Calculating Whole Wall R-Values on the Net

Jeffrey Christian is director of, and Jan Kosny is a staff scientist at, the Buildings Technology Center, Oak Ridge National Laboratory.

When fiberglass batts are pushed into place without cutting them to fit around electrical wires and outlet boxes, air pockets result.

Fastening the batt insulation's paper facer to the inside surface of each 2 x 6 stud, combined with installation techniques that result in rounded shoulders and cavity voids, add up to a typical worst case installation of these batts.

The air pockets created by the insulation batts' rounded shoulders, by themselves, did not greatly affect the test results.

Fiberglass batts are not always installed perfectly, but when that is the case, no air pockets exist.

Table 1. Comparing Clear-Wall and Whole-Wall R-Values Wall Type Clear-Wall R-Value Whole-Wall R-Value Standard 2 x 4 10.5 9.7 Standard 2 x 6* 16.5 13.6 2 x 6, batts installed perfectly 15.4 12.8 2 x 6, batts installed typically 14.1 11.7 2 x 6, batts installed with rounded shoulders 14.7 12.2 2 x 6, batts installed with rounded shoulders and 1-in gaps at top and bottom 14.0 11.4 Worst-case typical installation: rounded shoulders, 1-in gaps, and paper fastened to inside of 2 x 6 13.2 11.0 Steel frame C-stud 7.3 5.6 Steel frame wall with steel horizontal hat channeling 9.9 7.6 Steel frame wall with wood furring strips 11.3 8.2 Steel frame walls with wood frame 2 x 4s used at interface details 7.2 6.8 Steel and wood stud attached to 1/2-in OSB web 9.85 7.6 Steel frame with polyurethane insulation sprayed into cavity and fiberglass batts 11.1 8.2 Straw structural insulated panels** 16.5-16.7 15.7 Autoclaved concrete blockÝ  9.4 8.6 Insulating concrete formÝÝ  11.6-12.0 11.1 *Second 2 x 6 wall had slightly lower R-value due to presence of electric wiring and use of fiberglass insulation with lower resistivity (3.16 compared to 3.45 h · 12ft2 · °F/Btu · in). **Effective R-value (thermal mass) of straw SIP: 16.8-23.5, depending on climate. ÝEffective R-value (thermal mass) of concrete block: 12.1-16.8, depending on climate. ÝÝEffective R-value of ICF when thermal mass and airtightness is considered: 26-44, depending on climate.

At the Web site of Oak Ridge National Laboratory, calculating whole-wall R-values takes just eight simple steps.

The guarded hot box is a test apparatus that measures the thermal conductivity of full-size walls.

In the predesign stage, when homeowners or builders are weighing the relative merits of alternative wall systems, they must be able to accurately compare the R-values of those systems. To do this, they need to know not just the clear-wall R-value, which represents the wall containing the insulation, and the necessary structural members away from all interface details, but also the more representative whole-wall R-value (see Obtaining Whole-Wall R-Values). The whole-wall R-value accounts for all of the major thermal shorts at wall interfaces with windows, floors, ceiling, and other walls. Thermal shorts result in unwanted heat loss in the winter and heat gain in the summer. They also contribute to nonuniform interior surface temperatures, which can lead to ghosting and interior moisture condensation. Because it takes thermal shorts into account, the whole-wall R-value is almost always less than the clear-wall R-value.

Residential designers and building contractors generally understand how to take framing factors into account to calculate the whole-wall R-value of traditional dimensional wood frame walls. However, it is much more difficult to estimate the thermal performance of other wall systems accurately, particularly if several different system types are being considered at the building conceptual design stages.

Now the thermal performance of alternative wall systems can be compared fairly easily by anyone who has Internet access. An interactive calculation tool, Whole-Wall Thermal Performance Calculator, is available at www.ornl.gov/roofs+walls/whole_wall/wallsys.html. This tool accepts a simple description of custom building plans and enables the user to compare uniform whole-wall R-values of at least 40 different wall systems.

The Whole-Wall Thermal Performance Calculator draws on a database of whole-wall R-values generated through full-scale wall hot-box tests (see Table 1). More than 15 wall system manufacturers submitted 40 different wall systems for the hot-box tests. Each database entry is the result of testing that the Buildings Technology Center of Oak Ridge National Laboratory--a recognized, objective, and qualified third party--conducted in cooperation with the manufacturers of each wall system.

