Heat and temperature

Heat is the internal energy possessed by matter: a material or object with a high level of internal energy has a higher temperature than a body with a low level of internal energy. There are two components of heat to consider:

• * Sensible heat – the heat which results in a change in temperature of a material (‘sensible’ with the meaning of being sensed or felt)
• * Latent heat – the heat involved in changing the phase of a material, without affecting its temperature (e.g. converting liquid water to water vapour requires energy, while solidifying a liquid wax will release energy)

Most of this chapter addresses sensible heat: a use of latent heat is also explored (see: Phase change materials p.35).

Temperature is the fundamental measure of heat, and is measured in kelvin, indicated by K. The scale starts at 0 K – absolute zero – with the triple point of water (the temperature and pressure at which water can be gas, liquid and solid at the same time) at 273.16 K. One kelvin is defined as 1/273.16 of the thermodynamic temperature of the triple point of water.

For day-to-day we use the Celsius scale, which has the melting point of ice (at standard pressure) at 0ºC and the boiling point of water at 100ºC. For conversion between the two scales, a kelvin is equivalent to a degree Celsius, so subtract 273.16 to convert kelvins to degrees Celsius, and add 273.16 to convert degrees Celsius to kelvin. It is worth noting that there are still a few countries (in particular, the United States) where the Fahrenheit scale is in use as part of the imperial system of weights and measures.

Determining the temperature experienced by building occupants requires the consideration of two different components of heat:

• * The radiative temperature, resulting from radiation received and emitted (see: Radiation p.6);
• * The air temperature.

Those two values are combined, by a weighting method which includes the effect of air movement, to give the operative temperature that can be used for design purposes.

Establishing a suitable design temperature can be even more difficult, because a building occupant’s perception of whether a building is too hot or too cold can be affected by the prevailing outdoor temperature as well as factors such as dress codes.¹

Heat capacity and thermal mass

The application of heat to a material raises its temperature, and the extent of that rise depends on the material: some require a lot of energy to warm them, others require less. The amount of energy needed to raise the temperature of a material is expressed by its specific heat capacity, which is the energy needed to raise the temperature of one kilogram of the material by one kelvin (measured in J/kgK, joules per kilogram kelvin). As table 1.1 shows, many construction materials have similar specific heat capacities.

The thermal mass of a material is also affected by its density, so a more practical measure is the heat capacity, which is the energy required to raise the temperature of a layer of a certain thickness. The heat capacity of a layer is calculated by multiplying its density, specific heat capacity and thickness (in metres). Table 1.1 shows that mineral wool and dense concrete blocks have a similar specific heat capacity, but very different densities, so the heat capacity of a 100 mm layer of concrete block is nearly a hundred times that of a 100 mm layer of mineral wool.

Table 1.1 Heat Capacities of Common Materials

Heat transfer mechanisms

Where materials are hotter or colder than adjacent materials or surfaces there will be a transfer of heat from the hotter to the colder. There are three main mechanisms of heat transfer: conduction, mass transfer and radiation.

Conduction

Conduction heat transfer occurs as the molecules in solids, liquids and gases vibrate and collide against each other. Energy is passed from molecules with higher internal energy to those with lower internal energy, raising their temperature, which results in heat travelling through the material. Heat travels through different materials at different speeds: you can experience this when you touch a metal door handle and then the face of a wooden door. The metal handle feels colder, because it conducts heat away from your hand more quickly than the wood.

The rate of conduction heat transfer is affected by:

• * The proximity of molecules – in solids the molecules are packed more tightly together than those in liquids, so energy can pass more easily between them. In liquids the molecules are further apart, so fewer collisions occur. In gases the distance between molecules is even greater, resulting in less frequent collisions and even slower energy transfer.
• * The arrangement of the atoms (in solids) – for example, the atoms in metals are arranged in a tight lattice, which means the increased vibrations of the atoms produced by heating are transferred rapidly to neighbouring atoms.
• * The presence in some materials of free electrons – if some electrons within a material are not attached to atoms (typically, in metals) the electrons move when electronic or magnetic fields are applied; they move easily when their energy state is raised, so transferring energy through the material.

The rate at which heat travels through a material is expressed in its thermal conductivity, which is the amount of heat transferred per second through one metre of a material, with a one kelvin temperature difference between one side and the other. The unit of conductivity is W/mK (watts per metre kelvin). The conductivity of a material is often referred to as its lambda value, λ.

Most construction materials are composites, consisting of many elements amalgamated in different proportions. In addition to solid matter, construction materials also contain matter in gaseous and liquid state (timber, for example, typically has a moisture content between 10% and 20%). Thermal conductivities are therefore established by laboratory testing of samples.

Table 1.2 sets out the conductivities of a number of common construction materials. The range of thermal conductivities is huge; metals have conductivities 425–8000 times greater than those of thermal insulants. To put those values into context: in order to achieve the same performance as 100 mm of expanded polystyrene insulation you would need 420 mm of autoclaved aerated concrete, 3.67 m of concrete and 56.7 m of stainless steel.

Table 1.2 Thermal Conductivities for Common Construction Materials

Mass transfer

Heat is transferred in gases when molecules move by ‘mass transfer’ as a result of the mechanisms of convection and advection.

