Performance of 8 Cold-Climate Envelopes for Passive Houses

Master's Thesis, 2011

159 Pages, Grade: none


Table of Contents





Envelope Descriptions/Diagrams

Envelope Selection and Thermal Resistance (2D R-value modeling)

Thermal Bridges (THERM modeling)

Passive House Verification and House Model (PHPP modeling)

Hygrothermal Performance and Risk (WUFI modeling)

Life Cycle Environmental Impact Analysis (Athena LCA modeling)


Appendix A - Detailed Methodology 2D R-value modeling THERM modeling PHPP modeling WUFI modeling Athena LCA modeling

Appendix B-CommonMaterial Properties:ThermalResistance

Appendix C-CommonMaterial Properties:VaporPermeance

Appendix D - Energy Modeling Assumptions

Appendix E - WUFI Modeling Assumptions

Appendix F - Moisture Storage @80RH and 68°F (20°C)

Appendix G - Athena Modeling: Envelope materials and layer thicknesses

Appendix H - THERM Modeling: Thermal Bridge Details

Works Cited

Additional Sources


The author wishes to express sincere appreciation to the Center for Sustainable Building Research (CSBR) for their extended long-term support, and especially to William Weber, Senior Research Fellow at the CSBR, for all his work as advisor, thesis committee chair, and friend. A great deal of additional guidance, knowledge, and time was offered by Pat Huelman, Associate Extension Professor in the Department of Bioproducts and Biosystems Engineering, who first introduced the author to the wonderful world of building science. Special thanks are in order to Lucas Alm, Adjunct Assistant Professor, for stepping in as thesis committee member after the retirement of Steve Weeks, former Director of Graduate Studies at the College of Design.

The author also wishes to extend a tremendous thank you to the Research Centre on Zero Emissions Buildings at the Norwegian University of Science and Technology (NTNU), and Anne Grete Hestnes, Director, who graciously extended her assistance and the support of her center and university. Many thanks to the NTNU professors and SINTEF researchers who generously offered their time and support throughout the 2010/2011 academic year: especially Sivert Uvslokk, Arild Gustavsen, Stig Geving, and BeritTime. Tusentakk!

One final thank you to the U.S. Fulbright Program, for providing funding and the opportunity to study abroad at the Research Centre on Zero Emissions Buildings in Norway.

Abstract - "Performance of 8 Cold-Climate Envelopes for Passive Houses

While the design and construction ofenvelopesfor Passive House certified homes in central European climates is well developed and has achieved widespread acceptance and reliability, the same cannot be said in colder climate regions such as the upper Midwest (DOE climate zones 6 and 7) and Scandinavia. The objective ofthis research was to study some of the typical building performance issues relating to Passive House envelope construction for single family homes in cold climates by testing and developing a group of 8 envelope options. Typical issues include unfamiliarity with performance ofthermal bridge details, added embodied energy and carbon due to increased insulation and structure, and increased risk ofmoisture damage due to thicker, multi-layered assemblies and smaller drying potentials. The envelopes were tested and developed to meet set levels ofmoisture safety, life cycle energy and carbon impacts, and Passive House thermal bridging and energy performance requirements. The options chosen forstudy (listed below) were based on envelopes that have been usedfor Passive House and low-energy projects in cold climates, including Minnesota, Wisconsin, Maine, Illinois, Norway, and Denmark. A model single family home set in Minneapolis, MN was used as the basisfor comparison for a number of software analyses. Athena life cycle analysis software was used to determine embodied energy, carbon, and environmental impacts of the envelope types. WUFI hygrothermal modeling was used to determine moisture performance and risk relating to mold growth. The EN ISO 6946 2-D U-value calculation protocol was used in conjunction with the Passive House Planning Package (PHPP) to ensure that the energy efficiency requirements set by the Passive House Institute were met, while THERM software was used to determine the performance of a selection of thermal bridge details. Although significant variation wasfound in the performance of these eight types, all envelopes werefound capable of meeting the energy efficiency and thermal bridging requirements ofthe Passive House certification in a cold, Minnesota climate, while maintaining moisture safety, durability, and significant life-cycle energy and carbon savings. Thesefindings demonstrate that evenfor cold climates, a variety of envelope types can be usedfor certified Passive Houses.

Envelope options: 1) Advanced 2x6framing with interior cross strapping and exterior insulation, insulated with mineral wool, 2) Advanced 2x6framing, insulated with spray polyurethanefoam and exteriorrigidfoam, 3) Double 2x4 stud wall, insulated with blown cellulose, 4) I-joist (TJI) balloon framing, insulated with blownfiberglass, 5) Insulated concreteform wall (ICFs), using EPS insulation, 6) Concrete block wall, insulated with exterior mineral wool, 7) Massivtre/Structural engineered panel (SEP), insulated with exterior rigidfoam, and 8) Structural insulated panel (SIP), using EPS insulation. For comparison, a base option was also studied: Standard 2x6framing (16 inches on center) withfiberglass batt insulation.


This research investigates the characteristics of 8 Passive House envelope types in terms of the following topics: 1) insulation value, 2) thermal bridging, 3) overall energy performance, 4) moisture management, and 5) embodied energy and life cycle environmental impacts.

As one of the primary human activities, building has a tremendous impact on the environment. The construction and operation of buildings is estimated to contribute between 30% and 50% of the carbon dioxide and other global warming gases responsible for climate change1 (U.S. Green Building Council, 2011). Additional environmental impacts from buildings include habitat loss or disruption from mining and forestry activities, air and water pollution from the burning of fossil fuels for electricity and heat, and resource consumption and waste production associated with the fabrication of building materials and building construction, renovation, and demolition. In recognition of the impact buildings have on the environment, governments and organizations worldwide have recently developed a large number of sustainable building rating systems, guidelines, and certifications. The goal ofthese programs is to reduce the environmental impact of buildings while improving the indoor environmental quality for the occupants. Some of these programs are holistic in nature, such as the U.S. Green Building Council's LEED rating and certification system, which seeks to address many aspects of sustainable building design from indoor environmental quality, through site design and water use, to energy use. Other programs such as Passive House certification are strictly focused on a single environmental issue, most commonly energy use. This research is focused on Passive House certification and its implications for the design, construction, and durability of envelopes in very cold climates including northern North America and Scandinavia.

