Resources

Radon-Resistant Concrete Masonry Foundation Walls

INTRODUCTION

The U.S. Environmental Protection Agency (EPA) estimates that nearly 1 of 15 homes in the United States has elevated indoor levels of radon, the radioactive soil gas which is the second-leading cause of lung cancer (ref. 1). Fortunately, straightforward techniques exist to effectively reduce indoor radon levels.

Four factors contribute to radon entry into buildings: uranium, present in soils throughout the United States; soil permeability, which allows the radon to travel through the soil to the building’s foundation; pathways for radon entry, including cracks and plumbing penetrations; and lower air pressure inside the building, which draws radon inside. Radon-resistant construction techniques focus on controlling the pressure difference between the soil and indoor environment and on minimizing and sealing cracks and penetrations.

Concrete masonry’s versatility and inherent strength and durability characteristics make it especially well-suited for foundation walls . To adequately resist soil gas entry, concrete masonry walls must be designed and constructed to minimize cracking. References 2 through 5 provide excellent information on proper mortar joints, footing construction, wall bracing, backfilling, anchorage, waterproofing and structural design.

This TEK provides basic design guidance specific to radon resistant concrete masonry basement and crawl space construction based on Appendix F of the International Residential Code (IRC) (ref. 6) and EPA guidelines. Although these recommendations have been developed primarily for new low-rise residential construction, the same basic principles also apply to schools and other large buildings.

POTENTIAL FOR ELEVATED RADON LEVELS

It is most cost-effective to install radon-resistant features in homes with the greatest potential for high indoor radon levels. As a tool to help predict where this might occur, the EPA and U.S. Geological Survey have identified areas with high radon potential, based on indoor radon measurements, local geology and population densities (see http://www.epa. gov/radon/zonemap.html). State radon offices should also be consulted for more detailed local information. The EPA does not recommend soil testing as a method to predict the potential for elevated indoor radon levels.

The map assigns a zone number to each U. S. county based on potential indoor radon levels, as follows:

  • Zone 1: high potential (indoor levels greater than 4 pCi/L)
  • Zone 2: moderate potential (from 2 to 4 pCi/L)
  • Zone 3: low potential (less than 2 pCi/L)

BUILDING CODE REQUIREMENTS

The 2003 IRC Appendix F contains radon control methods for new home construction. Note that because these requirements are contained in an appendix, rather than in the body of the code, they become part of the local building code only when Appendix F is specifically adopted by the jurisdiction. The requirements in Appendix F are intended to apply to construction in Zone 1, based on the EPA map described above.

In addition to incorporating radon reduction methods into buildings in Zone 1, or where recommended by state radon offices, the EPA encourages builders to include these techniques in other areas, because high indoor radon levels could potentially occur in any area of the United States, and the installation of a passive venting system is very economical during initial construction, but becomes much more difficult as a post-construction mitigation technique

CONSTRUCTION REQUIREMENTS AND RECOMMENDATIONS

The basic approach to radon control in new buildings incorporates: soil depressurization to vent soil gases outside the building; designing the building’s heating, cooling and ventilation system to provide a slight positive pressure to prevent radon from being drawn inside; and sealing major radon entry routes. Note that several of these requirements are required or considered good practice for issues such as water penetration resistance and/or energy efficiency.

Sealing large openings and preventing large cracks is a key component of building radon-resistant foundations. However, field research has shown that attempting to seal all foundation openings is neither practical nor effective as a stand-alone radon prevention technique (ref. 1).

IRC Appendix F requires installation of a passive depressurization system for the soil beneath a home in Zone 1. This passive depressurization has been found to effectively reduce indoor radon levels by about half and, in most cases, to levels below the EPA action level of 4 pCi/L (ref. 1). Should high indoor radon levels exist with the passive depressurization system in place, the system can easily be upgraded to include an in-line fan to actively draw radon away from the foundation. The IRC includes several requirements that facilitate this possible upgrade. Although not required by the IRC, the EPA recommends that all homes be tested for radon after occupancy, and that mitigation measures be taken with readings at or above 4 pCi/L.

