Drainage, Ventilation Drying, and Enclosure Performance

Abstract

This paper explores the influence and role of both drainage and ventilation drying on the ability of enclosure assemblies to control moisture. Drainage is often the most direct method of removing water from within a wall (i.e., from exfiltration condensation or rain penetration), but it is often not sufficient to provide moisture control. Design approaches that rely solely on drainage to remove moisture from behind the outer layers or cladding ignore the significant quantities of moisture that can be stored in the outer layers of most enclosure walls. Most cladding systems have relatively low vapor permeability and therefore tend to restrict diffusive drying. Moisture trapped in or behind the cladding can be transported into the enclosure by solar-driven diffusion, especially in air-condition ed buildings. Rather than control vapor diffusion, a 6 mille vapor retarder close to the interior may, in many instances, exacerbate wetting and greatly retard drying.

Full text

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
0
Drainage, Ventilation Drying, and Enclosure Performance
By: J.F. Straube1 and E.F.P. Burnett2
ABSTRACT
This paper explores the influence and role of both drainage and ventilation drying on the ability of
enclosure assemblies to control moisture. Drainage is often the most direct method of removing
water from within a wall (i.e., from exfiltration condensation or rain penetration), but it is often
not sufficient to provide moisture control. Design approaches that rely solely on drainage to
remove moisture from behind the outer layers or cladding ignore the significant quantities of
moisture that can be stored in the outer layers of most enclosure walls.
Most cladding systems have relatively low vapor permeability and therefore tend to restrict
diffusive drying. Moisture trapped in or behind the cladding can be transported into the enclosure
by solar-driven diffusion, especially in air-conditioned buildings. Rather than control vapor
diffusion, a 6 mille vapor retarder close to the interior may, in many instances, exacerbate wetting
and greatly retard drying.
The role of within wall ventilation, especially for North American conditions, has not been well
researched and there does not appear to be any consensus with respect to the effect of ventilation
on drying. We have found that air flow behind the cladding (ventilation) can be an important
means of removing moisture stored within and behind vapor impermeable cladding. Calculations,
lab experiments, field monitoring and anecdotal evidence all show that ventilation can not only
improve the drying capacity of wall assemblies, it is sometimes necessary for proper performance.
Several years of temperature, humidity and moisture data collected from full-scale wall
assemblies installed in a natural exposure and test facility are used to demonstrate these points.
1 Research Engineer, Building Engineering Group, University of Waterloo, Waterloo, Ontario,
Canada, N2L 3G1.
2 Professor, Hankin Chair, Departments of Civil and Environmental Engineering and Architectural
Engineering, Pennsylvania State University, University Park, PA, USA, 16802-1408.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
1
INTRODUCTION
Moisture is one of the most important factors affecting building enclosure durability and
performance, especially in cold climates. The design of moisture-tolerant enclosures should
involve the simultaneous consideration and balancing of the potentials for wetting, storage, and
drying. Design guidelines may stress the avoidance of wetting, but the increase of safe moisture
storage capacity or drying potential can also improve the moisture tolerance of an assembly.
Drainage is usually regarded as the most important drying mechanism, and internal drainage has
recently received much attention with regard to walls clad with EIFS, wood siding, stucco, etc.
Screened and drained wall systems are widely recommended for all but the driest climates.
Drainage, however, does not necessarily remove sufficient moisture to ensure proper enclosure
performance: other drying mechanisms must be provided. One drying mechanism that has not
received the attention it is due is ventilation.
This paper presents a brief overview of how moisture is stored in hygroscopic materials and
typical screened and drained enclosure wall systems. Available moisture-removal mechanisms
will be discussed. Ventilation drying is examined in some depth, with the aid of theoretical
calculations, laboratory tests, and field measurements. Several important implications for
enclosure design and performance are presented and briefly discussed.
MOISTURE CONTROL
A logical approach to the development of a moisture-control strategy for enclosure assemblies
would assess the moisture storage and transport characteristics of the system as well as the wetting
and drying potentials. Moisture transport, wetting, storage, and drying are briefly discussed
below.
Moisture Transport
Drainage is a liquid flow mechanism driven by gravity. Capillary transport, driven by gradients in
suction stress, is another possible mechanism for transporting liquid moisture, although capillarity
can only redistribute moisture, not remove it, from the enclosure. Vapor diffusion, driven by
gradients in the vapor content of the air, and convection, driven by air pressure differences, are the
primary mechanisms transporting vapor.
Wetting
Wetting, theoretically speaking the increase in moisture content of a system, occurs by several
mechanisms. Vapor adsorbs to the internal surface area of porous materials, liquid is absorbed by
capillary attraction into cracks and pores, and liquid water and frost can adhere to surfaces.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
2
It needs to be emphasized that wetting is a dynamic process, and that the drying of one material
may occur by wetting of another.
Moisture Storage
Enclosure systems constructed of hygroscopic porous materials (e.g., wood, stucco, brick) can
store significant quantities of water. The capillary forces in such porous building materials will
continue to absorb water until the material’s moisture content reaches its capillary saturation
moisture content. Conversely, drainage cannot begin until the capillary saturation moisture content
is reached, or the rate of water deposition exceeds the rate of absorption.