Convection occurs when part of a volume of gas is heated: the molecules gain more energy and random motion increases, making the heated gas less dense. The heated gas is then displaced by cooler, denser gas, making it rise. The continued presence of a heat source will result in currents of heated gas rising and cooler gas sinking, transferring energy with the molecules. You can sense convection in gases by holding your hands above a central heating radiator. Air in contact with the radiator is warmed (mainly by conduction) and expands. It is then displaced by cooler denser air and forced upwards, as shown in Figure 1-01.

Figure 1-01 Convection Enables Heat from a Radiator to Warm the Building Interior

The rate of heat transferred by convection depends on the speed at which the gas moves (which is affected by its viscosity), the temperature difference, the direction of heat flow and the amount of energy required to raise the temperature of the gas (which is affected by its density and heat capacity).

Advection is the transfer of heat as the result of bulk movement of molecules in a gas driven by other forces, for example a difference in air pressure between the inside and outside of a building (see chapter 2 for a more detailed discussion of air movement).

Heat transfer through solids

The thermal performance of a layer of a material in a construction is expressed by its thermal resistance, which is the resistance to heat transfer of a specific thickness of a material. Thermal resistance is expressed in m²K/W (metre squared kelvin per watt): a higher thermal resistance is better for reducing the rate of heat transfer.

For solid materials, where conduction is the main mechanism of heat transfer, the thermal resistance is calculated by dividing the thickness of the material (in metres) by its thermal conductivity. The relationship between thickness and thermal resistance is straightforward: doubling the thickness of a material doubles its thermal resistance (see Figure 1-02). For the same thickness, a material with low thermal conductivity will give a higher resistance than a material with high thermal conductivity.

Figure 1-02 The Relationship between Thermal Resistance and Thickness

Heat transfer through air spaces

The thermal performance of air spaces within building elements is affected by all three heat transfer mechanisms.

Determining thermal resistances of air space

There is no simple relationship between the thermal resistance of an air space and its thickness because heat transfer across an air space involves all three mechanisms of heat transfer. Figure 1-03 shows the resistances of air spaces for upward, horizontal and downward heat flows. In each case the resistance increases with thickness, but reaches a plateau when the effect of convection heat transfer exceeds that of conduction heat transfer.

Figure 1-03 Air Space Resistance by Thickness

Heat transfer at surfaces of building elements

The overall rate of heat transfer through building elements is affected by the conditions at the interior and exterior surfaces of the elements. At interior surfaces, temperature-driven convection currents account for much of the heat transfer: the rate of transfer being greatest when the direction of transfer is up, rather than down. At exterior surfaces air movement produced by the wind is more significant than temperature-driven convection; in these circumstances, the large amount of air movement occurring at the surface means that conduction heat transfer has a negligible effect.

Radiation heat transfer also occurs at interior and exterior surfaces: the rate of transfer is determined by the balance of radiation emitted and received by the surface.

The thermal performances of interior and exterior surfaces are expressed as surface resistances which combine the effects of all mechanisms of heat transfer. Commonly used values are set out in table 1.3.

Radiation heat loss can also result in a substantial lowering of the external temperature of roofs at night. The rate of radiation heat transfer from a roof is the balance of radiation emitted from the surface and that absorbed from the sky and clouds. On clear, still nights the rate of radiation from the roof far exceeds the incoming radiation, resulting in rapid heat loss from the roof surface (commonly referred to as ‘night time radiative cooling’), which can be beneficial in a climate where cooling is required.

On well-insulated roofs, the rate of loss is far higher than the movement of heat from the building interior, resulting in a surface temperature substantially lower than the air temperature, which is why night-time radiative cooling can result in transient condensation underneath the roof covering (see chapter 3: External climate p. 92).

Heat and light

Light from the sun, including infrared, visible and ultraviolet light, is a form of radiation; when it reaches the surface of a building it can be reflected, absorbed and transmitted. Opaque materials will reflect or absorb radiation; as a material absorbs radiation it gains energy and its temperature will rise, the extent of that warming depending on the wavelength and intensity of the light, the absorptivity of the surface and the heat capacity of the material. As the surface warms, heat will be transferred by conduction towards the building interior.

Translucent materials, such as glass, reflect and absorb radiation, but they also transmit some infrared and visible light to the building interior. The amount of solar radiation which is transmitted depends on the characteristics of the glazing (see: Heat transfer through openings p. 16). The light will fall on the interior surfaces of the building, where a large part of its energy will be absorbed by the fabric, raising its temperature. Some of the energy will be re-radiated at longer wavelengths, transmitting the heat around the interior of the building, while some will be lost to the air inside the building. The overall effect of this solar gain will be to raise the building’s internal temperature.

Thermal insulation

Thermal insulation is used to reduce heat transfer through floors, walls and roofs. There are many different types of insulation, but virtually all work by trapping small pockets of low conductivity gas within their bulk. (The exception is the vacuum panel, consisting of microporous core which is evacuated then sealed in a gas-tight cover.) The main types of insulation are:

• * Closed-cell insulants – formed with bubbles of gas in a rigid matrix that may be made of plastic in the case of extruded polystyrene (XPS), polyurethane (PU), polyisocyanurate (PIR) and phenolic insulation, or glass in the case of foamed glass insulation. Aerogel – a very light, translucent solid – also consists of gas trapped within a rigid matrix.
• * Expanded bead insulation – typically in the form of expanded polystyrene (EPS), which is available bonded as boards or as loose beads.
• * Fibrous insulation – in which pockets of still air are formed by strands of material, which may be mineral fibre, glass fibre, cellulose (usually from recycled paper) or ‘natural’ materials such as sheep’s wool.

