The first article of this series discussed the problems with excessive air leakage, detrimental air leakage paths through the enclosure, and pressurization for certain building types and environments. In subsequent articles we presented materials and design strategies to minimize air leakage, and discussed how to quantify enclosure air leakage in existing buildings, and the challenges this poses. This article addresses the difficult question of how to predict the amount of air leakage in a new design, and the implications of inaccurate quantification of air leakage in new and existing buildings.
Underestimating air leakage could result in undersized mechanical equipment, but this is not typically the case since mechanical engineers employ safety factors in the design of cooling and, especially, heating systems. With the evolving implementation of sustainable design practices and the understanding of the effects of peak building energy on the required capacity of on-site equipment and the nation's energy infrastructure (i.e. the number of power plants), more sophisticated tools have gained prominence in sizing mechanical equipment and predicting building energy use.
By employing more advanced tools, there is a potential for smaller safety factors and, thus, smaller equipment. However, smaller safety factors coupled with inaccurate assumptions about building performance may cause problems. Complicating matters, building owners and designers often expect energy analyses developed during design to be absolute predictors of building energy use. This expectation is generally unrealistic due to the wide variation in actual versus assumed building operation. In addition, mischaracterizing air leakage rates will reduce the accuracy of energy models and may adversely affect design decisions that depend on whole building energy analysis results.
A theme throughout this series has been the importance of designing and understanding an air barrier as a system. In much the same way, the building enclosure itself is a system made up of many component assemblies, such as fenestration, insulation, waterproofing and air barriers. And, as discussed previously, the overall building is a system with far too many interdependent performance characteristics to allow for compartmentalization of our design evaluations.
Whole Building Energy Analysis
The use of whole building energy analysis (often referred to as "energy modeling") to evaluate design options is becoming increasingly widespread throughout the building design community, and particularly in high performance building design. Energy modeling allows users to include many more aspects of building design and operation than traditional HVAC sizing approaches. Naturally, it is very dependent on the accuracy of those inputs. No matter how simple or complex the tool, its performance relies on the user's understanding of the capabilities and limitations of that tool. The complexity and interdependence of building systems require that engineers constructing energy models understand how the systems interact and how the inputs and assumptions for individual systems relate to other systems and affect the predicted building performance as a whole.
Some practitioners in the building industry may expect energy models to be absolute predictors of building energy use. However, software developers (and most users) make no such claims. Still, the presence of engineers (or "modelers" who are not engineers or architects) who claim the ability to predict actual energy use has provided fodder for energy modeling's detractors. Though this article does not aim to examine - and certainly not adjudicate - the current deliberations in the industry over the proper role of whole building energy analysis, we think it is important to understand these tools have their shortcomings, but also their advantages. To evaluate design options, system control schemes and building operation considerations, whole building energy analysis is beneficial in its primary role as a comparative tool.
The assumed level of air leakage affects both the predicted energy performance of a building and what systems may appear attractive to a designer. That is, the relative performance of some energy efficiency measures is affected by the assumed air infiltration. The first item here is understood, and perhaps even intuitive, to the majority of designers. Mischaracterizing the air leakage performance of the enclosure in an energy model will likely affect the building energy use predicted by that model. The effect will be particularly pronounced in heating-dominated climates, where the effect of air leakage is more significant.
Probably less intuitive is the fact that incorrect air leakage assumptions, even if they are consistent across all analyses, can affect the predicted improvement or increase in energy use for the design option (or combination of criteria) being evaluated. At higher leakage rates, the heating and cooling requirements are higher. Since the absolute increase or reduction in energy use is similar under most conditions, the relative effect of air leakage will be less at higher leakage rates (the denominator in the equation increases while the numerator remains the same). Generally, however, the opposite problem seems to present itself in many energy models we have seen: The predicted air leakage rate is much lower than that of the actual building. This can result in overstated energy savings, which can affect the economic analyses for a project.
In most climates, the effect of inaccurate air leakage modeling on the evaluation of design options will not be significant enough to significantly influence the decision-making process. Existing buildings, as is often the case in energy modeling, present unique challenges. On existing building projects, we are often trying to evaluate and predict the reduction in energy use associated with "tightening" the building enclosure. In these cases, improvements in thermal (or solar heat gain) performance are often coupled to reductions in air leakage in the same energy efficiency strategy (e.g. replacing windows). The best approach is to evaluate the envelope modifications with a range of air leakage rates to gauge the sensitivity of the building's predicted performance to the assumed air leakage. We have found that utilizing the testing techniques discussed in the previous article to be effective in evaluating an existing building and envelope improvements. Whole building tests establish a baseline and component tests can be used to quantify the contribution of specific envelope areas to the overall leakage rate. By assuming a leakage rate for these components after upgrading the envelope, the whole building rate can be adjusted in the energy model.
Air Barriers in High Performance Building Design Standards
The first article in this series touched upon the rise in popularity of "green building" rating systems, such as LEED, and the increasing stringency of energy efficiency codes and standards, such as ASHRAE 90.1. We discussed the lack of air barrier requirements in most codes and standards and the absence of any credit in LEED rating systems for reducing air leakage. These systems - and the codes and standards themselves - rely heavily on whole building energy analysis to show compliance or to exhibit improved energy performance. Due to the issues discussed above, the energy savings associated with various energy efficiency measures may be misrepresented if air leakage is modeled inaccurately. That said, the effect this has on certification under LEED, while worthy of discussion, is a secondary concern. A good, continuous air barrier is essential to acceptable enclosure performance in high performance buildings. However, the supporting infrastructure for projects attempting to achieve a sustainable design has not been formalized. The benefits of air leakage must be properly included in these standards and programs, and the high performance building design process must account for air leakage in whole building energy analyses.
We have now developed an understanding of the importance of good air barrier design; outlined important design considerations, materials and methods; discussed approaches to quantifying and tracking enclosure air leakage in existing buildings; and presented the implications of inaccurate quantification or expectation of air leakage rates. We have touched upon the interaction between the enclosure and other building systems. We now must understand how a good air barrier changes overall building performance and how other systems need to be designed, constructed and controlled. The next article in this series will discuss the requirements for mechanical systems in tight buildings. We will focus primarily on how "business-as-usual" is often not an option and that code requirements may not be sufficient to provide an acceptable level of performance.
About the Authors
Sean O'Brien is a Senior Project Manager in the New York City office of Simpson Gumpertz & Heger Inc. Mr. O'Brien specializes in building science and building envelope performance, including computer simulation of heat, air, and moisture migration issues. He has investigated and designed repairs for a variety of buildings, from condominiums to natatoriums and art museums, and has published extensively on building science-related matters including moisture migration in masonry wall systems and condensation resistance of windows and curtain walls. He can be reached at email@example.com 
Michael Waite is an engineer in the New York City office of Simpson Gumpertz & Heger Inc. He specializes in the interaction between mechanical systems and the building enclosure. He has designed and investigated a wide range of building types and has focused primarily on building energy performance, building enclosure design, and thermal and hygrothermal performance building enclosures. He is a member of ASHRAE SSPC 90.1 and its Envelope Subcommittee, as well as several other industry organizations. He can be reached at firstname.lastname@example.org .