Good indoor air quality is necessary for maintaining health and high productivity. Consequently, sustainable building standards such as ASHRAE Standard 189.1 and building rating systems, such as U.S. Green Building Council’s (USGBC) Leadership in Energy and Environmental Design™ (LEED^®) program, place great importance on creating and maintaining acceptable IAQ. In fact, the LEED rating system was first developed to address IAQ concerns, and roughly one-quarter of the available credit points for new commercial buildings are still IAQ related. Preparers of such rating systems, like others, have struggled with how to characterize complex ventilation and infiltration issues. These issues are addressed in detail by many portions of this chapter; separate ASHRAE design guides, manuals, books, and standards; and the references cited; these sources also provide methods to demonstrate the effectiveness of various HVAC systems and techniques in providing good IAQ in residential, commercial, and other buildings. In all designs, care is needed to eliminate excessive ventilation (e.g., beyond that needed for IAQ or by an air-side economizer) to avoid inappropriately increasing energy use. Increasing the ventilation rate above that required by Standard 62.1, for example, does not necessarily increase the acceptability of the indoor air quality.

Air exchange of outdoor air with air already in a building can be divided into two broad classifications: ventilation and infiltration.

Ventilation is intentional introduction of air from the outdoors into a building; it is further subdivided into natural and mechanical ventilation. Natural ventilation is the flow of air through open windows, doors, grilles, and other planned building envelope penetrations. Mechanical (or forced) ventilation, shown in Figure 1, is the intentional movement of air into and out of a building using fans, ductwork, intake louvers, and exhaust grilles, for example.

Infiltration is the flow of outdoor air into a building through cracks and other unintentional openings and through the normal use of exterior doors for entrance and egress. Infiltration is also known as air leakage into a building. Exfiltration, depicted in Figure 1, is leakage of indoor air out of a building through similar types of openings. Like natural ventilation, infiltration and exfiltration are driven by natural and/or artificial pressure differences. These forces are discussed in detail in the section on Driving Mechanisms for Ventilation and Infiltration. Transfer air is air that moves from one interior space to another, either intentionally or not.

Figure 1. Two-Space Building with Mechanical Ventilation, Infiltration, and Exfiltration

Ventilation and infiltration differ significantly in how they affect energy consumption, air quality, and thermal comfort, and can each vary with weather conditions, HVAC system operation, and building use. Although one mode may be expected to dominate in a particular building, both must be considered in the proper design and operation of an HVAC system. Seasonal weather and other transient factors must be considered, as well.

Ventilation air is air used to provide acceptable indoor air quality. It may be composed of mechanical or natural ventilation, infiltration, suitably treated recirculated air, transfer air, or an appropriate combination, although the allowable means of providing ventilation air varies in standards and guidelines.

Modern commercial and institutional buildings normally have mechanical ventilation and are usually intended to be pressurized somewhat to reduce or eliminate infiltration. Mechanical ventilation has the greatest potential for control of air exchange when the system is properly designed, installed, and operated; it should provide acceptable indoor air quality and thermal comfort when ASHRAE Standards 55 and 62.1’s requirements are followed, although issues (e.g., unusually strong pollutant sources) can still result in unacceptable indoor environment conditions. Mechanical ventilation equipment and systems are described in Chapters 1, 4, and 10 of the 2020 ASHRAE Handbook—HVAC Systems and Equipment.

In commercial and institutional buildings, natural ventilation (e.g., through uncontrolled use of manually operated windows) may not be desirable from the points of view of energy conservation, comfort, security, or control of airborne pollen or other pollutants in some climates and locations. In commercial and institutional buildings with mechanical cooling and ventilation, an automatically controlled air- or water-side economizer may be preferable to operable windows for taking advantage of cool outdoor conditions when interior cooling is required. When moderate outdoor temperatures occur, an air-side economizer control scheme may not only increase the rate of ventilation but also operate the cooling equipment to optimize energy use (hybrid or mixed mode).

Infiltration may be significant in commercial and institutional buildings too, especially in tall, leaky, or partially pressurized buildings and in lobby and loading dock areas. The joint between roof decking and outer walls is often particularly leaky in commercial and other large buildings, and should be properly detailed, constructed, and inspected.

Figure 2. Simple All-Air Air-Handling Unit with Associated Airflows

In most of the United States, residential buildings have historically relied on infiltration and natural ventilation to meet their ventilation air needs. Neither is reliable for ventilation air purposes because they depend on weather conditions, building construction, occupants, and maintenance. Natural ventilation, usually through operable windows and screened doors, is more likely to allow occupants to control indoor airborne contaminants and interior air temperature, but it can have a substantial energy cost if used while the residence’s heating or cooling equipment is operating. Opened windows and doors also may lead to security, noise, or other concerns.

In place of or in addition to operable windows, small exhaust fans should be provided for localized venting of residential spaces with high pollutant levels or moisture (e.g., kitchens, bathrooms). Not all local building codes require that such exhaust be vented to the outdoors, but it is required by ASHRAE Standard 62.2. Instead, a local code may allow the air to be treated and returned to the space or to be discharged to an attic space. Poor maintenance of these recirculating treatment devices can make nonducted vents ineffective for ventilation purposes. Warm exhaust air can hold much moisture, so condensation in attics should be avoided. If not already required by code, consider venting attached garages and other storage spaces to the outdoors, as well.

Increasingly, building codes require general mechanical ventilation in residences. Heat recovery heat exchangers are popular for reducing energy consumption, especially in cold climates. Residential buildings with low rates of infiltration and natural ventilation, including most new buildings, require mechanical ventilation at rates given in ASHRAE Standard 62.2.

Figure 2 shows a simple air-handling unit (AHU) or air handler that conditions air for a building. Air brought back to the air handler from the conditioned space is return air (RA). The return air either is discharged to the environment [exhaust air (EA)] or is reused [recirculated air (CA)]. Air brought in intentionally from the environment is outdoor air (OA). Because outdoor air may need treatment to be acceptable for use in a building, it should not be called “fresh air.” Outdoor and recirculated air are combined to form mixed air (MA), which is then conditioned and delivered to the spaces served as supply air (SA). Any portion of the mixed air that intentionally or unintentionally circumvents conditioning is bypass air (BA). Because of the wide variety of air-handling systems, the airflows shown in Figure 2 may not all be present in a particular system as defined here. Also, more complex systems may have additional airflows.

In HVAC design, volumetric airflow rates Q are normally reported in cubic feet per minute (cfm). The incorrect term “volume” should not be used to describe airflow rates.

The outdoor airflow introduced to a building or zone by an air-handling unit can also be described by the outdoor air fraction X_oa, which is the ratio of the volumetric flow rate Q of outdoor air brought in by the air handler to the total supply airflow rate:

(1)

When expressed as a percentage, the outdoor air fraction is called the percent outdoor air. The design outdoor airflow rate Q_oa for a building’s or zone’s ventilation system is found by applying the requirements of ASHRAE Standard 62.1 or 62.2 to that specific building, occupancy, and HVAC system. The supply airflow rate Q_sa is that required to meet the thermal load. The outdoor air fraction and percent outdoor air then describe the degree of recirculation, where a low value indicates a high rate of recirculation, and a high value shows little recirculation. Conventional all-air air-handling systems for commercial and institutional buildings often have approximately 10 to 40% outdoor air.

100% outdoor air means no recirculation of return air through the air-handling system. Instead, all the supply air is treated outdoor air, also known as makeup air (KA), and all return air is discharged directly to the outdoors as relief air (LA), via separate or centralized exhaust fans or relief dampers and grilles. An air-handling unit that provides exclusively 100% outdoor air to offset air that is exhausted is typically called a makeup air unit (MAU).

When outdoor air via mechanical ventilation is used to provide ventilation air, as is common in commercial and institutional buildings and increasingly in residences, this outdoor air is usually delivered to spaces as all or part of the supply air. With a variable-air-volume (VAV) system, the outdoor air fraction of the supply air may need to be increased when supply airflow is reduced to meet a particular thermal load. In some HVAC systems, such as a dedicated outdoor air system (DOAS), conditioned outdoor air may be delivered separately from the way the spaces’ loads are handled (Mumma and Shank 2001).

Air movement within spaces affects the diffusion of ventilation air and, therefore, indoor air quality and comfort. Two distinct flow patterns are commonly used to characterize air movement in rooms: displacement flow and entrainment flow. Displacement flow, shown in Figure 3, is the movement of air within a space in a piston- or plug-type motion. Ideally, no mixing of the room air occurs, which is desirable for removing pollutants generated within a space. Air mixing does occur, however, to various degrees. A laminar-flow air distribution system that is intended to sweep air across a space with reduced turbulence and mixing may produce a high degree of displacement flow and thus more effective pollutant removal. The pollutants’ buoyancy and occupants’ thermal comfort are concerns when deciding on the intended direction of airflow.

Entrainment flow, shown in Figure 4, is also known as conventional mixing. Systems with ceiling-based supply air diffusers and return air grilles are common examples of air distribution systems that produce entrainment flow. Airborne pollutants are removed by dilution by the ventilation air that is delivered as all or part of the supply air. Entrainment flow with very poor mixing in the room has been called short-circuiting flow because much of the supply air leaves the room without mixing with room air. There is little evidence that properly designed, installed, and operated air distribution systems exhibit substantial short circuiting, although poorly designed, installed, or operated systems may short circuit (especially ceiling-based systems in heating mode) to a higher degree (Offermann and Int-Hout 1989).

Figure 3. Displacement Flow Within a Space

Figure 4. Entrainment Flow Within a Space

Theoretical perfect mixing occurs when supply air is instantly and evenly distributed throughout a space. Perfect mixing is also known as complete or uniform mixing; the air may be called well stirred or well mixed. This theoretical performance is approached by entrainment flow systems that have good mixing and by displacement flow systems that allow too much mixing (Rock et al. 1995). The outdoor air requirements given in the minimum ventilation rate in breathing zone table of ASHRAE Standard 62.1 assume delivery of ventilation air with perfect mixing within spaces. For more detailed information on space air diffusion, see Chapter 20.

Underfloor air distribution (UFAD or UAD), as shown in Figure 5, is a hybrid method of conditioning and ventilating spaces (Bauman and Daly 2003). Air is introduced through a floor plenum, with or without branch ductwork or terminal units, and delivered to a space by floor-mounted diffusers. These diffusers encourage air mixing near the floor to temper the supply air for thermal comfort. The combined air then moves vertically through the space, with reduced mixing, toward returns or exhausts placed in or near the ceiling. This vertical upward movement of the air is in the same direction as the thermal and contaminant plumes created by occupants and common equipment. The ventilating performance of UFAD systems is thus often between floor-to-ceiling displacement flow and uniform mixing.

Supply air that enters a space through a diffuser, grille, or nozzle is also known as primary air. An air jet is formed as this primary air leaves the supply air outlet. Secondary air is the room air entrained into the jet. Total air is the combination of primary and secondary air at a specific point in a jet and increases with distance from the outlet as is described further in Chapter 20. The term primary air is also used to describe supply air provided to fan-powered mixing boxes by a central air-handling unit.

Figure 5. Underfloor Air Distribution to Occupied Space Above (Rock and Zhu 2002)

For evaluation of indoor air quality and thermal comfort, rooms are often divided into two portions: the occupied zone and the remaining volume of the space. Often, this remaining volume is solely the space above the occupants and is referred to as the ceiling zone. The occupied zone is usually defined as the lowest 6 ft of a room, although layers near the floor and walls are sometimes deducted from it; when these deductions are made, the occupied zone is sometimes renamed the breathing zone. Ceiling and floor plenums are not normally included in the occupied or ceiling zones. Thermal zones are different from these room air zones, and are defined for HVAC subsystems and their controls.

The air change (or exchange) rate I compares airflow to the space’s volume and is

(2)

where

The air change rate has units of 1/time, usually h⁻¹. When the time unit is hours, the air change rate is also called air changes per hour (ACH), with units of h⁻¹. The air change rate may be defined for several different situations. For example, the air change rate for an entire building or thermal zone served by an air-handling unit compares the amount of outdoor air brought into the building or zone to the total interior volume. This nominal air change rate I_N is

(3)

where Q_oa is the outdoor airflow rate including ventilation and infiltration. I_N describes the outdoor air ventilation rate entering a building or zone. It does not describe recirculation or the distribution of ventilation air to each space in a building or zone.

For a particular space, the space air exchange rate I_S compares the supply airflow rate Q_sa to the volume of that space:

(4)

For a particular space or zone, I_S includes recirculated as well as any outdoor air in the supply air, and is used frequently in evaluating supply air outlet performance and space air mixing.

Timeconstants τ, which have units of time (usually in hours or seconds), are also used to describe ventilation and infiltration. One time constant is the time required for one air change in a building, zone, or space if ideal displacement flow existed. It is the inverse of the air change rate:

(5)

The nominal time constant compares the interior volume of a building or zone to the volumetric outdoor airflow rate:

(6)

Like the nominal air change rate, τ_N does not describe recirculation of air in a building or zone, or characterize the distribution of the outdoor air to individual spaces in a building or zone.

The space time constant compares the interior volume of a particular space to the total supply airflow rate to that space. The space time constant is the inverse of the space air change rate:

(7)

The space time constant includes the effect of recirculated air that is part of the supply air as well as that of outdoor air introduced to the space through the supply air. If infiltration is significant in a space, then the infiltration flow rate should be included when determining both the space air change rate and the space time constant.

When assessing time-varying ventilation in terms of controlling indoor air quality, the quantity of interest is often the temporal average rather than the peak. The concept of effective ventilation (Sherman and Wilson 1986; Yuill 1986, 1991) describes the proper ventilation rate averaging process. In this concept, the average (effective) rate is the steady-state rate that yields the same average contaminant concentration over the period of interest in the occupied space as does the actual sequence of time-varying discrete ventilation rates over the same period and in the same space. This effective rate is only equal to the simple arithmetic average rate when the discrete ventilation rates are constant over the period of interest and the contaminant concentration has reached its steady-state value. Simple arithmetic averaging of instantaneous ventilation rates or concentrations cannot generally be used to determine these averages because of the nonlinear response of indoor concentrations to ventilation rate variations.

An important constraint in the effective ventilation concept is that the contaminant source strength F must be constant over the period of interest or must be uncorrelated with the ventilation rate. These conditions are satisfied in many residential and commercial buildings because the emission rates of many contaminants that are controlled by whole-building ventilation systems vary slowly. Sherman and Wilson (1986) describe how to deal with pollutants that have stepped but otherwise constant emission rates. Pollutants such as carbon monoxide, radon, and formaldehyde, whose emission rates can be affected by ventilation, cannot be properly characterized with this concept and require more complex analyses. For constant-source-strength pollutants, the relationship between effective air change rate, effective ventilation rate, volumetric flow, source strength, average concentration, and time-averaged effective turnover time is given by

(8)

The time-averaged effective turnover time τ̅_e in Equation (8) represents the characteristic time for the concentration in the occupied space to approach steady state over the period of interest. It can be determined from a sequence of discrete, instantaneous ventilation air change rates I_i using the following (Sherman and Wilson 1986):

(9)

(10)

(11)

where

ASHRAE Standard 62.2 provides a set of factors to help calculate the annual effective air exchange rate.

