A warm flat roof, correctly designed, specified and built, is one of the simplest construction elements that can form part of the building envelope. Structural deck, vapour control layer, insulation, waterproofing; it’s the perfect example of a fabric first approach in action.
It’s a tried and tested solution that the industry knows works, not least because it keeps the roof structure at or around the same temperature as the building interior. That eliminates the possibility of condensation occurring, which could otherwise cause deterioration of the structure, and ensures that the roof performs for the building’s intended lifespan.
From a technical point of view, then, it is disappointing to still hear questions like, “Does the roof need a vapour barrier?”, highlighting that plenty of misapprehensions and misunderstandings remain about the ramifications of flat roof designs. Too often, designers and contractors alike are unaware of fundamental good-practice principles that ensure a roof performs.
Compromises can result from on-site constraints too, a good example being that of hybrid roofs. Imagine a flat roof constructed on timber deck and joists; it’s not surprising that people sometimes look at the space between the joists and wonder if they can use it for additional insulation. If there is a restriction on the thickness of insulation that can be accommodated above the structure then the temptation to utilise the extra space is even greater.
But if the basic warm roof construction is still followed before this extra insulation is placed, problems can arise. Not only is a vapour barrier, or vapour control layer (VCL), a vital component of the roof, it must also be correctly positioned: on the warm (internal) side of all the insulation.
There are numerous reasons why incorrect placement of insulation relative to the VCL should be avoided, perhaps best summarised in the foreword of BS 5250:2011 + A1:2016 Code of practice for the control of condensation in buildings.
Bearing in mind that occupants often fail to use buildings in the manner intended, be it by choice, lack of understanding or force of circumstance, designers are advised to err on the side of caution and adopt robust fail-safe solutions.
To that end, section H.2 of the standard lists three acceptable scenarios for the placement of insulation in a flat roof: cold roof, warm roof and inverted roof. Nowhere does it support the design and construction of hybrid flat roofs.
Some insulation manufacturers, however, are happy to advocate hybrid roof constructions. This is typically when it is considered that the balance of the two insulation thicknesses is ‘correct’, i.e. the layer of insulation between the joists does not have a higher thermal resistance than the layer of insulation installed over the deck.
A condensation risk carried out in accordance with the method detailed in BS EN ISO 13788 Hygrothermal performance of building components and building elements may appear to show no risk. But, for me, the fact that BS 5250 does not support the construction type means that any condensation risk analysis for a hybrid flat roof should be labeled as ‘Not recommended by BS 5250’.
Consider as well another reason to doubt the appropriateness of this analysis of a hybrid flat roof solution. Section 4.2 of BS 5250 says:
…BS EN ISO 13788 considers only the risks arising from the diffusion of water vapour through the building fabric; it does not take account of the much greater risk of condensation occurring as a result of air leakage, which transports water vapour through gaps, joints and cracks in the building fabric.
By its very nature, a hybrid roof that features insulation fitted between timber joists introduces the potential for air gaps. Section A.3 is clear that:
…the rate at which moisture is transported by air movement, where it occurs, is much greater than that of transportation by diffusion.
A condensation risk analysis cannot adequately allow for the standard of workmanship, and a high level of workmanship in itself would not mitigate potential issues.
If anybody is to make a judgement on the levels of workmanship, and whether it makes the construction method any more acceptable for a particular project, it should be the designer, building inspector or end user. Insulation manufacturers can advise on the interpretation of a condensation risk analysis for a hybrid roof, but first and foremost should promote the fundamental principles of good roof design outlined in BS 5250.
Considering one of those principles is to ensure the correct VCL position, some might ask if a hybrid roof is acceptable if the VCL is moved to ceiling level behind the internal plasterboard finish. While technically that might satisfy the basic requirements, practical thinking should be urged: are light fittings or other services going to be installed in the ceiling? Can the VCL’s continuity – vital to its performance – be guaranteed?
It only takes the next owner to want to change something and that VCL could be compromised as part of any works to the ceiling. Will anybody make sure it is restored to its original condition? If a VCL cannot be relied upon to mitigate the risk of condensation then the roof design cannot be considered appropriate, and there are precious few alternative solutions.
Despite the best efforts of many to promote the advantages of service voids – air spaces between ceiling and VCL where services can be installed and the ceiling altered without risking the integrity of the VCL – few seem willing to employ them. There are several reasons why: extra time and materials to construct, restrictions on headroom etc. , all of which serve to highlight why it is best to keep things simple.
One of the positive aspects of a warm roof is that the VCL can be installed with the confidence that it will perform for as long as the roof performs – hopefully for the life of the building. That’s what makes it a good example of fabric first construction, and why it should always be preferred over a hybrid alternative.