The Passive House concept grew from early experiments in North America with super-insulated homes. Learning from these early attempts, professors Bo Adamson of Sweden and Wolfgang Feist of Germany developed the first true Passive House in Darmstadt, Germany in 1990. In 1996, Feist decided to expand the concept into a practical approach to meet housing and energy needs2 (Klingenberg, Kernagis, & James, 2008). He founded the Passive House Insitute and initiated a certification scheme using the new Passive House Planning Package, an Excel-based energy simulation program. Today, Passive House certification has grown into one of the most popular, widely-used energy efficiency standards in Europe, where tens of thousands of homes and other small buildings have been built and certified to date. Enough certified Passive Houses have been built to generate considerable industry interest. Companies have developed a wide range of specialized building products including foundation systems, windows and doors, ventilation ducts, and air source heat pumps to meet the demands of the program. With large numbers of these products available and a construction industry that has achieved a level of familiarity and comfort with the required elements of certification, Passive Houses in central Europe are now beginning to achieve cost parity with standard methods of construction. The adoption of the Passive House certification standard has been slower in colder climates such as Scandinavia and much of North America, but is beginning to grow quickly, with hundreds of homes currently under review for certification in the United States.

Passive House certification is frequently touted as one of the most stringent energy efficiency standards in the world, commonly requiring an energy use reduction of anywhere from 65-85% compared to standard code construction. This level of efficiency is not arbitrary; rather, it was developed based on the notion that a sustainable civilization must exist on available renewable energy resources. After dividing this available pool of energy by the current population of the earth, and further subdividing between basic needs such as transportation, industry, and building/housing requirements, a primary energy budget of 120 kWh/m2/yr (38.1 kBtu/ft2/yr) was calculated for housing. This is based on the assumption of 35m2 (380ft2) of living space per person. For the purpose of the certification standard, the energy budget is applied worldwide irrespective of climate or other conditions. Although the validity of this approach can be debated, it has proven valuable in central European climates for a couple of reasons. First, the strict energy budget requires a steep reduction in energy use compared to standard construction, especially in regards to heating energy for the moderately cold European climate. The reduction in heating energy allows a small amount of heated ventilation air to satisfy the heating load of the building, eliminating the need for large radiators, boilers, or forced-air furnaces. This compensates for the increased cost of high-efficiency windows and doors, and higher levels of insulation. Second, energy use is reduced to the point where it becomes feasible in most situations to meet the remaining energy demand with renewable energy generated on site, typically with solar panels. This allows Passive House certified buildings to achieve net zero energy, while restraining the cost and size of expensive renewable energy systems.

This research was carried out primarily in Minnesota and Norway. In these cold climates, as in much of the Midwest and Scandinavia, the majority of energy use in homes and small buildings is for space heating. To achieve the level of energy efficiency required for Passive House certification, it is absolutely necessary to dramatically reduce space heating energy consumption. High-efficiency heating equipment is a must, but more critically, the heat load, or demand, must be reduced. This is accomplished primarily through improvements in the performance of a building's envelope, or external shell, such as high­performance triple-glazed windows, increased levels of insulation, thermal bridge-free construction, and airtight enclosures. To address these issues, the Passive House Institute has established some recommended guidelines:

“thermal bridge free construction" with Ψ values </= 0.01 W/m/K whole window U-values </= 0.8 W/m2/K (U-0.14, IP)

While most energy codes don't set performance standards to address heat loss from thermal bridges, window performance is regulated. Passive House windows provide roughly 2.5 times the insulation of standard windows in the United States, and 1.5 times that of standard windows in Norway. The Passive House Institute has also established some performance requirements: air tightness (infiltration) </= 0.6 air changes per hour (ACH) @50Pa specific space heat demand </= 15 kWh/m2/yr (4.75kBtu/sf/yr)

The requirement for air infiltration is roughly 4-5 times tighter than the standard level of air tightness achieved in typical residential construction in Minnesota3 (Sheltersource, 2002) and Norway. The requirement for space heating demand typically results in an 85% reduction in heat load compared to new single-family homes built to current Minnesota energy code. This dramatic reduction is generally achieved by meeting the other requirements and recommendations listed above, in addition to improving the insulation value of the envelope. Although the specific R-values for external walls, roof, and floor can vary according to the climate, achieving the 15 kWh/m2/yr target typically requires R­values 2-4 times higher than those required by energy codes in Norway and Minnesota.

Together, these envelope and space heating recommendations and requirements help a Passive House achieve the final requirement for overall energy use: specific primary energy demand </= 120 kWh/m2/yr (38.1 kBtu/sf/yr)

This requirement pertains to the building's total energy use including heating, cooling, lighting, appliances, etc, and guarantees that the worldwide energy budget calculated for housing is met by every certified Passive House.

As the need for dramatic reductions in the carbon intensity and energy use of buildings becomes clear, Passive House certification continues to grow and expand into colder climates. However, in very cold climates such as those found in Scandinavia and the northern parts of North America, high levels of insulation and airtight enclosures can pose difficulties in envelope design, assembly, and durability. For example, reducing heat flow through walls with increased insulation can reduce the energy available to dry out moisture within them. In addition, tighter envelopes can increase the potential for trapping moisture. A higher level of moisture fosters mold growth, decreasing indoor air quality (IAQ) and eventually leading to decay of the structure. Moreover, as the level of insulation is increased, the energy expended in manufacturing and construction grows. This extra “embodied energy" cuts into the energy savings the envelope is meant to provide. Finally, as insulation is increased, a greater proportion of heat is lost through thermal bridges4 (Christian & Kosny, 1996) - intersections or penetrations in the envelope where the insulation is not continuous. Without greater attention to insulating or minimizing these thermal penetrations, the effectiveness of the added insulation is reduced and Passive House certification could be jeopardized. Therefore, the performance of Passive House envelopes needs to be considered in light of the following topics: 1) insulation value, 2) thermal bridging, 3) overall energy performance, 4) moisture management, and 5) embodied energy and life cycle environmental impacts.