The age of air θ_age (Sandberg 1981) is the length of time t that some quantity of outdoor air has been in a building, zone, or space. The “youngest” air is at the point where outdoor air enters the building by mechanical or natural ventilation, or through infiltration (Grieve 1989). The “oldest” air may be at some location in the building or in the exhaust air. When the characteristics of the air distribution system are varied, age of air is inversely correlated with quality of outdoor air delivery. Units are of time, usually in seconds or minutes, so it is not a true efficiency or effectiveness measure. The age of air concept, however, has gained wide acceptance in Europe.

The age of air can be evaluated for existing buildings using tracer gas methods. Using either the decay (step-down) or growth (step-up) tracer gas method and assuming perfect mixing, the zone average or nominal age of air θ_age,N can be determined by taking concentration measurements in the exhaust air. The local age of air θ_age,L is evaluated through tracer gas measurements at any desired point in a space, such as at a worker’s desk. When time-dependent data of tracer gas concentration are available, the age of air can be calculated from

(12)

where C_in is the concentration of tracer gas being injected.

Because evaluation of the age of air requires integration to infinite time, an exponential tail is usually added to the known concentration data (Farrington et al. 1990).

Ventilation effectiveness is a description of an air distribution system’s ability to remove internally generated pollutants from a building, zone, or space. Air change effectiveness is a description of an air distribution system’s ability to deliver ventilation air to a building, zone, or space. The HVAC design engineer usually does not have knowledge or control of actual pollutant sources within buildings, so the minimum prescribed ventilation rates of ASHRAE Standard 62.1 define outdoor air requirements for typical, expected building uses. For most projects, therefore, air change effectiveness is of more relevance to HVAC system design than ventilation effectiveness. Various definitions for air change effectiveness have been proposed. The specific measure that meets local code requirements must be determined, if any is needed at all.

Air change effectiveness measures ε_I are nondimensional gages of ventilation air delivery. One common definition of air change effectiveness is the ratio of a time constant to an age of air:

(13)

The nominal air change effectiveness ε_I,N shows the effectiveness of outdoor air delivery to the entire building, zone, or space:

(14)

where the nominal time constant τ_N is usually calculated from measured airflow rates.

The local air change effectiveness ε_I,L shows the effectiveness of outdoor air delivery to one specific point in a space:

(15)

where τ_N is found either through airflow measurements or from tracer gas concentration data. An ε_I,L value of 1.0 indicates that the air distribution system delivers air equivalent to that of a system with perfectly mixed air in the spaces. A value less than 1.0 shows less than perfect mixing with some degree of stagnation. A value of ε_I,L greater than 1.0 suggests that a degree of plug or displacement flow is present at that point (Rock 1992).

An HVAC design engineer often assumes that a properly designed, installed, operated, and maintained air distribution system provides an air change effectiveness of about 1. However, the zone air distribution table of ASHRAE Standard 62.1 provides some estimates of effectiveness for operating in heating or cooling mode, and with various air distribution techniques. These values are then adjusted for commercial and institutional building design when the ventilation rate procedure (VRP) is used. If the IAQ procedure of Standard 62.1 is used, then actual pollutant sources and the air change effectiveness must be known for the successful design of HVAC systems that have fixed ventilation airflow rates.

ASHRAE Standard 129 describes a method for measuring air change effectiveness of mechanically vented spaces and buildings with limited air infiltration, exfiltration, and air leakage with surrounding indoor spaces.

Decay. The simplest tracer gas measurement technique is the decay method, also known as the step-down method. A small amount of tracer gas is injected into the space and is allowed to mix with the interior air. After the injection, F = 0 and then the solution to Equation (16) is

(17)

where C_o is the concentration of the tracer in the space at t = 0.

Equation (17) is generally used to solve for I by measuring the tracer gas concentration periodically during the decay and then fitting the data to the logarithmic form of Equation (17):

(18)

Like all tracer gas techniques, the decay method has advantages and disadvantages. One advantage is that, because logarithms of concentration are taken, only relative concentrations are needed, which can simplify calibration of concentration-measuring equipment. Also, the tracer gas injection rate need not be measured, although it must be controlled so that the tracer gas concentrations are within the range of the concentration-measuring device. The concentration-measuring equipment can be located on site, or building samples can be collected in suitable containers (e.g., grab bags) and analyzed elsewhere.

The most serious problem with the decay technique is imperfect mixing of tracer gas with interior air, both at initial injection and during decay. Equations (16) and (17) assume that the tracer gas concentration within the building is uniform at any particular time. If the tracer is not well mixed, this assumption is not appropriate and the determination of I is subject to errors. It is difficult to estimate the magnitude of errors caused by poor mixing, and there has been little analysis of this problem. Sometimes a two-zone model is applied to a room, and a mixing coefficient selected, to estimate the effect of poor mixing [e.g., Rock (1992)].

Growth. The growth or step-up method is similar to the decay method except that the initial tracer gas concentration is low and the injected tracer gas is increased suddenly during the test.

In the constant concentration technique, the tracer gas injection rate is adjusted to maintain a constant concentration within the building. If the concentration is truly constant, then Equation (16) reduces to

(19)

There is less experience with this technique than with the decay procedure, and an increasing number of applications for it exist (Bohac et al. 1985; Collet 1981; Fortmann et al. 1990; Kumar et al. 1979; Walker and Forest 1995; Walker and Wilson 1998; Wilson and Walker 1993).

Because tracer gas injection is continuous, no initial mixing period is required. Another advantage is that tracer gas injection into each zone of the building can be separately controlled; thus, the amount of outdoor air flowing into each zone can be determined. This procedure is best suited for longer-term continuous monitoring of fluctuating infiltration rates. One disadvantage is that it requires measurement of absolute tracer concentrations and injection rates. Also, imperfect mixing of the tracer and interior air causes a delay in the response of the concentration to changes in the injection rate.

In the constant-injection procedure, the tracer is injected at a constant rate, and the solution to Equation (16) becomes

(20)

After sufficient time, the transient term reduces to zero, the concentration attains equilibrium, and Equation (20) reduces to

(21)

Equation (21) is valid only when air change rate I and airflow rate Q are constant; thus, this technique is only appropriate for systems at or near equilibrium. It is particularly useful in spaces with mechanical ventilation or with high air change rates. Constant injection requires measurement of absolute concentrations and injection rates.

Dietz et al. (1986) used a special case of the constant-injection technique, using permeation tubes as a tracer gas source. The tubes release the tracer at an ideally constant rate into the building being tested, and a sampling tube packed with an adsorbent collects the tracer from the interior air at a constant rate by diffusion. After a sampling period of one week or more, the sampler is removed and analyzed to determine the average tracer gas concentration in the building during the sampling period.

Solving Equation (16) for C and taking the time average gives

(22)

where < … > denotes time average. Note that the time average of dC/dt is assumed to equal zero.

Equation (22) shows that the average tracer concentration < C > and injection rate F can be used to calculate the average of the inverse airflow rate. The average of the inverse is less than the inverse of the actual average, with the magnitude of this difference depending on the distribution of airflow rates during the measurement period. Sherman and Wilson (1986) calculated these differences to be about 20% for one-month averaging periods. Differences greater than 30% have been measured when occupant airing of houses caused large changes in air change rate; errors from 5 to 30% were measured when the variation was caused by weather effects (Bohac et al. 1987). Longer averaging periods and large changes in air change rates during the measurement periods generally lead to larger differences between the average inverse change rate and the inverse of the actual average rate.

Equation (16) assumes a single, well-mixed enclosure, and the techniques described are for single-zone measurements. Multizone measurement techniques address airflow between internal zones and between the exterior and individual internal zones (Fortmann et al. 1990; Harrje et al. 1985, 1990; Sherman and Dickerhoff 1989). These techniques are important when considering the transport of pollutants from one room of a building to another. A theoretical development is provided by Sinden (1978a). Multizone measurements typically use either multiple tracer gases for the different zones or the constant-concentration technique. A proper uncertainty analysis is essential in all multizone flow determination (Charlesworth 1988; D’Ottavio et al. 1988).

Stack pressure is the hydrostatic pressure caused by the weight of a column of air located inside or outside a building. It can also occur within a flow element, such as a duct or chimney that has vertical separation between its inlet and outlet. The hydrostatic pressure in the air depends on density and the height of interest above a reference point.

Air density is a function of local barometric pressure, temperature, and humidity ratio, as described in Chapter 1. As a result, standard conditions should not be used to calculate the density. For example, a building site at 5000 ft has air density that is about 20% less than if the building were at sea level. An air temperature increase from −20 to 70°F causes a similar air density difference. Combined, these elevation and temperature effects can reduce air density about 45%. Moisture effects on density are generally much less but can be significant if the change in elevation is great (e.g., in a natural draft cooling tower). Saturated air at 105°F has a density about 5% less than that of dry air at the same pressure.

Assuming the air temperature and humidity ratio are constant over the height of interest, the stack pressure decreases linearly as the distance above the reference point increases. For a single column of air, the stack pressure can be calculated as

(23)

where

For tall buildings or when significant temperature stratification occurs indoors, Equation (23) should be modified to include the density gradient over the height of the building.

Temperature, and thus air density differences between indoors and outdoors cause stack pressure differences that drive airflows across the building envelope; the stack effect is this buoyancy phenomenon. Sherman (1991) showed that any single-zone building can be treated as an equivalent box from the point of view of stack effect; if there is air leakage, follow the power law as described in the section on Residential Air Leakage. The building is then characterized by an effective stack height and neutral pressure level (NPL) or leakage distribution, as described in the section on Neutral Pressure Level. Once calculated, these parameters can be used in physical, single-zone models to estimate infiltration.

Neglecting vertical density gradients, the stack pressure difference for a horizontal leak at any vertical location is

(24)

where

Chastain and Colliver (1989) showed that, when there is stratification, the average of the vertical distribution of temperature differences is more appropriate to use in Equation (24) than the localized temperature difference near the opening of interest.

By convention, stack pressure differences are positive when the building is pressurized relative to outdoors, which causes flow out of the building. Therefore, absent other driving forces and assuming no stack effect within the flow elements themselves, when indoor air is warmer than outdoors, the base of the building is depressurized and the top is pressurized relative to outdoors; when indoor air is cooler than outdoors, the reverse is true. At some elevation in the building, with such conditions, the pressure indoors is equal to the outdoors: this height is the neutral pressure level.

Absent other driving forces, the location of the NPL is influenced by leakage distribution over the building exterior and by interior compartmentation. As a result, the NPL is not necessarily at the mid-height of the building; with effective horizontal barriers in tall buildings, it is also possible to have more than one NPL. NPL location and leakage distribution are described in the Combining Driving Forces and Neutral Pressure Level sections.

For a penetration through the building envelope for which (1) there is vertical separation between its inlet and outlet and (2) air inside the flow element is not at the indoor or outdoor temperature (e.g., in a chimney), more complex analyses than Equation (24) are required to determine the stack effect at any location on the building envelope.

When wind impinges on and flows around and over a building, it creates a distribution of static pressures on the building’s exterior surfaces that depends on the wind direction, wind speed, air density, surface orientation, and surrounding conditions. Wind pressures are generally positive with respect to the static pressure in the undisturbed airstream on the windward side of a building and negative on the leeward sides and roof. However, these pressures depend highly on wind speed, angle, turbulence, the surroundings, and building shape. Static pressures over building surfaces are almost proportional to the velocity head of the undisturbed airstream. The wind pressure or velocity head is given by the Bernoulli equation, assuming no height change or pressure losses:

(25)

where

C_p is a function of location on the building envelope and wind direction. Chapter 24 provides additional information on values of C_p.

Most pressure coefficient data are for winds approaching perpendicularly to upwind building surfaces. Unfortunately, for a real building, this fixed wind direction rarely occurs, and when the wind is not normal to the upwind wall, these pressure coefficients do not apply. Walker and Wilson (1994) developed a harmonic trigonometric function to interpolate between the surface average pressure coefficients on a wall that were measured with the wind normal to each of the four building surfaces. This function was developed for low-rise buildings three stories or less in height. For each wall of the building, C_p is given by

(26)

where

Because the cosine term in Equation (26) can be negative, its sign must be tracked. When cos(ϕ) is negative, subtract the value of the absolute of cos(ϕ) to the 3/4 power.

The measured data used to develop the harmonic function from Akins et al. (1979) and Wiren (1985) show that typical values for the pressure coefficients are C_p(1) = 0.6, C_p(2) = –0.3, and C_p(3) = C_p(4) = –0.65. Because of geometry effects on flow around a building, application of this interpolation function is limited to low-rise buildings of rectangular plan on flat, featureless sites, with the longest wall less than three times the length of the shortest wall. For less regular buildings or sites, simple correlations are inadequate and building-specific pressure coefficients are required; computational fluid dynamic models are often used. Chapter 24 discusses wind pressures for complex building shapes and for high-rise buildings in more detail.

The wind speed most commonly available for infiltration calculations is that measured at the local weather station, typically the nearest airport. This wind speed needs to be corrected for reductions caused by the difference between the height where the wind speed is measured and the height of the building, and reductions caused by shelter effects.

The reference wind speed used to determine pressure coefficients is usually the wind speed at the eave height for a low-rise building and the building height for a high-rise building. However, meteorological wind speed measurements are made at a different height, typically 33 ft for official weather stations, and at a different location than for the buildings of interest. The difference in terrain between the measurement station and the building under study must also be addressed. Chapter 24 shows how to calculate the effective wind speed U_H from the reference wind speed U_met using boundary layer theory and estimates of terrain effects.

In addition to the reduction in wind pressures caused by reduced wind speed, the effects of local shelter also act to reduce wind pressures. The shielding effects of trees, shrubbery, and other buildings within several building heights, horizontally, of a particular building produce large-scale turbulence eddies that not only reduce effective wind speed but also alter wind direction. Local geological features or gaps between neighboring large buildings can, at times, greatly increase wind velocity. Thus, meteorological wind speed data must be adjusted carefully when applied to specific buildings and their locations.

Infiltration rates measured by Wilson and Walker (1991) for a row of houses showed reductions in airflow rates of up to a factor of three when the wind changed direction from perpendicular to parallel to the row. They recommended estimating wind shelter for winds perpendicular to each side of the building and then using the interpolation function in Equation (27) to find the wind shelter for intermediate wind angles:

(27)

where

Although this method gives a realistic variation of wind shelter effects with wind direction, estimates for numerical values of wind shelter factor s for each of the four cardinal directions must be provided. Table 8 in the section on Residential Calculation Examples lists typical shelter factors. The wind speed used in Equation (25) is then given by

(28)

The magnitude of pressure differences found on the surfaces of buildings varies rapidly with time because of turbulent fluctuations in the wind (Etheridge and Nolan 1979; Grimsrud et al. 1979). However, using average wind pressures to calculate pressure differences is usually sufficient to calculate average infiltration values.

Operation of mechanical equipment, such as supply or exhaust systems and vented combustion devices, affects pressure differences across the building envelope and thus air change rates. Interior static pressure adjusts such that the sum of all airflows through openings in the building envelope plus equipment-induced airflows balances to zero. To predict these changes in pressure differences and airflow rates caused by mechanical equipment, the location of each opening in the envelope and relationship between pressure difference and airflow rate for each opening must be known. The interaction between mechanical ventilation system operation and envelope airtightness has been discussed for low-rise buildings (Nylund 1980) and for office buildings (Persily and Grot 1985a; Tamura and Wilson 1966, 1967a).