With regard to the latter point, it can be shown that many wetting mechanisms deposit water at
slow enough rates for most of the water to be absorbed. For example, condensation tends to
deposit moisture slowly. As a result, the material on which condensation occurs (e.g., brick
veneer, gypsum or waferboard sheathing) often has sufficient time to absorb the deposited
moisture. Driving rain deposition often occurs slowly enough for brick veneers and many stucco
finishes to absorb much of the water (Straube 1998). Therefore, it is reasonable to assume that in
many building enclosure wetting situations, a material must reach capillary saturation before a
sufficient volume of water will bead on the surface and thus allow drainage to occur.
The threshold moisture content level that corresponds to most moisture-related damage
mechanisms is often equivalent to that material’s moisture content when that material is in
equilibrium with an environment of approximately 80%RH (Ashton 1970, Baker 1969, Sereda
1975). At this relative humidity, both fungal growth and corrosion can be sustained, provided
temperature conditions are favorable. This is a first-order estimate, since wood may require
higher RH levels for decay fungi to act, and steel may corrode at lower RH levels. Nevertheless,
it is reasonable to use the moisture content of a material at 80%RH as a conservative threshold
level for performance problems.
Figure 1 is a plot of the moisture content (in mass percent) versus relative humidity for several
common building materials [Kuenzel 1997]. The difference between the capillary saturation
moisture content and the “safe” moisture content level at 80%RH is tabulated.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
3
0%
5%
10%
15%
20%
25%
30%
0% 20% 40% 60% 80% 100%
Relative Humidity
Moisture Content (M%)
concrete
brick
cement stucco
softwood
to 140%
"Safe" zone
Percent moisture content lost
between saturation and 80%RH
2.8%
22%
8.7%
125%
Figure 1: Sorption Isotherms of Some Cladding Materials
Provided that a smooth and unobstructed path exists, gravity drainage can remove the greatest
volume of water in the shortest time and hence can be one of the most important mechanisms for
moisture removal from within a building enclosure. However, even in perfectly constructed
envelopes, a significant volume of moisture cannot be drained. Regardless of its source, moisture
that enters an enclosure assembly can be stored in a variety of ways (Figure 2):
1. as water trapped at mortar dams in brick veneer walls or poorly drained portions of
other types of walls;
2. as droplets (or frost) adhered by surface tension to the backside of the cladding or front
side of the inner wall layers;
3. adsorbed or absorbed (i.e., retained by capillarity), in hygroscopic building materials
(especially brick, wood, fibrous insulation, paper, etc.) ; and
4. as vapor in the air.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
4
2
3
1
3
4
Screen
Drainage
opening
Airspace
Remainder
of wall
Figure 2: Moisture Storage in Screened and Drained Wall Systems
All of the moisture stored by the mechanisms listed above cannot be removed by drainage (below
the capillary saturation moisture content, water will not leave a material by gravity forces).
Therefore, it can be concluded that drainage, while necessary, is not sufficient to ensure a safe
moisture content -- a significant amount of moisture must be removed by other mechanisms in order
to reduce the moisture content to below the “safe” 80%RH level.
We also have reason to believe that drainage may not necessarily be the largest contributor to
moisture removal. For example, field monitoring of more than 20 well-built low-rise screened
and drained wall panels over a two-year period found only a few instances in which water was
measured draining from behind brick veneer, vinyl, or other drained cladding systems (Straube and
Burnett 1997). In the few instances in which a measurable amount of drainage did occur, the
amount of water collected was less than 50 ml/m2. The greater the driving rain exposure and the
lower the absorbtance of the cladding, the more often and the greater the volume of drainage.
Drying
Moisture is usually removed from within drained screened walls by several transport mechanisms
acting in series, often with phase changes. For example, water trapped in the stud space of a wall
may be directly removed by drainage, but it is much more likely that this liquid will be absorbed
by the wood (capillary transport), evaporated (phase change), and then leave the assembly in
vapor form by diffusion or convection
Moisture that is stored within porous or hygroscopic materials, such as wood siding and brick
veneers, or inner wall layers, such as expanded polystyrene (which can easily store several times
its own weight), waferboard and gypsum sheathing, can only be removed in vapor form. Water
stored in most cladding materials can be capillary transported to or near the exterior surface, here
it can leave by diffusion to the outdoor air. Alternatively, drying can proceed towards the interior

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
5
of the assembly, where the vapor can then be adsorbed by other hygroscopic materials, or can pass
through the assembly to the interior air. Materials within the assembly and not in capillary contact
with the cladding, can dry only by vapor diffusion through the cladding or towards the interior.
In summary, moisture can be removed from an enclosure wall (i.e., dry) in a variety of ways
(Figure 3):
1. gravity drainage of liquid moisture;
2. capillary transport to, and evaporation from, the outer surface of the screen;
3. diffusion and/or convection of water vapor outward through the screen, and inward into
the wall or building interior; and
4. convective flow of exterior air through the air space, (e.g., ventilation).