Note that for schools and other large buildings, EPA recommends installing an active depressurization system (i.e., with operational in line fan) during initial construction (ref. 7).

Basement Foundations

Figure 1 illustrates the requirements found in IRC Appendix F for installation of a passive sub-slab venting system for a residential basement. Note that these requirements apply to slab-on-grade foundations as well (see Figure 2).

The system is built on a layer of gas-permeable material, which allows the entire sub-slab area to be vented. The gas-permeable material should be placed under all floors that are in contact with the ground and are within the occupied spaces of the building. The soil gas-retarder membrane serves two purposes: to bridge any cracks that may occur in the slab, thereby preventing radon migration up through the slab at these points; and to prevent concrete as it is being placed from filling the voids in the gas-permeable material. Separate sheets should be lapped at least 12 in. (305 mm), but need not be sealed. In addition, the sheeting should fi t closely around pipes, wires and other penetrations.

The vent pipe extends from the gas-permeable layer to an exhaust point above the roof, allowing any radon that collects below the slab to be removed before it is pulled into the building. IRC Appendix F specifies a minimum vent pipe diameter of 3 in. (76 mm), although the EPA recommends a 4-in. (102-mm) vent pipe diameter to provide better passive venting (ref. 1). As an alternative to the vent pipe location shown in Figure 1, the IRC allows the vent pipe to be inserted directly into an interior perimeter drain tile loop or through a sealed sump cover, provided the sump is exposed to the gas permeable layer or connected to it through a drainage system.

Where interior footings or other barriers interrupt the gas-permeable membrane, each area should have its own vent pipe. These individual vent pipes can be run and exhausted separately, or they can be connected to a single exhaust pipe.

Although the IRC does not stipulate where or how the vent pipe should be run, the EPA offers the following guidance with the goal of inducing a natural upward draft in the vent pipe. The vent pipe should ideally have a vertical run, or, if this is not possible, elbows should be minimized, as these will restrict air flow. In cold climates, the vent pipe should be run in an interior wall. This will help keep the vent pipe warm, enhancing the natural stack effect. Locating the vent pipe in a cold exterior wall could hence make it less effective. In hot climates, the effectiveness of the passive stack depends more on wind, a hot attic and sun heating the pipe (ref. 1). For shallow roofs in hot climates, a higher exhaust point may improve the passive draw through the vent pipe.

The exhaust location requirements listed in Figure 1 help ensure that exhausted radon is not drawn back into the building, or an adjacent building, through a window or other opening.

The solid masonry top course helps resist radon entry from the cells of the masonry units into the habitable space above. This can be accomplished by using 100% solid units or by fully grouting the top course (mesh or other grout-stop device is installed below the course to contain the grout to the top course). Although not required by the IRC, full mortar head joints in this course will also help provide a continuous barrier. When a brick ledge is used, the solid course should be immediately below the brick ledge. As with other parts of the foundation, below grade penetrations and openings must be filled with polyurethane caulk or equivalent.

Several accommodations must be made for possible future fan installation, to convert to an active depressurization system in the event that high radon levels are recorded after occupancy. The first, that the vent pipe remain accessible through the attic or another area outside the habitable space (unless an approved roof-top electrical supply is provided for future use), ensures access for installing an in line fan. The second requirement is for installation of an electrical circuit box at the anticipated vent pipe fan location, typically in the attic, as well as in anticipated locations of system failure alarms.

In addition, IRC Appendix F requires:

  • all condensate drains must to be trapped or routed through nonperforated pipe to daylight,
  • sump pits to be covered with a sealed lid,
  • sumps used as a floor drain to have a lid with a trapped inlet,
  • ducts passing through or below the slab to be seamless (unless the air-handling system is designed to maintain continuous positive pressure in the ducts),
  • joints in ducts passing through or below the slab to be sealed,
  • homes to be constructed to minimize building depressurization, as otherwise required in Section M1601, Chapter 11 and R602.8.