Air exhausted from a building by a whole-building exhaust system must be balanced by increasing airflow into the building through other openings or the air-handling systems. As pressures vary, air leakage at some locations changes from outflow to inflow. When using makeup air and no dedicated exhaust, the situation is reversed and envelope inflows may become outflows. Thus, the effects of a mechanical system on a building must be considered. Depressurization caused by an improperly designed exhaust system can increase the rate of radon entry into a building and can interfere with proper operation of combustion device venting or other exhaust systems. Pollutant entry can be increased from garages and other attached storage spaces. Depressurization can also force moist outdoor air through the building envelope; for example, during the cooling season in hot, humid climates, moisture may condense in the building envelope and cause rust, rot, or mold. A similar phenomenon, but in reverse, can occur during the heating, and potentially humidifying, season in cold climates if the building is pressurized. Active pressure control is often recommended, as is proper use of moisture retarders, drainage, and drying of in situ building materials.

The interaction between mechanical systems and the building envelope also pertains to systems serving zones of buildings. Performance of zone-specific exhaust or pressurization systems is affected by leakage in partitions between zones as well as through exterior walls.

Mechanical systems can also create infiltration-driving forces in single-zone buildings. Specifically, some single-family houses with central forced-air duct systems have many distributed supply registers, yet only one central return grille. When insufficiently undercut internal doors are closed in these houses, large positive indoor-to-outdoor pressure differentials are created for rooms with only supply registers, whereas the room or hallway with the return grille tends to depressurize relative to the outdoors. This is caused by the resistance of the internal door undercuts, often partially blocked by carpeting, to flow from the supply register to the return; the magnitudes of the indoor/outdoor pressure differentials created average 0.012 to 0.024 in. of water (Modera et al. 1991). Balanced airflow systems with ducted air return and distributed grilles or adequately sized transfer grilles (where still allowed by fire code) reduce this effect significantly.

Building envelope airtightness and interzonal airflow resistance can also affect performance of mechanical systems. The actual airflow rate delivered by these systems, particularly ventilation systems, depends on the pressure differences they work against. This effect is the same as the interaction of a fan with its associated ductwork, which is discussed in Chapter 21 of this volume and Chapter 21 of the 2020 ASHRAE Handbook—HVAC Systems and Equipment. The building envelope and its leakage should be considered part of the ductwork in determining the pressure drop of the system.

Duct leakage can cause similar problems. Supply leaks to the outdoors tend to depressurize the building; return leaks from the outdoors tend to pressurize it. Keeping these ducts within the conditioned buildings, and sealing all ducts well with durable materials and high-quality construction methods, significantly reduces this problem.

Pressure differences caused by wind, stack effect, and mechanical systems are considered in combination by adding them together and then determining the resulting airflow rate through each building envelope. The airflows must be determined in this manner, as opposed to adding the airflow rates caused by the separate driving forces, because the airflow rate through each opening is not linearly related to pressure difference.

For uniform indoor air temperatures, the total pressure difference across each leak can be written in terms of a reference wind parameter P_U and stack effect parameter P_T common to all leaks:

(29)

(30)

where T is absolute air temperature in °R.

The pressure difference across each leak, with positive pressures for flow into the building, is then given by

(31)

where Δp_I is the pressure that acts to balance inflows and outflows, including mechanical system flows. Equation (31) can then be applied to every leak for the building with appropriate values of C_p, s, and H. Thus, each leak is defined by its pressure coefficient, shelter, and height. Where indoor pressures are not uniform, more complex and often numerical analyses are required.

The neutral pressure level (NPL) varies and is that height or heights in the building envelope where, at that particular instant, there is no indoor-to-outdoor pressure difference. Internal partitions, stairwells, elevator shafts, utility ducts, chimneys, vents, operable windows, and mechanical supply and exhaust systems complicate the prediction of NPL location. An opening with a large area relative to the total building leakage causes the NPL to shift toward the opening. In particular, chimneys and openings at or above roof height raise the NPL in small buildings. Exhaust systems increase the height of the NPL; outdoor air supply systems lower it.

Figure 6 qualitatively shows the addition of driving forces for a building with uniform openings above and below mid-height and without significant internal resistance to airflow. The slopes of the pressure lines are a function of the densities of the indoor and outdoor air. In Figure 6A, with indoor air warmer than outdoor and pressure differences caused solely by thermal forces, the NPL is at mid-height, with inflow through lower openings and outflow through higher openings. For the low air velocities typical in and around buildings, the direction of flow is always from the higher to the lower-pressure region.

Figure 6B presents qualitative uniform pressure differences caused by wind alone, with opposing effects on the windward and leeward sides. When temperature difference and wind effects both exist, the pressures caused by each are added together to determine the total pressure difference across the building envelope. In Figure 6B, there is no NPL because no locations on the building envelope have zero pressure difference. Figure 6C shows the combination, where the wind force of Figure 6B has just balanced the thermal force of Figure 6A, causing no pressure difference at the top windward or bottom leeward side.

The relative importance of wind and stack pressures in a building depends on building height, internal resistance to vertical airflow, location and flow resistance characteristics of envelope openings, local terrain, and the immediate shielding of the building. The taller the building and the smaller its internal resistances to airflow, the stronger the stack effect. The stack effect can be reduced by effectively sealing the building internally between floors, typically by gasketing elevator and stairway doors, and sealing pipe, duct, and electrical penetrations; these measures, when done by code-approved means, also typically reduce undesired smoke migration during fire events. Gasketing interior doors, especially those from exterior spaces or to elevator lobbies, in tall buildings can also help restrict air leakage paths.

The effect of mechanical ventilation on envelope pressure differences is more complex and depends on both the direction of ventilation flow (exhaust or supply) and the differences in these ventilation flows among the zones of the building. If mechanically supplied outdoor air is provided uniformly to each story, the change in the exterior wall pressure difference pattern is uniform. With a nonuniform supply of outdoor air (e.g., to one story only), the extent of pressurization varies from story to story and depends on internal airflow resistance. Pressurizing all levels uniformly has little effect on pressure differences across floors and vertical shaft enclosures, but pressurizing individual stories increases the pressure drop across these internal separations. Pressurizing the ground level is often used in tall buildings in winter to reduce negative air pressures across entries; vestibules and revolving doors are also used to limit air leakage. Vestibules may also be used for elevator lobbies and stair towers to reduce air and smoke movement vertically through tall buildings.

Figure 6. Distribution of Indoor and Outdoor Pressures over Height of Building

Available data on the NPL in various kinds of buildings are limited. In tall buildings studied by Tamura and Wilson (1966, 1967b), the NPL varied from 0.3 to 0.7 of total building height. For houses, especially those with chimneys, the NPL is usually above mid-height. Operating a combustion heat source that vents to the outdoors raises the NPL further, sometimes above the ceiling (Shaw and Brown 1982).

Compartmentation of a building also affects the NPL location. Equation (24) provides a maximum stack pressure difference, given no internal airflow resistance. The sum of pressure differences across the exterior wall at the bottom and top of the building, as calculated by these equations, equals the total theoretical draft for the building. The sum of actual top and bottom pressure differences, divided by the total theoretical draft pressure difference, equals the thermal draft coefficient. The value of the thermal draft coefficient depends on the airflow resistance of exterior walls relative to the airflow resistance between floors. For a building without internal partitions, the total theoretical draft is achieved across the exterior walls (Figure 7A), and the thermal draft coefficient equals 1. In a building with airtight separations between each floor, each story acts independently, its own stack effect being unaffected by that of any other floor (Figure 7B). The theoretical draft is minimized in this case, and each story has its own NPL.

Real multistory buildings are neither open inside, nor airtight between stories. Vertical air passages, stairwells, elevators, and other service shafts allow airflow between floors. Figure 7C represents a heated building with uniform openings in the exterior wall, through each floor, and into the vertical shaft at each story. Between floors, the slope of the line representing the indoor pressure is the same as that shown in Figure 7A, and the discontinuity at each floor (Figure 7B) represents the pressure difference across it. Some of the pressure difference maintains flow through openings in the floors and vertical shafts. As a result, the pressure difference across the exterior wall at any level is less than it would be with no internal flow resistance.

Figure 7. Compartmentation Effect in Buildings

Maintaining airtightness between floors and from floors to vertical shafts is a way to control indoor/outdoor pressure differences because of the stack effect and, therefore, infiltration and exfiltration. Good separation is also conducive to proper operation of mechanical ventilation and smoke management systems. However, care is needed to avoid creating pressure differences that could prevent egress doors from opening in an emergency. Tamura and Wilson (1967a) showed that when vertical shaft leakage is at least two times envelope leakage, the thermal draft coefficient is almost one and the effect of compartmentation is negligible. Measurements of pressure differences in three tall office buildings by Tamura and Wilson (1967b) indicated that the thermal draft coefficient ranged from 0.8 to 0.9 with ventilation systems off. Modern internal sealing techniques should result in much less vertical leakage.

The design, operation and maintenance of a building’s ventilation system and envelope, as well as other factors, can significantly affect the building’s potential vulnerability to extraordinary threats, which range from intentional releases of chemical or biological agents inside or outside a building, to releases of chemicals in industrial or transportation accidents, to natural disasters. ASHRAE (2003) addresses several key steps to manage risk from extraordinary incidents, including

Persily (2004) details how ventilation affects buildings’ vulnerability to airborne chemical and biological releases, as well as some strategies for using ventilation, particularly involving airtightness and pressurizing the building interior to protect against outdoor releases, to increase the level of building protection against such incidents. Persily et al. (2007) evaluate retrofit options for protecting buildings from airborne threats; approaches considered include enhanced particle filtration, sorbent-based gaseous air cleaning, ventilation system recommissioning, building envelope airtightening, building pressurization, relocation of outdoor air intakes, shelter-in-place (SIP), isolation of vulnerable spaces such as lobbies, and system shutdown and purge cycles. The filtration and air cleaning options have the advantage of always being operational as long as the systems are properly designed, installed, and maintained. However, the lack of standard test methods is a critical issue in application of some air-cleaning technologies. Building envelope air sealing and pressurization can be quite effective in protecting against outdoor releases as long as effective filtration against the contaminant of concern is also in place. Protection provided by operational changes such as system shutdown and purging depends heavily on timing; if timing is inappropriate, occupant exposure may be increased. Isolating vulnerable zones and other system-related modifications depend on building layout and system design, and careful implementation is necessary for effectiveness under the range of conditions that exist in buildings. Finally, many retrofits can also increase energy efficiency and improve indoor air quality, which should be included in a life-cycle cost comparison of different options to the degree possible. Chapter 59 of the 2019 ASHRAE Handbook—HVAC Applications addresses extraordinary events further.

Air exchange also can affect a building’s thermal load by altering performance of the envelope insulation system. Airflow through insulation can decrease the thermal load through heat exchange between infiltrating or exfiltrating air and the insulation. Conversely, air moving in and out of the insulation from the outdoors can increase the thermal load. Experimental and numerical studies demonstrate that significant thermal coupling can occur between air leakage and insulation layers, thereby modifying the heat transmission in building envelopes. In particular, research (Bankvall 1987; Berlad et al. 1978; Lecompte 1987; Wolf 1966) shows that convective airflow through air-permeable insulation in an envelope assembly may degrade its effective thermal resistance. This R-value degradation occurs when outdoor air moves through and/or around the insulation within the wall cavity and returns to the outdoors without reaching the conditioned space. A literature review by Powell et al. (1989) summarized the findings about air movement effects on the effective thermal resistance of porous insulation under various conditions. The effect of such airflow on insulation system performance is difficult to quantify, but should be considered. Airflow within the insulation system can also decrease the system’s performance because of moisture condensation in and on the insulation.

Even if air flows only through cracks instead of through the insulation, the actual heating/cooling load from the combined effect of conduction and airflow heat transfer can be lower than the heating/cooling load calculated by Equation (33). This reduction in total heating/cooling load is a consequence of the thermal coupling between conduction and convection heat transfer and is called infiltration heat recovery (IHR). Using a computer simulation, Kohonen et al. (1987) found that the conduction/infiltration thermal interaction reduced total heating load by 15%. Several experimental studies (e.g., Claridge and Bhattacharyya 1990; Claridge et al. 1988; Liu and Claridge 1992a, 1992b, 1992c, 1995; Timusk et al. 1992), using a test cell under both steady-state and dynamic conditions, found that the actual energy attributed to air infiltration can be 20 to 80% of the values given by Equation (35). Judkoff et al. (1997) measured heat recovery in a mobile home under steady-state conditions, and found that up to 40% heat recovery occurs during exfiltration through the envelope. Buchanan and Sherman (2000) performed two- and three-dimensional computational fluid dynamics (CFD) simulations to study the fundamental physics of the IHR process and developed a simple macro-scale mathematical model based on the steady-state one-dimensional convection-diffusion equation to predict a heat recovery factor. Their results show that the traditional method may overpredict the infiltration energy load. Using physical experiments, ASHRAE research project RP-1169 (Ackerman et al. 2006) showed that thermal resistances are affected by infiltration and exfiltration, but, on a net basis, the IHR effect can be neglected.

Heating and cooling degree-days (HDDs and CDDs) are a simple way to characterize the severity of a particular climate. Heating and cooling degree-day values are based on sensible temperature data, but infiltration loads are both sensible and latent. Infiltration degree days (IDDs) more fully describe a climate and can be used to estimate heat loss or gain from infiltration in residences (Sherman 1986). Total infiltration degree-days is the sum of the heating and cooling infiltration degree-days and is calculated from hour-by-hour weather data and base conditions using weather weighted by infiltration rate. The selection of base conditions is an important part of the calculation of the IDDs. ASHRAE Standard 119 lists IDDs for many locations with a particular set of base conditions.

Natural ventilation openings include (1) operable windows, clerestories, doors, and skylights; (2) roof ventilators; (3) stacks; and (4) specially designed inlet or outlet openings such as OA louvers and EA grilles, ventilation penthouses or shafts, chimneys, windcatchers, and roof monitors.

Operable windows transmit light, and provide ventilation when opened. They may open by sliding vertically or horizontally; by tilting on horizontal pivots at or near the center; or by swinging on pivots at the top, bottom, or side. The type of pivoting used is important for weather protection and affects airflow rate. Exterior doors, often with insect screens, can also provide a path for natural ventilation; as with operable windows, security concerns must be considered.

Roof ventilators provide a weather-resistant air outlet. Capacity is determined by the ventilator’s location on the roof; the resistance to airflow of the ventilator and its ductwork; the ventilator’s ability to use kinetic wind energy to induce flow by centrifugal or ejector action; and the height of the draft.

Natural-draft or gravity roof ventilators can be stationary, pivoting, oscillating, or rotating. Selection criteria include appearance, ruggedness, corrosion resistance, stormproofing features, dampers and operating mechanisms, noise, cost, and maintenance. Natural ventilators can be supplemented with power-driven supply or exhaust fans; their motors need only be energized when the natural exhaust capacity is too low. Gravity ventilator dampers can be manual or controlled by thermostat or wind velocity.

