or extensive damage, such as roof cladding becoming dislodged in high winds, and the design should recognise this risk.
Sources of durability information include:
lexperience in the use of traditional materials embodied in Codes of Practice, textbooks and trade association publications;
lcertificates assessing performance of products issued by the British Board of Agrément and other authoritative bodies;
lpublications from the Building Research Establishment and other research bodies, eg BFRC/Napier Assessment of Lifespan of Flat Roofs in 1993;
lpredictions of service life of products provided by manufacturers, eg case studies of roofs covered with the same materials for 20 years and more.
Condensation theory The air all around us contains moisture in the form of a colourless vapour that we cannot see. When the air gets warmer, more water vapour can be held in the air. When the air is cooled, the amount of water vapour that air can hold is reduced, so that there comes a point when it can no longer retain all of the vapour. The water changes back from a gas to a liquid, that we call 'condensate'. The temperature at which condensation starts to form when the air is cooled is known as the 'dew point'.
Within a room, condensation starts to form on any cold surfaces that are at a temperature below the dew point. We regularly see this on glass windowpanes on cold frosty mornings.
Design life assumptions It is important to note that the prediction of durability is subject to many variables and cannot be exact. British Standards do not deal with many aspects of obsolescence, such as when buildings or parts of buildings are replaced or altered to suit different needs or change of use.
Accelerated testing of components by itself can seldom be used to give an accurate basis for predicting service life and is not usually feasible for large assemblies of components. Also, relevant test certificates are not always available from manufacturers and may have to be obtained by testing for a specific project.
Future changes in weather patterns such as more extreme temperatures, stronger winds or more intense rainfall could affect the durability of a roof, although confident predictions of climate change are difficult to make.
Causes of deterioration The life of a roof component will depend upon whatever acts on it, such as water, heat or foot traffic. In many instances deterioration is not the result of action by a single agent, but the effect of a combination of agents. Frequently the action of one is dependent upon the presence of others. For example, an increase in temperature of 10°C can double the rate of a chemical reaction causing deterioration. Moisture is probably the agent with the greatest influence on the durability of building materials. Many types of deterioration, such as rusting of unprotected steel, can take place only if moisture is present. Understanding the agents that can affect future service life can lead to improvements at the design and specification stage, resulting in a longer service life.
Controlling condensation Controlling condensation will prolong the life of the roof and the internal fabric.
Condensation theory is not new and should form part of basic training for building designers. What is new is the introduction of European Standards and test methods, together with tighter regulations to reduce energy loss from buildings. Building control officers are now actively applying Part L of the Building Regulations, and at least one legal action against a major housebuilder is in progress.
Condensation can occur in buildings when water vapour condenses, either on exposed building surfaces where it supports mould growth, or hidden from view within building elements and known as 'interstitial condensation'. Condensation and mould problems are widespread, affecting about 15 per cent of homes in the United Kingdom to some degree. The dampness and associated mould growth can be distressing to occupants and are a major cause of respiratory allergies. The British Standard Code of Practice for the Control of Condensation in Buildings was published on 1 November 2002 and supersedes BS 5250: 1989, which is now withdrawn.
Buildings at risk Some buildings suffer from condensation problems more than others do. A warehouse storing dry goods presents the minimum risk, whereas a building enclosing a swimming pool presents the highest risk. A commercial roof system that is suitable for a warehouse is therefore unlikely to be satisfactory for a pool hall. This is a basic fact of which everyone in the roofing and cladding industry should be aware.
The new version of BS 5250 is helpful in introducing for the first time a definition of five different types of buildings, as shown in Table 1. Unfortunately an upperbound condition is not given for 'special buildings'. In future, for all building specifications, the building designer should identify in which humidity class the building or parts of the building will be.
System designers and merchants selling vapour-control layers can then design appropriate roof constructions and supply the correct materials.
Risk analysis The basic formulae and method for predicting condensation formation have not changed. However, the standard values for thermal conductivity and vapour resistivity of different materials have been updated to comply with CIBSE Guide A3 and BS EN 12524. Manufacturers should now quote the .
90/90 value for the thermal conductivity of their product. This is the value that 90 per cent of the materials delivered wi ll exceed.
Future analysis of condensation risk, including the use of computer programs with defined material values, should use the latest published information.
A common problem that has arisen in recent years is that condensation risk analysis programs have predicted the formation of condensation, although to only a small degree, such as 5g per square metre after 60 days of cold weather. This would be equivalent to a film of water 0.005mm thick, which is practically nothing. The new version of BS 5250 is of assistance in assessing the effects of condensation by providing new information. These descriptions put the rate of condensation into context and were based on research commissioned by the Metal Cladding Roofing Manufacturers Association and undertaken by the Building Research Establishment at East Kilbride.