1
2
4
3
Figure 3: Moisture Removal in Drained Screened Walls with Ventilated Air Spaces
Drained wall systems without vents, face-sealed and perfect barrier walls and solid walls will, of
course, have fewer possible drying mechanisms.
A layer of 6 mille (0.15 mm) thick polyethylene is often placed just outside the interior finish,
either because building codes requires it or because it is deemed good practice. The water-vapor
permeance of 0.15 mm polyethylene sheet is about 3.4 ng/Pa·s·m2 (ASHRAE 1997). This
permeance is so low, and its location so close to the interior, that little diffusive drying into the
building can be expected. Therefore, drying of walls with an interior polyethylene vapor barrier
can only proceed outwards, i.e. to the exterior, or to the interior via air leakage (convection).
Significant amounts of moisture redistribution from outer layers to inner or vice versa may, of
course, still occur.
Outward diffusive drying of the inner layers of a wall will be greatly retarded in walls that have a
cladding with high vapor resistance. For example, the water-vapor permeance of 90 mm thick
brickwork is 45 ng/Pa·s·m2 (ASHRAE 1997); in Canada this qualifies it as a Type 2 vapor barrier.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
6
Other types of finishes such as cement-based stucco, plywood siding, natural stone veneers, and
some synthetic exterior finish systems may also have high vapor resistances.
Therefore, to increase the drying capacity, and thereby the moisture tolerance, of an enclosure
system it follows that:
• the vapor permeance of inner layers should be high enough to allow inward drying while still
controlling outward-acting wintertime diffusion condensation, and
• the vapor permeance of the cladding should be increased or some other means found to allow
vapor to leave via the exterior of the enclosure.
The first conclusion suggests that “more is not necessarily better” when choosing a vapor diffusion
retarder for some wall systems in some climates. Although this first conclusion clearly has
important implications, the focus of this particular paper is the second conclusion, specifically the
contention that ventilation may be beneficial to enclosure performance by providing a means of
allowing water-vapor to exit via the exterior of the enclosure.
VENTILATION DRYING
This section of the paper develops the physics of ventilation drying and demonstrates a means of
assessing the influence of ventilation flow on wall drying. Most of the physics are developed
more fully in Straube and Burnett (1995) or Straube (1998).
In theory, ventilating the space behind the cladding with outdoor air offers two major benefits:
• the flow of relatively dry outside air allows convective drying of all surfaces lining the air
space (e.g., the inside face of the cladding and the outside face of the inner wall layers), and
• water vapor diffusing through the inner wall layers can bypass the vapor diffusion resistance of
the cladding and be carried directly outside.
Thus, ventilation could increase the drying potential of walls, especially in assemblies that either
store significant amounts of water in their outer layers or have a cladding with high vapor
resistance.
The heat capacity of air is so limited that little heat can be carried out of the air space by
ventilation (unless there are very large and fast airflows). In most enclosure walls, ventilation
will not affect the insulation value of the air space for the majority of the time as long as the
insulation (e.g., insulating sheathing, batt) is protected from wind washing. Very small airflows
can, however, transport significant quantities of moisture if they act for long enough. Because the
air space in many walls is usually warmer and contains more moisture than the outdoor air, even
small ventilation flows over many days have the potential to remove useful amounts of moisture.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
7
Forces Driving Ventilation Flow
A combination of wind pressure differences, thermal buoyancy and moisture buoyancy drive
ventilation flow. The provision of vent openings at both the top and bottom of the air space will
generally promote the most ventilation flow because these vent locations take maximum advantage
of both buoyancy forces and spatial wind pressure variations.
Wind pressure is probably the most important force driving ventilation flow. For most locations,
the wind exceeds 1 m/s 80 to 90% of the time, but the average wind velocity is generally quite low
(3 to 4 m/s at 10 m above grade). Although low-rise houses are often protected from wind effects
(both by neighboring buildings and their location close to the ground), mid- and high-rise buildings
are usually fully exposed to the wind. Measurements on low-rise buildings (Straube and Burnett
1995) show that average wind pressures driving ventilation can be expected to be in the order of 1
Pascal. The average pressure will fall in a wide range between 0.1 and 10 Pascals, depending on
the geometry and size of the building, the location and distance between vents, and wind speed and
wind direction
Increasing temperature and/or water vapor content will decrease the density of air; these changes
in density generate buoyancy effects that can drive ventilation airflow. Measurements of solar
heating and outward heat flow in winter cause the air space of typical brick veneer walls to be an
average of at least 3 to 5 ºC above ambient over the entire year (Straube and Burnett 1997). Daily
variations of 10 to 30 ºC above ambient can be expected if the enclosure is exposed to the sun.