The thermal performance of an insulant depends on how well it addresses the three mechanisms of heat transfer:

• * Conduction through the solid material – the rate of conduction depends on the type of material and its density, with lower densities reducing the amount of material through which heat can be conducted.
• * Conduction through the gas – the conductivity of the gas is key to overall performance, with air having a conductivity of 0.025 W/mK, argon 0.016 W/mK and pentane 0.014 W/mK.
• * Convection – for closed-cell insulants, convection is insignificant, because the isolation of the cells precludes convection currents. However, convection is an issue in fibrous insulants, because differential temperatures across the insulation generates convection currents. Where the direction of heat transfer is upward (in a heat-out roof or heat-in floor), convection currents will increase the rate of transfer.
• * Radiation – there is some radiative heat transfer across the air spaces within insulation, with the rate of transfer dependent on the surface characteristics of the cells or fibres. The only insulants which specifically address radiative transfer are graphite-enhanced polystyrene insulation (EPS and XPS), which have lower conductivities as a result of the graphite granules reducing radiative transfer.

Table 1.4 gives typical thermal conductivities for common insulants.

Table 1.4 Typical Conductivities of Thermal Insulation

To ensure that the values realistically represent the performance of the fabric, the stated thermal conductivity of an insulant should account for:

• * Variations in production – The performance of insulation will inevitably vary within and between production batches. Plotting the measured conductivity from different batches will typically result in a bell curve (Figure 1-04). Using typical, or mean, values of conductivity results in half the samples having a better conductivity than that quoted, and half the samples being worse. To avoid overestimating installed performance European thermal insulation standards require thermal conductivities to be quoted as lambda 90/90 (λ_90/90) values, which means 90% of samples have thermal conductivity equal to or better than the value quoted, with a 90% level of certainty.
• * Long-term change in performance – The thermal conductivity of closed-cell plastic insulation materials (such as PIR, PU and XPS) will change over time, initially as oxygen and nitrogen in the atmosphere migrate into the insulation and change the balance of gases within the cells, and later as the blowing agent migrates to atmosphere. (One reason foil facings are used on PIR boards is to reduce gas migration and improve long-term performance.) The performance of fibrous insulation will also change, for example as a result of deterioration of the chemicals which bind the fibres together. Calculations of long-term performance need to use the thermal conductivities which reflect the deterioration of performance over time: these are commonly referred to as ‘aged’ conductivities.
• * Variations with temperature – The thermal conductivity of insulation increases with temperature, mainly as a result of the higher conductivity of the blowing agent or air in the insulant. Fibrous insulants will also experience increased internal convection, further increasing the thermal conductivity. Where the in-use temperature of insulation is significantly different from the temperature at which the thermal conductivity was obtained, the conductivity or thermal resistance should be adjusted for temperature.²
• * Moisture – Water is a substantially better conductor of heat than air or other gases (thermal conductivity 0.60 W/mK, compared with 0.025 W/mK for air and 0.014 W/mK for carbon dioxide). Consequently any water absorbed by an insulant will increase its thermal conductivity (i.e. reduce its capabilities as an insulator). Where insulation is to be installed in damp conditions – for example in foundations – the potential for water absorption should be taken into account when calculating the amount/type of insulation material needed.

Figure 1-04 Determining the Quoted Thermal Conductivity of Insulation

Thermal bridging

Many constructions contain a layer of material which is regularly interrupted by a second material, for example, insulation interrupted by the timber studs, or mortar joints in aircrete blockwork: this is referred to as thermal bridging (sometimes incorrectly termed ‘cold bridging’). The interrupting material generally has a higher thermal conductivity than the main material, so the rate of heat transfer will be higher than if there were no bridging, and the U-value (see: Heat transfer through building elements p. 13) will be higher. (An exception to this is mortar bridging in stone walls, where the mortar has a lower conductivity that the stone) Figure 1-05 shows the effect on the temperatures through a wall where a layer of insulation is bridged by a structural support.

As well as affecting the rate of heat transfer, thermal bridging can also affect surface temperature, increasing the risk of surface condensation (see chapter 3: Controlling surface condensation p. 84).

Figure 1-05 Temperature Changes Caused by Thermal Bridging

Heat transfer through building elements

The rate of heat transfer through a building element is dependent on the performance of the materials and air spaces and surfaces of which it is composed. In the UK and Europe the standard measure of the thermal performance of a building element is the thermal transmittance, commonly referred to as the U-value. The U-value is the rate of heat transfer through one square metre of an element’s surface, with a temperature difference of one kelvin between one side and the other. The units are W/m²K (watts per metre squared kelvin). Under identical conditions, an element with a low U-value will transmit less heat than an element with a high U-value.