Crawl Space Foundations

Crawl space foundations require preventive measures similar to those for basements, or must be provided with an approved mechanical crawl space ventilation system.

The IRC requirements for passive depressurization in a crawl space are illustrated in Figure 3. The major differences between the crawl space and basement requirements are:

  • a requirement for natural ventilation per IRC Section R408,
  • without a slab, it is difficult, if not impossible, to seal radon out at the crawl space floor, so sealing takes place at any penetrations through floors above crawl spaces as well as at access doors and other openings or penetrations between the crawl space and an adjoining basement,
  • air-handling units located in crawl spaces must be sealed to prevent air from being drawn into the unit, and
  • ducts in crawl spaces must have seams and joints sealed.

Combination Foundations

Buildings with combination basement/crawl space or slab ongrade/crawl space foundations are required to have a separate radon vent pipe for each foundation area. The vent pipes can either be connected to a single vent that exhausts through the roof, or each can be exhausted separately (ref. 6).

In addition, the EPA recommends that special care be taken at points where the different foundation types meet, because additional soil gas entry routes typically exist at these locations.

REFERENCES

  1. Building Radon Out, EPA/402-K-01-002. Environmental Protection Agency, 2001.
  2. Concrete Masonry Basement Wall Construction, TEK 03-11, Concrete Masonry & Hardscapes Association, 2001.
  3. Allowable Stress Design of Concrete Masonry Foundation Walls, TEK 05-01B, Concrete Masonry & Hardscapes Association, 2001.
  4. Strength Design of Reinforced Concrete Masonry Foundation Walls, TEK 15-02B, Concrete Masonry & Hardscapes Association, 2004.
  5. Preventing Water Penetration in Below-Grade Concrete Masonry Walls, TEK 19-03B, Concrete Masonry & Hardscapes Association, 2012.
  6. International Residential Code for One- and Two-Family Dwellings. International Code Council, 2003.
  7. Radon Prevention in the Design and Construction of Schools and Other Large Buildings, EPA 625/R-92/016. U. S. Environ-mental Protection Agency, 1994.

Control of Air Leakage in Concrete Masonry Walls

INTRODUCTION

Energy efficiency in buildings has become increasingly important. Whether complying with newer energy codes or gaining recognition for sustainable building practices, reducing the overall energy usage in new and existing buildings continues to be a leading consideration for design teams.

Many methods are employed to increase building energy efficiency. One consideration is reducing air leakage through the building envelope. In addition to the negative impact on a building’s energy efficiency (due to the loss of conditioned air via exfiltration and/or the introduction of unconditioned air via infiltration), air leakage in buildings can also impact moisture control, indoor air quality, acoustics and occupant comfort.

Reduced air leakage is one area where masonry walls excel compared to other wall types when proper design criteria are applied. This TEK reviews available information on masonry wall air leakage, reviews the most recent code criteria, presents concrete masonry wall assemblies that meet this criteria, and provides general guidance on improving the control of air leakage in masonry walls.

AIR LEAKAGE

Air leakage consists of air infiltration from the exterior into the conditioned spaces of buildings and/or exfiltration of conditioned interior air out of buildings. Although under a pressure differential air can pass directly through many materials, air leakage occurs primarily through a myriad of cracks, gaps, improperly designed or constructed joints, utility penetrations, junctions between wall and window and door frames, junctions between wall and roof assemblies, and other avenues.

Historically, air leakage has been the primary source of building ventilation. Because it is uncontrolled and weather-dependent, however, the direct result of air leakage is an increase of energy consumption to maintain space conditioning. Recognition of this increased energy consumption has caused air leakage to be regulated by code for many newer commercial buildings.