A natural-draft roof ventilator should be positioned so that it receives full, unrestricted wind. Turbulence created by surrounding obstructions, including higher adjacent buildings, impairs a ventilator’s ejector action. Inlets can be conical or bell-mouthed to increase their flow coefficients. The opening area at any inlet should be increased if screens, grilles, or other structural members cause flow resistance. Building air inlets, most effectively installed at the lowest levels, should be larger than the combined throat areas of all roof ventilators.

Stacks or vertical flues should be located where wind can act on them from any direction. Without wind, stack effect alone removes some air from the rooms with inlets. Much practical experience with natural ventilation from before the adoption of modern air-conditioning appears in publications by ASHVE (a predecessor society to ASHRAE).

In buildings that use natural ventilation, floor-to-ceiling heights are often increased well beyond the normal 8 to 10 ft. Higher ceilings, as seen in buildings constructed before air conditioning was available, allow warm air and contaminants to rise above the occupied portions of rooms. Air is then exhausted from the ceiling zones, and outdoor air is introduced near the floors; a degree of floor-to-ceiling displacement airflow is thus desirable when using natural ventilation during the cooling season.

The ventilation airflow rate required to remove a given amount of heat from a building can be calculated from Equations (33) and (35) if the quantity of heat to be removed and the indoor/outdoor temperature and humidity ratio differences are known.

The relationship describing the airflow through a large intentional opening is based on the Bernoulli equation with steady, incompressible flow. The general form that includes stack, wind, and mechanical ventilation pressures across the opening is

(36)

where

The discharge coefficient C_D is a dimensionless number that depends on the geometry of the opening and the Reynolds number of the flow.

Aspects of wind that affect the ventilation rate include average speed, prevailing direction, seasonal and daily variation in speed and direction, terrain, and local obstructions such as nearby buildings, hills, trees, and shrubbery. Liddament (1988) reviewed the relevance of wind pressure as a driving mechanism. A multiflow path simulation model was developed and used to illustrate the effects of wind on air change rate.

Average wind speeds may be lower in summer than in winter; directional frequency is also a function of season. Natural ventilation systems are often designed for wind speeds of one-half the seasonal average. Equation (37) shows the rate of air forced through ventilation inlet openings by wind or determines the proper size of openings to produce given airflow rates:

(37)

where

Air intakes should be placed in exterior high-pressure regions, and air reliefs should be placed in exterior low-pressure regions, but because of wind variations, these static locations will, at times, not be optimal. Other considerations include flow control when wind speed is high, and security. Air intakes should face directly into the prevailing wind. If they are not advantageously placed, flow will be less than that predicted by Equation (37); if intakes are unusually well placed, flow will be slightly more. Desirable air relief locations are (1) on the leeward side of the building directly opposite the intake; (2) on the roof, in the low-pressure area caused by flow separation; (3) on a side perpendicular to the windward face, where low-pressure areas occur; (4) in a dormer on the leeward side; (5) in roof ventilators; or (6) by stacks. Chapter 24 gives a general description of the wind pressure distribution on a building, but transient computational modeling will likely be required.

If building internal resistance is not significant, flow caused by stack effect can be expressed by

(38)

where

Equation (38) applies when T_i > T_o. If T_i < T_o, replace T_i in the denominator with T_o, and replace (T_i – T_o) in the numerator with (T_o – T_i). If there is thermal stratification, use an average temperature for T_i. If the building has more than one opening for natural ventilation’s airflow, the relief and intake areas are considered equal. The discharge coefficient C_D accounts for all viscous effects such as surface drag and interfacial mixing.

Estimating ΔH_NPL is difficult for naturally ventilated buildings. If one window or door represents a large fraction (approximately 90%) of the total opening area in the envelope, then the NPL is at the mid-height of that aperture, and ΔH_NPL equals one-half the height of the aperture. For this condition, flow through the opening is bidirectional (i.e., air from the warmer side flows through the top of the opening, and air from the colder side flows through the bottom). Interfacial mixing occurs across the counterflow interface, and the orifice coefficient can be calculated by (Kiel and Wilson 1986):

(39)

If enough other openings are available, airflow through the opening will be unidirectional, and mixing cannot occur. A discharge coefficient of C_D = 0.65 should then be used. Additional information on stack-driven airflows for natural ventilation can be found in Foster and Down (1987).

The greatest flow per unit area of openings is obtained when the intake and relief areas are equal; Equations (38) and (39) are based on this equality. Increasing the relief area over intake area (or vice versa) increases airflow but not in proportion to the added area. When openings are unequal, use the smaller area in Equation (38) and add the increase as determined from Figure 8.

Figure 8. Increase in Airflow by Increasing Area of One Opening

Several general guidelines should be observed in designing for natural ventilation. Some of these may conflict with other climate-responsive strategies such as using orientation and shading devices to minimize solar gain, with building codes that encourage compartmentalization to restrict fire and smoke movement, or with other design considerations.

System selection

Building and surroundings characteristics

Opening locations

Opening characteristics

Methods and tools have been developed in recent years to move the art of designing natural ventilation systems beyond the application of simple rules of thumb to engineered design (Axley et al. 2002; CIBSE 2005; Emmerich et al. 2012).

Successful application of purely natural ventilation systems for cooling may be very limited in hot or humid climates, such as in much of the United States, by thermal comfort issues and the need for reliability. However, hybrid (or mixed-mode) ventilation systems or operational strategies offer the possibility of saving energy in a greater number of buildings and climates by combining natural ventilation systems with mechanical equipment (Emmerich 2006). The air-side economizer is one form of hybrid ventilation control scheme, and enjoys wide use in commercial, industrial, and institutional buildings in appropriate climates. The report of the International Energy Agency’s (IEA) Annex 35 describes the principles of hybrid ventilation technologies, control strategies, design and analysis methods, and case studies (Heiselberg 2002). Integrated multizone airflow and thermal modeling is recommended when designing natural and hybrid ventilation systems (Axley 2001a; Dols et al. 2014; Li and Heiselberg 2003).

A building’s envelope leakage can be measured with pressurization testing, commonly called a blower-door test. A fan pressurization test is relatively quick and inexpensive, and it characterizes building envelope airtightness independent of weather conditions; some jurisdictions now mandate that buildings be tested and rated. In this procedure, a large, variable-flow fan or blower is mounted in a door or window and induces a large, roughly uniform pressure difference across the building shell [ASTM Standards E779 and E1827; Canadian General Standards Board (CGSB) Standard 149.10; ISO Standard 9972]. The airflow required to maintain this pressure difference is then measured. The leakier the building is, the more airflow is necessary to induce a specific indoor/outdoor pressure difference. The airflow rate is generally measured at a series of pressure differences ranging from about 0.04 to 0.30 in. of water. Depressurization tests are also used.

The results of a pressurization test, therefore, consist of several combinations of pressure difference and airflow rate data. An example of typical data is shown in Figure 9. These data points characterize the air leakage of a building and are generally converted to a single value that is reported as the building’s airtightness. There are several different measures of airtightness, most of which involve fitting the data to a curve describing the relationship between the airflow Q through an opening in the building envelope and the pressure difference Δp across it. This relationship is called the leakage function of the opening. The form of the leakage function depends on the geometry of the opening. Background theoretical material relevant to leakage functions may be found in Chastain et al. (1987), Etheridge (1977), Hopkins and Hansford (1974), Kronvall (1980), and Walker et al. (1997).

Figure 9. Airflow Rate Versus Pressure Difference Data from Whole-House Pressurization Test

Openings in a building envelope are usually not uniform in geometry, and wind varies, so generally flow never becomes fully developed. Each opening in the building envelope, however, is often described by Equation (40), commonly called the power law equation:

(40)

where

Sherman (1992a) showed how the power law can be developed analytically by looking at developing laminar flow in short pipes. Equation (40) only approximates the relationship between Q and Δp. Measurements of single cracks (Honma 1975; Kreith and Eisenstadt 1957) show that n can vary if Δp changes over a wide range. Additional investigation of pressure and flow data for simple cracks by Chastain et al. (1987) indicated the importance of adequately characterizing the three-dimensional geometry of openings and the entrance and exit effects. Walker et al. (1997) showed that, for the arrays of cracks in a building envelope over the range of pressures acting during infiltration, n is constant. A typical value for n is about 0.65. Values for c and n can be determined for a building by using fan pressurization testing.

In some cases, the predicted airflow rate is converted to an equivalent or effective air leakage area as follows:

(41)

where

All openings in the building shell are combined into an overall opening area and discharge coefficient for the building when the equivalent or effective air leakage area is calculated. Some users of the leakage area approach set C_D = 1. Others set C_D ≈ 0.6, which is the discharge coefficient for a sharp-edged orifice. The air leakage area of a building is, therefore, the area of an orifice with an assumed value of C_D that would produce the same amount of leakage as the building envelope at the reference pressure.

An airtightness rating, whether based on an air leakage area or a predicted airflow rate, is generally normalized by some factor to account for building size. Normalization factors include floor area, exterior envelope area, and building volume.

With the wide variety of possible approaches to normalization and reference pressure difference, and the use of the air leakage area concept, many different airtightness ratings are used. Reference pressure differences include 0.016, 0.04, 0.10, 0.20, and 0.30 in. of water. Reference pressure differences of 0.016 and 0.04 in. of water are advocated by researchers because they are closer to the pressure differences that actually induce air exchange and, therefore, better model the opening’s flow characteristics. Although this may be true, they are below the typical range of measured values in the test; therefore, predicted airflow rates at 0.016 and 0.04 in. of water are subject to significant uncertainty. This uncertainty and its implications for quantifying airtightness are discussed in Chastain (1987), Modera and Wilson (1990), and Persily and Grot (1985b). Round-robin tests by Murphy et al. (1991) to determine the repeatability and reproducibility of fan pressurization devices found that subtle errors in fan calibration or operator technique are greatly exaggerated when extrapolating the pressure versus flow curve down to 0.016 in. of water, with errors as great as ±40%, mainly because of fan and meter calibration errors at low flow.

Some common airtightness ratings include the effective air leakage area at 0.016 in. of water assuming C_D = 1.0 (Sherman and Grimsrud 1980); the equivalent air leakage area at 0.04 in. of water assuming C_D = 0.611 (CGSB Standard 149.10); and the airflow rate at 0.20 in. of water, divided by the building volume to give units of air changes per hour (Blomsterberg and Harrje 1979).

Air leakage areas at one reference pressure difference can be converted to air leakage areas at another reference pressure difference by

(42)

where

Air leakage area at one reference pressure difference can be converted to airflow rate at some other reference pressure difference by

(43)

where

Flow coefficient c in Equation (40) may be converted to air leakage area by

(44)

Finally, air leakage area may be converted to flow coefficient c in Equation (40) with

(45)

Equations (42) to (45) require assumption of a value for n, unless n is reported with the measurement results. When whole-building pressurization test data are fitted to Equation (40), the value of n generally is between 0.6 and 0.7. Therefore, using a value of n in this range is often reasonable.

Fan pressurization measures a building’s leakage behavior that ideally varies little with time and weather conditions. In reality, unless wind and temperature differences during the measurement period are sufficiently mild, pressure differences induced by weather during the blower door test cause measurement errors. Modera and Wilson (1990) and Persily (1982) studied the effects of wind speed on pressurization test results. Several experimental studies also showed variations of about 20 to 40% over a year in the measured airtightness in the homes studied (Kim and Shaw 1986; Persily 1982; Warren and Webb 1986).

Figure 10 summarizes envelope leakage measured in North American housing (Sherman and Dickerhoff 1998) and from several European and Canadian sources (AIVC 1994). This figure shows the large range of measured envelope tightness but also shows typical and extreme values in the housing stock.

ASHRAE Standard 62.2 establishes air leakage performance levels for residential buildings. These levels are in terms of effective annual average infiltration rate Q_inf, which is based on normalized leakage area NL:

(46)

where

where

Figure 10. Envelope Leakage Measurements

The fan pressurization procedure discussed in the section on Envelope Leakage Measurement allows whole-building air leakage to be measured. The location and size of individual openings in building envelopes are extremely important because they influence the air infiltration rate of a building as well as the envelope’s heat and moisture transfer characteristics. Additional test procedures for pressure-testing individual building components such as windows, walls, and doors are discussed in ASTM Standards E283 and E783 for laboratory and field tests, respectively.

Dickerhoff et al. (1982) and Harrje and Born (1982) studied air leakage of individual building components and systems. The following points summarize the percentages of whole-building air leakage area that they found associated with various components and systems. Values in parentheses include the range determined for each component and the mean of the range.

Walls (18 to 50%; 35%). Both interior and exterior walls contribute to the leakage of the structure. Leakage can occur between the sill plate and the floor or foundation; through cracks below the bottom of the gypsum wallboard, around and through electrical boxes, and through plumbing penetrations; and into the attic at the top plates of walls. Holes drilled through the top plates into the attic for passage of wiring are often unsealed.

Ceiling Details (3 to 30%; 18%). Leakage across the top ceiling of the heated space is particularly insidious because it reduces the effectiveness of insulation on the attic floor and contributes to infiltration heat loss. Ceiling leakage also reduces the effectiveness of ceiling insulation in buildings without attics. Recessed lighting, plumbing, and other penetrations leading to the attic are some particular areas of concern, as are intentional openings such as access hatches or for whole-house fans.

Forced-Air Heating and/or Cooling Systems (3 to 28%; 18%). The location of the heating or cooling equipment, air handler, or ductwork in conditioned or unconditioned spaces; the venting arrangement of a fuel-burning device; and the existence and location of a combustion air supply all affect air leakage. Modera et al. (1991) and Robison and Lambert (1989), among others, found that the variability of leakage in ducts passing through unconditioned spaces is high, the coefficient of variation being about 50%. Field studies also showed that in situ repairs can eliminate one-quarter to two-thirds of the observed leakage (Cummings and Tooley 1989; Cummings et al. 1990; Jump et al. 1996; Robison and Lambert 1989). The 18% contribution of ducts to total leakage significantly underestimates their effect because, during system operation, pressure differentials across duct leaks are approximately ten times higher than typical pressure differences across envelope leaks (Modera 1989; Modera et al. 1991) and result in large (factors of two to three), changes in ventilation rate (Cummings et al. 1990; Walker 1999; Walker et al. 1999).

Windows and Doors (6 to 22%; 15%). More variation in window leakage is seen among window types (e.g., casement versus double-hung) than among new windows of the same type from different manufacturers (Weidt et al. 1979). Windows that seal by compressing the weather strip (casements, awnings) often show significantly lower leakage than windows with sliding seals. Leakage around the frames of windows also can be significant if not properly sealed, typically with controlled-expansion foam, during their installation. The door between a residence and an attached garage, or an enclosed hallway in multiunit housing, also needs gasketing to reduce air leakage as well as smoke and other pollutants’ movement.

Fireplaces (0 to 30%; 12%). When a fireplace is not in use, poorly fitting dampers allow indoor air to escape. Glass doors may not seal the fireplace structure more tightly than a closed damper does. Chimney caps or fireplace plugs (with signs that warn they are in place) effectively reduce leakage through a cold fireplace but may not be allowed by fire code. The gap between a metal fireplace insert and the surrounding wall, tiles, or brick is often another leakage path if not properly sealed.

