What’s in a U-value calculation?
U-values are an important indicator of building performance. So why is there such limited understanding of what calculations show? This article was first printed in Issue 12 of Insulate Magazine, read it in full here.
Mainly because the standards describing how to calculate U-values are complex and inaccessible. For the majority, there is no reason to try and understand them – but the knock-on effect of that is assumptions.
Assumptions that any calculation committed to paper has been done accurately; that the piece of paper is automatically correct. If the result is the right number then all is well. If it shows a higher number then suddenly there are problems to solve.
There are a variety of software packages for calculating U-values. In theory, they are all created equal as they are based on the same standards and calculation methods, but their user- and reader-friendliness can vary. The images accompanying this article are from one widely used program and illustrate what a good level of information looks like. Other software will display it differently, but the important thing is knowing what to look for.
(1) Project details and construction type
In an ideal world, every U-value calculation should be for a particular project. Of course, specifications get repeated from one job to the next, so it’s easy to understand why generic calculations might be relied upon, but relating the calculation to a project leaves no room for doubt as to what has been specified and what should be constructed on site.
Stating the application or type of construction also makes clear to the reader what exactly is being calculated – just in case a lack of detail or confusing terminology elsewhere means it is not entirely obvious.
(2) Material layers
This section provides the greatest scope for ensuring the proposed construction can be well understood. The build up is listed layer by layer, with a description of the material and/or a product reference where necessary. If proprietary products are not being used then the description should reference a formal standard from which acceptable material data has been obtained.
It can be easy to open a catalogue of materials and unthinkingly import layers with bland descriptions. Sometimes extra clarification isn’t required, but the quality of a calculation is significantly enhanced when the layer descriptions include appropriate detail.
For example, if the person doing the calculation has to make an assumption, or factor in a worst-case scenario, it can be described here. Again, it shows that the calculation is specific to the project in question; that some attempt has been made to interpret the specifier’s intentions and make the calculation a fair representation of what is likely to be constructed.
(3) Material data
Although insulation materials contribute the bulk of thermal performance in any calculation, it’s still necessary to take account of every layer in the build up. The calculation also has to factor in the direction of heat flow (i.e. downwards, horizontal, upwards) for the purposes of establishing surface and airspace thermal resistances.
Working out the thermal resistances of materials requires their thickness and thermal conductivity. Sometimes a calculation doesn’t display these values – such as when the layer is a very thin membrane that has no meaningful thickness and no declared thermal conductivity.
The order in which materials are listed is not important for calculating the U-value. It does matter, however, for the accuracy of any accompanying condensation risk analysis – and for aiding the readability of the calculation. For anybody comparing the calculation to a design specification or on-site construction, it is common sense and good practice that calculations reflect the correct sequencing of layers.
(4) Bridging details
Calculating U-values is easiest when a construction comprises simple, uninterrupted material layers of consistent thickness – but layers often have to be interrupted by other materials. Accounting for bridging is important, especially when it’s an insulation layer and the disparity in thermal performance of the two materials is significant.
This section states what percentage of the layer is bridged and by what material. But it might not mean anything to the reader. It’s not obvious what “12.5% timber bridging” represents in reality, but if the layer description says the material is bridged by 50mm wide timber joists or studs at 400mm centres, then the world suddenly makes more sense!
(5) Result and corrections
And so we reach the headline of the whole document! A figure for the predicted loss of heat energy (in Watts, W), per square metre of the construction fabric, per degree of temperature difference (in Kelvin, K) between inside and outside.
But what are the numbers that accompany the result?
Upper and lower limit thermal resistances refer to the two methods of calculating heat paths through the construction, as described by the Combined Method. These two values are averaged, then a reciprocal taken, to give the U-value.
Within the confines of mathematically modelling construction build ups, certain corrections can be applied to reflect the realities of installing insulation. In any calculation, you may see values associated with the following:
- Uf – corrections for mechanical fixings penetrating the insulation layer.
- Ug – corrections for air gaps in the insulation layer
- Up – correction for compression of insulation in built up metal roofing and cladding.
- Ur – correction for rainwater cooling on inverted roofs.
- Urc – corrections for rails and/or brackets supporting rainscreen cladding.
We don’t have the space to delve into these corrections, but if they impact on the calculation sufficiently then they cause the result to be changed. It’s important to ensure that comparable corrections have been included when judging one calculation against another, otherwise the comparison is not a fair one.
(6) Application-specific information
In the last issue of Insulate we looked at standards that define how U-values are calculated. The Combined Method is fairly limited in scope, so other standards supplement it with calculation methods for ground floors, anything involving steel components etc. Where these other standards are employed by the calculation software, the relevant information is displayed in this section.
One point worth noting for ground floor constructions: while it’s tempting to think that perimeter upstand insulation provided to screeds and slabs would be classed as edge insulation, for the purposes of a U-value calculation it isn’t. It is there to treat the thermal bridge, so could be referenced in the layer description to show it has been thought about (e.g. ‘Reinforced concrete slab with perimeter upstand insulation to address thermal bridging’).
Anybody calculating U-values should be armed with as much information as possible about the intended materials, and should ask lots of questions where relevant information is missing. Judging the veracity of calculations is difficult for many, but communication is key. Knowing that a design has been accurately calculated, or that what is being constructed reflects what a calculation shows, can go a long way to helping address performance gap issues.