Practical measurement of the thermal performance of houses

The ‘Loughborough In-Use Heat Balance’ is a new method for evaluating the thermal performance of houses developed by researchers from the Building Energy Research Group (BERG) at Loughborough University. The method has been developed with the aim of making widespread measurement of the thermal performance of houses a practical reality. The method addresses the ‘performance gap’, where the actual thermal performance is significantly worse than predicted. Accurate assessment of building performance would provide a route towards quality assurance, allow targeting of lower performing houses for remedial works, and provide more certain estimates of the energy savings resulting from retrofit.

Almost a third of all energy use and associated CO2 emissions in the UK occurs in houses, and of this around two thirds is used for space heating. However, the UK housing stock is among the oldest in the world, with relatively low thermal efficiency. This is not surprising, as much of the stock was built before the links between human activity and climate change were well understood, and therefore thermal performance was not a priority in the design and construction of dwellings. These factors make housing a clear opportunity for emissions and energy demand reduction, which is particularly pertinent in view of the emissions reduction targets set out in the 2008 Climate Change Act.

Given the rather low demolition rate in the UK, and with new house construction rates well below the required level, retrofitting of existing houses will necessarily form the largest part of any actions to address this opportunity for emissions reduction. A critical component to help design successful policies in this area is accurate assessment of the thermal performance of houses, commonly defined in terms of the heat loss coefficient (HLC) with units of Watts per Kelvin (w/k).

Determining the thermal performance of houses has proven to be a major challenge that has yet to be satisfactorily addressed. In the UK’s Standard Assessment Procedure (SAP), the performance of houses is calculated using an elemental approach, summing the laboratory measured performance of each element of the building. However, evidence gathered using whole house and elemental measurements has shown that estimates of thermal performance are routinely inaccurate, with actual heat loss almost always larger than was predicted.

What is really surprising is that these differences are not small. Actual performance in some cases has been found to be up to 120% worse than predicted, with large variations observed between seemingly similar houses. The two clear lessons are that all houses are different, and that heat loss tends to be higher than expected – creating the motivation for an alternative measurement solution.

At present, the most commonly used whole-house measurement technique in the UK is the ‘co-heating test’. The method involves measuring the energy required to heat the interior of a house to a controlled elevated level, commonly 25 Degrees Celcius, and normalising for the weather conditions during the test. Co-heating tests carried out by Leeds Beckett University have provided much of the evidence for the performance gap and their experimental method is the one commonly followed.

In its current form, the co-heating test, the use of which has been primarily limited to research applications, requires a house to be empty for a period of at least two weeks and can only be carried out during winter months, as it requires a period of cold and preferably dull weather. Furthermore, an extended period of elevated internal temperature has been found to cause damage in newly built houses through the rapid drying out of materials.

Thermal image showing very different levels of heat loss for adjoining semi-detached houses

The Loughborough In-Use Heat Balance has been developed to address these issues as a part of the work of PhD Research Student Richard Jack, a member of the London-Loughborough Centre for Doctoral Training in Energy Demand ( The In-Use Heat Balance uses broadly the same analytical approach as co-heating, with measurements informing an energy balance in the house. The key advance made by the new method is that it allows occupants to remain in their home and behave almost entirely as normal. The energy inputs, typically electrical and from a gas boiler in UK houses, are measured to form one half of the balance, while the internal and external temperature is measured to form the other. The method might be imagined as a leaky bucket – where the rate of water input is measured to find the sum of the leaks. Finally, analytical techniques are applied to account for the influence of the weather conditions during the test, particularly the heat gains due to solar irradiation. The measurements can be taken in a discreet manner in order to have little impact on the occupants’ use of the house.

The method has been tested in unoccupied test houses at Loughborough University, using synthetic occupancy to test the effect of variables such as window opening, hot water use and heating practices, and has shown close agreement with the results of co-heating tests. In a project carried out with the Energy Technologies Institute (ETI), the method has also been used to test the performance of a house before and after a retrofit. In this application the results of the in-use heat balance and co-heating tests agreed to within 6% both before and after retrofit.

At present, a minimum monitoring period of 3 weeks is recommended in order to allow robust normalisation for differing weather conditions and the uncertainty of the measurement is estimated to be ±15%. In practice both of these values will be dependent upon the performance of the house and the weather conditions experienced during the monitoring. The effect of these factors is the subject of ongoing research and, with better understanding, it is likely that both the monitoring period and the uncertainty can be reduced. Information to aid this better understanding is currently being gathered, with the Loughborough In-Use Heat Balance being applied in five occupied homes in conjunction with the ETI as a part of their novel whole-house retrofit approach, which was documented in issue 98 of Innovation and Research Focus.

To develop the method further, a greater body of tests carried out in different house types with different occupants is required, ideally with matched co-heating measurements to provide a baseline comparison.

If you can provide data and would like to take part in this research, or would like further information, please contact the team at Loughborough University, School of Civil & Building Engineering (email Richard Jack at: or Dennis Loveday at:


Download article

Download IRF100

Read IRF100 e-book