The U-value of an element is the inverse of the total thermal resistance (U = 1/R).³ Figure 1-06 shows the effect of increasing the thickness of thermal insulation in a construction: the resistance increases steadily, while the U-value drops rapidly to start with (adding 50 mm of insulation to an existing 50 mm improves the U-value from 0.6 W/m²K to 0.3 W/m²K), until it reaches a point where additional insulation makes very little difference (adding 50 mm to an existing 250 mm only improves the U-value from 0.12 W/m²K to 0.10 W/m²K). When looking to improve the thermal performance of a building’s fabric it is important to compare the benefit which would be achieved by adding the same amount of insulation to different elements.

Figure 1-06 Thermal Resistance and U-Values for Increasing Thicknesses of Insulation

Heat transfer through the ground

Heat transfer through elements in contact with the ground – ground floors and basement walls and floors – is affected by the conductivity and heat capacity of the ground and the ground temperature. The temperature varies on an annual cycle, and also varies across the floor; in heat loss conditions, the ground beneath a floor will be warmer at the centre than at the perimeter (see Figure 1-07).

Figure 1-07 Temperature Profile below a Ground-Bearing Slab

The Combined Method

The combined method calculates U-values using (a) the resistances of the layers in the construction and (b), for each thermally bridged layer, the fractional areas that represent the proportions of each material in the layer.

The calculation proceeds by determining the thermal resistance of the construction in two ways (R_max and R_min), as shown in the following diagrams.

R_max is the total resistance calculated from heat paths through the construction (referred to in standards as the ‘upper limit of thermal resistance’). The calculation involves the resistance and fractional area of each heat path, as illustrated here for a plan section through a timber-framed wall, showing two heat paths and their fractional areas.

R_min is the total resistance calculated from the resistances of individual layers (referred to in standards as the ‘lower limit of thermal resistance’), taking account of bridging in the layer, as illustrated in this section of timber-framed wall.

In any construction containing thermally bridged layers R_max will (as the names suggest) be bigger than R_min: the actual total resistance lies somewhere between the two and, for the purposes of the calculation, is taken to be their average. The U-value is then the reciprocal of the total resistance.

The full combined method includes adjustments for mechanical fasteners and air leakage, as well as modifications for tapered insulation layers and inverted roofs.

Although the overall rate of heat loss through a large floor will be greater than that through a smaller floor of the same construction, the rate of heat loss per square metre will be less for the larger floor, giving a lower average U-value. A modified version of the combined method calculation takes account of these effects by considering the ratio between the floor’s perimeter and its surface area: it also allows for heat loss through the sub-floor void in suspended ground floors.

Heat transfer through openings

The thermal performance of glazing in openings (doors, windows or rooflights) and curtain wall systems depends on the performance of the glazing and the framing. The ‘glazing’ of a building may, of course, not be glass, but polycarbonate, aerogel or some hitherto undiscovered substance. For the sake of simplicity I will refer to it as glass.

The glazing is subject to conduction, convection in gas-filled cavities and radiation heat transfer in both directions. (The effect of radiation heat transfer in the form of solar gain is discussed below: Controlling heat transfer through the fabric. p. 17) Glass itself has a high thermal conductivity (1.0 W/mK, comparable with that of concrete blockwork), so most of the thermal performance in a double- or triple-glazed unit comes from the air spaces between the panes of glass. Cavity widths are limited to prevent convection.

The thermal performance of the glazing may be given as a centre pane U-value, which is the U-value measured perpendicularly through the centre of the glazing, taking no account of the effects of glazing spacers and edge seals.

Heat transfer within the frame takes place by conduction and, in hollow frames, by convection. Convection transfer in hollow frames is addressed by dividing the frame into smaller chambers which may sometimes be filled with foam insulation.

The overall U-value for an opening takes account of the performance of the glazing, spacers and frame. The U-value will vary according to the relative proportions of glazing and framing so the performance is often quoted for a standard window size and configuration.

The U-value of a window only addresses conduction heat transfer, so alternative measures, such as the Window Energy Rating (and the related Door Energy Rating), have been developed to express overall performance, taking account of conduction heat transfer, air leakage and the beneficial effects of solar gain. The Window Energy Rating is expressed in A–G bands, with A being the most energy efficient.

Heat transfer at junctions

A significant amount of heat transfer occurs at the junctions between building elements (e.g. the wall–floor junction) and at the edges of openings (e.g. the window jamb). Heat transfer through such junctions is expressed by the linear thermal transmittance or ψ-value (psi-value), which is the rate of heat transfer through one metre length of the junction, with a temperature difference of one kelvin between one side and the other. (The effect that the additional heat loss has on the temperature at the junction is calculated using the ‘temperature factor’, which is discussed below: chapter 3, Box: Temperature factor p. 88) The units are W/mK (watts per metre kelvin).

Calculating the linear thermal transmittance for a junction involves numerical modelling. For many junctions it is sufficient to model in a two-dimensional cross-section, but for complex constructions, particularly those which include steel framing, three-dimensional modelling is required.

The selection of thermal insulation

The use of thermal insulation is essential for reducing heat transfer through the elements that separate the conditioned interior of a building from outside. When selecting an insulant for a particular application there are six main criteria to consider:

• * Thermal performance
• * Thickness
• * Moisture
• * Strength
• * Air movement
• * Installation

Other considerations – beyond the scope of this book – might include embodied carbon, a requirement for local sourcing and price. Suffice to say that the better an insulant performs in the aspects listed above, the more it will cost.