Exhaust Vents from Conditioned Spaces (2 to 12%; 5%). Exhaust ducts going from conditioned spaces to the outdoors frequently have either no dampers or dampers that do not close properly. The gap between the duct and the wall or roof penetration often needs sealing, as well.

Diffusion Through Walls and Ceilings (<1%). Compared to infiltration through holes and other openings in the structure, diffusion through well-constructed walls and ceilings is not an important flow mechanism. At 0.02 in. of water, the permeability of building materials produces an air change rate of less than 0.01 ach by wall diffusion in a typical house. Proper installation of an air-retarding membrane can help minimize this air leakage path.

Component Leakage Areas. Individual building component leakage areas vary widely from house to house. Typical variability for an individual component is about a factor of 10, depending on the component’s construction and installation. Testing should establish the installed leakage of a component in critical applications such as energy-efficient manufactured housing.

Leakage distribution is particularly important in multifamily apartment buildings. These buildings often cannot be treated as single zones because of the internal resistance between apartments. Moreover, leakage between apartments varies widely, from very small for well-constructed buildings with air/moisture retarders and well-sealed wall and floor penetrations between units, to as high as 60% of the total apartment leakage in older, tall apartment buildings of masonry construction (Diamond et al. 1986; Modera et al. 1991). RDH (2013) reviewed the current state of airtightness in multifamily buildings, including testing requirements and techniques, performance targets, and current airtightness levels.

New Buildings. It is much easier and more cost-effective to build a tight building than to tighten an existing building. Elmroth and Levin (1983), Eyre and Jennings (1983), Marbek Resource Consultants (1984), and Nelson et al. (1985) provide information and construction details on airtight building design for houses. The economic paybacks for basic “weatherization” improvements are often the most attractive of all energy conservation measures in buildings where no actions have been taken previously.

A continuous air infiltration retarder, formerly known as an air barrier, can be an effective way to reduce air leakage through walls, around window and door frames, and at joints between major building elements. Take particular care to ensure its continuity at all wall, floor, and ceiling joints; at window and door frames; and at all penetrations of the retarder such as electrical boxes, plumbing connections, and utility service penetrations. Joints in the air/vapor retarder must be lapped and sealed. Plastic vapor retarders installed in the ceiling should be tightly sealed to the vapor retarder in the outer walls and should be continuous over the interior walls. A seal at the top of the interior walls reduces leakage into the attic; the plate on top of the studs generally gives a poor seal. In cold climates, the air infiltration retarder can be installed either on the inside of the wall framing, in which case it usually functions as a vapor retarder as well, or on the outside of the wall framing, in which case it should have a permeance rating high enough to allow diffusion of water vapor from the wall. For a discussion of moisture transfer in building envelopes, see Chapters 25 and 26.

Existing Buildings. Air leakage paths must first be located before the envelope of an existing building can be tightened. As discussed earlier, air leakage into and out of buildings is caused by not only windows and doors but also a wide range of unexpected and unobvious construction defects. Many important leakage routes can be very difficult to find. A variety of techniques developed to locate leakage paths are described in ASTM Standard E1186 and Charlesworth (1988).

Once leakages are located, they can be repaired with materials and techniques appropriate to the size and location of the leak. Diamond et al. (1982), Energy Resource Center (1982), Harrje et al. (1979), and many websites and online videos include information on airtightening or “weatherization” in existing residential buildings with caulking, sealing, weatherstripping, and use of door sweeps, for example. With these procedures, air leakage of residential buildings can be reduced dramatically: anywhere from 5% to more than 50%, depending on the extent of the tightening effort and the experience of those doing the work (Blomsterberg and Harrje 1979; Giesbrecht and Proskiw 1986; Harrje and Mills 1980; Jacobson et al. 1986; Verschoor and Collins 1986). Much less information is available for airtightening large, commercial buildings, but the same general principles apply (Parekh et al. 1991; Persily 1991); the joint between corrugated floor or ceiling decking and the exterior walls is often poorly sealed or not at all, and is often hidden from easy view.

The most fundamental function of a house is to provide shelter from outdoor conditions. A first response to poor outdoor air quality is to go indoors. Closing windows and other air intakes and turning off exhaust fans reduces air exchange with the outdoors, decreasing the immediate intrusion of outdoor air into the home. However, because no home is perfectly airtight, closing doors and windows does not eliminate intrusion. Because all indoor air ultimately comes from outdoors, indoor air quality eventually comes to dynamic equilibrium with outdoor conditions in an idealized, unoccupied building. The tighter the building, the longer the time needed to come to equilibrium.

The delay time (the time it takes to completely change the air in a building) is determined by the ventilation rate. The effectiveness of sheltering within the home thus depends on envelope tightness. For a home with 0.35 air changes per hour, the delay time is roughly 3 h. For a very tight house without mechanical ventilation, the delay time can easily be twice as long. Most houses in the United States are leakier and thus could have a delay time of about one hour (Sherman and Matson 1997).

Reactive gases in outdoor air, such as ozone, can be decreased to some degree by the building envelope. For other outdoor contaminants, the building envelope serves to delay, not reduce, their introduction into the indoor environment. Such a delay is not very helpful at reducing exposures to outdoor contaminants that persist over days, but can be an effective strategy for short-duration (e.g., less than a few hours) sources. In houses without indoor sources, ozone levels tend to be higher in houses that do not have air conditioners than in those with air conditioners; ozone levels also are higher when windows are open than when they are closed (Weschler 2000). For outdoor exposure times shorter than the delay time, the house serves as a reservoir of cleaner air of that particular pollutant. After the outdoor contaminant is gone, windows can be opened to flush out pollutants that entered or were internally generated during the exposure period.

Simply going indoors may not be sufficient for highly unusual but potentially lethal events. Chemical spills or fires, explosions, bioterrorism, or similar toxic air pollutant releases can temporarily create dangerous outdoor conditions that render other air quality issues insignificant. With sufficient warning, occupants should leave the vicinity, but the unexpected nature of these events means that the only viable alternative may be to shelter in place.

This strategy may work for short-term releases of airborne toxins. Homes are often too leaky to provide the protection needed for longer-duration events, but individual rooms can be temporarily sealed to become safe havens. A safe haven should be chosen to have as little contact as possible with outer walls, and preferably be on the side of the house farthest downwind from the source. Impermeable tape can be used to seal leaks, cracks, seams, register grilles, and doors, with thick plastic sheeting used to span larger gaps (Sorensen and Vogt 2001). If such a shelter has an air change rate of 0.15 h⁻¹ with the house, it can take 4 to 6 h for contaminated outdoor air to affect significantly the safe haven. With very well-sealed, occupied spaces, the air quality inside degrades because of the occupants and materials within, especially rapidly for smaller spaces with higher occupant density. Suffocation is remotely possible, but thermal stress and odor accumulation are more typical.

Where emergencies are somewhat more likely, engineered safe rooms with adequate HVAC systems are needed. In such cases, a safe haven can be designed in advance with thermal conditioning and a highly efficient particle/gas-phase filtration system capable of providing several hours, days, or weeks of protection (Ormerod 1983). A short-term safe haven might be effectively combined with another emergency shelter (e.g., tornado, hurricane, civil defense) to reduce cost.

When considering how much whole-house ventilation should be supplied, both typical and unusual sources of indoor pollution should be controlled first to limit the required ventilation rate. This can be done either by mitigating the source itself or by using local exhaust to extract contaminants before they can mix into the indoor environment. Typical sources that should be considered include the following.

Clothes Dryers and Central Vacuum Systems. Clothes dryer exhaust is heavily laden with moisture and laundry by-products such as flammable lint and various gaseous contaminants. Many moisture problems have been traced to clothes dryers vented indoors or to attics, crawlspaces, garages, or other inappropriate locations. Exhaust from clothes dryers, which is typically about 150 cfm, must be vented directly to the outdoors. Similarly, central vacuum systems must be vented directly outdoors to exhaust the finer particles that pass through their filters.

Combustion. Water and carbon dioxide are always produced during combustion of hydrocarbons in air. Dangerous compounds are created, as well. All these products of combustion must be vented directly outdoors, preferably using sealed combustion or direct-vent equipment. Venting must meet or exceed all applicable codes. For buildings with naturally aspirated combustion appliances, excessive depressurization of the building by exhaust systems must be avoided to eliminate backdrafting. Consider a depressurization safety test, such as described in ASTM Standard E1998 or CGSB Standard 51.71. Fireplace combustion products should be isolated from the occupied space using tight-fitting doors and outdoor air intakes, when necessary. Flues and chimneys must be designed and installed to disperse combustion products well away from air intakes and operable windows, for example. Chapter 35 of the 2020 ASHRAE Handbook—HVAC Systems and Equipment has more information on venting systems. Exhaust hoods and fans for ranges and cooktops must vent outdoors.

Carbon monoxide is one of the most unwelcome indoor contaminants, and in significant concentration, poses an imminent threat to life. It can come from virtually any source of combustion, including from motorized vehicles and generators (Emmerich et al. 2013). Even combustion appliances that meet manufacturers’ specifications can interact with the building and emit carbon monoxide. At least one carbon monoxide alarm meeting safety standards such as CSA Standard 6.19 should be installed near sleeping areas in each dwelling, including each unit of multifamily residential buildings, that has combustion appliances (e.g., fireplaces, stoves, furnaces, water heaters) within the pressure boundary, or has attached garages or storage sheds. Carbon monoxide alarms also should be considered for nonresidential buildings: poisonings have occurred in many building types, including hotels, motels, stores, restaurants, nursing homes, dormitories, laundromats, and schools.

Garages. Garages and storage spaces often contain many sources of contaminants. Doors between them and occupied spaces must be well sealed with gaskets or weatherstripping, including their sills, and possibly be self closing. Depressurized sections of HVAC systems for the regularly occupied portions of residences, such as the systems’ air handlers or return or intake ducts, should not be located in garages. If such sections must pass through garages, they must be well sealed and never have registers or grilles installed in the garages or storage spaces; if such spaces need conditioning, they must use their own isolated systems. Take care to ensure that there is a good pressure barrier between the garage and the occupied space, typically using an air/moisture retarder such as heavy polyethylene as well as excellent sealing of the door between. Carbon monoxide sources may be present in garages, so pressure barriers, fire-rated compartmentation, and ventilation of attached residences are life-safety measures. Separate ventilation systems that slightly depressurize attached garages and storage spaces and that exhaust directly outdoors should be considered, especially when these support spaces are tightly constructed or are in cold climates. Several studies (Batterman et al. 2006; Emmerich et al. 2003; Fugler 2004) of contaminant sources and transport in garages found that, in some cases, significant fractions of infiltration air enter houses from attached garages, and that modern residential garages are tighter than older garages to reduce energy consumption and to improve comfort; garages were previously assumed to be leaky enough to avoid many related IAQ problems.

Particulates. The ventilation system should be designed such that return and any outdoor air is well filtered before passing through the thermal conditioning components of the HVAC system. Pressure drops associated with this filtration must be considered in the design and installation of the air-handling system. Particulate filters or air cleaners should have a minimum efficiency of 60% for 0.125 in. particles, which is equivalent to a MERV 6 designated filter according to ASHRAE Standard 52.2.

Microbiologicals. Because ventilation can increase the source as well as removal rates of various air pollutants, it is, at best, moderately effective at reducing exposures to many airborne microbiologicals. Ventilation can, however, be part of the moisture balance that is critical to retarding fungal growth on surfaces and spores released into the air, depending on indoor and outdoor conditions.

Radon and Other Soil Gases. Buildings are exposed to gases and water that migrate through the soil into occupied spaces through cracks or other leaks. Soil gases vary with time and conditions, and can contain toxins from pesticides, landfills, fuels, or sewers, but the highest-profile pollutant in this category is radon and its radioactive-decay-produced “daughters.” Source control measures, such as differential pressure control and ground-to-building airtightening, are far more effective than ventilation mechanisms at controlling exposure to soil gas in buildings. See Chapter 11 for more information.

Volatile Organic Compounds (VOCs). VOCs are ubiquitous in modern life. Examples of products that emit VOCs include manufactured wood products, paints, stains, varnishes, solvents, pesticides, adhesives, wood preservatives, waxes, polishes, cleansers, lubricants, sealants, dyes, air fresheners, fuels, plastics, copy machines, printers, tobacco products, perfumes, cooking by-products, and dry-cleaned clothes. Whenever possible, VOCs and other toxic compounds should be stored outside the occupied space in loosely constructed or ventilated enclosures such as garden sheds or detached garages, and away from occupied buildings’ ventilation intakes and operable windows. When unusual amounts of such compounds are present, consider using additional ventilation should be considered (e.g., extra ventilation during and for a while after construction projects). Reduced-VOC-emitting products are often available.

Outdoor Air. Outdoor air may at times contain unacceptably high levels of pollutants, including ozone, pollen, carbon monoxide, particulate matter, odors, or toxic agents. At such times, it may be impossible to provide acceptable indoor air quality using solely outdoor air, and increased ventilation rates can actually decrease indoor air quality. In locations where this problem may be anticipated, provide automatic or manual controls to allow temporary reduction of the ventilation rate. If these events are frequent, consider a means to more effectively clean recirculated air during these times.

After source elimination, the single most important source control mechanism in dwellings is local exhaust. Kitchens, utility rooms, bathrooms, and other spaces designed to allow specific contaminant releases should be provided with local exhaust that is vented to the outdoors. Workshops, recreation rooms, smoking areas, art studios, greenhouses, and hobby rooms may also require local ventilation and/or air cleaning to remove contaminants generated by the activities involved. Contaminants of concern should be evaluated to determine how much additional ventilation is required. Many of these rooms can be adequately ventilated by following the requirements for kitchens or bathrooms. If unvented combustion appliances must be used, rooms with these appliances should also meet general ventilation requirements for kitchens, because such appliances generate significant amounts of moisture and, often, ultrafine particles, even when operating properly.

Mechanical exhaust (i.e., a fan with ducting to the outdoors) with adequate makeup air is the preferred method for providing local ventilation. Normally, such an exhaust system is designed to operate intermittently under manual control to send contaminated air outdoors when occupants recognize the need for ventilation. However, in many circumstances, a continuous, lower-flow-rate exhaust can work as well. Specific provisions for intermittent versus continuous exhaust rates are typically included in ventilation standards such as ASHRAE Standard 62.2; some example information is provided here.

Continuous Local Mechanical Exhaust. A continuously operating mechanical exhaust system is intended to operate without occupant intervention. This exhaust may be part of a balanced whole-building mechanical ventilation system. The system should be designed to operate during all hours in which the dwelling is occupied. Override control, to provide higher exhaust rates when needed, should be provided. The minimum delivered ventilation should be at least that given in Table 1.

Intermittent Local Mechanical Exhaust. An intermittently operating local mechanical exhaust is intended to be operated as needed by the occupant. Shutoff timers, occupancy controls, multiple-speed fans, and switching integral with room lighting are helpful, provided they do not impede occupant control. The minimum airflow rate should be at least that given in Table 2.