Thermal buoyancy pressures can be found from (Straube and Burnett 1995):
∆P = 3465 · ∆h · ( 1
T1
T
amb −− )(1)
where,
∆P is the pressure difference driving ventilation flow [Pa]
∆h is the difference in height between vents [m]
Tamb is the exterior ambient temperature [K]
T is the temperature in the air space [K]
Average pressures of the order of 1 Pascal can be expected due to the combined effects of
moisture and temperature buoyancy. Moisture buoyancy is a small (i.e., ∆P < 1 Pa) but sometimes
important contributor to ventilation pressures.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
8
Ventilation Flow
The amount of ventilation airflow can be found using standard fluid mechanics given the driving
pressures and the physical characteristics of the enclosure. There are two major flow-resisting
mechanisms: friction with the sides of the airflow path (the air space) and the restriction of air
flow through the vents.
The roughness of the air space sides is not very important to flow in most practical situations, but
the partial blockage of the air space by mortar fins, strapping, furring, bulging insulation, displaced
building paper, etc., can be very important. Wide air spaces are suggested as a practical means to
overcome these potential blockages. In wall systems with discrete vents (e.g., masonry veneers),
the vents themselves impose most of the resistance to airflow. Increasing the vent area results in a
directly proportional increase in the airflow through the airspace of such systems. European open-
jointed panel cladding systems generally permit an order of magnitude more air flow than typical
masonry veneer wall systems because of the large vent areas (more than 1% of wall area) and
clear cavities used in the former. Airflow through clear cavities of 12 mm (i.e., a commonly
specified dimension) behind drained-screened stucco and EIFS systems is expected to lie
somewhere between brick veneers and open-jointed cladding.
A review of the literature, simple calculations, and some field measurements of ventilation
pressures (Straube and Burnett 1995) confirm that the flow generated by typical driving pressures
(0.5 to 2 Pascals) can be expected to be in the order of 0.2 - 2 m
3/h per m
2 of cladding. (These
values naturally depend on the vent area and the depth and degree of blockage of the air space.)
Field measurements of well-vented wall systems (i.e., vent areas of more than 1% of wall area)
show that such systems typically experience flow velocities of 0.05 to 0.2 m/s (Jung 1985, Popp et
al. 1980, and Kuenzel et al. 1983). Schwarz (1973) and Uvsløkk (1988) both found higher
average velocities behind well-vented cladding panels with continuous slotted vents.
Although large vent areas are recommended to increase ventilation flow, it is presumed here that
the cladding is not part of the air barrier system. Most modern walls with drainage openings fit
this description, and larger vent areas will not compromise the air tightness of the wall system.
European codes are generally more specific regarding the size and location of vents and require
much higher vent areas than North American codes. Most of the relevant wall cavity ventilation
research has been conducted in Europe. Despite the extensive use of ventilated cladding systems
in Europe, the benefits, drawbacks, and mechanics of ventilation flow have not been clearly
defined. Moreover, very little work has been focused on masonry veneer wall systems.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
9
Predicting Ventilation Drying
Given a knowledge of the quantity and quality (i.e., temperature and moisture content) of the
ventilating air, the maximum drying capacity can be estimated. However, several simplifying
assumptions must be made:
1. the air in the space is well mixed, i.e., the moisture content is constant over the whole
air space;
2. the rate of drying is controlled by the rate of ventilation flow, not the rate of
evaporation from the material along the side of the air space; and
3. the drying process does not modify temperature conditions.
Field monitoring of various wall systems has shown that the first assumption is quite accurate
under most conditions. Because the vapor permeance of air is so high, it is difficult for large
gradients of air moisture content to form in clear air spaces. This assumption is no longer valid
under high flow conditions near the inlet vent because the rate of diffusive redistribution within the
space is less than the convective vapor flux under these conditions.
The second assumption is also valid as long as the ventilation flow rate is low and the sides of the
air space are wet (i.e., the rate of evaporation is greater than the ventilation drying rate). As the
moisture content of the material surfaces falls significantly below capillary saturation, this
assumption becomes progressively less accurate. However, materials that have a high moisture
diffusivity and vapor permeance fit this assumption well. It follows that calculations based on this
assumption are peak drying rates, or drying rates when the air space sides are saturated; it is
precisely these conditions that one is trying to alleviate by means of ventilation drying.
The validity of the third assumption depends on the drying rate. At low ventilation rates, the
specific heat capacity of air is too low to change the temperature conditions of the air-space air or
sides. At low drying rates, the amount of latent heat required to evaporate moisture is very small
and has little effect on temperatures. Very high drying rates, such as would occur during a sunny
period immediately after a rain event, might depress the temperature. This assumption limits the
accuracy of calculations to ventilation drying during extreme events, i.e., the third assumption is
valid most of the time.
In summary, the three assumptions listed above are valid for low ventilation flows (i.e., those
typically experienced) and airspaces that have wet materials (i.e., those walls that require drying).
Example Calculation
Consider a well built brick veneer wall system with a 50 mm air space and open head joint vents
spaced at 600 mm on center, both at the top and the bottom of the air space. Assume that a layer of
12.7 mm OSB (Oriented Strand Board) sheathing (density 700 kg/m3) has been saturated by

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
10
exfiltration condensation over the course of a winter.