Reducing the effect of thermal bridging

The effect of thermal bridging on an element’s thermal performance depends on:

• * The relative conductivities of the two materials in the bridged layer – where the conductivities are close (e.g. mortar (0.94 W/mK) and brick (0.77 W/mK)) bridging will have a negligible effect, but the effect will be substantial where insulation layers (typically 0.022–0.040 W/mK) are bridged by timber (0.13 W/mK) or steel (15 W/mK);
• * The relative proportions of the two materials – Figure 1-08 shows how the U-value of a timber-framed wall changes if the proportion of timber bridging the insulation layer is increased (note that 15% bridging is the default percentage for such calculations).

Ideally, thermal bridging should be avoided, but there are many constructions, such as those with timber and steel framing, in which space constraints make it inevitable. In such cases the effects of thermal bridging may be mitigated by:

• * Increasing the thickness of the bridged insulation layers
• * Reducing the proportion of bridging, (e.g. large-format aircrete blocks with thin adhesive joints give 1.3% bridging compared with 6.7% for standard-format blocks laid in mortar)
• * Replacing some insulation between framing with a continuous layer of insulation over the face of the framing
• * Using proprietary components to form thermal breaks and reduce heat transfer: this is particularly important where steel brackets and fasteners penetrate insulation layers

Figure 1-08 The Effect of Bridging Percentage on Thermal Performance

Improving the thermal performance of air spaces

The rate of radiation heat transfer across an air space depends in part on the emissivity of its surfaces, so it is possible to reduce the rate of transfer by forming one or more of the surfaces of the air space from materials with low surface emissivities (typically 0.03–0.05) such as:

• * Reflective metal facings of PIR insulation boards
• * Membranes with metalised surfaces
• * Encapsulated aluminium foil membranes
• * Multi-foil membranes

Table 1.5 shows the increase in thermal resistance of a 25-mm wide cavity when one or both surfaces have low emissivity. As with any cavity, the direction of heat transfer affects the thermal resistance. When assessing the long-term performance it is important to use an emissivity value which takes account of the effects of dust and contamination of the surface and the consequent reduction in thermal resistance. Note also that low-emissivity surfaces are also reflective: in some situations an anti-glare coating is required to reduce the risk to installers (and the coating will also increase the emissivity).

Although low-emissivity surfaces do improve the thermal resistance of an air space, in virtually all cases thermal insulation to the same thickness would provide an even higher resistance. It is therefore sensible to use low-emissivity surfaces only where there are designed air spaces in the construction, rather than deliberately introducing an air space.

Controlling heat transfer through openings

When we consider heat transfer through openings we must first address heat transfer which occurs by a combination of conduction, convection and radiation, and then consider the transfer of solar radiation and light.

Heat loss

The rate of heat transfer through openings is typically six to eight times greater than that of the elements in which they sit. This is hardly surprising, given that the thermal conductivity of glass is similar to that of concrete and it is installed in thin layers.

Heat transfer can be reduced by using double or triple glazing in which two or three panes of glass are separated by narrow air spaces. (Although wider cavities would reduce conduction heat transfer, they would simultaneously increase convection heat transfer.) Further reductions in heat transfer may be made by filling the cavity with a gas that has a lower conductivity than air, such as argon and krypton.

The effect of glass on radiation heat transfer depends on the wavelength of the radiation (see chapter 5: Light and materials p. 21). Visible light and infrared light are transmitted, but radiation with longer wavelengths, such as that emitted by the internal surfaces of the building, is largely absorbed by the glass.

Applying a low-emissivity coating on one of the internal surfaces of double or triple glazing can reduce the amount of radiation transmitted across the cavity (in the same way as a lowemissivity surface does in a wall cavity).

There are two main types of low-emissivity coating:

• * Soft coat – multiple thin layers of silver applied after manufacture of the glass. Soft coatings give emittances of 0.05–0.10, so are more effective than hard coatings, but they are susceptible to moisture and physical damage.
• * Hard coat – modified tin oxide coatings applied as part of the manufacturing procedure. Hard coatings give emittances of 0.10–0.20. They are more resistant to damage than soft coat, but are also more expensive.

The thermal performance of existing windows can be improved by the application of lowemissivity window films. Table 1.6 summarises the typical range of U-values which can be obtained with different glazing configurations.

Table 1.6 Typical Thermal Performance of Windows (Wood Frame, All Units With 12 Mm Gap)

Of course, other components contribute to the overall thermal performance, including the spacers of double- or triple-glazed units and the framing members. Metal frames present a particular challenge, given the high thermal conductivity of steel and aluminium: the frames should be formed with thermal breaks, typically of neoprene rubber, to isolate the interior components from the exterior components and so reduce conduction heat transfer.

Reducing heat transfer at junctions

The reduction in elemental U-values over recent decades has resulted in heat-transfer at junctions forming a much greater proportion of overall heat transfer. For example, in a dwelling built to meet the regulatory standards of 1995, junction heat loss would account for roughly onefifth of fabric losses, but junction heat loss would account for two-fifths of losses in the same dwelling built to meet 2013 standards. Consequently, any serious attempt to reduce heat transfer must attend to junctions.