Alternatives. Cleaning recirculated air can sometimes be substituted for local exhaust, if it can be shown to be effective and reliable in removing contaminants of concern. Natural ventilation is not generally a suitable method for local exhaust and ventilation air needs in many climates and spaces. Using natural ventilation can cause reentrainment problems when contaminated exhaust or exfiltrating air reenters the building. In milder climates, natural ventilation may be acceptable when the contaminant of concern is related to odor rather than health or safety. Purpose-designed passive exhaust systems have shown acceptable ventilation in some European settings, and may be considered in lieu of mechanical systems. Axley (2001b) discusses evaluation and design of passive residential ventilation systems further.

Although control of significant sources of pollution in a dwelling is important, whole-house ventilation through centrally introduced, conditioned, and distributed outdoor air may still be needed. Each dwelling should be provided with outdoor air according to code adopted by the local jurisdiction; general guidance according to ASHRAE Standard 62.2 is given in Table 3 of this chapter. The rate is the sum of the Area-Based and Occupancy-Based columns. Design occupancy expectations in the United States can be based on the number of bedrooms as follows: first bedroom, two persons; each additional bedroom, one person. Additional ventilation should be considered when occupant densities exceed one person per 250 ft².

Natural whole-house ventilation that relies on occupant operation should not be used to make up any part of the minimum total whole-house ventilation air requirement. However, because occupancy and sources vary significantly, the capacity to ventilate above minimum rates can be provided by operable exterior openings such as doors and windows.

Ventilation air should be provided to each habitable room through mechanical and natural air distribution. If a room does not have a balance between air supply and return or exhaust, pathways for transfer air should be provided. These pathways may be door undercuts, transfer ducts with grilles, or simply grilles where ducts are not necessary; however, building codes must be checked, because compartmentation requirements for limiting fire and smoke movement may restrict transfer air means or require rated fire and smoke dampers, for example.

In houses without central air handlers, special provisions to distribute outdoor air may be required. Rooms in which occupants spend many continuous hours, such as bedrooms, may require special consideration. Local and whole-house ventilation equipment should be chosen to be energy efficient, easy to maintain, reliable, durable, and quiet. Consider using heat recovery, especially in cold climates.

Occupant comfort, energy efficiency, ease of use, service life, initial and life-cycle costs, value-added features, and indoor environmental quality should be considered when selecting a strategy and system. Poorly designed, installed, or maintained HVAC and related systems can be potential contributors to poor indoor air quality. For example, occupants may not use the ventilation systems as intended if operation results in thermal or acoustical discomfort or excessive energy use. The resulting lack of ventilation might produce poor indoor air quality. Therefore, careful design, construction, commissioning, operation, and maintenance is necessary to provide optimum effectiveness.

All exhaust, supply, or air handler fans have the potential to change the pressure of an indoor space relative to the outdoors. High-flow fans, such as in the air handler and with some cooking exhausts, can cause significant depressurization, particularly in tightly constructed homes. Considering these effects is essential in design. Excessive depressurization of the living space relative to outdoors may cause backdrafting of combustion appliances and increased migration of contaminants such as radon or other soil gases, car exhaust, or insulation particles into the living space. Depressurization can also result in increased moisture intrusion into building cavities in warm, moist climates and may cause more structural damage and fungal growth. Pressurization of the living space can cause condensation in building cavities in cold climates, also resulting in structural damage. Excess pressure can best be prevented by balancing ventilation and exhaust systems and tightly sealing duct systems. In addition, adequate pathways must be available for all return air to the air-handling devices.

Occupant activities, operation of exhaust fans, and leaky ducts may depressurize the structure. Options to address backdrafting concerns include

The system must also be designed, built, operated, and maintained in a way that discourages growth of biological contaminants. Typical precautions include sloping condensate drain pans toward the drain, keeping condensate drains free of obstructions, keeping cooling coils free of dirt and other obstructions, maintaining humidifiers, and checking for and eliminating any cause of moisture inside ducts. Additional information on residential ventilation system design and IAQ control may be found in ASHRAE Guideline 24.

These models of residential infiltration are based on statistical fits of infiltration rate data for specific houses. They use pressurization test results to account for house airtightness and take the form of simple relations between infiltration rate, an airtightness rating, and, in most cases, weather conditions. Empirical models account for envelope infiltration only and do not include intentional ventilation. In one approach, the calculated air change rate at 0.20 in. of water based on a pressurization test is simply divided by a constant approximately equal to 20 (Sherman 1987). This technique does not account for the effect of infiltration-driving mechanisms on air exchange. Empirical models that do account for weather effects have been developed by Kronvall (1980), Reeves et al. (1979), and Shaw (1981).

The latter two models account for building air leakage using the values of c and n from Equation (40). The only other data required are wind speed and temperature difference. These empirical models predict long-term (one-week) infiltration rates very well in the houses from which they were developed; they do not, however, work as well in other houses because of the building-specific nature of leakage distribution, wind pressure, and internal partitioning. Persily (1986) and Persily and Linteris (1983) compared measured and predicted house infiltration rates for these and other models. The average long-term differences between measurements and predictions are generally about 40%, although individual predictions can be inaccurate by 100% or more (Persily 1986; Walker and Wilson 1998).

Multicell models of airflow in buildings treat buildings as a series of interconnected zones and assume that air within each zone is well mixed. Several such models have been developed by Allard and Herrlin (1989), Etheridge and Alexander (1980), Feustel and Raynor-Hoosen (1990), Herrlin (1985), Liddament and Allen (1983), Walton (1984, 1989), and Walton and Dols (2003). They are all based on a mass balance for each zone of the building. These mass balances are used to solve for interior static pressures in the building by requiring that inflows and outflows for each zone balance to zero. The user must provide information describing building envelope leakage, values to account for wind pressure on the building envelope, temperatures for each zone, and any mechanical ventilation airflow rates. Wind pressure coefficient data in the literature, air leakage measurement results from the building or its components, and air leakage data from research can be used as estimates. These models not only solve for whole-building and individual zone air change rates, but also determine airflow rates and pressure differences between zones. These interzone airflow rates are useful for predicting pollutant transport within buildings with well mixed zones. Chapter 13 has more details on multizone airflow and IAQ modeling.

Several procedures have been developed to calculate building air change rates that are based on physical models of the building interior as a single zone. These single-zone models are only appropriate for buildings with minimal internal resistance to airflow, and are therefore inappropriate for large, multizone buildings. Some models of this type have been developed by Cole et al. (1980), Sherman and Grimsrud (1980), Walker and Wilson (1998), and Warren and Webb (1980). The section on Residential Calculation Examples uses both basic and enhanced models (Bradley 1993; CHBA 1994; Hamlin and Pushka 1994; Palmiter and Bond 1994; Walker and Wilson 1998).

The basic model uses effective air leakage area A_L at 0.016 in. of water, which can be obtained from a whole-building pressurization test. The enhanced model uses pressurization test results to characterize house air leakage through leakage coefficient c and pressure exponent n. The enhanced model improves on the basic model by using a power law to represent envelope leakage, including a flue as a separate leakage path, and having separate wind effects for houses with crawlspaces or basements instead of slab-on-grade foundations.

For both models, the user must provide the wind speed, the temperature difference, information on the distribution of leakage over the building envelope, a wind shelter parameter or local shielding for the basic model, and a terrain coefficient. The predictive accuracy of the enhanced model can be very good, typically ±10% when parameters are well known for the building in question (Palmiter and Bond 1994; Sherman and Modera 1986; Walker and Wilson 1998). All models’ results are sensitive to the user-provided data, and some of these parameters are often quite difficult to determine.

Simplified physical models of infiltration solve the problem of two natural driving forces (wind and stack) separately and then combine them in a process called superposition. Superposition is necessary because each physical process can affect internal and external air pressures on the building’s envelope, and can cause interactions between physical processes that are otherwise independent. An exact solution is impossible, because (1) detailed properties of all the building leaks are unknown and (2) leakage is a nonlinear process. Therefore, most modelers have developed a simplified superposition process to combine stack and wind effects. Sherman (1992b) compared various superposition procedures and derived a generalized superposition equation involving simple leakage distribution parameters, and showed that the result is always subadditive. Typically, only 35% of infiltration from the smaller effect can be added to the larger effect. Depending on the specific building and conditions, that percentage could go as high as 85% or as low as zero. Walker and Wilson (1993) compared several superposition techniques to measured data. Sherman, as well as Walker and Wilson, found quadrature, shown in Equation (47), to be a robust superposition technique:

(47)

The following sections discuss how superposition is combined with calculation of wind and stack flows to determine total flow.

Basic Model. The following calculations are based on the Sherman and Grimsrud (1980) model, which uses the effective air leakage area at 0.016 in. of water. This leakage area can be obtained from a whole-building pressurization test. Using effective air leakage area, the airflow rate from infiltration is

(48)

where

Table 4 shows values of Cs for one-, two-, and three-story houses. The value of wind coefficient Cw depends on the local shelter class of the building (Table 5) and the building height. Table 6 shows values of Cw for one-, two-, and three-story houses in shelter classes 1 to 5. In calculating values in Tables 4 and 6, the following assumptions were made:

Example 2.

Estimate the infiltration at winter design conditions for a two-story house in Lincoln, Nebraska. The house has effective air leakage area of 77 in² and volume of 12,000 ft³, and the predominant wind is perpendicular to the street (shelter class 3). The indoor air temperature is 68°F.

Solution: The 99% design temperature for Lincoln is –2°F. Assume a design wind speed of 15 mph. From Equation (48), with C_s = 0.0299 from Table 4 and C_w = 0.0086 from Table 6, the airflow rate caused by infiltration is

From Equation (2), air change rate I is equal to Q divided by the building volume:

Example 3.

Predict the average infiltration during a one-week period in January for a one-story house in Portland, Oregon. During this period, the average indoor/outdoor temperature difference is 30°F, and average wind speed is 6 mph. The house has volume of 9000 ft³ and effective air leakage area of 107 in², and it is located in an area with buildings and trees within 30 ft in most directions (shelter class 4).

Solution: From Equation (48), the airflow rate caused by infiltration is

The air change rate is therefore

Example 4.

Estimate the average infiltration over the heating season in a two-story house with volume of 11,000 ft³ and leakage area of 131 in². The house is located on a lot with several large trees but no other close buildings (shelter class 3). Average wind speed during the heating season is 7 mph, and the average indoor/outdoor temperature difference is 36°F.

Solution: From Equation (48), the airflow rate from infiltration is

The average air change rate is therefore

Enhanced Model. This section presents a simple, single-zone approach to calculating air infiltration rates in houses based on the Walker and Wilson (1998) model. The airflow rate from stack and wind infiltration is

(49)

(50)

where

In calculating tabulated values of C_s, C_w, and s, the assumptions were

The following examples use typical values for terrain factors, house height, and wind speed measurement height, wind speed multiplier G given in Table 7, and use a relationship based on equations found in Chapter 24.

Example 5.

Estimate the infiltration at winter design conditions for a two-story slab-on-grade house with a flue in Lincoln, Nebraska. The house has a flow coefficient of c = 4370 cfm/(in. of water) ⁿ and a pressure exponent of n = 0.67 (corresponding to effective leakage area of 77 in² at 0.016 in. of water). The building volume is 12,000 ft³. The 97.5% design temperature is −2°F, and design wind speed is 15 mph.

Solution: For a slab-on-grade two-story house with a flue, Table 8 gives C_s = 0.001478 (in. of water/°F)ⁿ and C_w = 0.001313 (in. of water/mph²)ⁿ. The house is maintained at 68°F indoors. The building wind speed is determined by taking the design wind speed U_met and applying by the wind speed multiplier G from Table 7:

From Table 5, the shelter class for a typical urban house is 4. Table 9 gives the shelter factor for a two-story house with a flue and shelter class 4 as s = 0.64. The stack flow is calculated using Equation (49):

The wind flow is calculated using Equation (50):

Substituting Q_s and Q_w into Equation (47) gives Q = 126 cfm = 7560 ft³/h. From Equation (2), air change rate I is equal to Q divided by the building volume:

Example 6.

Estimate the average infiltration over a one-week cold-weather period for a single-story house with a crawlspace in Redmond, Washington. The house has a flow coefficient of c = 6890 cfm/(in. of water)ⁿ and a pressure exponent of n = 0.6 (corresponding to effective leakage area of 107 in² at 0.016 in. of water). The building volume is 9000 ft³. During this period, the average indoor-to-outdoor temperature difference is 29°F, and wind speed is 6 mph. The house is electrically heated and has no flue.

Solution: For a single-story house with no flue, C_s =0.000891 (in. of water/°F)ⁿ. For a crawlspace, C_w = 0.001074 (in. of water/mph²)ⁿ. From Table 7, for a one-story house, G = 0.48.

Table 9 gives shelter factor s = 0.50 for a house with no flue and shelter class 4. Stack flow is calculated using Equation (49):

Wind flow is calculated using Equation (50):

Substituting Q_s and Q_w into Equation (47) gives Q = 48 cfm = 2880 ft³/h. From Equation (2), air change rate I is equal to Q divided by the building volume:

Example 7.

Estimate the infiltration for a three-story house in San Francisco, California. The house has a flow coefficient of c = 8740 cfm/(in. of water)ⁿ and a pressure exponent of n = 0.67 (corresponding to effective leakage area of 155 in² at 0.016 in. of water). The building volume is 14,200 ft³. The indoor/outdoor temperature difference is 9°F and wind speed is 10 mph. The house has a flue and a crawlspace.

Solution: For a three-story house with a flue, C_s = 0.001791 (in. of water/°F) ⁿ. For a crawlspace, C_w = 0.001295 (in. of water/mph²)ⁿ. From Table 7, for a three-story house, G = 0.67.

The prevailing wind blows along the row of houses parallel to the street, so the house has a shelter class of 5. Table 9 gives the shelter factor for a three-story house with a flue and shelter class 5 as s = 0.43.

Substituting Q_s and Q_w in Equation (47) gives Q = 83 cfm = 4960 ft³/h. The air changes per hour are

Significant infiltration and mechanical ventilation often occur simultaneously in residences. The pressure difference from Equation (31) can be used for each building leak, and the flow network (including mechanical ventilation) for the building can be solved to find the airflow through all the leaks while accounting for the effect of the mechanical ventilation. However, for simplified models, natural infiltration and mechanical ventilation are usually determined separately and require a superposition method to combine the flow rates.

Sherman (1992b) compared various procedures and derived a generalized superposition equation that involves simple leakage distribution parameters. The result is always subadditive. For small, unbalanced fans, typically only half the mechanically introduced outdoor airflow contributes to the total, but this fraction can be anywhere between 0 and 100%, depending on leakage distribution. When the makeup air fan’s flow rate is large, infiltration may be ignored, because the building becomes either neutrally or positively pressurized.

In special cases when the leakage distribution is known and highly skewed, it may be necessary to work through the superposition method in more detail. For example, in a wind-dominated situation, a makeup air fan has a much bigger effect than an exhaust fan on changing the total ventilation rate; the same is true for houses with high neutral pressure levels in cold climates. For the general case, when details are not known or can be assumed to be broad and typical, the following superposition gives good results:

(51)

The preceding sections on estimating infiltration in low-rise residences represent current analytical techniques typically used for research and remediation purposes, but most small residential buildings are designed and constructed without direct involvement of ventilation engineers. Contractors, who typically prepare these buildings’ designs, are required to follow mandates in various codes and standards, and they apply experience-based rules of thumb when determining, for example, exhaust needs. Often, leaky buildings or air quality problems result. Research and experience have shown that tightening building envelopes and using mechanical ventilation with heat recovery can improve indoor air quality and reduce energy consumption. Retaining the services of a ventilation engineer before construction begins is advisable in some situations.