Ventilation of tile and slate roofs One of the methods for reducing the amount of condensation that can form within a roof is to provide ventilation to the outside of the building. Within the updated BS 5250: 2002, detailed guidance is given on the application of design principles.
Section 8.4 of the standard describes the minimum ventilation openings required to ensure air movement through roof voids, transferring water vapour safely out of the building, and allowing the roof to 'dry out'.
The minimum free air spaces defined should be maintained through ventilation products and roof voids alike, with particular attention paid to potential restrictions at changes in roof slope and at junctions with walls. If condensation does occur, it will normally be on the underside of the roof underlay, whatever the type.
The new version of BS 5250 gives guidance on the use of vapour-permeable underlays. Clause b) on page 27 allows for the use of vapour-permeable underlays without ventilation between the underlay and insulation, provided that there is a ventilated air space above the underlay and below the tile/slate. The minimum size of counter batten for ventilation purposes should be not less than 25mm, and ventilation openings are necessary to each and every counter batten void at both low and high levels. Hartington Conway has introduced new slate and tile vents for these applications.
The provision of ventilation openings to the perimeters of roofs of all types may increase the risk of water entry, particularly on shallow slopes of exposed buildings.
Ventilation openings also add complexity to the detailing and add cost. However, as clearly set out in the latest version of BS 5250, there remains a need to provide adequate roof ventilation to overcome the formation of interstitial condensation, especially over buildings with high internal humidities. Curing condensation problems on a roof after it has been built without adequate ventilation is not straightforward and can become costly. The best advice is to follow the guidance given in the latest version of BS 5250.
Getting lead right When detailed and fitted correctly, lead sheet has proved to be a reliable long-life material, with outstanding durability even in the most severe weather conditions. It remains popular for flashing and weathering applications, where it can be cut easily and fitted using only a few simple hand tools. Its extreme malleability makes it ideal for dressing over complicated shapes, particularly the multi-curved contours of many roofing tiles.
Earlier this year the Lead Sheet Association published a new book, Rolled Lead Sheet - A Complete Manual. It is an update of the previous three-part Lead Sheet Manual, revised to conform to the latest British Standards. The book has many colour illustrations and is available from the Lead Sheet Association at £40.
Cashing in on failure When there is a problem with a roof there are usually lessons to be learnt, although as an industry we are often slow to recognise what the lessons are and to take corrective actions. In the past senior members of the industry reaching retirement have been encouraged to write down and record their experiences and findings. Carl Cash is, with respect, one such 'old soldier' who has more than 40 years of experience in the roofing industry in North America. He has recently written a book entitled Roofing Failures, in which he shares his practical experiences, employing the wit and humour for which he has become renowned in presentations to conferences over many years.
The 253-page book is divided into two parts: an introduction that explores the characteristics of different types of roofing systems and defines 'failure', and a second part that gives a collection of case histories.
The book has a full index, together with appendices listing trade names and useful websites. It is not a dry textbook to be put on the shelf. It contains a wealth of common sense that is easy to read in short sections.
A chapter is dedicated to special roofing features at intersections. Flashings are perhaps the most critical area of any roofing system because they are where stress is likely to be concentrated; they are also the area that requires the greatest time, attention and skill from the designer as well as the roofing tradesman. It is the first place to look for any problems, including wear and dislocations.
Each of the chapters in the second part of the book contains a situation or case history that illustrates a type of roofing failure.
The names of the participants, locations and the events have been altered to 'protect both the innocent and the guilty.' They are allegories intended to be illustrations of positions or situations to avoid, preventing a repetition of the failures. At the end of each case study a set of questions is posed, such as who was responsible for the fault and how could it have been prevented? As with any good quiz book, the answers are given in the back, albeit in an abbreviated form that encourages the readers to think for themselves.
In drawing to a conclusion, Cash recognises that the cases presented show that there are repetitive elements and he goes on to identify corrective actions that could have prevented the failure from taking place, or at least minimised the event's impact.
The principal actions, described in more detail in the book, are to:
lhave an effective peer review;
luse systems that have a successful track record;
luse inspectors hired by the owner to oversee the installation;
lbuy competence in your supplier, designer and contractor.
Roofing Failures does not set out to make the reader a roofing systems designer or specifier. Instead the intent is to present an overview of the subject so that the reader can have at his command the knowledge to ask intelligent questions, and thus avoid some of the problems observed by Cash and his colleagues over the years. The book is published by Spon Press at £40.