Reducing air leakage rates, however, can pose potentially adverse health effects due to stale and polluted air by reducing the air exchanges that dilute contaminants. Mechanical ventilation systems are usually required to satisfy air exchange requirements that have historically been met by uncontrolled air leakage. Although there is an added cost with a designed mechanical ventilation system, it is theoretically offset by the energy savings associated with the reduced air leakage. Heat recovery or energy recovery units (HRV/ERV) can be used to reduce the amount of space conditioning required to condition the fresh air. These systems should be designed carefully, however, as some research shows that the energy consumed by operating the HRV/ERV systems could exceed the cost of conditioning the fresh air (ref. 1).

Studies have shown that air leakage in buildings can be difficult to accurately predict and measure (ref. 2). Prediction and measurement of air leakage rates in walls has been the subject of study by both U.S. and international researchers. U.S. results have focused primarily on the wood stud wall construction with fibrous insulation common to home building. International research has looked at masonry walls as well as wood frame walls, because masonry is the traditional European construction method.

AIR LEAKAGE LOCATIONS

A key issue when addressing air leakage is the significant difference between air leakage at discreet sites, such as at member junctions and at door and window openings where caulking and sealing is at issue, versus the diffuse air leakage that can occur directly through a wall assembly. Chapter 16 of the ASHRAE Fundamentals Handbook (ref. 3) includes the results of residential air leakage studies that show that the largest source of air leakage occurs through wall cracks, joints and utility penetrations. Other major leakage sources include leakage around doors and windows, ceiling penetrations and utility penetrations to the attic, and the HVAC system. The same studies showed that diffusion through walls was less than 1%; i.e., compared to infiltration through holes and other openings, diffusion through walls was not an important flow mechanism in residential buildings. These data are illustrated in Figure 1.

AIR LEAKAGE CRITERIA

To reduce air leakage rates, air barrier systems are sometimes designed and installed as part of the building envelope. Alternatively, the thermal envelope can be designed and detailed to perform as an air barrier system. Current building codes (ref. 4) do not stipulate quantitative requirements for air barriers, but instead require that the exterior envelope be sealed to minimize the infiltration/exfiltration of air through both commercial and residential building envelopes.

The 2012 International Energy Conservation Code (IECC) (ref. 5) and some local jurisdictions, however, have adopted performance requirements for the control of air leakage in commercial buildings. The 2012 IECC provides three levels of compliance, applying to air barrier materials, air barrier assemblies, or the whole building. These commercial air barrier criteria apply only to buildings in Climate Zones 4 through 8. The compliance criteria are (only one of these criteria need to be satisfied):

  • a building material intended to serve as an air barrier must have an air permeance of less than 0.004 cfm/ft2 at a pressure differential of 1.57 lb/ft2 (0.02 L/s-m2 at 75 Pa),
  • an assembly of materials intended to serve as an air barrier, such as a concrete masonry wall assembly, must have an air leakage rate of less than 0.04 cfm/ft2 at a pressure differential of 1.57 lb/ft2 (0.2 L/s-m2 at 75 Pa), or
  • a building must have an air leakage rate of less than 0.4 cfm/ft2 at a pressure differential of 1.57 lb/ft2 (2.0 L/s-m2 at 75 Pa).

Also contained within the code are several “deemed-to-comply” materials and assemblies. The following masonry-related materials and assemblies are included in this list and are therefore considered to comply with the code:

  • fully grouted concrete masonry (although listed as a material, this compliance option is more accurately deemed an assembly),
  • as a material, portland cement/sand parge or gypsum plaster with a minimum thickness of 5 /8 in. (16 mm),
  • as an assembly, portland cement/sand parge, stucco or plaster with a minimum thickness of 1 /2 in. (13 mm), and
  • concrete masonry walls coated with one application of block filler and two applications of a paint or sealer coating.