If exterior conditions are 7 ºC and 85% RH (vapor pressure: 851 Pa), the outdoor contains about
6.6 g per m
3. If the sun shines on the wall, the air space temperature can rise to at least 20 ºC
above the outdoor temperature for 6 to 8 hours. The humidity in the air space will also be nearly
100%RH (as it must be if the materials lining the sides of the air space are saturated); air at 27 ºC
can store 25.8 g per m3. The difference of about 20 g per m3 is the amount that can be removed by
ventilation. As discussed earlier, ventilation flows of 0.2 to 2 m3/m2·h might be expected in such
a wall (Straube and Burnett 1995). This flow rate is so small that it generates flow velocities of
only 2.6 to 26 mm/s in a 2.4 m high air space. Over an 8 hour period at a flow rate of 1 m3/m2·h,
the moisture content of the materials lining the airspace could drop by as much as 160 g; this could
reduce the moisture content of the OSB by almost 2%. The temperature would also drop a few
degrees because of the heat of evaporation.
Diffusive drying of the sheathing can be calculated in a similar manner. If the sheathing is at 27 ºC
and 100%RH (3567 Pa), drying by diffusion would be:
(3567- 851)Pa × 46 ng/Pa·s·m2 × 3600 s/hr × 8 hrs = 3.6 g/m2
In this simple example, ventilation drying would remove more than 40 times the amount of stored
moisture as diffusive drying.
Equivalent Vapor Permeance
The simple example calculation shown above is still somewhat unwieldy, and does not permit the
easy comparison of the effects of ventilation. Although the tabulated values for cladding materials
such as metal and vinyl indicate these materials are perfect vapor barriers, cladding systems made
of these materials are clearly not vapor impermeable. For these types of cladding materials (vinyl
siding, metal panels), the satisfactory performance of wall assemblies can only be explained by the
ventilation, albeit exceedingly small, of the airspace, often through small unintentional openings. It
would also be useful to have a permeance value for ventilated brickwork that can be used in one-
dimensional calculations (such as those outlined in the Handbook of Fundamentals or in computer
models such as MOIST, MATCH, and WUFI).
Using the assumptions listed earlier, it is possible to determine the combined or effective vapor
permeance for an enclosure layer, which includes the effects of both diffusion and airflow. The
mass of water in air can be found from a form of the ideal gas law:
wv = PV
RT
v
v
⋅⋅
⋅⋅ (2)

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
11
where wv is the mass of water (kg),
Pv is the vapor pressure of water (Pa),
V is the volume of air (m3),
Rv is the gas constant for water vapour (461.5 J/kg·K), and
T is the temperature (K).
For a difference in vapor pressure, assuming well-mixed air in the air space and a small
temperature difference between the air streams, the mass of water transported by an air volume
exchange is:
∆wv =
∆
∆
PV
RT
v
v
⋅⋅
⋅⋅ (3)
If the temperature difference is not small and accuracy is important, Equation 2 would need to be
evaluated at each temperature.
The property that defines the amount of diffusive water vapor transport across a material layer is
called the water-vapor permeance. For a unit change in vapor pressure and volumetric flow rate,
Equation 3 yields a system property that can be considered to be the equivalent vapor permeance
of the cladding due to ventilation airflow. Using a parallel flow model, the combined effect of
diffusion and ventilation can be modeled as a combined equivalent permeance.
For a ventilation flow rate of 0.00028 m3/m2·s (1 m3/m2 ·hr), a vapor pressure difference of 1 Pa,
and a mean temperature of 15 ºC, the mass of water transferred will be:
∆ww = )15273(5.461 00028.01 +⋅
⋅
· 1012 ng/kg
= 2100 ng/Pa·s·m2.
This value of permeance is over 40 times that of a 90 mm brick masonry veneer -- this is the same
conclusion reached in the previous example. Such calculations indicate that, at the very least,
small rates of ventilation can play a very important role in bypassing the vapor resistance of the
cladding. Even with a ventilation rate of only 0.1 m3/m2 · hr, the transfer of vapor out of the cavity
by mass transport is likely to be four to five times greater than that by diffusion alone.
The air velocity in a cavity 2.5 m high and 50 mm deep necessary to generate 1 m
3/m2 · hr of
airflow is 0.014 m/s. Compare this velocity to the measured velocities (of 0.05 to 0.5 m/s)

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
12
referenced earlier. A velocity of 0.0014 m/s is so small that it is exceptionally difficult to
measure and the pressures necessary to generate this small flow rate are generally considered so
small as to be insignificant (i.e., ∆P << 1 Pascal). Ventilation drying may have been dismissed in
much of the literature because of the difficulty of measuring such small velocities (e.g., less than
about 0.2 m/s) and pressures. However, the preceding examples confirms that very small
ventilation rates can have a significant influence on the actual vapor permeance of the cladding
system, and thus on the drying performance of the wall assembly.