The purpose of this research was to evaluate a group of eight envelope types which have been used for Passive House and low energy projects in very cold climates, equivalent to DOE climate zones 6-7. The envelopes were selected, developed, and tested for performance in each of the topic areas mentioned above. The key objective was to determine which envelope types could be designed for moisture safety in these climate conditions while simultaneously meeting the demands of Passive House certification and providing life cycle savings in energy and carbon emissions. This research extends similar work that has been done for Passive House envelopes in central European climates (contained in resources such as the Passive House Bauteilkatalog) to envelope types in more common use throughout North America and Scandinavia.

Each of the 5 topics forms a section in this research paper. Within each topic, the performance of all eight envelope types was analyzed using an appropriate software program and/or standardized techniques. Each topic generally begins with a discussion of relevant terms and concepts. This is followed by a brief description of methodology. (A more detailed discussion on methodology can be found in Appendix A.) Performance results, analysis, and comparison with the base case envelope generally conclude each section. The topics are arranged as follows:

1) Selection of envelope types from case studies and calculation of thermal resistance using ENISO 6946:2007, a standardized 2-dimensional U-value calculation technique. 2) Thermal bridge analysis for a selection of thermal bridge details, with heat loss values calculated using Therm software, version 6.3. 3) Verification of Passive House-level energy performance, using a basic model home in Minneapolis, MN as input for the Passive House Planning Package (PHPP) version 2007, the energy modeling software used forofficial Passive House certification. 4) Modeling of hygrothermal (moisture) performance and assessment of mold growth risk, using Wärme Und Feuchte Instationär (WUFI) Pro software, version 5.1. 5) Life-cycle performance in terms of energy and carbon emissions, and modeling of additional environmental impacts, using Athena Environmental Impact Estimator, version 4.1.

As previously mentioned, the particular envelopes chosen for this research were selected and developed from cold-climate Passive House and low energy case studies. These envelope types generally work as a system, meaning that above grade walls and roof function together to create a continuous layer of insulation and an air tight boundary. Thus, above grade walls are paired with a matching roof construction to complete the above grade envelope. For example, SIP panel walls are paired with a SIP panel roof, I-joist-framed walls are paired with an I-joist roof, and concrete ICF walls are paired with a concrete roof. In terms of analyzing the research results, the intention of this pairing was to understand the full implications of each particular construction type.

In the case of the ICF envelope, the above grade walls could be extended below grade with very little change. In most cases, however, the above grade envelope was not suitable for below grade use and was paired with a different assembly. The below grade walls for most envelope types consisted of a poured concrete wall 8 inches thick (203mm), insulated with 12 inches (305mm) of rigid type II EPS and finished with an interior layer of gypsum board. Inall cases, the bottom of the external envelope was a 4 inch (102mm) thick concrete floor slab, insulated with 14 inches (356mm) of high density type IX EPS foam. The envelope types selected for analysis were as follows:

1) Advanced Frame with Cross Strap

“Advanced" 2x6 stud wall with interior cross strapping and exterior insulation, insulated with mineral wool. Roof - "Cold attic" light frame wood truss construction, with interior cross strapping, insulated with blown cellulose and a layer of mineral wool batt insulation between the strapping.

Note - "Advanced framing" refers to framing with studs 24 inches on center (601mm) in line with roof trusses, single top and bottom plates, 2-stud corners with clips for drywall attachment, and insulated headers using metal brackets for support rather than additional "jack" studs.

2) Advanced Frame with SPF

"Advanced" 2x6 stud wall, insulated with spray polyurethane foam (SPF) and unfaced exterior rigid polyisocyanurate foam. Roof - "Cold attic" light frame wood truss construction, insulated with a 1 inch layer of SPF and blown cellulose on top.

3) Double Stud

Double 2x4 stud wall, insulated with blown cellulose. Roof - "Cold attic" light frame wood truss construction, insulated with blown cellulose.

4) TJI Frame

I-joist (TJI) balloon frame wall, insulated with high-density blown fiberglass. Roof-TJI roof construction, insulated with high density blown fiberglass (contains loft space).

5) ICF

Insulated concrete form (ICF) wall with additional exterior EPS insulation. Roof- Flat pre-cast hollow- core concrete plank construction, insulated with exterior polyisocyanurate foam.

6) Mass wall

6 inch concrete block wall, insulated with exterior mineral wool. Roof - "Cold attic" light frame wood truss construction, insulated with blown cellulose.

7) SEP panel

Massivtre/Structural engineered panel (SEP) wall, insulated with exteriorfoil-faced polyisocyanurate foam. Roof-Wood truss construction, insulated with exterior foil-faced polyisocyanurate foam (contains loft space).

8) SIP panel

Structural insulated panel (SIP) wall, using integral EPS insulation, with additional unfaced exterior polyisocyanurate foam. Roof- SIP panel roof, with additional unfaced exterior polyisocyanurate foam (contains loft space).

In addition, a non-Passive House base option for comparison, based on current MN energy code:

9) Base case (standard frame)

Standard 2x6 framing 16 inches (400mm) on center with fiberglass batt insulation between the studs. Roof - "cold attic" light frame wood truss construction, insulated with blown cellulose.