The last option is justified based on research completed in the early 2000s. More recent research has documented additional options for materials and coatings to allow concrete masonry assemblies to comply with the maximum assembly air leakage requirement of 0.04 cfm/ft2 at a pressure differential of 1.57 lb/ft2 (0.2 L/s-m2 at 75 Pa). Although not included explicitly in the code, these tested assemblies can be approved under IECC Section 102, Alternate Materials, as meeting the intent of the code. The testing is described in the Masonry Wall Assemblies section below, and the results are summarized in the Guidelines section on page 7.

The 2012 IECC also lists the following materials as acceptable air barrier materials (ref. 5). Any one of these can be used in conjunction with concrete masonry construction, as shown in Figures 2 and 3.

  • extruded polystyrene insulation board with a minimum thickness of 1/2 in. (13 mm) with joints sealed,
  • foil-backed polyisocyanurate insulation board with a minimum thickness of 1/2 in. (13 mm) with joints sealed,
  • closed-cell spray foam insulation with a minimum density of 1.5 pcf (2.4 kg/m3) with a minimum thickness of 1 1/2 in. (36 mm),
  • open-cell spray foam insulation with a density between 0.4 and 1.5 pcf (0.6 – 2.4 kg/m3) with a minimum thickness of 4 1/2 in. (114 mm), and
  • gypsum wallboard with a minimum thickness of 1/2 in. (13 mm) with joints sealed.

MASONRY WALL ASSEMBLIES

Multi-Wythe Walls

Multi-wythe concrete masonry assemblies have a variety of options available for compliance with the commercial building air leakage requirements listed above. In addition to the deemed-to-comply options, there are many proprietary air barrier materials and accessories available. Most air barrier materials are some type of coating, which is usually applied to the cavity side of the back up wythe. In addition, some types of spray-applied insulation or rigid insulation (with sealed joints) can be used as an air barrier, as illustrated in Figure 2.

Single Wythe Walls

The available options for single-wythe concrete masonry assemblies are illustrated in Figure 3. Solid grouting is available, as well as coating with a paint, sealer, or block filler. Additionally, exterior wallcoverings and interior wall finishes offer solutions, such as parge coating, stucco, plaster, various insulations and gypsum wallboard. Note that paints, sealers or block fillers are effective when applied to either the interior or exterior surface of the concrete masonry. Hence, when a coating is specified, architectural finishes need not be compromised by the coating.

Concrete Masonry Air Leakage Testing

Research sponsored by CMHA and the NCMA Education and Research Foundation (refs. 6, 7) has documented additional concrete masonry wall assemblies that can meet the air barrier assembly requirements of 0.04 cfm/ft2 at a pressure differential of 1.57 lb/ft2 (0.2 L/s-m2 at 75 Pa). The results are summarized below. See References 6 and 7 for full descriptions of the assemblies and test results.

Commercial-Grade Latex Paint

One project (ref. 6) tested the effects of commercial-grade latex paint on the air leakage rate of concrete masonry wall assemblies. The walls were ungrouted except at the four edges (which were grouted solid to isolate air permeance to a 1 m2 test surface). The research employed a modified ASTM E2178, Standard Test Method for Air Permeance of Building Materials (ref. 8), because there is no standardized test procedure specifically suited for testing concrete masonry assemblies. Three wall sets were built using plain gray concrete masonry units, each with different concrete mix designs, then tested for air leakage.

The wall sections were painted with a typical commercial-grade latex paint (28% solids content by volume), then the air leakage rate was re-measured. The research documented that the air leakage rate decreased as the paint thickness increased: it was determined that the air leakage rate of the wall was inversely proportional to the thickness of the paint applied.

While surface texture was not directly measured in this study, it is believed that the surface texture of smooth-faced concrete masonry units affects the ability of the coating material to develop a continuous coating, which is important for reducing air leakage rates through assemblies.

