05.05 Condensation risk

Categories: Thermal Properties

Introduction
Heat transfer and condensation risk are linked. The occurrence of surface condensation depends on the lowest temperature on the surface of a component and the relative humidity of the adjacent air. The temperature distribution within the component in turn depends on the heat transfer through the component.

Condensation risk is one of the more difficult aspects of building performance to quantify. Condensation can occur under steady or transient conditions, and depends on the amount of water vapour present in the air, and the temperature of the building fabric.

Condensation poses three problems: the first is that if condensation occurs on an exposed surface then it is visible, and visible condensation on a regular basis may lead to a perception of poor performance. Secondly, condensation may lead to mould growth or staining or corrosion of surfaces on which it occurs, or onto which it runs - corrosion is a particular concern where it is unlikely to be seen until too late, whereas staining is a problem on visible exposed surfaces. Finally, condensation may lead to a breakdown or degradation of the performance of insulation systems.

BS 5250 provides guidance as to the causes and control of condensation in traditional forms of building envelope. In many facades condensation risk may be assessed by the methods laid out in BS 5250. However, in some modern forms of construction, particularly curtain walling systems, BS 5250 does not appear to be directly applicable, and the designer is left with the problem of assessing condensation risk in complex facades.

BS 5250 is properly titled Control of condensation in buildings. There will always be some conditions under which condensation occurs; this can be acceptable if those conditions either occur infrequently or are of short duration. In practice the designer should aim to eliminate long term condensation under normal conditions and to make provision for short term condensation under sensible extremes. However, many specifications unrealistically demand that condensation must never occur under some very extreme conditions.

For facade types such as curtain walling the assessment of condensation risk usually takes the form of an assessment to determine the air temperature and relative humidity at which condensation will first appear for given environmental temperatures. The quantity or extent of condensation is not always considered, even though this may be negligible.

Additional information relating to condensation risk is given in BS 5250, BS 8200 and BS 6229. The Building Regulations Approved Document F also deals with issues relating to condensation, putting in place requirements for proper ventilation. Compliance with the Building Regulations should be the principal target of the designer, but additional provisions may be necessary to control condensation to an acceptable level. It should be noted however that no guarantee can ever be given of the total elimination of condensation under all conditions.

Condensation risk assessment for any facade depends on a knowledge of two quantities - the amount of water vapour in the bodies of air either side of the facade and the temperature variation across the facade. Each of these can be considered separately.
 

Water vapour
All naturally occurring volumes of air contain some moisture in the form of water vapour. The quantity of water vapour in any body of air is limited by the principles of thermodynamics to an amount termed the ‘saturation’ amount.

The amount of water vapour present in a body of air may be stated as an absolute value using three concepts - moisture content, which is the mass of water vapour per unit dry mass of air, vapour pressure, which is the partial pressure exerted by the water vapour within the air, or dew-point temperature, which is the temperature to which the body of moist air would need to be lowered, without changing the amount of water vapour, to reach a saturation condition. The saturation condition is then defined by either a saturation moisture content or a saturation vapour pressure. Each of these parameters is a function of air temperature only, and does not vary with atmospheric pressure. Tables, charts and formulae exist by which the saturation properties of moist air may be determined for a wide range of temperatures. Appendix A of BS 5250 gives a psychrometric chart relating vapour pressure and moisture content to temperature, and the CIBSE Guide Parts C1 & 2 give formulae and extensive tables of the same properties.

This image shows the relation between the partial water vapour saturation pressure and the temperature.

Every point on the curve indicates a situation for which the air is saturated by water vapour. For the combinations of temperature and partial water vapour pressure which lie underneath the curve (e.g. point P), the air is not saturated by water vapour.

For point P the vapour saturation pressure can be reached in two ways:

• by raising the partial water vapour pressure at a constant temperature (isobaric): when reaching the point Q the air is saturated, further supply of water vapour will lead to condensation,
• by reducing the temperature at a constant water vapour pressure (isothermal): at point S the air is also saturated by water vapour, further reducing of the temperature will lead to condensation. The temperature belonging to the point S is called the dew-point temperature of the situation in point P.