A more detailed, pictorial description ofthe envelope types developed and analyzed forthis research is shown on the following pages. As much as possible, the envelopes use real building products and material dimensions.

illustration not visible in this excerpt

Figure 1) Advanced Frame with interior cross strapping and mineral wool.

illustration not visible in this excerpt

Figure 2) Advanced Frame with spray polyurethane foam (SPF).

illustration not visible in this excerpt

Figure 3) Double stud frame with blown cellulose.

illustration not visible in this excerpt

Figure 4) TJI Frame with blown high-density fiberglass.

illustration not visible in this excerpt

illustration not visible in this excerpt

Figure 6) Mass wall with mineral wool batts

illustration not visible in this excerpt

Figure 7) Massivetre/Structural Engineered Panel (SEP) with foil-faced polyisocyanurate.

illustration not visible in this excerpt

Figure 8) Structural Insulated Panel (SIP)

illustration not visible in this excerpt

Envelope Selection and Thermal Resistance (2D R-value modeling)

Part A) Research location and climate comparison

illustration not visible in this excerpt

Figure 10) DOE climate zone map of the U.S. Passive House case studies were taken primarily from climates zones 6 and 7.

As Passive House certification expands out of central Europe into Scandinavia and North America, envelope types change with different building traditions and availability of resources. Thus, it was necessary to establish a set of envelopes based on Passive House case studies from these particular regions and climate zones. The climate zones of interest in this research correspond primarily to IECC (International Energy Conservation Code) climate zones 6 and 7, although several Passive House case studies were also selected from climate zone 5. In terms of heating degree days, equivalent climate regions were included from Scandinavia.

Since this research was carried out in Minnesota and Norway, particular emphasis was placed on climate conditions and Passive House case studies in these places. A comparison of climate in Minnesota and Norway shows that the variation in climate across Norway is greater than in Minnesota, with a wider spread of heating degree days (both colder and warmer temperatures), and precipitation (both more and less annual accumulation). This is likely due to Norway's greater north-to-south length, proximity to the ocean, and varied topography and elevation. However, many regions in Norway are quite similar to the climate in Minnesota, in terms of heating degree days and precipitation. For example, much of the interior of southern Norway experiences similar winter temperatures and yearly levels of precipitation. These are conditions that can be found across much of the rest of Scandinavia, where the warming influence from the ocean is not as great.

illustration not visible in this excerpt

Figure 11) Base 65°F (18°C) heating degree days in Norway and the Midwest. Maps are of equal scale. Several Norwegian cities are plotted on the Midwest map according to their average yearly HDD count.

illustration not visible in this excerpt

Figure 12) Yearly average precipitation in Norway and the Midwest. Maps are of equal scale.

In general, levels of solar radiation are much lower throughout Norway and the rest of Scandinavia than in Minnesota. This is due to Scandinavia's polar latitude. Coastal Norway also experiences significantly increased cloud cover. While the Minneapolis/St. Paul metropolitan area lies along the 45°N latitude (equivalent to Venice, Italy), Oslo is very close to the 60°N latitude. Thus, Minneapolis/St. Paul receives roughly 1.5x more solar radiation than Olso - 428 kBtu/ft2/yr on average compared to roughly 285 kBtu/ft2/yr, respectively (1350 kWh/m2/yr to 900 kWh/m2/yr).

With the exception of available solar radiation, the general similarity in climate between Scandinavia and the Midwest suggests that Passive House envelopes in one region may be suitable for use in the other region. That is, an envelope that satisfies Passive House energy efficiency requirements while exhibiting safe moisture behavior in one region might meet the same performance standards in the other region. This assumption is discussed and analyzed in greater detail in section 3, Passive House Verification and House Model.

Part B) Passive House case studies and development of target R-values

A number of Passive House and low-energy house case studies were examined from Minnesota, Wisconsin, Maine, Illinois, Norway, and Denmark. Heating degree days (climate data), envelope types and layers, R-values, and energy performance was noted. The eight envelope types analyzed in this report were selected from these case studies.

The survey of envelope types shows that despite significant variation - especially in the types of insulation products - the use of stud wall framing is quite common in both Scandinavia and the Midwest. A variety of stud types and framing techniques are used for Passive Houses and low energy houses in both regions, including double stud walls, “advanced framing" with 2x6 studs (48 mm x 148mm), and I-joist, or TJI, balloon framing. The prevalence of timber framing is likely due to the availability of large timber resources and the presence of a strong timber industry. These factors likely also spurred innovation in the use of structural engineered wood products. For example, both Norway and Minnesota have developed solid wood structural panels for use as wall and roof elements. The panels are designed to take the entire structural load - performing the function of studs, sheathing, and interior finish. In Norway, they take the form of thick, 5.5 inch (140mm) plywood panels, termed "Massivtre", with high-grade finished surfaces and rougher core layers. In Minnesota, the panels are thinner, at 1.5 inches (38mm) thick, and take the form of a high-density OSB product termed "SEP", for structural engineered panel. In both cases, all of the insulation is applied to the exterior of the panels.

Envelope types from Denmark tend to mimic Passive House envelopes developed for use in central Europe (albeit with higher levels of insulation), with frequent use of lightweight concrete block walls and other masonry products. However, a series often Passive House-certified "Comfort Houses" (Komfort Husene) sponsored by the insulation manufacturer Isover also feature innovative wood construction including Massivtre panels.

A couple types of envelope systems featuring integral EPS insulation are unique to the U.S. These include insulated concrete forms (ICF) and structural insulated panels (SIP). These products use rigid EPS foam as a structural material. In the case of ICF construction, the EPS acts as a stay-in-place form for poured concrete walls. With SIP construction, a thick core of EPS is bonded to outer skins of OSB, forming a rigid panel able to act as the roof and wall structure, sheathing, and insulation. These panelized systems are not currently popular in Scandinavia, most likely because concern exists over the close proximity of the EPS to interior environments and the potential for toxic gases when EPS is exposed to fire.