The results of this research indicate that the air leakage rate of 12-in. (305-mm) concrete masonry walls can be reduced to 0.04 cfm/ft2 or less at a pressure differential of 1.57 lb/ft2 (0.20 L/s-m2 at 75 Pa) by applying between 3.3 and 14.6 mils (0.084 and 0.371 mm) of commercial-grade latex paint for concrete masonry units with a smooth textured surface and a coarse textured surface, respectively.

High-Quality Latex Paint

More recent research (ref. 7) evaluated the effects of four additional coatings: a high-quality latex paint, masonry block filler, water repellent surface coatings, and gypsum wallboard. The concrete masonry units used in this study were also plain gray, medium-weight “utility” type units with a fairly open surface texture (see Figure 4). The use of integral water repellent admixtures was also investigated.

The latex paint used in this project was a high quality retail paint, with a 28% solids content by volume and 47% solids content by weight. To evaluate this paint, a single coat was applied with an average dry film thickness of 1.28 mil (0.033 mm). The paint reduced the air leakage rate by 94%, to a calculated average air leakage rate of 0.011 cfm/ft2 (0.05 L/s-m2), well below the assembly requirement of 0.04 cfm/ft2 (0.2 L/s-m2).

The results indicate that when a high quality latex paint is used, a single coat is all that is necessary to create a continuous coating and provide the required barrier to air flow.

Masonry Block Filler

The block filler evaluated was a water-based masonry primer designed for use on concrete and concrete masonry surfaces. This material is typically used as a base primer coat on concrete and masonry surfaces in preparation for painting. It is a thicker coating material than latex paint, designed to fill pores and surface imperfections in masonry walls. Based on information provided by the manufacturer, this material has a 46% solids content by volume and 55% solids content by weight.

A single coat of block filler was applied with an average dry film thickness of 2.10 mil (0.053 mm). The air leakage rate was reduced by 86% due to the presence of the block filler coating, to 0.011 cfm/ft2 (0.05 L/s-m2). This result is well below the air barrier assembly requirement of 0.04 cfm/ft2 (0.2 L/s-m2).

Gypsum Wallboard

A set of assemblies was also evaluated for air leakage after installing ½ in. (12.7 mm) gypsum wallboard to simulate a single-wythe assembly with a drywall-finished interior.

When the gypsum wallboard was tested by itself, it had an air permeance below the air barrier material requirement of 0.004 cfm/ft2 (0.02 L/s-m2). When the concrete masonry assembly was tested with wallboard attached, it was evident that the performance of the assembly was dominated by the air permeance of the wallboard, as very little air leakage was measured, and the results were below the 0.004 cfm/ft2 (0.02 L/s-m2) requirement for an air barrier material.

Water-Repellent Surface Coatings

Because many single-wythe concrete masonry assemblies use some type of water repellent surface coating, these coatings may be an efficient way to reduce air leakage rates. Both a silane/siloxane and an acrylic microemulsion water repellent coating were evaluated.

While both water repellent coatings reduced the air leakage rate of the assemblies, the reduction was not sufficient to comply with the 2012 IECC air barrier assembly requirements for commercial buildings.

Integral Water Repellents

The effect of an integral water repellent in concrete masonry units and masonry mortar was also evaluated. Integral water repellents in concrete masonry units can improve the compaction of the unit, leading to a slightly tighter concrete matrix and, in some cases, a more uniform surface texture.

The tested set of concrete masonry assemblies contained an integral water repellent admixture at an appropriate dosage to produce water repellent characteristics.

Compared to the assemblies without an integral water repellent, the addition of integral water repellent decreased the air leakage rate by 28% on average. This decrease is likely due to a slightly tighter pore structure resulting from the use of the integral water repellent. The decrease in leakage rate, however, was not sufficient to reduce the assembly air leakage rate to levels that comply with the 2012 IECC.

CONCRETE MASONRY COMPARED TO FRAME CONSTRUCTION

Typical masonry construction does not include some of the leakage sites common in frame walls. Masonry walls do not have sole plates (sills), because the wall is a continuous assembly from the footing up. The top of a masonry wall is typically a tie-beam or bond beam. Trusses or rafters are set to a plate attached to the top course of masonry. Quality caulking and sealing are important at the ceiling finish edge. Sealing is also required at attic access ways, as well as around any wall penetrations.