There are two main constraints which affect junction heat loss:

• * The geometry of some junctions results in a greater exposed heat loss surface on the exterior than on the interior (Figure 1-09), resulting in additional heat loss. For such junctions it is therefore not possible to eliminate the additional heat loss, only reduce it.
• * Structural requirements often prevent the formation of a continuous layer of insulation across the junctions. For example, Figure 1-10 shows a conventional wall junction, where the loadbearing structure of the wall prevents the insulation in the ceiling from connecting with the insulation in the wall.

Figure 1-09 Plan Section of a Wall Junction Showing the Greater External Surface Area

Figure 1-10 Cross-Section of the Wall-Ceiling Junction at a Gable

Within those constraints, it is possible to minimise the heat loss at junctions by following four rules of thumb:

• * Minimise the number of junctions – the use of several different construction types to form the thermal envelope will produce more junctions, more discontinuities and more heat loss.
• * Set door and window frames to overlap the insulation plane, preferably rebating them into the full thickness of the insulation.
• * Specify lintels carefully. Steel lintels transfer heat rapidly through walls because of their high thermal conductivity. Use well-insulated lintels and consider using two separate lintels where the wall contains a cavity.
• * Overlap insulating layers to reduce bridging paths. This is particularly useful to address wall/floor junctions and wall/roof gable junctions (see Figure 1-10) where the structural constraints prevent continuity of insulation.

Good thermal detailing will also produce a more airtight structure and so contribute to reducing heat transfer by air leakage (see chapter 2: Infiltration p. 48).

Air infiltration

Air movement by ventilation and by infiltration will also move heat into and out of the building. As will be seen in chapter 2 it is important to maintain good internal air quality: the best strategy is to eliminate unintended air movement and deliberately supply the necessary clean air, adopting a strategy of ‘build tight, ventilate right’.

Heat Loss in Party Walls

In the UK a common method of preventing sound transmission between semi-detached or terraced houses is to construct the party wall from two leaves of masonry or timber framing with a cavity between them (see chapter 4: Isolation p. 111). The cavity in the party wall usually extends for the full height of the dwelling, terminating at the underside of the roof. In many houses, the party wall cavity is continuous with the external wall cavity, in order to reduce sound transmission around the edges of the party wall.

Until recently, building regulations in the UK presumed that there was no substantial heat loss through party walls because the dwellings on either side were likely to be heated to a similar temperature. However, research carried out by Leeds Metropolitan University (2007) showed that cavity party walls produced substantial heat loss. The result should not have been unexpected, because it only confirms the normal movement of air in a cavity.

Heat travels by conduction through each leaf of the party wall from the dwelling into the cavity, where it warms the air in the cavity (as illustrated below). The warmed air rises up the cavity, passing out to the atmosphere at the roof junction, and is replaced by cold air drawn in from the connected cavities in the external walls. The overall result is a continuous flow of cold air into the cavity, producing high levels of heat loss.

The solutions are to avoid wall cavities altogether (and rely on other mechanisms to reduce sound transmission) or to fill the cavity with fibrous insulation and seal around its perimeter. Holistic thinking about all parts of the wall’s performance is crucial.

Heat Loss in Cavity Party Walls

Steady state method

The κ-value (kappa-value) is the amount of energy required to raise the temperature of a square metre of the building element by one kelvin, and is expressed in kJ/m2K (kilojoules per metre squared kelvin). It is the sum of the heat capacities of the layers of which the element is composed, excluding material which lies:

• * More than 100 mm from the internal surface (Figure 1-11a) – for example, in an external wall the κ-value would include the thermal mass of the internal finish and some of the masonry. This limit is imposed because only those parts of the element closest to the surface are involved in daily temperature variations.
• * More than midway through the element (Figure 1-11b) – for example, only the first 40 mm of an 80 mm thick partition wall would be included in the calculation. This prevents the overestimation of the thermal mass of the internal elements when both surfaces are assessed individually.
• * Beyond thermal insulation (Figure 1-11c) – for example, if a wall has internal insulation behind a plasterboard lining, the κ-value would include only the plasterboard.

Table 1.7 shows the κ-values of common constructions. Masonry and concrete constructions have much higher κ-values than timber-framed constructions, except where internal insulation isolates the thermal mass of that masonry.

κ-values are generally used in month-based steady state analysis of energy demand, such as the UK’s national calculation methodologies, SAP and SBEM.

Table 1.7 Typical K-values

Figure 1-11 Limits on Measurement for Calculating Kappa Values

Dynamic method

The dynamic approach to thermal mass, commonly referred to as the ‘admittance method’, uses the density, specific heat capacity and thermal conductivities of materials to calculate measures of thermal mass, which can then be used to assess the thermal performance of elements on an hourly basis.

The dynamic performance of materials can be expressed by:

• * Thermal diffusivity – a measure of how rapidly heat travels in a material. Under daily temperature cycles, heat will travel further into materials with higher diffusivity, which means they will be more effective for cyclical storage of heat.
• * Thermal effusivity – the ‘thermal inertia’ of a material, that is its capacity to absorb and release heat. A material with a high effusivity will be able to store and release a large amount of heat.

Table 1.8 shows that brick and dense concrete have higher diffusivities and effusivities than aircrete, while the thermal insulants have diffusivities in the same range as masonry, but have effusivities which are substantially lower. Brick and concrete will therefore be able to store much more energy than thermal insulants.