From the figure is clear that all situations, lying on the line PS, have the same dew-point temperature: the dew-point temperature depends on the partial water vapour pressure and is independent of the temperature.

There are two parameters by which the relative moisture content of a body of air may be expressed - percentage saturation and relative humidity:

The percentage saturation of a body of air is the ratio of the actual moisture content to the saturation moisture content for the same air temperature, expressed as a percentage.

The relative humidity of a body of air is the ratio of the actual water vapour pressure to the saturation vapour pressure for the same air temperature, expressed as a percentage.

A saturated air mass has a relative humidity and percentage saturation of 100%, and a dry air mass has a relative humidity and percentage saturation of 0%.

The relationship between all of the parameters defined above is known and readily tabulated or charted, such as in the examples in the CIBSE Guide Parts C1 & 2.

For any facade, with a moist body of air on each side, if there is a difference in water vapour pressure across the facade then there will be a flow of water vapour through the facade with a corresponding vapour pressure gradient between the two extremes.

A mass of air at a temperature of 0°C and 100% relative humidity contains the same amount of water vapour as the same mass of air at 20°C and 26% relative humidity. A body of warm air at a moderate relative humidity usually has a higher moisture content, and so a higher vapour pressure, than a body of cool air at a higher relative humidity.

The rate of vapour flow between a point of high vapour pressure and a point of low vapour pressure depends upon the vapour flow resistance of the path between the points; even a solid material may contain pores which result in a low vapour resistance.

If a barrier to vapour flow exists somewhere in a facade between a humid atmosphere and a dry atmosphere then water vapour will diffuse into the facade, up to the barrier, and settle at a pressure that is equal to the vapour pressure in the atmosphere to that side of the vapour barrier. If no vapour barrier is present then a steady flow may be established and a vapour pressure gradient will form through the facade. At each point in the facade the allowable vapour pressure (the saturation, or dew-point, condition) is limited by the point temperature. For each point in the facade the actual water vapour pressure can be converted to a dew-point temperature, which is the temperature at which the same vapour pressure would represent the saturation state. If the local point temperature is below the local dew-point temperature then condensation may occur at that point. Installing a vapour barrier has the effect of raising the dew-point on the moister side of the vapour barrier, and so will increase the risk of condensation at all points to the moist-side of the vapour barrier - similarly there will be a decrease in the risk of condensation at all points on the dry-side of the barrier.

Note that in tropical climates or in air-conditioned buildings during summer the higher moisture content may be outside the building and the use and position of a vapour barrier may need to be considered very carefully.

Materials such as glass, metals and rubber are usually non-porous, and so facades in which the visible surface is made up entirely of these materials (for example, a stick-system curtain wall) could be naturally vapour-tight unless unsealed gaps are left in the construction. In this case guidance should be sought as to the likely water vapour pressure variation through the facade.

The services engineer should be able to estimate the water vapour content of the air within a building, by comparing rates of generation of water vapour (depending on the usage of the building) to ventilation rates and rates of diffusion of water vapour through the facade. Appendix C of BS 5250 outlines a procedure for estimating the temperature and relative humidity levels in a building or room, and it is important to note that direct moisture diffusion through the building facade is often insignificant - it is ventilation which dominates the transfer of moisture to or from a building. If a building is vapour-tight and poorly ventilated then moisture levels will be higher, and condensation is much more likely to occur during cold weather. A vapour barrier, or a vapour-tight facade, should only be used in conjunction with good natural or mechanical ventilation. BS 5720 and BS 5925 give guidance on achieving good ventilation.

