Andrew Lundberg, thermal modelling expert at the Association of Thermal Modellers details the challenges faced in accurately assessing building fabric performance, hailing competency as the key…
When it comes to delivering low-energy buildings in a bid to produce homes which come with both increased comfort and lower running costs for occupants, the race is well and truly on. Any designer or specifier would likely confess to sometimes struggling to keep up with the myriad of new products entering the market claiming to deliver what their predecessors or competitors never could – even lower-energy buildings. Running in parallel with these challenges is also the need to keep one’s eye on regulatory compliance, something which has essentially become a numbers game, particularly when it comes to Part L1a for new dwellings.
Amid the rapid development of renewable heat and electricity producing products, it is often easy to forget that the best approach to energy efficiency is not to produce required energy in an efficient way, but to limit the required amount of energy in the first place. This requires first and foremost, a fabric-first approach, which can have a great effect on final heat demand. SAP assessors, who may or may not be members of the design team, are having to find new ways of inputting building data in order to demonstrate regulatory compliance. The risk then becomes one whereby the SAP assessor has to input a value which the actual building, by design, cannot achieve.
One area receiving increased focus is thermal bridging. However it is still too little understood by the industry. What is a thermal bridge? How is it assessed? Who should assess it, and what qualifications should they have?
Thermal bridges exist in every single building ever built. Anywhere that otherwise uniform heat flow through the building fabric is affected by a change in the fabric’s properties, a thermal bridge is presented. This could be junctions of the various building elements, such as wall, floor, roof & window intersections, or any location whereby the building fabric is penetrated by a single point item such as a beam resting on a wall. Even in a straight wall with an embedded structural component at one point, a thermal bridge is present due to the variation in heat flow around that component, even if the U-value is maintained at the same value across the entire structure.
Thermal bridges are assessed in accordance with a national convention document produce by the BRE (Building Research Establishment) entitled BR497: Conventions for calculating linear thermal transmittance and temperature factors. This is carried out using numerical analysis software by means of finite element analysis. As the name suggests, the assessment criteria focuses both on determining the excess energy being lost at junctions, as well as surface temperatures which occur at junctions and what the subsequent risk of mould growth and surface condensation may be for various building types. The former criterion will result in changes to energy consumption and directly affect heating bills, the latter criterion will determine whether mould is likely to form at a specific location under standard conditions. So we must remind ourselves that mould growth doesn’t happen by coincidence…it’s an inherent design property of our buildings. It therefore makes sense to focus firstly on eliminating mould growth, and secondly on reducing excess heat loss across the junction. The thermal bridge assessor needs to have a keen understanding of both phenomena.
So who is the thermal bridge assessor, who determines their competence and where can one find one? Under the latest edition of Approved Document L1a, the competent assessor is someone who has completed training in the software tool that they are using and has achieved results within the range of accuracy in assessing the validation cases in the aforementioned document BR497. No further requirements of the assessor are presented in the document. Previous mentions of a national government approved assessors register have been removed from the latest iteration, so the onus on finding a ‘competent’ assessor lies with whoever is responsible for providing the thermal bridge values for the junctions to the SAP assessor, or the SAP assessor themselves, where non-standard values are being used. The questions being asked of anyone claiming competency therefore should be, at least, “have you completed formal training in thermal bridge assessment?”, and “can you demonstrate that you’ve completed the validation cases from BR497 within the stated tolerances?” Once these questions are answered, the competent thermal bridge assessor is in a position to deliver accurate assessments of junctions, advise on necessary changes to junction design to reduce heat loss, thereby reducing heat losses via thermal bridging, improving SAP values, and eliminating mould growth risk.
In determining energy loss via thermal bridging, three distinct approaches are presented in the Approved Document L1a. Firstly, the building can be designed in accordance with the DCLG Approved Construction Details or another government-approved source involving independent assessment of the construction method. This allows the psi-value for each junction to be taken from table K.1 of the SAP 2012 document. Simple? Well almost too much so. The table presents one psi-value for each junction type, e.g. wall/floor junction with insulation over slab, and is applicable to any construction type and over any range of U-values. The reality is that an external wall corner in a timber frame will likely have a very different psi-value compared to that of an external wall corner built in masonry cavity wall construction and so on. Furthermore, changes to U-values of planar elements will result in changes to the psi-value for the same junction. So although one may satisfy building control with this method of accounting for thermal bridging, it would be folly to think that it’s in any way an accurate account of a junction’s performance. Catalogues of thermal bridges are also being produced by some product manufacturers or private organisations, however, if one is truly interested in an accurate determination of building fabric performance, one should look closely at the background to any published psi-values. Many of these sources have psi-values which are applicable over any range of U-values below a liberal value, making them vague and ambiguous.
Where a junction is not constructed in accordance with the ACD, a default psi-value for the equivalent junction should be used from table K.1. These values are exactly double those of the equivalent ACD psi-value, which means that improvements to the building design elsewhere will have to compensate for the fact that this bespoke junction hasn’t been assessed by a thermal modeller, in order to maintain compliance. This will almost certainly have cost implications for the overall design & build. Another option is to enter a global psi-value (y-factor) of 0.15W/m2K, which in theory encompasses the combined heat losses from all thermal bridges in one figure. This is certainly an approach best avoided, indeed the compensatory measures required elsewhere in the design due to its use could be beyond sensible or reasonable. To put this figure in context, if a building is designed with an average elemental U-value of 0.15W/m2K, and a thermal bridge y-factor of 0.15W/m2K is also used, essentially it is being stated that 50% of building fabric heat losses are due to thermal bridging alone. This is in almost any standard building a gross over-estimation.
And so we return to the concept of the competent thermal modeller. Their integration into the design team from the outset can ensure that junctions are designed in a manner which reduces excess heat loss & eliminates mould growth whilst also maintaining build-ability, reducing the need for expensive compensatory measures elsewhere in the design, and ensuring accurate estimation of building fabric performance. Until we are at a stage of assessing every design by modelling, or available catalogues of details and their respective psi-values take a quantum leap, we are in energy terms thinking one dimensionally, with our estimations of performance in the second or third dimension being at best an uneducated guess.
Andrew Lundberg
Accredited Thermal Modeller under the Irish NSAI Thermal Bridge Assessors scheme, member of the Association of Thermal Modellers, lecturer in Thermal Modelling at the Dublin Institute of Technology
Passivate
Tel: +44 208 144 6946
andrew@passivate.ie