The concept of equivalent vapor permeance allows for a quantitative assessment of the importance
of ventilation airflow to drying. Equivalent conductances, or surface films, are widely used to
model convective and radiative heat transfer as conductive heat flow. The equivalent vapor
permeance allows convective vapor flow to be modeled as diffusive flow. The important role of
the sun and the wind is also explicitly incorporated in the assumptions that need to be made for
these calculations.
Note that the existing research into attic and crawlspace ventilation is not directly related to
enclosure wall ventilation. Attics have much greater air volumes, less rain penetration and
absorption, higher measured rates of ventilation, and tend to be affected by night-sky cooling.
Crawlspace ventilation flows are smaller than attic flows but the airspace temperature is not
increased by solar radiation. In fact, the hygrothermal state of crawlspaces is often greatly
influence by the ground conditions. Ventilation wetting is often possible in crawlspaces.
FIELD MONITORING
The Building Engineering Group (BEG) has been conducting full-scale field testing of enclosure
wall systems since 1988. The primary objective in most of the several projects completed has
been the study of hygrothermal performance of wall systems common to colder climates. In
particular, means of minimizing rain penetration, methods of ensuring drainage and predicting
driving rain, the mechanics of ventilation drying, and the drying of built-in moisture have been
studied.
To support the theoretical studies of ventilation drying, BEG conducted measurements of
ventilation pressures, wall temperatures, and air vapor content in full scale walls exposed to the
natural environment in several projects (Straube and Burnett 1995, Straube and Burnett 1997).
Well over 50 wall panels involving 25 different types of wall systems have been monitored. To
illustrate the role of ventilation drying, two different brick veneer wall systems will be examined.
Figure 4 presents a simplified horizontal cross section of each.
Test panels for walls A and B were built following current practice for masonry-clad, framed wall

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
13
systems in Canada. The brick veneer on these panels was built with great care to ensure that the
30 mm wide air space (slightly larger than the nominal 25 mm typically provided) was kept clear
of mortar dams, bridges, and droppings. Panel A employed mineral fiberboard insulation (48
kg/m3 density) on exterior gypsum sheathing. Panel B incorporated insulating Type III extruded
polystyrene sheathing and sheathing paper.
Common to all panels was an 85 mm clay brick veneer with open head joints at 600 mm on center,
top and bottom, 38x89 mm wood or steel framing (single top and bottom plates with studs at 400
mm on center) filled with low-density batt insulation, a 0.15 mm (6 mille) polyethylene vapor
retarder (M = 3.4 ng/Pa·s·m2) and painted gypsum board interior finish. The combination of
interior drywall and polyethylene was confirmed to be airtight by testing.
The gypsum sheathing in Wall A was vapor permeable (M > 2000 ng/Pa·s·m2). Wall B employed
an asphalt-impregnated perforated sheathing paper (M = 300 to 700 ng/Pa·s·m2 depending on RH)
and extruded polystyrene insulating sheathing (M = 40 ng/Pa·s·m2).
30 air space
50 mineral fibre board insulation
exterior gypsum sheathing
with sealed joints
30 air space
sheathing paper
32 extruded polystyrene
Wall B
Wall A
Figure 4: Simplified Wall Cross-Sections
Painted interior drywall and 6 mille poly were common to both systems.
One panel of wall type A was built and installed facing east. Four panels of wall type B were
built and one panel was installed facing north, south, east and west. The 1.2 m wide and 2.4 m
high full-scale panels were installed in the natural exposure and test facility, the Beghut, located on
the University of Waterloo campus in South-western Ontario, Canada.
Each panel was instrumented with 12 to 15 temperature sensors, 3 to 6 pairs of Delmhorst pins
near the center of the studs and plates (for measuring wood moisture content), and 4 to 6 relative
humidity transducers. A special base detail allowed cavity drainage to be intercepted and
measured. The panels were installed in July or August and exposed to the environment for more
than 24 months. The sensors were read every 5 minutes and average values were stored. The
interior conditions were maintained at 50±5% relative humidity and 21±1 ºC.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
14
Panel A was monitored for one year with its vents open, and for one summer with its vents sealed
air tight. Drainage from the weep holes was always intercepted, collected, and measured. The
four Wall B panels were continuously monitored over the same period.
Results
Figure 5 is a plot of the framing moisture content of each of the panels over a 12 month period.
The line labeled “vented” is for 1996, and that labeled “unvented” is for 1997. The plot of wall
panel B is for 1997 (1996 was essentially the same). During the summer of 1996 the moisture
content of the vented Wall A climbed to almost 15%. This is not a dangerous level, but clearly
shows that summertime wetting could occur in the Wall A. Wall B exhibited no such wetting
because of the relatively vapor-resistant exterior sheathing. It is clear that sealing the vent
openings on June 1, 1997 (Day 211) had a significant impact on the moisture content of Wall A.