In the absence of any stated moisture levels some guidance is given in Tables 6 and 7 of BS 6229 and this may be used for the UK. The values given, and the corresponding dew-point temperatures, are:
 

A common failing of specifiers is that a relative humidity is stated but with no reference to the air temperature for which the relative humidity applies - relative humidity alone is not an absolute statement of moisture content and must be related to a temperature if the moisture content of the air is to be determined. Note also that the risk of condensation depends upon the temperature distribution of the component or facade, but that the temperature distribution does not depend on the moisture content of the air (except in the specific case that a porous material has become water-saturated, but this should generally be avoided).
 

Temperature variation
The temperature variation across a facade is a function of the local heat transfers through the various components of the facade. The temperature variation is usually established under steady-state conditions (measurements and hand-calculations all work on a steady-state basis, as do the majority of computer-based analyses - computer-based predictions of dynamic temperature variation usually require such a degree of simplification that any detail of the structure is lost and local point temperatures cannot be predicted with any accuracy). For simple, layered, components the temperature distribution may be determined by hand-calculation, but for more complex components with significant two- and three-dimensional elements the assessment of heat transfer requires more advanced methods.

To determine the temperature variation across a facade requires only that a representative sample of the facade be placed between two environments at known temperatures and allowed to reach a steady-state. This is also the procedure used to obtain the U-value of the facade, and point temperatures are usually obtained as a result of a heat transfer analysis (see above). Measurement gives the most accurate idea of the temperature distribution on the facade. Detailed calculation methods (computer simulation) give the assessor the option of locating the minimum temperature on a facade, but may not predict point temperatures to a high level of accuracy, although U-values might be predicted to within 5%.

Computer simulation might be used to obtain a first estimate of point temperatures, and to locate the coldest parts of the facade.  If a condensation risk analysis suggests that the performance of the facade is borderline then a measurement of heat transfer and point temperatures may be required. Standard for specifying and assessing for heat transfer (the U-value) (CWCT, 1998) gives guidance on issues relating to heat transfer assessment.

It is assumed in the discussion below that an assessment of the facade has been performed and that a set of point temperatures is available. The assessment should have allowed for interactions between components, and any cold bridges within the facade, and the facade should be assessed as it is to be used - i.e. taking account of any third-party additions such as dry linings, blinds or sills at windows.

Point temperatures may have been assessed for different environmental temperatures than those to be used in the condensation risk analysis. It is however possible to modify point temperatures resulting from an assessment to allow for different environmental temperatures:

Treating a component as a complex network of thermal resistances it has been shown that if each resistance is independent of temperature (the real variation of thermal resistance with temperature is generally small and can be neglected) then the temperature at any point within the component is in a constant relationship to the warm-side and cold-side temperatures, such that the temperature factor

c  =  ( T_point - T_cold ) / (T_warm - T_cold )

is fixed for any given point, regardless of the temperatures T_warm and T_cold that are used in the assessment (see Han, Khusinsky and Crooks, 1992).

As an example assume that the temperature  T_point = 4.5^oC has been determined at some point in a facade using T_warm = 20^oC and T_cold = 0^oC for the assessment. If the facade is now used in a location where T*_warm = 21^oC and  T*_cold = -4^oC  then

c  =  ( T_point - T_cold ) / (T_warm - T_cold )  =  ( 4.5 - 0 ) / ( 20 - 0 )  =  0.225

and

T*_point  =  T*_cold + c ( T*_warm - T*_cold )  =  - 4 + 0.225 ( 21 - (- 4))  =  1.625^oC

This simple conversion eliminates the need to determine point temperatures for a range of environmental temperatures. The results of a standard heat transfer (U-value) assessment can therefore be used to predict condensation risk under a wide range of environmental conditions.

Condensation risk assessment falls into two categories - assessing the risk of surface condensation, and assessing the risk of interstitial condensation. These are each considered below.
 

Surface condensation for steady state conditions
Surface condensation takes place when the surface temperature of construction elements is lower than the dew-point temperature of the environment.

A typical sequence for a surface condensation risk assessment for steady state conditions is:

• Establish the temperature variation across the surface of the facade.
• Establish the vapour pressures in the bodies of air either side of the facade, and convert the vapour pressures to dew-point temperatures.
• Identify each point on each exposed surface where the actual temperature is below the dew-point temperature.
• Assess whether the risk of condensation is high enough to warrant preventative or remedial action.