A selection of case studies illustrating use of the eight envelope types analyzed in this report are shown in the summary charts below. Note that the envelopes analyzed in this report were not taken directly from the case studies. Rather, general envelope types were selected, then specific insulation materials, thicknesses, cladding types, vapor retarders, etc were modified in an attempt to meet specific target R-values, levels of moisture safety, and life-cycle savings in energy use and carbon emissions.

illustration not visible in this excerpt

The average R-value of the exterior walls, roofs, and floor slabs in this selection of case studies is R- 62.9 (U-0.090, SI), R-83.8 (U-0.068, SI), and R-67 (U-0.085, SI), respectively. (These averages do not include Wilder House 1, which was not designed as a Passive House.) These averages were used as the basis for choosing “target R-values". Ultimately, the R-values chosen as targets for the envelopes in this study were set at R-60 (U-0.095, SI), R-80 (U-0.071, SI), and R-60 (U-0.095, SI) for the walls, roofs, and floor slabs respectively. The target values were adjusted down slightly from the case study averages for a couple of reasons. First, the majority of Passive House case studies surpassed the level of energy performance required for certification by significant amounts. Second, the basic model house design used for Passive House energy verification in this study incorporated an earth-sheltered walkout basement, which lowers the heat load for nearly half of the conditioned house volume. The target R-value for the below grade walls in the house model was set at R-50 (U- 0.114). The adequacy of the target R-values was checked early on by running a test energy model in PHPP, usingthe house model in a Minneapolis, MN climate.

Part C) Terms and concepts in stud wall R-value calculations

Frequently, the energy performance of exterior walls is compared based on the R-value of the insulation, evaluated independently from the wall it is installed in. This simplistic comparison greatly overestimates the true energy performance of envelope assemblies with repetitive thermal bridges, such as stud walls. It does not take into account the thermal bridging effect of the studs, which create shortcuts for heat around the insulation. The percentage of wall area occupied by studs or other framing members is frequently referred to as the “framing factor", and is often as high as 25% in standard U.S. residential construction5 (Syed & Kosny, 2006). This amounts to a significant fraction of wall area that is essentially un-insulated. Additional thermal bridges are also created by traditional framing techniques at wall corners, window and door penetrations, and connections to the roof and foundation.

To take these effects into account, the Oak Ridge National Laboratories Building Technology Center has developed an alternative approach to comparing wall R-values6 (Kosny & Christian, 1996). The "center of cavity" R-value calculates the R-value at a point in the wall containing the most insulation. It is the R-value typically referenced in U.S. residential building codes, but remains an inaccurate estimate for the true energy performance of framed walls with large numbers of repetitive thermal bridges. The "clear wall" R-value looks at a larger section of the wall and includes the effects of studs or other repetitive thermal bridges. The clear wall R-value is a more accurate estimate of a framed wall's true R-value. Most technical R-value calculation procedures such as EN ISO 6946 and the simpler parallel paths method calculate the clear wall R-value. If an assembly has no repetitive thermal bridges, then its clear wall R-value will equal its center of cavity R-value. Finally, the "whole wall" R-value considers the design of the entire wall, including details at corner, roof and foundation intersections, and window/door penetrations. However, whole wall R-values were not calculated for this report. Rather, the thermal bridges typically included in the whole wall R-value were calculated independently, as individual thermal bridges. This is in accordance with Passive House certification protocol.

Typically quoted framing factors for stud wall construction often include estimates of additional framing necessary for bearing points, exterior wall corners, intersections with interior walls, etc. For advanced framing, these additional framing members are strictly minimized, but are still present. To account for this, the framing factor for all walls utilizing advanced 2x6 framing was increased slightly from a calculated clear wall percentage of 9% (based on an 8 foot high wall (2.44 m) with single top and bottom plates and 24 inches on center (610 mm) stud spacing) to 12%. If present, additional wood members such as interior cross strapping was also included in the framing factor calculation. The framing factor for the TJI wall was also increased by 3% from a calculated clear wall percentage of 12% to a total of 15%. The framing factor used for SIP walls included bottom and top plates spaced 8 feet (2.44 m) apart and 5/8" thick plywood splines at 12 feet on center spacing (3.66 m). The framing factor used for the base case 2x6 stud wall using standard framing (16 inches on center (406 mm) with double top plates and 3-stud corners) was 25%, based on several nationwide studies (U.S.) reported in the Journal of Building Physics7 (Syed, 2006).

The framing factors used for roof assemblies were somewhat smaller due to the absence of top and bottom plates or any type of additional framing. The framing factor used for roofs with light frame roof trusses (cold attics) was 6%, based on a roof truss spacing of 24 inches on center (610 mm).

The framing factor for the TJI roof was slightly higher at 9.6% due to the width oftop and bottom flanges, while the framing factor for SIP roofs was 2.7%, accounting for spline connections between panels.

Part D) R-value: Methodsofcalculation

The R-value calculation protocol used in this report is EN ISO 6946:2007, as required by Passive House certification. It is a 2-D R-value calculation method, meaning that it takes into account lateral heat flow within a wall, in addition to perpendicular heat flow. This makes it more accurate than the simpler parallel paths method frequently used in the U.S. The R-values are clear wall R-values, which include the effects of thermal bridging created by the framing factors discussed above. A more detailed description of the R-value calculation protocols can be found in Appendix A, Detailed Methodology, 2D R-value modeling.

Note that since the envelopes were developed using real products - including accurate material dimensions, tested R-values taken from product information sheets, and common construction practices - the final calculated R-values rarely meet the targets exactly. Certain materials, such as wood studs, were deemed so ubiquitous that specific product information was not required, although standard material dimensions and thermal properties were still assumed. A list of R-values for all materials used in this report can be found in Appendix B, Common Material Properties, Thermal Resistance.