Commercial Buildings

Measured air leakage rates from existing commercial buildings constructed during or after 1980 have been compiled (ref. 9). From this data, 84% of the masonry buildings included had measured whole-building air leakage rates of less than 2 cfm/ft2 at a pressure differential of 1.57 lb/ft2 (10 L/s-m2at 75 Pa). In comparison, only 30% of frame-walled buildings had measured whole building air leakage rates of less than 2 cfm/ft2 at a pressure differential of 1.57 lb/ft2 (10 L/s-m2 at 75 Pa) (it should be noted that none of these buildings were constructed to meet an air tightness standard). The reported leakage rates were normalized by the above grade area of the building envelope. The data were compiled from various references, and represent a range of climates and building types, making it difficult to draw definite conclusions. The results do indicate, however, that existing masonry buildings tend to have much lower air leakage rates than existing frame-walled buildings.

Residential Buildings

The air leakage rates of masonry walls have also been researched widely in Europe by such groups as the Air Ventilation and Infiltration Center in England (ref. 10). Results from detailed air leakage work performed in Finland show that concrete masonry and lightweight concrete (panelized) walled homes had much lower air leakage rates than wood frame structures (ref. 11). Figure 5 illustrates these differences, comparing older wood frame houses averaging 7.3 air changes per hour (ACH) at 50 Pa to more modern site-built wood frame houses averaging 8.5 ACH, with a very wide range of values. Prefabricated wood element (panelized) houses were better at 6.0 ACH. Both concrete masonry and lightweight concrete houses, however, had roughly one-half the air change rate of the average panelized wood frame homes.

Proper sealing of components into masonry rough openings may be more important than reducing air leakage through masonry assemblies. Dr. Hiroshi Yoshino of Japan’s Tohoku University investigated Japanese housing air leakage (ref. 12) in a broad comparison with data from other nations. He ranked data points from his own research and other investigators into air tightness categories. He observed that some concrete multi-family housing was so air tight that indoor air quality and condensation problems resulted, and ventilation was required. Concrete masonry houses of “air-tight” construction ranked among the best in Japan for air tightness. Several of the other Japanese reports he cited also showed concrete and concrete masonry houses to have lower air leakage rates than typical Japanese frame houses.

Belgian researchers used a sequential technique in masonry homes to examine incremental air leakage measures (ref. 14). Figure 6 shows the progression of air change rates at 50 Pa from “normal construction,” which evidently assumes no air leakage reduction measures, to a masonry wall with all windows, doors and penetrations sealed and weather stripped. Sealing just these items resulted in about 87% less air leakage. The largest improvements are seen after sealing the door and window frames to their respective rough openings, which agrees with the data in ASHRAE (ref. 3). The Belgian findings also agree with a statement in a compendium of European air leakage results which states: “The critical details from the point of view of air-tightness are associated with the (quality of) formation of openings in masonry walls…” (ref. 14).

IMPACTS ON MOISTURE

When an air barrier material is required, its placement can be critical to controlling moisture and hence to wall durability.

First, because air movement can carry a significant amount of moisture into or through a building assembly, and second because the air barrier can act as a vapor retarder. Note that an air barrier is designed to control the movement of air both into and out of the building envelope, whereas a vapor retarder is designed to restrict the diffusion of water vapor through building materials and subsequent condensation. Because a vapor retarder can also inhibit drying, the need for a vapor retarder varies with climate, construction type and building use.

Although the functions of air barriers and vapor retarders differ, in some cases one component can serve both needs. In designs where one material is installed to control both air and water vapor movement, it is important that the material is continuous to provide the required level of air tightness. Where separate airflow and vapor retarders are installed, care must be taken to ensure that the air barrier cannot cause moisture condensation. This can be accomplished through the choice of vapor-permeable materials or through proper placement.