This procedure requires that the temperature variation be assessed for the facade as it will be used, and that realistic assessments are made of the temperature and relative humidity of the environments on either side of the facade.

This procedure will usually be performed on the assumption that conditions are steady. It should be the aim of the designer to at least avoid regular condensation under typical conditions. The assessment should not be based on summer design temperatures on the warm-side of the facade with winter design temperatures on the cold-side of the facade, and should use cold-side temperatures that are reasonable when compared to meteorological data - using a cold-side air temperature of -20°C for a condensation risk assessment is somewhat pointless if the actual air temperature for the location in question only falls below 0°C for one night on average every year. The CIBSE Guide Part A2 gives guidance on how to identify representative cold-side air temperatures.

This image shows typical point temperatures and temperature contours for a curtain walling transom with a low-E double glazing unit above and an insulated panel below, as predicted using a detailed calculation method (this analysis is discussed in more detail Guide to good practice for assessing heat transfer and condensation risk for a curtain wall (CWCT, 1998)).

The assessment has been based on environmental temperatures of 20°C and 0°C, and the temperature contours have been plotted at 1.8, 6.1, 9.3, 12.0 and 14.4°C, which are the dew-points for relative humidities of 30, 40, 50, 60 and 70% at 20°C air temperature. From the temperature contours it is possible to identify the part of each surface (exposed or within cavities) that will experience condensation - for example the 9.3°C temperature contour passes entirely within the facade components and so condensation will not occur on visible exposed room-side surfaces below 50% relative humidity. Similarly the 14.4°C temperature contour only breaks the surface of the facade on the warm-side glazing gasket and on the glazing unit, indicating that surface condensation will only occur over a small part of the facade even at 70% relative humidity.

Note that if the warm-side and cold-side environmental temperatures are changed then each contour can be re-labelled by putting the contour temperature through the conversion described above. For example, if the environmental temperatures are changed to 22° C and -2° C then the c-value of the 14.4° C temperature contour is

c  =  ( T_contour - T_cold ) / (T_warm - T_cold )  =  ( 14.4 - 0 ) / ( 20 - 0 )  =  0.72

and the new temperature of the contour is

T*_contour  =  T*_cold + c ( T*_warm - T*_cold )  =  - 2 + 0.72 ( 22 - (- 2))  =  15.28^oC

which corresponds to a relative humidity of 65.7% at 22° C air temperature.

It cannot be emphasised enough that the component must be assessed as it will actually be used if the assessment is to yield valid information about condensation risk. Furthermore it is not always sensible to use the U-value of a component to determine an ‘average’ surface temperature, as the minimum surface temperature may be several degrees cooler than the average. Also it does not follow that if a component such as a curtain walling frame gives satisfactory condensation performance with an insulated glazing unit (say with a centre-glazing U-value of 1.8 W/m²K) then it will give better condensation performance with any infill panel which has a better centre-panel U-value. The frame performance depends on the edge detail of the infill, which may have worse relative performance for the infill panel than for a glazing unit.

The Guide for assessment of the thermal performance of Aluminium curtain wall framing (CAB, 1996) also discusses condensation risk assessment, using the U-value of a frame to determine a mean surface temperature. Again it is possible to assess the risk of condensation using any combination of internal and external temperatures.
 

Surface condensation predictions for dynamic conditions
Although the condensation risk assessment may show that long term condensation will not be a problem it might still be the case that condensation will occur under some short term conditions, usually of elevated humidity or reduced warm-side temperature; an example might be the rapid reduction in air temperature after office heating is turned off at night - if the ventilation rate is reduced at the same time then the vapour pressure may remain high, whilst the building fabric becomes progressively cooler and condensation then occurs. This effect is more difficult to predict, and requires some subjective decisions about the rate of fabric cooling and the variation in humidity levels. Due to those changes in the moisture production, e.g. the presence of people or the moisture content of the ventilation air, the partial water vapour pressure and so the dew-point temperature of that environment change as a function of time.