Part E) R-value: Results

The following chart shows clear wall R-values for the above grade walls and corresponding roofs, both calculated using EN ISO 6946:2007. The chart also lists center of cavity R-values, which are used to gauge the influence of thermal bridges within a clear wall section of the wall. This is done by comparing the clear wall R-value to the center of cavity R-value. Finally, the chart lists wall thickness, which is used to calculate the R-value per unit of thickness, a measure of the insulating efficiency of the wall.

illustration not visible in this excerpt

Figure 13) R-value results for wall and corresponding roof assemblies

The R-values calculated for wall and roof assemblies show little variability because they cluster around the target values - R-60 (U-0.095, SI) for walls, and R-80 (U-0.071, SI) for roofs. However, the SIP panel envelope was designed so that a lower roof R-value was compensated with a higher wall R-value. This allowed the use of stock material widths for the additional exterior polyisocyanurate insulation, and kept the exterior insulation to a single layer, easing construction.

In terms of wall thickness and efficiency (R-value provided per unit thickness), the results show greater variability. While most wall types cluster around 19 to 20 inches in thickness (483mm to 508mm), including cladding, several types are significantly thicker or thinner. Walls incorporating concrete are the thickest and provide the lowest R-value per unit thickness. This is due to the layer of concrete, which provides almost no insulation value. Wood framed walls perform better, since the structural layer of wood framing can be insulated and wood is not as conductive as concrete.

The thinnest walls are generally those that use the greatest amount of high R-value insulation, such as spray polyurethane foam (SPF) and polyisocyanurate. The thinnest and most efficient wall is the SEP panel wall. Since the structural layer is extremely thin - only 1.5 inches (38mm) thick - and the insulation material is highly insulating foil-faced polyisocyanurate - up to R-6.5/inch (λ-0.022, SI) - the wall is not much thicker than a standard stud framed wall. It provides an R-value per inch of 5.0, almost 50% higher than the group average of 3.4 (excluding the base case).

illustration not visible in this excerpt

Figure 14) Wall thickness including cladding. For this graph only, the SIP panel wall R-value was lowered to R-62.5 (similar to the other Passive House wall assemblies), eliminating 1 inch (25.4mm) of exterior insulation. This provided an even R-value basis for comparison of wall thickness.

The effect of repetitive thermal bridges such as studs on an assembly's R-value was analyzed by comparing the center of cavity R-value to the clear wall R-value. By dividing the clear wall R-value by the center of cavity R-value, a percentage representing the reduction in R-value due to thermal bridging was calculated. This is essentially a ratio comparing an assembly's R-value with thermal bridges to its R-value without thermal bridges.

illustration not visible in this excerpt

Figure 15) The effects of repetitive thermal bridges on clear wall R-value for each wall option. (Roof options not included.)

This analysis shows that frame wall options perform somewhat worse than solid wall options in terms of the assembly's own internal thermal bridging. (Note that results can be different when whole-wall details are introduced, such as intersections ofwall, roof, and floor, or penetrations around windows and doors.) The base case standard frame option has the largest reduction in R­value, 24.5%, since the studs are closely spaced and there are no continuous layers of insulation to break their thermal conduction. The I-joist frame option also performs relatively poorly for the same reason, but does minimize thermal bridging through the “studs" by using a very thin profile - the web of the I-joist. To compensate for its increased thermal bridging, the I-joist wall requires insulation with a higher thermal resistance compared to the double stud wall. (The I-joist assembly requires dense-pack fiberglass at R-4.35/inch to achieve the R-60 target, while the double stud assembly uses blown cellulose at R-3.8.inch to meet R-60 with the same wall thickness. This gives a sense for the impact of the thermal bridging due to the frame.) The advanced frame with SPF and cross strap options use a thick layer of exterior insulation to help break the thermal conduction of the studs. The cross strap option improves on this further by isolating the interior of the studs with another thin layer of insulation. The cross strap option exhibits the smallest reduction in R-value of all the frame wall assemblies, at 5.6%. With the exception of the SIP panel option, the solid wall assemblies exhibit no loss in R-value due to repetitive thermal bridges. The SIP option does exhibit some thermal bridging due to the presence of top and bottom plates and occasional plywood splines at panel connections. Nevertheless, it outperforms all the frame options with an R-value reduction of 4.4%.

These results show that the target R-values for wall and roof assemblies can be met with all the envelope options studied, but that some types are significantly more efficient, providing a greater R­value per unit of thickness while minimizing the effects of repetitive thermal bridges. These types include SIP panels and SEP panels. In addition, SEP panel construction provides a thin profile, which may be of importance in high-density downtown areas where thick walls consume valuable floor space. The advanced frame with SPF envelope option also provides a thin profile and a high R-value per unit of thickness, but it is characterized by a relatively high thermal bridging effect. This means that at least 10% of the insulation value of the spray foam is lost due to thermal bridging through the studs - which is an expensive loss given that closed cell SPF is one of the most expensive insulation products utilized in this study.

Thermal Bridges (THERM modeling)

Part A) Terms and concepts in thermal bridge analysis

Thermal bridges generally come in two types, structural and geometric. A structural thermal bridge occurs when one of the layers in a building assembly is not continuous. For example, a structural thermal bridge occurs where the insulation in an exterior wall is interrupted by a penetration such as a window frame, rim joist, stud, or merely a change from one assembly to another - as at the junction of a concrete block wall and stud wall. The second type of thermal bridge, a geometric thermal bridge, occurs at corners - as at the corner of a roof and exterior wall. Even if a corner assembly is perfectly continuous with no interruption or change in materials or thickness, a geometric thermal bridge still occurs in that location. However, corners are frequently characterized by a combination ofgeometric and structural thermal bridges because corners typically have additional structural elements (such as studs) that reduce or interrupt the thickness of insulation.

Heat loss through a thermal bridge is measured by a psi (Ψ) value, which is somewhat similar to a U- value. Just as a U-value is multiplied by the area of the wall or roof surface to calculate the total heat loss, a Ψ value is multiplied by the length of the thermal bridge to calculate the total heat loss.

