Closing the Performance Gap in Non-residential buildings
January 23rd, 2018
There is an increasing awareness in the construction industry that buildings do not perform as expected in terms of energy and carbon emissions, an issue which has become known as the “performance gap”.
For example, in a recent study by Innovate UK, it was shown across a range of non-domestic buildings that carbon emissions were 3.8 times high than design estimates, and that energy use varied widely. There are a range of possible causes for the performance gap in buildings. Here we outline a number of issues we have experienced and how they may be overcome in the design and construction of buildings.
For most new buildings a computer model will be produced to demonstrate compliance with Part L of the building regulations. This involves the use of approved software which calculates the carbon emissions of the building, which must be less than or equal to a target generated by the software. This provides a means of checking compliance with building regulations, but is problematic for predicting actual building energy consumption and emissions. The models contain many generic (and unalterable) assumptions about the building use and internal heat gains, and often simplistic methods of calculating system performance. We often find that models will predict heating demands lower than Passivhaus buildings, when we know the real building demand could be more than five times this amount.
Building-specific energy modelling is needed to be able to reasonably estimate building performance. The modelling must take into account the specifics of building use, appliances, fabric and building services performance. However, like all models, the accuracy of the results is limited by unknowns. Calculating lighting energy use, for example, is relatively straightforward, whereas predicting vending machine energy performance may turn out to be a wild goose chase! Some other areas of uncertainty, like the impact of thermal bypass in insulation, or thermal bridging, are discussed further below. Where uncertainties are identified, it is useful to identify the potential impact through sensitivity tests.
A key feature of the Passivhaus approach is that many uncertainties are simply designed out or eliminated by careful construction. Not surprisingly, Passivhaus buildings can be successfully modelled using an excel spreadsheet.
Insulation & Thermal Bypass
The energy performance of building elements such as walls is calculated by reference to the “U-value”, which describes the amount of heat passing through 1m2 of surface. This value is theoretical, and assumes that the insulation has been fitted perfectly over the element. Thermal bypass describes a number of mechanisms by which heat can bypass the insulation layer, and results from imperfections in the installation. Additional heat loss is created when there are gaps between insulation boards or behind insulation boards, or in the worst-case scenario, both.
Some specific examples include;
- Poorly cut or fitted insulation boards, resulting in gaps
- Mortar snots in masonry walls, resulting in gaps behind the insulation
- Poorly fitted substrate (e.g., cement particle board)
- Fixings and brackets in the insulation layer which prevent insulation boards being laid flat
- Unsealed party wall cavities
The effect of thermal bypass can be significant- potentially resulting in heat losses being triple what was calculated. In many cases the effect cannot be quantified- so it is best to avoid as far as possible.
How can thermal bypass be avoided?
- Where possible use flexible insulation products which when fitted tightly will compress and expand to fill gaps.
- Design for air-tight and wind-tight construction- preventing air movement greatly reduces transfer of heat.
- Simplify junction details to make fitting insulation easier- avoid difficult angles and kinks in the thermal line
- Robust specifications for installation quality (see below)
- Training of installers to understand the issue
- Development of installation methods for each type of insulation product
- Cutting tools suitable for the insulation type
- Site monitoring of insulation fitting
- Thermography to spot problems (although it’s normally too late by this point).
Thermal bypass Examples: Left: Gaps between insulation boards allow air and heat to bypass the insulation layer. Right: insulation boards not joined at the corner resulting in thermal bypass. Gap behind board on left will also allow air circulation.
In non-residential developments, thermal bridging is rarely given attention, despite building regulations requirements for “continuous insulation”, calculations of junction heat losses, and a system for site inspections. In practice, the impact of thermal bridging is largely ignored, with only a nominal allowance made for it in the energy model.
The impact of thermal bridging becomes more significant for building’s with high levels of insulation, which would include buildings constructed to the latest Part L2A standards. Based on a simple 5x5m ground floor room, we calculated that with high levels of insulation, heat loss from thermal bridges could account for 28% of the total for the space, and therefore not considering them could result in underestimating the heat demand by 39%.