More detailed information on vapor retarders in concrete masonry walls can be found in TEK 06-17B, Condensation Control in Concrete Masonry Walls (ref. 13).

DISCUSSION

Air leakage measurements indicate that properly constructed concrete masonry walls may have better natural resistance to air leakage than typical frame construction. If a further reduction in air leakage rates is required, various options are available. Retrofits for reducing air leakage in concrete masonry construction are straightforward, because fewer dissimilar joints are involved. Also stucco, paints and mastics tend to be less expensive than new sheathing, polymer papers, etc.

GUIDELINES

The following concrete masonry wall assemblies are considered to meet an air leakage of less than 0.04 cfm/ft2 (0.20 L/s-m2) at 75 Pa,
either by prescriptive code requirements or as demonstrated through laboratory testing.

By prescriptive IECC criteria (ref. 5):

  • Fully grouted concrete masonry.
  • Concrete masonry with a portland cement/sand parge, stucco or plaster with a minimum thickness of 1/2 in. (13 mm).
  • Concrete masonry walls coated with one application of block filler and two applications of a paint or sealer coating.
  • By laboratory testing (refs. 6, 8):
  • 12-in. (305-mm) concrete masonry sealed with at least two coats of commercial-grade latex paint.
  • 8-in. (203-mm) concrete masonry coated with a single coat of high quality latex paint.
  • 8-in. (203-mm) concrete masonry coated with a single coat of masonry block filler.

It can be reasonably assumed that compliance would also be achieved by applying these coatings to walls having a larger thickness than those tested.

When coatings such as paint or block filler are called for, they can be applied to either the interior or exterior side of the concrete masonry, so any masonry architectural finishes need not be compromised.

REFERENCES

  1. Sherman, Max H. and Iain S. Walker, LBNL 62341. Energy Impact of Residential Ventilation Norms in the United States, Lawrence Berkeley National Laboratory, 2007.
  2. Carr, D. and J. Keyes, Component Leakage Values and their Relationship to Air Infiltration, Steven Winter Associates, 1984.
  3. 2009 ASHRAE Handbook – Fundamentals. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., 2009.
  4. International Energy Conservation Code. International Code Council, 2006 and 2009.
  5. International Energy Conservation Code. International Code Council, 2012.
  6. Biggs, David T., Air Permeance Testing of Concrete Masonry Wall Assemblies, FR06. National Concrete Masonry Research and Development Laboratory, January 2008.
  7. Assessment of the Effectiveness of Water Repellents and Other Surface Coatings on Reducing the Air Permeance of Single Wythe Concrete Masonry Assemblies, MR36. Concrete Masonry & Hardscapes Association, 2010.
  8. Standard Test Method for Air Permeance of Building Materials, E2178-03. ASTM International, 2003.
  9. Emmerlich S. J., T. McDowell, W. Anis, Investigation of the Impact of Commercial Building Envelope Airtightness on HVAC Energy Use, NISTIR 7238. National Institute of Standards and Technology, 2005.
  10. Air Ventilation and Infiltration Center, Old Bracknell Lane West, Bracknell, Berkshire, RG12 4AH, Great Britain.
  11. Kohonen, R., S. Ahvenainen and P. Saarnio. Review of Air Infiltration Research in Finland, Air Infiltration Review Vol. 6, No. 1, 1984.
  12. Yoshiro, Dr. H. Overview of Air Infiltration in Japan, Air Infiltration Review. Vol. 5 No. 3, May 1984.
  13. Condensation Control in Concrete Masonry Walls, TEK 06-17B, Concrete Masonry & Hardscapes Association, 2011.
  14. Caluwaerts, P. and P. Nusgens. Overview of Research Work in Air Infiltration and Related Areas in Belgium, Air Infiltration Review. Vol. 5 No. 1, 1983.