ASTM Standard E779 and CGSB Standard 149.10 include methods to measure the airtightness of building envelopes of single-zone buildings. Although many multizone buildings can be treated as single-zone buildings for testing by opening interior doors or by inducing equal pressures in adjacent zones, these standards provide no guidelines for dealing with problems arising in tall buildings, such as stack and wind effects. Tall buildings require refinement and extensions of established procedures because they have obstacles to accurate measurement not present in small buildings, including large envelope leakage area, interfloor leakage, vertical shafts, and large wind and stack pressures. Chapter 4 in the 2015 ASHRAE Handbook—HVAC Applications discusses tall buildings and their challenges in detail.

For conducting a pressurization test in a large building, the building’s own air-handling equipment sometimes can be used to induce test pressures, as described in CGSB Standard 149.15. In other cases, a large fan is brought to the building to perform the test, as described by CIBSE Standard TM23. ASHRAE RP-1478’s final report (Anis and Brennan 2014) discusses how to test the envelope leakage of large commercial buildings.

In the past, building envelopes of large commercial buildings were often assumed to be quite airtight. Tamura and Shaw (1976a) found that, assuming a flow exponent n of 0.65 in Equation (40), air leakage measurements in eight Canadian office buildings with sealed windows ranged from 0.120 to 0.480 cfm/ft². Persily and Grot (1986) performed whole-building pressurization tests in large office buildings that showed that pressurization airflow rate divided by building volume was relatively low compared to that of houses. However, if these airflow rates are normalized by building envelope area instead of by volume, the results indicate envelope airtightness levels similar to those in typical North American houses. The same study also looked at eight U.S. office buildings and found air leakage ranging from 0.213 to 1.028 cfm/ft² at 0.30 in. of water. This means that the office buildings’ envelopes were leakier than expected. Typical air leakage values per unit wall area at 0.30 in. of water were 0.10, 0.30, and 0.60 cfm/ft² for tight, average, and leaky walls, respectively (Tamura and Shaw 1976a).

Emmerich and Persily (2014) summarized available measured airtightness data for almost 400 buildings, including about 70 constructed between 2000 and 2010. The average air leakage for the buildings was 20% tighter than the average for the 228 buildings included in a similar 2011 analysis. The data were analyzed to determine factors that affect airtightness (e.g., building type, height). Recent additions to the database, previously reported by Emmerich and Persily (2005), include numerous buildings constructed to meet the specifications of sustainable building programs such as U.S. Green Building Council’s LEED^® rating system, as well as buildings designed and constructed with air retarders. The overall average airtightness reported was 0.72 cfm/ft² for six-sided (i.e., including slabs and basement surfaces) envelope surface areas at 0.0109 psi. The data show only weak trends related to year of construction, height, floor area, wall construction, or building type, but do demonstrate that buildings designed and constructed with attention to airtightness are much tighter than typical commercial buildings. The analysis found that the 79 buildings with air retarders had an average air leakage almost 70% less than the average for the 290 buildings not specified as having air retarders, thus demonstrating the critical need to design and construct commercial buildings with dedicated air retarders to support sustainable building design. However, the wide variation among the measures taken to limit or reduce air leakage among these buildings and the lack of detailed descriptions of the air retarders make it difficult to predict a specific level of airtightness that will result from use of a specific air retarder approach.

In the United States, commercial building construction practices are addressed by various standards, codes, and green building program requirements, and Table 10 summarizes some of the relevant air leakage limits from these requirements. Both ASHRAE Standards 90.1 and 189.1 require continuous air barriers (CABs) for most commercial buildings. Since 2010, Standard 90.1 requires the CAB to meet either a material tightness limit (0.004 cfm/ft² under a pressure differential of 0.30 in. of water) or an assembly tightness limit (0.04 cfm/ft² under a pressure differential of 0.30 in. of water), but it does not include a whole-building tightness limit or a requirement for whole-building pressurization testing. The building commissioning requirements of Standard 189.1-2014 include a whole-building test demonstrating the building meets a tightness limit of 0.25 cfm/ft² under a pressure differential of 0.30 in. of water or the implementation of a rigorous envelope commissioning program.

The 2012 International Energy Conservation Code^® (IECC^®) (ICC 2012a) has requirements with options for a CAB with material or assembly tightness or a whole-building test. The 2012 International Green Construction Code^® (IgCC^®) (ICC 2012b) includes the same requirements as the 2012 IECC but also includes a whole-building testing requirement consistent with the U.S. Army Corps of Engineers (USACE 2009) value. Many U.S. state building codes currently or will include requirements for continuous air barriers either through reference to IECC, IGCC, ASHRAE Standard 90.1 or 189.1, or their own independent requirements; a current list of requirements is available at www.airbarrier.org.

Since 2009, the USACE has required that conditioned buildings be built or retrofitted to include a continuous air barrier to control air leakage through the building envelope (USACE 2009). The specification requires whole-building testing with a maximum leakage of 0.25 cfm/ft² at 0.30 in. of water based on the six-sided building enclosure area (including slab and subgrade walls). The average tightness for a set of 285 new and retrofitted USACE buildings was reported to be 0.18 cfm/ft² (Zhivov 2013). Also, the U.S. General Services Administration (GSA 2010) since 2010 requires all new U.S. federal buildings for the Public Buildings Service to include an air barrier with the whole building having an air leakage rate of not more than 0.39 cfm/ft² at 0.30 in. of water.

Grot and Persily (1986) also found that eight recently constructed office buildings had infiltration rates ranging from 0.1 to 0.6 ach when outdoor air intakes were deactivated. Infiltration rates exhibited varying degrees of weather dependence, generally much lower than that measured in houses. Infiltration in commercial buildings can have many negative consequences, including reduced thermal comfort, interference with proper operation of mechanical ventilation systems, degraded indoor air quality, moisture damage of building envelope components, and increased energy consumption. These results suggest strongly that commercial buildings’ envelopes require tighter construction, and that continuous air barrier systems should be used in the enclosures of all conditioned buildings. Since 1997, the Building Environment and Thermal Envelope Council of the National Institute of Building Sciences has sponsored several symposia on air barriers for buildings in North American climates. Others have also published articles on the importance of limiting air leakage in commercial buildings (Anis 2001; Ask 2003; Fennell and Haehnel 2005).

Envelope leakage in commercial buildings also depends on HVAC system operation. Often, commercial buildings and their HVAC systems are in operation during normal daytime business hours but switch into “unoccupied” operation at nights and on weekends and holidays. If pressurized while their HVAC systems operate, infiltration is often very low or even eliminated in buildings with tight envelopes. However, in unoccupied mode, this pressurization is often lost, so infiltration and potentially moisture intrusion may be significant at times.

In large buildings, air leakage associated with internal partitions becomes very important. Walls, doors, floors, ceilings, plenums, chases, transfer ducts and grilles, and elevator and stairway enclosures are the major separations of concern in these buildings. Their leakage characteristics are needed in multizone models to predict infiltration through exterior walls and airflow patterns in a building. These internal resistances are also important in the event of a fire to predict smoke movement patterns and evaluate smoke management systems.

Table 11 gives air leakage areas calculated at 0.30 in. of water with C_D = 0.65 for different internal partitions of commercial buildings (Klote et al. 2012). Figure 13 presents examples of measured air leakage rates of elevator shaft walls (Tamura and Shaw 1976b), the type of data used to derive the values in Table 11. Consult Chapter 53 of the 2019 ASHRAE Handbook—HVAC Applications for performance models and applications of smoke control systems. As with the exterior shell, making significant effort to seal unintentional openings in internal partitions in large buildings can improve thermal comfort and energy efficiency, as well as improve the building’s performance in a fire.

Figure 13. Air Leakage Rates of Elevator Shaft Walls

Leakage paths, intentional or otherwise, at the top of elevator shafts are often equivalent to orifice areas of 620 to 1550 in². Air leakage rates through stair tower and elevator doors are shown in Figure 14 as a function of average crack width around the door. Sealing elevator and stair doors well, and possibly using indoor vestibules for them, can reduce air as well as smoke flow rates significantly. Air leakage areas associated with other openings in commercial buildings are also important for air movement calculations. These include interior doors and partitions, suspended ceilings in buildings where space above the ceiling is used for air supply or return, and other components of the air distribution system.

Door infiltration depends on the type and use of door, room, and building, and on air speed and pressure differentials. In residences and small buildings where doors are used infrequently, a closed door is assumed, and air exchange can be estimated based on air leakage through cracks between door, frame, and sill. Where exterior door are used frequently, airflow through them increases significantly as door-opening frequency increases. Consider using vestibules or revolving doors for high-frequency applications.

Figure 14. Air Leakage Rate of Door Versus Average Crack Width

Automatically swinging, sliding, rotating, or overhead doors are a major source of air leakage. They are normally installed where large numbers of people use the doors or where bulk goods or vehicles are transported through the doorways. These doors stay open longer with each use than manually operated doors. Air leakage through an automatic door can be reduced by installing a vestibule. However, pairs of automatic doors on the inside and outside of a vestibule normally have overlapping open periods, even when used by only one person at a time. Therefore, it is important that designers include airflow through automatic doors when calculating heating and cooling loads in adjacent spaces.

To calculate the average airflow rate through an automatic door, the designer must consider the area of the door, the pressure difference across it, the discharge coefficient of the door when it is open, and the fraction of time that it is open. Obtaining the discharge coefficient is complicated by the fact that it changes as the door opens and closes.

To simplify this calculation, ASHRAE research project RP-763 (Yuill 1996) developed Figure 15 to combine the discharge coefficients of doors as they open and close with the fraction of time that doors are open at a particular level of use. This figure presents an overall airflow coefficient as a function of the number of people using a door per hour. To obtain the average infiltration rate through an automatic door, multiply this coefficient by the door’s opening area and by the square root of the pressure difference between the outdoor and indoor air at the door’s location. The pressure difference across a door in a building is difficult to predict accurately and depends on wind pressure on the building, stack effect caused by the indoor/outdoor temperature difference, and effects of air-handling system operation. It also depends on leakage characteristics of the building’s exterior walls and of internal partitions.

Figure 15. Airflow Coefficient for Automatic Doors

Figure 16. Pressure Factor for Automatic Doors

Two simplified design methods are presented here. The first method uses practical assumptions to determine design values for R_p, the square root of the pressure difference across the automatic door, given in Figure 16. The second method requires explicit calculation of envelope pressures.

In Figure 16, airflows shown for outdoor air temperatures of 80 and 100°F, represented by dotted lines, are outward flows. They intercept the vertical axis at a lower point than the other lines because wind pressure coefficients on the building’s downwind face, where the greatest outward flows occur, are lower than on the upward face. In many buildings, air pressure in the building is controlled by varying the flow rate through return or exhaust fan(s) or by controlling the relief air dampers. These systems are usually set to maintain a pressure slightly above ambient in the lobby, but in a large building, multiple sensors may be used to regulate air pressure on each floor independently, for example. Subtracting the interior pressure maintained from the wind pressure gives the net pressure for estimating airflow through an exterior door.

Method 1. For the first method, the infiltration rate through the automatic door is

(52)

where

Method 2. Airflow Q is

(53)

where

To find Δp, it is necessary to find the pressure differential created by both wind and stack effect. To give the largest possible pressure difference across the door, there are no interactions between the two natural pressures:

(54)

where

Example Calculations.

Find the maximum possible winter infiltration through an automatic door on the ground floor of a 20-story building. The area of the door is 36 × 84 in. = 3024 in² = 21 ft². Each floor is 13 ft high. Approximately 300 people per hour pass through the door. The design wind conditions are 15 mph, indoor temperature is 70°F, and outdoor temperature is 20°F. The airflow coefficient from Figure 15, using the line for doors without vestibules, is approximately 920 cfm/[ft² · (in. of water)^0.5].

Method 1:

The pressure factor from Figure 16 is 0.5 in. of water^0.5. Equation (52) gives the door’s airflow as

Method 2:

The worst possible case for wind surface pressure coefficient C_p at any point and in any position on the ground floor of the building is inferred from the figures in Chapter 24 to be about 0.75. Using this in Equation (25), together with the specified wind speed, results in p_w = 0.082 in. of water. Assume that H is one-half the door height (42 in.). To have maximum pressure across the door, assume the neutral pressure plane is located halfway up the building such that

Substituting these values into Equation (24) gives Δp_s = −0.19 in. of water. This is the maximum stack pressure difference given no internal resistance to airflow. To find the actual stack pressure difference, it is necessary to multiply this by a draft coefficient. For this example, the coefficient is assumed to be 0.9, which is the highest value that has been found for tall buildings. Therefore, Δp_s = 0.9(–0.19 in. of water) = –0.17 in. of water. The total pressure is then Δp = 0.082 − (−0.17) = 0.252 in. of water. Substituting into Equation (53),

If the building has a vestibule, the airflow coefficient is read from Figure 15 using the line for doors with vestibules, and it is approximately 626 cfm/ft² · in. of water^0.5, reducing airflow to 6600 cfm into the building.

Standard-sized revolving doors are often used where there is high use, but typically not where people would often be carrying large bags or pushing carts. Dols et al. (2014) applied a coupled building energy and airflow simulation tool to simulate airflow through entry doors in a prototype medium office building created by Ng et al. (2012) and found that using a vestibule reduced infiltration through the building entrance by 23%. More information on predicting infiltration rates through standard and extra-large revolving doors is still needed.

Air curtains are jets of air projected across envelope openings with the intention of reducing air exchange and the entrance of dust and insects, for example. They are commonly applied to loading dock doorways and high-use building entrances. Performance of air curtains strongly depends on factors such as jet characteristics, wind, and building pressurization. More discussion on air curtain performance is available in the research literature and in Chapter 20 of the 2020 ASHRAE Handbook—HVAC Systems and Equipment.

Per ASHRAE Standard 62.1, the design ventilation rate is determined starting with a table of minimum ventilation requirements for different space types. These requirements are expressed as an outdoor airflow rate per occupant or per unit floor area, or often both, depending on space type. These ventilation rates are based on air pollutants generated by people, activities, and building materials and furnishings. The rates are then adjusted for various parameters (e.g., multiple zones, type of room air distribution).

The HVAC designer faces several challenges in designing an air distribution system to deliver outdoor air to building occupants. The first is to determine whether the outdoor air is acceptable for use, and to design a system for cleaning the air if it is not acceptable. A second is to design an air intake and distribution system that will deliver the required level of outdoor air to the occupied portions of the building, and not just admit it to an air handler. This outdoor air must be delivered not only at design conditions, but throughout the year. The task is complicated by weather-related variations in indoor/outdoor pressure difference. Other complications include pressure variations caused by building components such as intermittent exhaust fans or dirty filters, and probably most significantly by supply flow variations associated with variable-air-volume (VAV) systems (Janu et al. 1995; Mumma and Wong 1990). This delivery issue is related to the discussion in the section on Air Change Effectiveness.