Table 1.8 Dynamic Thermal Performance of Common Construction Materials

The dynamic performance of elements is assessed by analysing a notional 24-hour heating and cooling cycle to determine the key properties of admittance, decrement and surface factor.

Admittance

Admittance, Y, (W/m²K, watts per metre squared kelvin) is a measure of how rapidly heat will pass between the surface of an element and the interior of the building. Admittance is most strongly affected by the part of the element nearest the surface. Figure 1-12 shows the U-values and admittances for different thicknesses of three typical materials: one dense (concrete); one medium weight (blockwork); and one lightweight (insulation). The calculations are based on the values in table 1.9. Although the dense and medium weight materials have much higher admittances than the lightweight material, in each case the admittance is practically constant beyond a certain thickness. This stability justifies the use of the 100-mm thickness limit in the steady state κ-value calculations.

Table 1.9 Properties of Materials in Figure 1-12

	DENSITY (kg/m3)	CONDUCTIVITY (W/mK)	SPECIFIC HEAT CAPACITY (J/kg m3)
Concrete	2500	1.5	1000
Blockwork	1000	0.25	1000
Insulation	25	0.03	1000

SOURCE: BASED ON DATA FROM BS EN ISO 10456 AND CIBSE GUIDE A3.

Table 1.10 shows admittance values for typical constructions. Those with higher admittances will absorb more heat from the occupied space than those with lower admittances, and so will be better for reducing cooling loads (see: Solar gain p. 31).

Figure 1-12 U-Values and Admittances for Three Common Materials

Table 1.10 Typical Dynamic Thermal Mass Properties For Walls

Surface factor and surface delay

The surface factor, F, compares the cyclical variation in heat flow from a surface to the cyclical variation into the surface. (It is ridiculous that the decrement factor is f and the surface factor is F. Sometimes building physics seems designed to be confusing!) The surface factor is used to assess the effect of solar radiation and internal gains on internal surfaces. Structures with high admittances will have low surface factors. The time lag between the peak flows is defined by the time factor, ψ (psi).

Figure 1-13 Daily Variation in Temperature through a Wall

Mass-enhanced U-values

U-values are usually calculated independently of thermal mass, but if it is likely that the exterior temperature will fluctuate above and below the internal temperature of a building over the course of a day, a more accurate picture of performance can be gained by taking thermal mass into consideration (rather than using the basic steady state U-value calculation).

As the balance of internal and external temperature changes, so the direction of heat flow through the element will change. At night, when the external temperature is lower, heat transfer will be from inside to outside, but the transfer will be slowed down by the thermal mass of the structure. During the day, if the internal temperature is lower than outside, the direction of heat transfer will reverse, and some of the heat stored within the construction will travel back to the building interior. Taken together, those effects will give a lower effective U-value.

The improvement in performance produced by a mass-enhanced U-valve is directly related to the decrement delay: where there is a wide diurnal temperature variation, a structure with higher thermal mass will transfer less heat and so be more energy efficient.

Solar gain

In temperate climates solar gain is generally beneficial during the heating season because it can supply part of the heating demand. However, it can be problematic in other seasons – potentially raising internal temperatures to uncomfortable levels. In other climate zones solar gain may be problematic in all seasons. But irrespective of whether solar gain is beneficial or problematic, there are three fundamental considerations:

• * The amount of solar radiation at the site, which is determined by the building’s location
• * The amount of that radiation reaching the openings, which is determined by the orientation of the building and the shading of the openings
• * The amount of solar radiation which passes through the openings to the interior, which is determined by the characteristics of the glazing (see above: Controlling heat transfer through openings p. 20)

In some projects it will also be necessary to control indirect solar gain through the opaque fabric.

Location

The primary determinant of the amount of solar radiation a building can receive during a year is its latitude (see chapter 5: The sun).

The amount of solar radiation received on the facets of a building will vary with their orientation. Figure 1-14 shows the relative amount of solar energy falling on vertical surfaces at different compass directions for a site at 55ºN (e.g. northern England/southern Scotland). The heating season values in this example show a large bias towards SE-S-SW compared with NW-N-NE, when the sun is low on the horizon. The differences are not as great during the summer when the sun has the greatest arc. Solar radiation will also be more intense during the summer, because the sun is higher in the sky.

Figure 1-14 Relative Proportion of Solar Energy by Compass Direction

Data derived from SAP 2012

Seasonal differences are more marked at more northerly/southerly latitudes: at the equator the sun is directly overhead at noon, with little variation in its path or day length through the year. The amount of available solar radiation will also be affected by the climate, as cloud cover will reduce the amount of radiation reaching the ground.

Orientation and shading

The proportion of the available solar radiation a building receives will be affected by the orientation of openings and their shading. For a building with the solar potential shown in Figure 1-14 the maximum benefit of solar gain in winter would be obtained by setting a high proportion of the area of openings to face between south-east and south-west. However, adopting that orientation pattern might not be beneficial over the whole year, because the reduction in heating energy demand produced by solar gain in the winter might be outweighed by cooling energy demand during the summer.