Therefore, when considering dynamic heat transfer in building constructions with regard to surface condensation on thermal bridges, it is interesting to evaluate both the dynamic surface temperatures and the dynamic dew-point temperatures.

However, a condensation risk analysis under steady conditions will indicate where condensation will first appear, and the designer should ensure that those parts of a facade most at risk from condensation, or condensation run-off, do not use materials that may be damaged by contact with water. It should also be noted that good ventilation of rooms will help to keep vapour pressures down.

Another situation where condensation may occur under transient conditions is clear night sky radiative cooling. On a clear night the sky has a very low temperature, where infra-red radiative heat losses are concerned, and an exposed surface may fall to a temperature below that of the cold-side air. This cooling effect may occur very rapidly if the external surface has a low heat capacity, and so the surface temperature may fall below the dew-point temperature of the cold-side air. This effect is encountered with buildings clad using profiled sheet metal systems. In this case the surfaces at risk should be resistant to standing water; although this will be the case for surfaces normally exposed to rainfall it should not be assumed that the reverse of any cladding panels will have been suitably protected.

The absolute moisture content of the air in a room depends on the moisture production in that room. When moisture production changes, e.g. the production is raised because more people are present in the room, the absolute moisture content can be calculated using the following formula:

x_i  =  x₀ + DF ( 1 - e^-nt ) /n

and

x₀  =  x_e + F₀ / n

with

x_i       indoor absolute moisture content at t>0s (kg/m³)
x₀      indoor absolute moisture content at t=0s (kg/m³)
x_e      outdoor absolute moisture content (ventilation air ) (kg/m³)
F₀     moisture production at t=0s (kg/m³s)
DF   change in moisture production (kg/m³s)
n        ventilation rate (m³/h)
t         time (s)
 

From the absolute moisture content the partial water vapour pressure can be calculated, using the following formula:

p  =  p_A R_D x_i/ R_A r

p      partial water vapour pressure (Pa)
p_A    atmospheric pressure (1.013 x 10⁵ Pa)
R_D    water damp gas constant (461.5 J/kgK)
R_A    air gas constant (286.9 J/kgK)
r      volumic mass (0.84 kg/m³)
 

The dew-point temperature belonging to a partial water vapour pressure can be calculated from:

T_d  =  ( 4030 / (18.95 - ln ( 0.01 p ) ) ) - 235

with

T_d    dew-point temperature (degrees C)

Using the formulas above the absolute moisture content, the partial water vapour pressure and the dew-point temperature are calculated as a function of the moisture production, the ventilation rate and time.

However, in reality the situation might be more complex: the moisture production in e.g. an office room will rise and fall dependent on the rate of occupation by people and the moisture content of the ventilation air, which might vary in time. The course of the absolute moisture content and the dew-point temperature will even be more complex when assuming a variable value of the outdoor absolute moisture content and the ventilation rate. Nowadays there is software on the market, using a dynamic heat transfer models, to predict condensation risk considering both the surface temperature of building constructions and the dew-point temperature as a function of time.
 

Avoiding surface condensation
Surface condensation can be reduced in two ways - improve the resistance of the facade to heat transfer or reduce the water vapour pressure in the adjacent air.

Improving resistance to heat transfer does not mean reducing the average U-value of the facade - condensation occurs at the point on the surface of the facade where heat transfer occurs most readily. It is therefore necessary to improve the resistance to heat transfer at that point. A lower surface temperature generally indicates the presence of a cold-bridge. The assessment of cold-bridging and guidance on reducing cold-bridging are given in the Building Regulations Approved Document L.