Many commercial and institutional buildings have multiple-zone, recirculating ventilation systems wherein one or more conventional air handlers condition a mixture of outdoor and recirculated air (supply air) to more than one ventilation zone. Each zone may have a different outdoor air fraction required by ASHRAE Standard 62.1, but each air handler can typically only provide one outdoor air fraction. Therefore, the zone that requires the greatest outdoor air fraction (the critical zone) defines the outdoor air intake rate of the air handler. Consequently, all other zones receive more outdoor air than their minimum requirement. These zones can be considered to have unused ventilation air, which could be returned to the air handler and recirculated; thus, the outdoor air fraction at the air handler could be reduced while still meeting the needs of the critical zone. A method to address multiple-zone recirculating systems, based on the system ventilation efficiency equation (SVEE), is provided in the ventilation rate procedure of Standard 62.1. ASHRAE-sponsored research experimentally tested the validity of this equation and confirmed that the SVEE is a valid predictor of ventilation distribution in a building (Yuill et al. 2012).

Secondary Path Systems. Some systems circulate unused ventilation air through paths other than through the central air handler. Common examples include transfer fans and fan-powered boxes. Warden (1995) suggested that Standard 62.1 should allow for the increased distribution efficiency that is possible with these systems, and presented a generalized SVEE that includes secondary air paths. A form of this equation is given in Equation (A-3) of ASHRAE Standard 62.1-2010.

Equation (A-3) depends on the ventilating ability of the secondary air E_r. The standard previously described E_r in terms of the proportion of average system return air directly recirculated from the critical zone. Yuill et al. (2008) argued that the formulation of Equation (A-3) implicitly defines E_r as a descriptor of the vitiation of the secondary air, and presented a formal definition of E_r, a version of which was adopted for Standard 62.1. Under this definition, if secondary air is drawn from an area that is less contaminated than the building average, values of E_r greater than 1 are possible. Yuill et al. (2012) measured E_r in several locations in an office building and found results ranging from 0.74 to 1.01. Results in the same building could have ranged from 0.14 to 1.13 if the fan-powered boxes had been located differently.

Relatively few measurements of as-built office building ventilation performance have been conducted, and those data generally have not used consistent measurement methods or involved representative collections of buildings. The U.S. Environmental Protection Agency (EPA) Building Assessment Survey and Evaluation (BASE) study involved indoor environmental measurements, including ventilation, in 100 randomly selected office buildings using a standardized protocol (EPA 2003). Persily et al. (2005) analyzed the BASE data and found that outdoor air ventilation rates measured using duct traverses at air handler intakes were higher than might be expected, with a mean of about 117 cfm per person. However, these elevated values are partially explained by low occupant density (mean of about four persons per 1000 ft²) and high outdoor air fractions (mean of about 35%). Considering only values that correspond to minimum outdoor air intake, the mean ventilation rate was 23 cfm per workstation. About one-half the ventilation rates under minimum outdoor air intake were below 20 cfm per person. Another key outcome of this study is documentation of measured airflow rates that are quite different from their design values. This finding highlights the need for good system commissioning and maintenance to achieve design intent. Designing and configuring systems to encourage regular maintenance by providing easy access to key system components is also important.

The example building is about 8 mi northeast of downtown Atlanta, and is close to a major highway and its access roads. Atlanta’s climate is hot and humid in the summer, and has relatively mild winters. The average annual outdoor air temperature is about 60.6°F and the heating degree-days per year, base 65°F (HDD₆₅), are about 3265 (Rock 2005). From Chapter 14, the winter 99% design outdoor air (OA) temperature is 23°F, whereas the 1% cooling dry-bulb temperature is 91°F, with a mean coincident wet-bulb temperature of 74°F. The 99.6 and 0.4% design wind speeds are 12 mph in the winter and 9 mph in the summer, both from the northwest. Warm and humid winds also travel north from the Gulf of Mexico, and occasionally the wind is from the Atlantic Ocean from the southeast.

The approximately 30,500 ft² building is a two-story, flat-roofed, slab-on-grade commercial office building with a substantial roof overhang in each direction. Materials and construction quality are average commercial grade. Double-paned windows, and similar spandrel glass, are fixed in their metal curtain wall frames; all windows are nonoperable. The remaining portions of the exterior walls are brick. There are relatively few exterior doors, as described later in this example. The building is surrounded by black asphalt driveways, a parking lot, and some vegetation. The nearby highway is across a parallel two-lane access road, to the northwest.

The building is occupied during normal weekday business hours, and occasionally for special weekend events. Night and weekend thermostat setbacks are used. On the perimeter of the building are mostly single-person offices and conference rooms. The core of the building is mainly open plan with cubicle workspaces, as well as various support rooms, restrooms, two stair towers, and an elevator. There is a large mailroom on the first floor and a lunchroom on the second. Occupant density is high during workdays. The overhead fluorescent lighting is typical of such office buildings, and there are significant computing, printing, and copying equipment loads. Smoking is not allowed in the building.

The building is owner occupied. Owners generally have long-term interests in minimizing costs, and in maximizing indoor air quality and thermal comfort so that workers’ productivity is high.

For this example, assume a conventional all-air overhead HVAC system, and that the building is well sealed. Consequently, a slightly positive overall building pressurization is assumed during occupied hours. Because water condensation in the exterior envelope of the building is possible, air pressurization should be as low as is practical, and continuous air and moisture retarders should be installed. As a more expensive alternative to slight pressurization, the automatic control system could actively manage the dampers’ positions and fans’ operation to maintain an average neutral pressurization, relative to the outdoors. In either case, a good assumption is that infiltration is minimized, the windows and spandrel glass are fixed and well sealed, and the exterior doors are normally kept closed. During high-wind conditions beyond design, windward perimeter spaces may have some infiltration loads, but under nonpeak outdoor temperatures, a well-zoned HVAC system should have enough capacity to handle these extra loads. If both the OA temperature and wind are extreme, then these upwind perimeter spaces may become slightly uncomfortable. These extreme conditions are expected to occur only a few hours in a typical year.

Spaces with exterior doors can experience significant infiltration loads when people enter and leave. First-floor vestibules on the north and south sides of the building help limit this infiltration through the two main entrances. The double doors from stair tower #2 have infrequent use, and a high level of thermal comfort in stair towers is not typically expected. Thus, brief infiltration surges in stair tower #2 are deemed acceptable. However, the doors from the parking lot to the mailroom are frequently used by staff for shipping, receiving, entrance, and egress, and infiltration loads on the vestibules and mailroom are of concern. Many designers choose to ignore these extra loads in pressurized buildings, because they are transient and not easily characterized; the systems’ capacities are likely sufficient to minimize uncomfortable conditions in these spaces. In this example, however, the HVAC designer is concerned about summertime airborne moisture, especially in the mailroom where books and other publications are stored, because strong, humid, southerly winds easily overcome a slight indoor pressurization when the large doors to the parking lot are open.

This chapter and many of its supporting references describe detailed methods for estimating infiltration or air leakage. Typically, pressure differences, openings’ coefficients, and hour-by-hour weather data are required to perform these transient calculations, usually using a computer program separate from that used for thermal load calculations. For HVAC design purposes for a building similar to the example, an air change rate of unconditioned outdoor air through infiltration, per space, expressed in air changes per hour or airflow rate (in cfm) is of more immediate use. Either value is then entered into the load calculation program. Unfortunately, accurate air changes per hour are difficult, if not impossible, to predict, so design estimates must be made. For example,

Either 1.0 ach or 22 cfm of infiltration is then used as input for the load calculation program for this space. The 1.0 ach assumption was made by the designer during on-site observation that these particular manually operated exterior doors have low usage. If passage rates were known, Yuill’s (1996) flow rate estimation method would have been used instead.

In practice, this back entrance from the parking lot on the southeast side of the building is the primary means of entrance and egress, and as such, the estimated infiltration for it is increased to 2.0 ach, compared to the north vestibule’s 1.0 ach.

In colder U.S. climates, it is common practice for low-cost commercial buildings to have only space heating, and not cooling, in stair towers and vestibules. However, for this building in the Southeast, the designer decided to provide cooling for these vestibules. Thus, the estimated infiltration rates are applied to both the heating and cooling load calculations for these spaces. The building’s mailroom, which also has exterior doors, is to be heated and cooled, too.

Even though the mailroom has only a single layer of doors to the outdoors, and not a vestibule, the designer estimated the infiltration at a lower rate (0.5 ach) than those for the vestibules. This is because of the mailroom’s large interior volume relative to its exterior doorway’s area.

Note that no estimate of air changes will be accurate at all times; this portion of HVAC design is still largely an art because of the many unknowns and variability of weather and building use. For improved energy conservation, all exterior doors must be extremely well weatherstripped and have automatic closers. A sign indicating that the doors should be kept closed when not in use should be placed on the mailroom’s exterior doors. High-quality gaskets and sealants for the windows and spandrel glass are also required to minimize infiltration.

(This section assumes that ANSI/ASHRAE Standard 62.1-2010 has been adopted into the local building code without modification.) At least 10 rooms require direct, powered air exhaust: two restrooms per floor, a darkroom, three designated photocopy spaces, and two janitors’ closets. The restrooms have three flushable fixtures each, so from Table 6-4 of Standard 62.1-2010, with intermittent use, each restroom requires

Also from Table 6-4, the darkroom on the second floor needs

Similarly, the designated photocopy areas need 0.50 cfm_ea/ft², so

The two small janitors’ closets, one on each floor, also require exhaust:

These local exhaust airflow rates are then entered into the load calculation program. They are room loads, attached to each particular space, and are not combined and entered as systems-level loads. The load calculation program evaluates the room loads, appropriately combines them, and then finds the systems-level loads for various peak hours.

Some local code authorities amend the requirements of Standard 62.1, or have not yet adopted the most current version, so significant deviations from these examples are possible. For example, in much of the United States, janitorial closets and photocopy rooms have not been required to have local exhausts. Standard 62.1-2010 recognized that these spaces can be significant sources of airborne pollutants, and some direct exhaust from each of them can be very beneficial for improving indoor air quality.

(This section also assumes that ANSI/ASHRAE Standard 62.1-2010 has been adopted into local code without changes.) Ventilation air is needed to maintain acceptable indoor air quality. The example building is well sealed, natural ventilation is not used, and no credit for any infiltration is taken toward ventilation air requirements, as is typical for conventional commercial buildings. Thus, minimum ventilation air required by Standard 62.1 is provided mechanically through the building’s AHUs. Because smoking is not allowed in the building, no extra ventilation for environmental tobacco smoke (ETS) is needed. However, considering outdoor air pollution from the major highway nearby as well as metropolitan Atlanta’s smog, some outdoor air pretreatment may be considered later in the design process.

Standard 62.1 has two methods for determining needed ventilation airflow rates: the performance IAQ procedure (IAQP), and the prescriptive ventilation rate procedure (VRP). Most HVAC designers of conventional buildings with normal occupancies and outdoor air conditions use the VRP, which is appropriate for this example building.

Required ventilation air, which is solely conditioned outdoor air, is admitted to this building through two air-handling units; each AHU serves one floor. Flow rates of outdoor air are input values for, and carried through to the results of, the load calculation simulation. Energy needed to condition the outdoor air ultimately is a systems-level load, because all of this ventilation air is conditioned by the AHUs before its introduction to the building.

Commercial load calculation programs often provide suggested values of ventilation airflow rates and occupancy schedules, but may not have been updated to reflect the latest VRP requirements and procedures of Standard 62.1. As such, it is difficult to present an example here; instead, a sample check using some assumed values for the first-floor executive director’s office is given. It is assumed that this room is a separate thermal zone because of its use and its location on the southwest corner of the building and thus two afternoon solar exposures.

Executive Director’s Office

The supply airflow rate was estimated using 300 ft²/ton, a sensible heat factor of 0.9, a cooling supply (55°F) to room (75°F) air temperature difference of 20°F, 12,000 Btu/h per ton of cooling, and the sensible heat equation 1.1 × cfm_sa × ΔT. From Table 6-1 of Standard 62.1-2010, the office’s population P can be estimated as

In this case, however, there is only one regular occupant of the space. The needed ventilation airflow rate to the breathing zone V_bz is then found from the table as follows:

where

Note that Standard 62.1’s VRP includes a building component R_aA_z, as well as the traditional per-person people component.

Because this is a conventional office building, with ceiling plenums and no raised floors, overhead air supply and return is assumed. The cooling mode, not heating, is dominant in this and most other U.S. office buildings that have high internal heat gains and well-sealed and insulated envelopes. From the standard’s Table 6-2, with ceiling supply of cool air, the zone air distribution effectiveness E_z is estimated as 1.0. From Equation (6-2) of the standard, the design zone outdoor airflow rate V_oz is then

But this is still not the amount of outdoor air that must be conditioned by the air handler: the rate must be adjusted for inefficiencies and recirculation in the air distribution system.

Because single-duct VAV with terminal reheat air distribution systems were initially planned by the designer, Standard 62.1’s multiple-zone recirculating systems adjustment is needed. For this thermal zone, the primary outdoor air fraction Z_p for its VAV terminal unit and downstream is

However, for VAV systems, the minimum expected primary airflow rate should be used. In this case, 412 cfm is the peak design airflow rate. Designers often assume about 30% of this peak flow as the minimum in VAV systems, so for this space, 412 × 0.3 = 124 cfm. The adjusted primary outdoor air fraction is then

The preceding calculations need to be performed for every thermal zone on each air handler. Then, for each system, the highest primary outdoor air fraction is used to estimate the air distribution systems’ ventilation effectiveness; the 62.1 user’s manual (ASHRAE 2010b) includes a spreadsheet for doing these calculations. Increasingly, load calculation programs include the necessary routines and data.

For the purposes of this example, 0.16 is assumed to be the maximum Z_p, so, from Table 6.3 of Standard 62.1, the system ventilation efficiency E_v is 0.9. If, instead, the standard’s Appendix A method for determining E_v were used, a value closer to 1.0 for perfect mixing would likely result for this example’s conventional overhead all-air cooling system. Table 6-3’s value of 0.9 is likely somewhat conservative, but is obtained quickly for design purposes.

Next, the uncorrected outdoor air intake flow rate V_ou is needed; Standard 62.1’s Equation (6-6) includes diversity factor D to adjust the people component of the flow rate. All zones’ flow rates are needed to perform this calculation. For this example, the uncorrected outdoor air intake flow rate for the first floor’s AHU was estimated from floor area, an occupancy of 5 people per 1000 ft², and 20 cfm per person, and is assumed to be 1525 cfm. The adjusted outdoor air intake flow rate V_ot for this AHU is then

After load calculations are complete, these assumed airflow rates can be replaced with actual values for each zone, and the outdoor airflow rate can be updated. Repeating the load calculations may be necessary. The final value of the adjusted outdoor air intake flow rate is then reported on the AHU’s equipment schedule so that testing, adjusting, and balancing (TAB) personnel and others can use this information to ensure that the system admits the desired flow rate of ventilation air. The information is also used to select air cleaners, dampers, coils, ducts, and fans.

For more examples on determining ventilation air rates for commercial buildings, see the user’s manual for Standard 62.1 (ASHRAE 2010b). For low-rise residential buildings, consult ASHRAE Standard 62.2 and its user’s manual (ASHRAE 2010a).