0
5
10
15
20
0 90 180 270 360
Days From Nov. 1
Moisture Content (%)
Wall A - unvented
Wall B
Wall A- vented
Figure 5: Framing Moisture Content vs. Time

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
15
0
10
20
30
40
960711 960712 960713 960714 960715 960716 960717 960718
Date
Air Moisture Content (g/m3)
Air Space
Exterior Air
Studspace
Vented
0
10
20
30
40
970711 970712 970713 970714 970715 970716 970717 970718
Date
Air Moisture Content (g/m3)
Air Space
Exterior Air
Studspace
Unvented
Figure 6: Hourly Moisture Content of Wall A - Vented (top) vs. Unvented (bottom)

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
16
The temperature and relative humidity measurements were used to calculate the moisture content of
the air in the air space. Over the summer period, the average moisture content of the exterior air in
1996 was 9.6 g/m3. Over the summer period in 1997, the average exterior air content was 9.1
g/m3. The average moisture content of the air in the airspace of the well-vented Wall A was 10.9
g/m3: about 1.3 g/m3 higher than the exterior. During the following summer when Wall A was
unvented, the moisture content in the airspace was 13.1 g/m3, 4 g/m3 or 44% above that of the
exterior. The well-vented east-facing Wall B exhibited airspace moisture content of 1.0 g/m3
(11%) above the exterior during the same period. These average values suggest that venting the
airspace had a significant effect on the moisture content of the air in the air space.
The moisture content of the air was also examined on an hourly basis. Figure 6 compares the
moisture content of the air in the studspace, air space, and exterior during a typical week during
July for Wall A when it was vented and when it was unvented. The moisture content in the air
space is clearly much more closely coupled to the exterior in the vented case than the unvented
case. Water vapor that is driven off hygroscopic materials, especially the brickwork, by solar
heating enters the air space air but is unable to leave by ventilation in the unvented wall.
In the unvented case, the influence of the high air moisture content is reflected in the studspace of
the wall because of the vapor-permeable sheathing. The high moisture content level of the air
space occasionally resulted in condensation on the polyethylene. This condensation and the
prolonged high relative humidity resulted in the observed higher moisture content.
The importance of the temperature difference between the outside air and the air space air was
discussed earlier. Figure 7 presents a plot of the relative distribution of hourly average exterior
air temperature and cladding temperature for a three month summer period (a total of about 2800
data points). Because of solar heating, the brickwork temperature is higher on average (by almost
7 ºC) and more variable than the exterior. The temperature difference between the cladding and
the ambient air significantly increases the potential for drying from both faces of the cladding.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
17
0.00
0.05
0.10
0.15
0.20
0 10 20 30 40 50
Temperature (C)
Relative Frequency
Brickwork
Exterior Air 23.5
16.6
Note: East-facing
red brickwork
Average
Figure 7: Distribution of Hourly Average Brick Veneer and Exterior Air Temperature over
Summer Period
It is the temperature difference between the air space and the exterior air that is of particular
interest for ventilation drying. Figure 8 compares the distribution of this temperature difference
for the air space in Wall A when the wall was vented and unvented. Two important points are
illustrated by this data. First, the average temperature in the air space is considerably higher than
the exterior air temperature (by about 5 ºC). Secondly, the difference between the air space
temperature of the vented and unvented wall configuration is practically negligible. Hence, the
assumption that ventilation air flow will not cool the cladding is valid for a well-vented brick
veneer. Because the vapor carrying capacity of air is non-linearly related to temperature, the 10%
of the time that the air space is more than 12 ºC above the exterior has a disproportionate effect on
the drying potential.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
18
0.00
0.05
0.10
0.15
0.20
0.25
0.30
-5 0 5 10 15 20 25 30
Temperature (C)
Relative Frequency
Vented
Unvented 5.0
5.4
Note: Airspace behind
east-facing red brickwork
Average
Figure 8: Distribution of the Difference between Air Space and Exterior Air Temperature of
Vented and Unvented Walls
INWARD VAPOR DIFFUSION
The wetting exhibited by Wall A is an example of warm weather, solar-driven condensation
caused by the evaporation and inward vapor transport of moisture stored in the enclosure.
Although widely recognized in the research community (Wilson 1963, Sandin 1991, Hens 1995),
the control of inward vapor transport through enclosures in cold and temperate climates is rarely
considered by most design professionals. Ventilation appears to be one method of controlling this
type of wetting; another is the use of more vapor resistant layers (the approach of Wall B). Both of
these methods are used (perhaps unwittingly) in many walls and are the reason that inward
diffusion is not often seen as a serious problem. However, if ventilation is restricted and the wall
has very little vapor resistance between the cladding and an inner vapor retarding layer (e.g.,
poly), problems can arise.
We are of the opinion that some moisture problems ascribed to water penetration may, in fact, be
due to inward vapor drive condensation (e.g., “water penetration” of vapor permeable housewraps
may be exacerbated by inward drives). Some professionals are advocating the use of heavier, less
permeable building papers (e.g., 30 pound felt) or multiple layers of the same. These building

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
19
papers often have sufficient vapor resistance to limit or avoid warm weather condensation but may
reduce outward drying of moisture from other wetting mechanisms. Ventilation can also be used to
control inward vapor drives by allowing the vapor to escape to the exterior by convection.