The change in the angle at which solar radiation reaches the earth (see chapter 5, Figure 5-03 p. 122) can be used to maximise the benefit of solar gain in winter, while minimising gain in the summer. For example, as Figure 1-15 shows, in the winter (when the sun is at its lowest), solar radiation will penetrate deep into a building, while in the summer, (when the sun is higher), direct solar radiation can be blocked by shading the opening with an overhang (such as a deep soffit on a pitched roof) or brise soleil. External blinds will also reduce solar gain, but internal blinds and curtains will be less effective, because the solar radiation will already have reached the interior of the building.

Figure 1-15 Shading an Opening with an Overhang to Control Solar Gain to the Interior

Reducing heating demand

Figure 1-16 is a simplified schematic diagram showing the relationship between the main factors which go towards ensuring the building is maintained at the design temperature. (One simplification is the omission of energy for artificial lighting, which is affected by the sizing of the windows, but is, in turn, linked to the rate of heat loss through the openings and the amount of solar gain.) Reducing the heat losses on the left-hand side of the diagram, by minimising heat transfer through the fabric and through air movement, will reduce heating demand. However, on the right-hand side of the diagram, maximising the beneficial effect of solar gain will reduce the heat demand.

Figure 1-16 Factors Affecting Heat Demand

The building needs adequate controls to ensure that the heating system responds promptly to the changing temperature inside the building. The complexity of the heating system and its controls will, necessarily, vary with size and complexity of the building and the size of the acceptable temperature range. (However, the design and specification of heating systems and their controls is beyond the scope of this book.)

Finally, to optimise energy demand it is crucial to consider the role of thermal mass.

Reducing cooling demand

Figure 1-17 is a simplified schematic diagram showing the relationship between the main factors associated with overheating. On the left-hand side the climatic conditions result in heat transfer into the interior through air movement and through the fabric, while solar radiation results in heat gains through openings and through the fabric.

Figure 1-17 Factors Affecting Cooling Demand

The risk of overheating can be mitigated, and the need for mechanical cooling reduced or even eliminated by:

• * Minimising solar gain through the opaque fabric and openings
• * Reducing heat transfer through the fabric from outside to inside⁶
• * Using ventilation systems to remove warmed air from the building and introduce cooler air (see chapter 2: Air for moderating temperature)
• * Making use of the thermal mass of the fabric

The rest of this section considers the measures available to minimise solar gain, and how thermal mass may be employed to reducing overheating.

Phase change materials

When a solid material is melted, energy is required to raise the material to its melting point, but a much greater amount of energy is used to make the phase change from solid to liquid. That energy, which is known as latent heat, may be provided deliberately or be drawn from the materials surroundings: for example, an ice cube dropped in a drink will melt as it gains energy from the liquid by conduction; it will also cool the drink.

The same principle can be applied to reduce the overheating of buildings through the use of phase change materials (PCMs). These are typically paraffin waxes or salt compounds which are engineered to melt at a specific temperature. They are supplied in sheet format or incorporated into gypsum board or a similar finishing material.

A PCM behaves as a normal material as long as the internal temperature does not reach its melting point. If the internal temperature exceeds the melting point, the PCM will absorb heat from the space as it melts, so limiting further temperature rise in the space. In effect, the PCM provides additional thermal mass at temperatures above its melting point. Correctly configured a PCM can provide the same thermal mass as five times its thickness of concrete.

Figure 1-18 shows the temperature profile of a room with and without a PCM installed. The temperature with the PCM never reaches the same peak as that without the PCM, but also it never reaches the same low point. This is because once the room temperature drops below the melting point of the PCM it will start to solidify and will release heat back into the room.

Figure 1-18 The Effect of a Phase Change Material on Internal Temperature

For a PCM to be used successfully:

• * The melting point must be matched to the predicted internal conditions: too low and it will increase the heating load; too high and it will not be activated
• * There must be adequate night-time cooling to solidify and ‘recharge’ the PCM for the following day

PCMs are only beneficial for buildings with low thermal mass: by reducing overheating they reduce or even eliminate the cooling load.

Modelling whole building energy performance

Analysis of building energy performance during the design process should be carried out using software which considers the internal and external conditions of the building, its size and layout, the thermal performance of the fabric and the performance of the building services (including the efficiency, responsiveness of the controls), including any renewables. There are two main approaches to analysis:

• * Steady state methods, which typically model the building’s performance on a monthly basis,⁷ by calculating the energy required to maintain the design conditions for that period. They use steady state values for thermal mass (see chapter 3: Steady state analysis).
• * Dynamic simulation modelling (DSM) which models the building’s state at hourly intervals, with one hour’s results forming the starting point for the next. DSM uses dynamic thermal mass values (see: Dynamic method in Assessing thermal mass) and is more accurate than steady state methods. DSM requires more data on internal and external environmental conditions.

The predicted energy performance derived from modelling will never exactly match the actual performance of the building when in use: in practice the actual performance is invariably worse than the predicted performance. Part of that ‘performance gap’ will be the inevitable result of simplifications built into the model and assumptions made about the occupants’ behaviour.

However, larger discrepancies can result from:

• * Inaccurate data input during analysis
• * Poor site practices during construction
• * Incorrectly set controls during use
• * Poor communication with building users about the operation of the building

Addressing those issues is, inevitably, more difficult than identifying them.