Geometric thermal bridges are related to the difference between an envelope's interior versus exterior surface area - the thicker the envelope, the greater the difference between interior and exterior surface area. If heat loss is calculated based on the exterior surface area of an envelope, the magnitude of heat loss is overestimated. Likewise, if heat loss is calculated based on the interior surface area, heat loss is underestimated. Since the walls of Passive Houses are often very thick, the difference in heat loss between an envelope measured with external and internal dimensions is quite large. The use of geometric thermal bridge values helps to correct for this discrepancy. When an envelope's interior dimensions are used to calculate heat loss, the Ψ value at a corner is often positive, representing additional heat loss which accounts for the underestimation. When an envelope's exterior dimensions are used, the Ψ value at a corner is often negative, representing a subtraction in heat loss to account for the overestimation. Following Passive House conventions, all thermal bridges were calculated based on exterior dimensions. This meant that most geometric thermal bridges had a negative Ψ value.

Both geometric and structural thermal bridges are taken into account for Passive House energy loss calculations. These thermal bridges can be further classified into linear thermal bridges (such as the thermal bridge along the length of a rim joist or exterior corner), point thermal bridges (such as at screw penetrations or 3-dimensional corners), and repetitive thermal bridges (such as studs and other repetitive structural bridging elements like cross strapping or splines in a SIP wall). Following Passive House conventions, repetitive structural thermal bridges were not calculated as distinct thermal bridges. This is because the heat loss resulting from this form of thermal bridging was included in the 2-D R-value calculations of the wall and roof assemblies. Also following Passive House conventions, point thermal bridges were not considered as they usually make up a very small proportion of total heat loss. Thus, only linear thermal bridges (both structural and geometric types) were calculated independently and entered into the heat loss calculations in the PHPP.

In accordance with Passive House guidelines, “thermal bridge free" details were designed for the majority of structural and geometric thermal bridges. In Passive House terms, this meant that the thermal bridge details had Ψ values </= 0.01 W/m/K. In some cases, this guideline was not met. For these situations, reasoning is provided in the following discussion.

Part B)Thermal bridges investigated

Technically, thermal bridge Ψ values are not required for Passive House energy loss calculations as long as the details are considered "thermal bridge free". However, in regions where Passive House certification is a relatively new building standard and thermal bridge free constructions details have not been widely developed or adopted, this becomes a challenging question. How does one know if a detail is "thermal bridge free"? For this reason, a variety of thermal bridge details for each envelope type were tested and developed to meet the Passive House guideline. In addition, inclusion of these Ψ values in heat loss calculations reduced the overestimation of heat loss caused by the use of external dimensions, increasing the accuracy of the energy model and easing the achievement of Passive House energy performance requirements.

Thermal bridges that were analyzed included:

1) exterior wall corners above grade 2) foundation wall corners below grade (bermed side of basement) 3) exterior wall corners with foundation wall (walk-out side of basement) 4) wall to roof corners at gable walls 5) wall to roofcorners at side (bearing) walls 6) roof peak (cathedral ceilings only) 7) rim joist on foundation wall 8) rim joist on above grade wall 9) basement floor slab to foundation wall intersection below grade (bermed side of basement) 10) basement floor slab to above grade wall intersection at grade (walk-out side of basement)

Note that thermal bridges at penetrations (namely, around windows and doors) were not modeled individually for each envelope. Rather, for the purpose of heat loss calculations, the standard worst case Ψ value for window installation given in the PHPP software (0.04 W/m/K) was assumed. This represents a window installed with no extra attention paid to minimizing thermal bridges at the window frame.

Part C) Thermal bridges: Methods of calculation/modeling

Ψ values were calculated using THERM 6.3 software to model heat flow through linear thermal bridges. Above grade thermal bridges, such as exterior wall corners and the intersections of wall and roof, were modeled exposed to an ambient air temperature of 0°F, while below grade thermal bridges, such as the intersection of the foundation wall and basement floor slab, were modeled exposed to a ground temperature of 40°F. The below grade thermal bridges were modeled without any earth layers, following Passive House conventions. Perimeter thermal bridges (at grade level) included an earth layer and were modeled according to conventions described in the Passive House Institute's Protokollband Nr. 16, "Wärmebrückenfreies Konstruieren". Thisconvention incorporates a 2.5m x 2.5m block of earth intersecting the model near the corner of the wall and floor slab. A more detailed description ofthis protocol and the methodology used to calculate Ψ values can be found in Appendix A, Detailed Methodology, THERM Modeling. The R-values assumed for materials in the THERM models matched those used throughout the rest of the report. A list of R-values can be found in Appendix B, Common Material Properties, Thermal Resistance.

Part D) Thermal Bridges: Results

As the thickness of an envelope's insulation grows, the heat lost through thermal bridges makes up a greater proportion of overall heat loss. Eventually, increasing insulation thickness becomes less important than addressing thermal bridges. For example, research done by Oak Ridge National Laboratory shows that the percentage of heat lost through thermal bridging in a typical insulated 2x4 stud frame wall (89mm thick) is approximately 10%. (This calculation considers whole-wall thermal bridge details such as those included in details 1-10 listed above, rather than the thermal bridges in a clear wall, such as repetitive studs and top/bottom plates.) That percentage goes up to 16% for a typical 2x6 stud frame wall (140mm thick)8 (Christian, 1996). Extrapolating these results to a typical high R-value Passive House envelope suggests that thermal bridging would quickly make up the majority of heat loss through the envelope, unless steps are taken to address thermal bridges. However, following the Passive House Institute's "thermal bridge free" construction guidelines, the heat loss resulting from thermal bridges should remain at a negligible level, roughly 1-2% oftotal heat transmission through the envelope.

The following chart contains Ψ values calculated with THERM, listed by envelope type for each of the 10 thermal bridge locations described above. The average Ψ value for the thermal bridge location is listed in the far right column, while the bottom row lists the average Ψ value for each envelope type.

illustration not visible in this excerpt

Figure 16) Thermal bridge Ψ values for envelopes, given in W/m/K. Large negative values represent the best performance, large positive values represent the worst performance. Ψ values </= 0.01 W/m/K qualify as “thermal bridge free" construction, and are recommended for Passive House certification. Ψ values > 0.01 W/m/K are shown in red.