Reducing the water vapour pressure can be achieved by venting or by ventilation. This is a recurring theme of BS 5250 and BS 8200, and this is the approach suggested by Approved Document F of the Building Regulations. The difference between venting and ventilation is a subtle one - a vent is a small opening through which weak air movement occurs, whereas ventilation openings are large and significant air movement occurs. Vents are appropriate where drafts are to be avoided, and ventilation is appropriate where significant air movement is desirable. The ventilation system may be mechanical, where moisture is removed from the air by some dehumidification process, or passive, where a window is simply left open; venting is usually a continuous background process.

If properly undertaken an assessment of heat transfer should generate data that can be used to assess condensation risk. Condensation risk is discussed in more detail in Standard for specifying and assessing for condensation risk (CWCT, 1998), and the use of heat transfer and condensation risk analyses is demonstrated in Guide to good practice for procedures for assessing glazing frame U-values and Guide to good practice for assessing heat transfer and condensation risk for a curtain wall (both CWCT, 1998).
 

Interstitial condensation
Condensation within the building facade is more difficult to predict. BS 5250 gives guidance for assessing water vapour movement through a simple layered wall, and for assessing the risk of condensation, but this same guidance cannot be readily applied to more complex curtain walls.

The problem with assessing interstitial condensation risk in a modern facade such as a curtain wall is that many of the materials used have a very high resistance to the transmission of water vapour, so that the principal means of water vapour diffusion is through the openings and crevices at the joints between the various components. It is difficult to quantify these openings, and so it is difficult to determine how much water vapour will diffuse into the facade.

As an example consider the curtain wall of this image.

The warm-side surface is either aluminium, glass or rubber, all of which have a very high resistance to vapour transmission. If the joints are perfectly sealed then the warm-side surface of the curtain wall would be a vapour barrier, and by a similar argument the cold-side surface might also be a vapour barrier. The cavity in the frame section, at the edge of the glazing unit, between the warm-side and cold-side surfaces, therefore has the potential to be a sealed cavity, but it also has the potential to be open either to the room environment or to the outside environment. Workmanship thus becomes a critical issue in ensuring that moist air does not find its way into the cavity.

Clearly the designer should consider features which ensure the vapour-tightness of the warm-side surface, such as one-piece frame gaskets around the perimeter of the glazing and moulded gaskets to seal between the various frame sections, whilst at the same time the potential cold-side vapour barrier could be deliberately broken to allow venting, which might be normal for pressure-equalised or drained-and-ventilated systems in any case.

Now consider a modification to this curtain wall, where a dry-lining is added to the warm-side of the curtain wall by a third party, as shown in this image.

The dry-lining, and the new cavity between the dry-lining and the curtain wall, reduce heat loss through the curtain wall, but will also reduce the temperature on the warm-side of the insulated panel. Furthermore the dry-lining is probably constructed from materials with a lower resistance to vapour transfer and is unlikely to be vapour-tight. The risk of condensation on the back of the curtain wall is now much higher.

A vapour-tight coating on the dry-lining will reduce the diffusion of water vapour through the dry-lining, but this will be of little consequence if the curtain wall still provides a vapour barrier - the cavity between the dry-lining and the curtain wall could saturate with water vapour at a vapour pressure close to that of the warm-side environment. Regular condensation may occur as a result, and this could result in significant damage to fixings and insulation materials. It is important in this case to seek expert opinion as to the likely effects of condensation within the cavity.

Note also that whilst applying a non-porous coating to the dry-lining surface may give short term improvements but the coating may become damaged in future, and the need to maintain the integrity of coatings should be considered. An alternative solution to the condensation problem introduced by the dry-lining may be to ventilate the cavity between the dry-lining and the curtain wall to the cold-side environment. However, this might incur greater heat loss, and may also allow wind-loads to be transferred to the dry-lining. BS 5250 and BS 8200 both give guidance for the size of vents and ventilation openings and should be consulted for advice.

The only sensible way to assess condensation risk in a complex facade is to assume that the higher vapour pressure will exist throughout the facade. The assessor then determines the dew-point temperature corresponding to the higher vapour pressure and identifies all of those points which are at a lower temperature - it is always possible to draw a temperature contour through the facade corresponding to this dew-point temperature, and this will at least allow an assessment of where drainage needs to be provided and which materials are at risk from damage.