Fire prevention One of the major problems in roof contracting at present is the increased cost of employers' and public liability insurance, particularly for contractors working with hot bitumen and naked flames. Over the past decade, Icopal has been researching and developing alternative bituminous waterproofing systems that can be laid without a naked flame. Earlier this year, it launched the patented FireSmart Roofing Membrane System, which apparently is the only roofing membrane accredited by the Loss Prevention Certification Board. The FireSmart Gun is capable of creating high temperatures, satisfactory to lay SBS modified products efficiently. This gun has been developed using turbine technology, and works on propane gas with a 110V electric power supply. Unlike traditional hot air guns, only hot gas comes out of the nozzle with no oxygen. This impedes any ignition of combustible material during membrane application, reducing the risks of a fire developing.
One of the difficulties facing surveyors of existing buildings constructed w ith metal composite panels with foam insulation is how to identify the original product and its fire rating. Kingspan Insulated Panels has introduced a new approach to labelling roof, wall and ceiling panel systems by adding markings that are readily visible in ultra-violet light. The discreet markings are applied near to panel edges in an invisible, ultraviolet sensitive ink, providing critical details regarding the date and time of manufacture, and whether they are one of Kingspan's Firesafe LPCB and Factory Mutual insurerapproved systems.
Tile and slate codes In the UK there are perhaps half a dozen principal Codes of Practice that give guidance and recommendations on how roofs should be designed and built. One such British Standard, BS 5534: the Code of Practice for Slating and Tiling, has been revised recently. The new edition was published in June of this year and will supersede the old BS 5534 Part 1: 1997 that is to be withdrawn on 1 January 2004. The new edition is a full revision of the British Standard and has grown in size from 78 to 133 pages.
A new section covers proprietary tiles and artificial slates. Where manufacturers' recommendations vary from those given in the British Standard, the designer is reminded to seek evidence that the product is fit for its intended purpose, and that the product should only be used if the evidence is available.
Chapter 5 has been revised to 'reflect the new knowledge and experience on rain resistance and wind load resistance'. The calculation for the minimum head-laps and side-laps in slating was previously based on angles of creep. This has been replaced with a new method based on a series of factors, c and E 1, giving a parabolic curve that is a better approximation to the actual drainage path between two slates. The effective result of the change is that minimum head-laps for double-lap fibre-cement and natural slates have been revised, as summarised in the table below.
Designers should be aware that, for shall ow roof slopes in severely exposed areas, the minimum head-laps have been increased by up to 23mm. This could increase the number of slates required on a roof by 5 per cent.
The revision to BS 5534 brings the British Standard up to date with current practice and the latest European Standards. It follows the trend of increasing the number of calculations that someone in the manufacture/design/construct team will need to undertake. To assist specifiers, Marley Roofing Products and Eternit Slates now offer the Marley Assured Roofing Specification scheme, that gives a 15-year design performance indemnity cover on all roofs constructed using the company's products. Lafarge Roofing has recently introduced the 'DesignMaster 4' CD that has more than 2,500 roof designs for pitchedroofing specification. These developments help to ensure that the roofing industry can offer trained and experienced technical staff who know how to use the new British Standards with confidence, enabling us to design and build roofs that will last.
Keith Roberts is a chartered civil and structural engineer based in Oxfordshire who specialises in roofing and cladding. He can be contacted at www. robertsconsulting. co. uk Acryp 1500 Alwitra 1501 Conway Icopal 1502 Hartington 1503 Keston 1504 Kingspan Panels 1505 Lafarge Roofing 1506 Lead Sheet Association 1507 Matthew Hebden 1508 Marley Roofing 1509 MCRMA 1510 Rigidal 1511 Sandtoft 1512 READER ENQUIRIES Enquire at www. ajplus. co. uk/ajdirect Humidity Building type Design internal conditions class Air temperature humidity point pressure 1 Storage 15°C 50% °C 0.9 2 Offices, 20°C 50% °C 1.2 3 Dwellings with low occupancy 20°C 60% 2°C 1.4 4 Dwellings with high occupancy 20°C 70% 5°C 1.7 Sports halls, kitchens, canteens, buildings heated with unflued gas heaters 5 Special 25°C >55% 5°C >1.7 Laundries, breweries, swimming pools Table of internal humidity classes, and design conditions derived from BS 5250: 2002 for external temperatur e of 0°C Roof Minimum head lap (mm) pitch Moderate exposure (<56.5 l/m 2)Severe exposure (>56.5 l/m 2) Old standard New standard Old standard New standard >45°,< 75° 65 54 65 69 40° 65 60 75 76 35° 75 67 75 86 30° 75 77 75 98 27.5° 85 83 90 106 25° 90 91 100 16 22.5° 100 101 115 128 20° 115 113 130 143 Table showing minimum head laps for double-lap slates, taken from BS 5534: 1997 and BS 5534: 2003. Note that narrow slates may require longer head-laps