CONCLUSIONS
It should be evident that drainage is a necessary water removal mechanism in screened walls.
However, because a significant quantity of moisture can be stored in hygroscopic materials,
drainage may not be sufficient to remove all potentially damaging moisture from rain penetration
and exfiltration condensation. Diffusive drying and ventilation drying are therefore important
mechanisms for removing moisture that is inevitably stored in a drained-screened wall assembly.
Ventilation drying can theoretically bypass the vapor resistance of the cladding and thereby
improve enclosure drying. The role of the sun and wind must be accounted for in any realistic
assessment of ventilation drying. Although ventilation flow rates are very small and difficult to
measure, field measurements confirm the promise of ventilation drying. Much more fieldwork
needs to be conducted employing carefully designed experiments.
The interaction of ventilation and inward vapor drives was also demonstrated. Since diffusive
drying to the inside can be important, the current practice of installing very low permeance vapor
diffusion retarders needs to be questioned. For many climates and many types of wall systems,
wintertime wetting by diffusion is an insignificant wetting mechanism.
ACKNOWLEDGMENTS
This research is part of the In-Service Performance of Enclosure Walls Project. The support,
technical input and funding of the seven corporate partners (Brampton Brick, Canada Brick,
Celfortec, Durisol Materials, Owens-Corning Canada, Roxul Insulations, and Sto Corporation),
has been invaluable and is gratefully acknowledged. The Ontario Government provided funding
through the University Research Initiative Fund. The Vents and Ventilation Drying Project was
supported by the Canada Mortgage and Housing Corporation. The assistance and technical review
of CMHC’s project manager, Pierre-Michel Busque was most helpful.
REFERENCES
ASHRAE, 1997. 1997 ASHRAE Handbook - Fundamentals, Atlanta: American Society of Heating
Refrigerating, and Air-Conditioning Engineers, Inc.
Ashton, H.E., CBD 124 Biological Attack on Organic Materials, NRCC, Ottawa, April, 1970.
Baker, M.C., CBD 111 Decay of Wood, NRCC, Ottawa, March, 1969.
Hens, H. and Fatin, A., 1995. Heat-Air-Moisture Design of Masonry Cavity Walls: Theoretical
and Experimental Results and Practice. ASHRAE Transactions, CH-95-3-2, pp. 607-626.

From : Proc. Of Thermal Performance of Exterior Envelopes of Buildings VII, 1998, pp. 189-198.
J.F. Straube and E.F.P. Burnett jfstraube@uwaterloo.ca
20
Jung, E. "Dauerstandverhalten von Verblendziegelmauerwerk unter Witterunsbeansphruchung und
Auswirkungen von Kerndämm-Maßnahmen, " Baustoffindustrie, No. 6, November, 1985, pp. 185-
188.
Künzel, H.M., 1997. Material Databank of WUFI Version 2.0 Hygrothermal Modeling Software,
collected by Fraunhofer Institut for Building Physics, Holzkirchen, Germany.
Künzel, H., Mayer, E., 1983. Untersuchung über die notwendige Hinterlüftung an
Außenwandbekeidung aus großformatigen Bauteilen. Schriftenreihe 3/1983: Bundesminister für
Raumordnung, Bauwesen, und Städtebau , Bonn. Germany.
Popp, W., Mayer, E., Künzel, H., 1980. Untersuchungen über die Belüftung des Luftraumes
hinter vorgesetzten Fassadenbekleidung aus kleinformatigen Elementen. Forschungsbericht B
Ho 22/80: Fraunhofer Institut für Bauphysik, , Holzkirchen, Germany.
Sandin, K., 1991. Skalmurskonstruktionens fukt- och temperaturbetingelser. Rapport R43:1991
Byggforskningsrådet, Stockholm, Sweden.
Schwarz, B., 1973. Witterungsbeansphruchung von Hochhausfassaden. HLH Bd. 24, Nr. 12, pp.
376-384.
Sereda, P.J., CBD 170 Atmospheric Corrosion of Metals, NRCC, Ottawa, 1975.
Straube, J.F. Burnett, E.F.P., 1997. In-Service Performance of Enclosure Walls. Volume 1:
Summary Final Report. Building Engineering Research Report: University of Waterloo.
Straube, J.F. Burnett, E.F.P., 1995. Vents, Ventilation Drying, and Pressure Moderation.
Building Engineering Research Report for Canada Mortgage and Housing Corp.: Ottawa.
Straube, J.F., 1998. Moisture Control in Enclosure Walls, Ph.D. thesis. Waterloo, Ont. Canada:
University of Waterloo.
Uvsløkk, S., 1988. The Importance of Wind Barriers for Insulated Wood Frame Constructions.
Proc. of Symposium and Day of Building Physics, Lund University, August 24-27, 1987,
Swedish Council for Building Research, pp. 262-267.
Wilson, A.G., "Condensation in Insulated Masonry Walls in the Summer", Proc. Of RILEM/CIB
Symposium, Helsinki, 1965, pp. 2-7.