As well as increasing energy consumption, thermal bridges also create cold surfaces which can result in condensation forming internally, or within the fabric. Internal dampness creates ideal conditions for mould growth, with the associated health effects on occupants.
How can thermal bridging be reduced?
- We would take a ‘eliminate, reduce, calculate’ approach. Where possible, thermal bridging should be eliminated by ensuring insulation continuity (a building regs requirement!). Where thermal bridging is inevitable, try to achieve some measure of insulation, or use thermal breaks. Where thermal bridges remain, their effect should be calculated using numerical modelling to generate a φ-values which can be fed into the energy model.
- As a rule of thumb, a thermal bridge can be ignored if two thirds of the insulation thickness is maintained.
- For steel structures, structural elements should be kept on the warm side of the insulation (and generally the air barrier too). The conductivity of metal is over a thousand times that of insulation, and if penetration through the insulation layer will greatly increase heat losses.
- Glazing should ideally positioned in the middle of the insulation layer.
- Window and door heads should be detailed to minimise bridging (e.g., thermally broken lintels).
- Consider using timber for structural elements instead of steel. Timber is much less conductive than metal, and can be formed into insulated cassettes e.g., to construct a parapet.
- Simplify the thermal line. Overhangs or external shading could be supported independently of the internal structure to avoid bridging the insulation layer.
- Where structural zones are insulated (e.g., timber frames, or steel SFS systems), additional elements (fixings, studs, noggins etc) may be installed that are not shown on the drawing, increasing thermal bridging. Better to be pesimistic, or to avoid insulating structural zones!
- Avoid complex 3-dimensional thermal bridges! These require specialist software to analyse and much better to eliminate the bridge.
- Cladding fixings and wall ties can add significant heat loss. Lower heat loss products are available such as basalt wall ties, or thermally broken brackets.
Heat losses for a 5x5m ground floor room, for different insulation standards. Thermal bridge heat losses based SBEM technical manual.
Results from a numerical thermal model of a ground to wall junction. Colours show the temperature through the junction.
Thermal bridging examples. Left: heat loss through the lintel above the window. Right: heat loss under the external walls around the perimeter of a building.
Controls for mechanical systems and lighting, if set up correctly, are vital for minimising building energy consumption, by providing energy-consuming services only when needed. Control strategies such as weather compensation of boilers, or free cooling in ventilation systems, can reduce energy consumption by optimising the operation of the plant itself.
However, it is tempting for services engineers to push for increasing levels of complexity in control systems, in pursuit of greater and greater energy savings. In reality such complexity is more likely to lead to poor performance, as the chances of set up problems increases, and the end user is left with a system they can neither understand or operate correctly. A balance is needed between simplicity and sophistication, which will be different for each project and application.
Some key questions to ask might be;
- Can standard control strategies be used rather than complex bespoke ones?
- What are the big energy consumers in the building and how can the controls reduce these?
- Do the energy savings justify the complexity (and cost) of the controls?
- What level of expertise does the client have to operate the control system?
- Would providing packaged controls with each plant item be a more appropriate than a full BMS?
- Has the appropriate level of user control been provided?
- Has sufficient monitoring been provided to identify issues?
It is tempting to add increasing complexity to control systems to squeeze out as much energy reduction as possible, but the reality might be very different. As complexity increases, so does the chance of errors in design, setup, commissioning, and operation, which inadvertently increase energy consumption.
Data, data, data
It takes time in new buildings to optimise performance in terms of energy and environmental control, as well as flushing out latent defects. The availability of data from a BMS system is crucial for first quantifying performance issues and defects, and then monitoring subsequent improvements.
To provide such data, the BMS must be set up to log the required parameters, and then to store the data somewhere. Normally the BMS hardware itself will have limited storage capacity, so a suitable location on a server or PC hard drive must be found. Once data is obtained it must be analysed and compared against expectations to build a clear picture of the issues.
 – Approved document L2A, 2013.
See the Service page for more information on how we can help you design buildings and optimise them for comfort and energy efficiency.
Back to News & Knowledge