Nowadays there is software on the market that can predict interstitial condensation for two- and three-dimensional situations. The calculation is based on the Cammerer method (similar to method explained in BS 5250 for one-dimensional situations) for two- and three-dimensional constructions. The method predicts that there will be a condensation risk in the zones where the vapour pressure is higher than the saturated vapour pressure. Basically it works on the following principles:

• From the external and the internal temperatures the node temperatures of the triangulation grid can be obtained by a finite element or different method.
• From the node temperatures the corresponding saturation vapour pressures can be calculated in every node using the following equations:

 

 
 
 

For -30^oC £ T £ 0^oC   :  P_s = 611 e^( f(t))   where f(t) = 82.9x10^-6 T - 288.1x10^-6 T² + 4.403x10^-6 T⁴

and

For 0^oC £ T £ 40^oC   :  P_s = 611 e^( f(t))   where f(t) = 72.5x10^-6 T - 288.1x10^-6 T² + 0.790x10^-6 T⁴
 

• Simularly to the first step, the finite element software for predicting the temperatures (using the thermal conductivity ? of the materials), can also be used to calculate the vapour pressure in every node (using the vapour resistance factor ? of the materials). The vapour pressures of the internal and external environment can then be used instead of the environmental temperatures.
• At those places where the vapour pressures are higher than the saturated vapour pressures there is a potential risk for condensation.
• The vapour pressures in the previous step may be unrealistic because they are higher than the saturated vapour pressures. This is characteristic of the Cammerer method. In reality, the vapour pressure is not higher than the saturated vapour pressure so they can be assumed to be equal. To equalize those values, a more refined procedure (method of Glaser) can be followed (derived from the Cammerer method) and the amount of condensing water expected can be assessed.
• This normal Glaser method (described above) allows to predict only the amount of condensation in the interfaces between the material layers. What happens to the condensation is not taken into account. Nevertheless, in practice the condensation will wet the material layers themselves. This is the case for two groups of materials:
• Capillary materials will absorb condensation
• Non-capillary, non vapour-tight materials containing air pores (such as plastic foams) can be humidified by condensation

Avoiding interstitial condensation
It is simpler to design to reduce the risk of condensation in a complex facade than it is to predict the risk. The following simple rules will reduce the risk of condensation in a complex facade:
 

• The insulating elements of the facade should be placed as near as possible to the cold-side, ensuring that as much of the facade as possible is at a high temperature and likely to be above the dew-point temperature. Note that an air-filled cavity is an insulator, and so placing a feature such as a dry-lining on the warm side of a vapour-tight construction, such as a curtain wall, is likely to increase the risk of condensation in a location where it cannot be seen.

• If any layer of the facade is to be vapour-tight then it should be the warm-side surface, ensuring that all of the cold-side of the facade is at the lower vapour pressure of the cold-side air.

• Use materials that decrease in vapour resistivity in going from warm to cold. This ensures that the vapour pressure, and hence the dew-point temperature, falls rapidly from the warm-side of the construction to the cold. A vapour barrier should not be added unless calculations clearly show that condensation would otherwise occur, and consideration should be given to improving the thermal properties of the wall to avoid condensation in the first instance.

• Use materials that decrease in thermal conductivity in going from warm to cold. This ensures that the temperature of the facade falls least rapidly from the warm-side of the construction.

• If there are cavities within the facade then vent or ventilate them to the cold-side air, in accordance with the guidance of BS 5250 and BS 8200, but be aware of the implications for heat transfer through the facade if a cavity is ventilated.

• Avoid constructions which under excessive conditions allow moisture to become trapped between impervious layers. The gradual build up of moisture can result in decaying materials and fixings.

It is now possible to be more specific in assessing and reducing interstitial condensation risk, other than for the simple layered walls considered in BS 5250. However, by following some simple rules condensation risk can be minimised.