Examining Ψ values by location (reading horizontally in the above chart), it is possible to recognize the difference between geometric and structural thermal bridges. Geometric thermal bridges are characterized by the first 6 thermal bridge locations. The Ψ values of these thermal bridges are strongly negative due to the use of external dimensions and the considerable thickness of the Passive House envelopes. (Again, the negative Ψ values help to correct for over-estimation of heat loss caused by the use of external surface areas for heat loss calculations.) Generally for geometric thermal bridges, the greater the envelope thickness, the more strongly negative the Ψ value. For example, the mass wall has the thickest envelope and the lowest geometric thermal bridge Ψ values. In addition to envelope thickness, structural elements disrupting the continuity of insulation at the corner also affect the Ψ value. As much as possible, the continuity of insulation layers should be maintained around corners, with as few penetrating structural elements as possible. For example, the TJI frame and double stud envelope have the same wall thickness and R-value, but the Ψ value is lower for the double stud wall corner indicating better performance (see THERM model snapshots below).

illustration not visible in this excerpt

Figure 17) THERM models of double stud wall corner (left) and TJI frame wall corner (right). The double stud wall corner has no structural bridging elements, while the TJI frame wall corner has several. Both walls have the same R-value and thickness. The double stud corner performs better, with a Ψ value of -0.058 W/m/K, while the TJI corner has a Ψ value of -0.051 W/m/K.

Simple structural thermal bridges (locations 7 and 8) are characterized by positive Ψ values. It is possible to keep Ψ values below the recommended 0.01 W/m/K, but care must be taken to design a detail that is relatively free of bridging elements. This is easier to do if the envelope assembly is continuous, such as at a rim joist attached to balloon framing (see TJI frame detail 8). If the envelope is not continuous, such as at the intersection of above grade wall/foundation wall at the rim joist, Ψ values increase. This is reflected in the higher average Ψ value for thermal bridges at detail 7 compared to detail 8. This situation can be exacerbated if the insulation layers of the adjoining walls are offset from each other (see THERM model snapshots below). Manipulating the width of interior and exterior insulation and adjusting the alignment of the walls to re-center the insulation layers on top of one-another improves the Ψ value.

illustration not visible in this excerpt

Figure 18) THERM models of SEP rim joist (left) and SIP rim joist (right) on foundation wall. The SEP panel's exterior foam is centered above the foundation wall's exterior foam, resulting in uniform thermal gradients and a low Ψ value, 0.003 W/m/K. The SIP panel is not centered above the foundation wall's exterior foam, resulting in distorted thermal gradients and a higher Ψ value, 0.009 W/m/K.

Thermal bridges at the foundation are perhaps the most challenging to eliminate. Despite the presence of a strong 90° corner - with its tendency to drive Ψ values well below 0 - the necessity for a continuous frost wall extending below frost depth can create difficulties. The frost wall or footing tends to disrupt the continuity of insulation as it wraps around the corner from wall to floor slab. This results in Ψ values that are often positive despite the geometric nature ofthe thermal bridge. To investigate this issue, several styles of footing and wall/floor slab connection were modeled.

One common technique at this detail is to eliminate the frost wall by using a shallow, frost- protected foundation. This type of monolithic floor slab/foundation is common in Scandinavia and its popularity is growing in the U.S.9 (National Association of Home Builders, NAHB, 2004). Rather than a deep frost wall that extends below frost depth, this style of foundation relies on a “frost skirt" or horizontal layer of insulation to keep the ground beneath the footing from freezing. By eliminating the frost wall, insulation can more easily wrap around the wall/floor corner. Unfortunately, it is often difficult to keep the insulation layers aligned as they transition from wall to floor slab, unless the wall has a very thick layer of exterior insulation. (See THERM model snapshots below.) This lack of alignment combined with the thickened concrete profile at the edge leads to very poor Ψ values.


1 U.S. Green Building Council. (2011). Buildings and Climate Change. Accessed August 2011 from

2 Klingenberg, K., M. Kernagis, and M. James. (2008). Homes for a Changing Climate. Larkspur, CA: Low Carbon Productions.

3 Sheltersource, Inc. (2002). Evaluating MN Homes Final Report. Prepared for the MN Department of Commerce.

4 Christian, J.E. and J. Kosny. (1996). Thermal Performance and Wall Ratings. Oak Ridge National Laboratory.

5 Syed, A. and J. Kosny. (2006). Effect of Framing Factor on Clear Wall Rvalue for Wood and Steel Framed Walls. Journal of Building Physics, v 30 (n. 2), 163 – 180.

6 Kosny, J. and J. Christian. (1996). Whole Wall Thermal Performance. Oak Ridge National Laboratory.

7 Syed, 2006.

8 Christian, 1996.

9 National Association of Home Builders, NAHB. (2004). Revised Builder’s Guide to Frost Protected Shallow Foundations.

Excerpt out of 159 pages


Performance of 8 Cold-Climate Envelopes for Passive Houses
University of Minnesota - Twin Cities  (College of Design)
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
12299 KB
thermal bridges, hygrothermal, LCA, life cycle analysis, environmental impact, WUFI, Therm, PHPP, envelope, fabric, building skin, SIP, structural insulated panel, ICF, insulated concrete form, Massivtre, double stud, advanced framing, I-joist, SEP, structural engineered panel, cold climate, Passivhus, R-value, U-value, insulation, passive solar, zero energy, embodied energy, embodied carbon, Passive House, Passivhaus
Quote paper
Rolf Jacobson (Author), 2011, Performance of 8 Cold-Climate Envelopes for Passive Houses, Munich, GRIN Verlag,


  • No comments yet.
Look inside the ebook
Title: Performance of 8 Cold-Climate Envelopes for Passive Houses

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free