The Building Conservation Directory 2022

42 T H E B U I L D I N G C O N S E R VAT I O N D I R E C T O R Y 2 0 2 2 C AT H E D R A L C O MM U N I C AT I O N S is therefore a balance to strike between reducing demand (by improving the thermal performance of our buildings) and transitioning our infrastructure. Within this balance there is arguably room to leave some of our most sensitive assets largely untouched and accept the energy related costs of our national enjoyment of unique buildings. The rest, however, must be upgraded as part of our response to the climate crisis. METHODS OF ASSESSMENT Careful assessment of a building’s thermal performance is necessary to identify both the most effective retrofit approach and the safest, as energy retrofitting exposes all buildings to some risks. Many of these are associated with moisture because the movement of heat is closely linked with that of water. Historic buildings generally manage moisture in very different ways to modern buildings, so materials, detailing and techniques must be carefully selected and designed to minimise these risks. Observations Most assessments, of whatever sort, begin with simply looking at the building and maybe at drawings if they are available. A skilled surveyor, from any discipline, can discern a great deal from careful observation. When we look carefully, buildings tell us their stories, which can be as useful as they are fascinating. A simple but very accurate approach to assessing any building’s thermal performance is to inspect energy bills, and sometimes other data, for example sub- metering information that is sometimes available in larger commercial buildings with more complex systems. The meter never lies, and a simple table or bar graph of monthly figures can be revealing. We can get very rough estimates of, for example, the energy used for hot water as opposed to space heating. This is a simple form of ‘disaggregation’ – estimating what proportion of energy goes to different end- uses – which is extremely useful because it helps us understand how to allocate resources for the best result. It is also typical to ‘normalise’ the data by dividing it by the floor area so the information can be compared with buildings of a different size. Modelling and simulation Calculations, modelling and simulation can play an important role in the assessment of thermal performance. These can range from simple u-value calculations on individual elements, to energy models of entire buildings or even stocks of buildings, perhaps on the books of a housing authority. It is, however, essential to bear in mind that, in the words of the statistician George Box, ‘all models are wrong but some models are useful’. A pertinent example of how wrong models can be in this context is the persistent over-estimation of solid-wall U-values, with significant consequences when it comes to retrofit. In any case, it is important to consider the various mathematical models as tools in our assessment toolbox; they are all designed to do a particular job or range of jobs, and they all have their own strengths and weaknesses. In selecting the right tools, it is helpful to categorise them. They might consider individual elements such as walls, or the entire fabric of a building and its systems. Another distinction is static versus dynamic; a related factor is the specific physical processes the model approximates, and the way it does this. For example, there are several ways of modelling the heat flow through a wall, varying in sophistication. It is not always the case that more complicated is better. Typically, we do not know the precise thermal properties of the materials, and variations in the build-up of the wall such as unexpected air pockets or mortar snots in the cavity can put our estimates way out. Such factors will not be overcome by using a more rigorous model, because it is impractical to measure the presence of such irregularities. It is better to be approximately correct than precisely wrong. The static versus dynamic distinction is most common with respect to whole- building models, which may assume steady flux of heat (often on a month-by-month basis) or on an hour-by hour basis. This additional resolution in the time domain can be useful but can sometimes hinder the process because there are many more inputs to be configured, which is time consuming and can obscure some basic assumptions. On the other hand, dynamic tools can integrate much more detail with respect to aspects such as intermittent heating, solar gain and more sophisticated models of thermal comfort. In the context of historic buildings that can have quite different heating regimes to modern dwellings, this can lead to more nuanced conclusions in the hands of a professional who is knowledgeable in both simulation and historic buildings. One obvious benefit of modelling is that changes to the building can be ‘tested’ quickly and cheaply and with no Roof insulation work in an unlisted building – dormer windows are notoriously difficult to treat within the space available. The evolution of temperature and humidity sensors, clockwise from left: 1) a traditional hygrometer (perhaps 1950s), whirled like a rattle to simulations measure dry bulb and wet bulb temperature; 2) a data logger (perhaps 1990s) to which several temperature and humidity and other sensors can be connected; and 3) a modern sensor/logger. Dynamic modelling of the heat flow through the external envelope of a building helps us to understand its thermal performance, but it cannot take into account irregularities in historic structures.