These tests generally validated the results from thermal modeling we also conducted, but occasionally generated unexpected findings. For example, we found that a hand-stuccoed straw bale wall had an R-value of 16, rather than the assumed R-60, which had been based on limited thermal resistivity measurements (see Refining Straw Bale R-Values, HE Mar/Apr '99, p. 13). Computational fluid dynamic modeling revealed that this lower R-value resulted from internal convection in the discontinuous gaps between the stucco and the straw bales and between the drywall and the straw bales. We had a second straw bale wall built that was mechanically stuccoed, virtually eliminating the air gaps. Testing of this wall showed that it had an R-value of 26.

Another result that was noteworthy, if not exactly surprising, was our finding that the whole-wall R-value of a 2 x 6 wood frame wall with R-19 fiberglass batts installed with rounded shoulders, 2% cavity voids, and the paper facer fastened to the inside surface of each stud was only 11. This whole-wall R-value derived from this worst-case typical installation of batts represents a 42% reduction from the R-19 value that the consumer may expect, based on the fiberglass batt's label. The seemingly insignificant insulation installation errors and thermal shorts resulting from interface details accumulate to significant impacts.

The database encompasses a fairly wide spectrum of wall materials. These include structural insulated panels (SIPs) with a compressed straw core, steel C-stud frame (2 x 6 and 2 x 4), metal frame walls with different types of sprayed foam insulation, various metal stud geometries, structural straw bale walls, and an insulating concrete form (ICF) wall.

1. Connect to the home page, www.ornl.gov/roofs+walls/.
2. Select Whole-Wall Thermal Performance Calculator from the list of interactive calculators.
3. Select a type of wall system--for example, steel stud.
4. This step is optional. View and download a copy of the interface details used for that system. These AutoCAD drawings are available both in line format and rendered to fill in objects on the drawings.
5. Select wall options such as stud spacing, finishes, insulation, and sheathing.
6. Select either a standard house with specified dimensions or a custom house.
7. If you selected a custom house, specify the building perimeter, ceiling height, number of sides on the building, and window and door areas. This information is needed to weight the thermal performance of interface details properly, relative to the clear-wall performance.
8. After one more page, the results will be displayed to show the clear-wall and whole-wall R-values. You can repeat the process to make as many alternative wall system comparisons as you like.

Obtaining Whole-Wall R-Values The ranch house, whose floor plan and elevation is shown above, serves as an optional reference house. The clear-wall R-value represents the thermal performance of the area of the wall containing insulation and only the necessary structural members. This clear-wall R-value does not take into account the impact on a wall's performance of the interface details. The interface details are the wall connections to such other envelope components as the wall-corner connection, the wall-floor connection, the wall-ceiling connection, the window surround, and the door surround. The whole-wall R-value gives a more accurate assessment of the actual thermal performance of a wall assembly because it reflects the weighted thermal performance of the total clear-wall area and the actual number of envelope interface details for any given user-input building plan and wall elevations. For instance, if a corner detail area has proportionally more highly conductive structural material than insulation, when compared to the clear-wall area, the whole-wall R-value will decrease relative to the clear-wall R-value. The percentage difference between the whole-wall R-value and the clear-wall R-value is an excellent metric to describe the severity of thermal shorts that exist in a wall system. The lower the percentage difference between these two R-values, the less thermal shorting exists. The complete whole-wall rating procedure provides a means to compare the performance of wall systems with respect to the following five elements: thermal shorts; exterior envelope thermal mass benefit, for walls with more mass than conventional 2 x 4 dimensional wood frame; air-tightness relative to typical wood frame opaque wall construction; moisture control; and sustainability to account for the relative total life-cycle environmental impacts of different wall systems. The procedure used tests the entire opaque wall portion of a residential building. The National Fenestration Rating Council provides the thermal-performance label for windows and doors. For illustrative purposes, a standard house is used to select the quantity of each interface detail and to present a set of comparable results. The reference house shown below is a 1,540 ft2 ranch-style home with four wall-to-wall corners, seven windows, and two doors. The one-story wall has a 164-ft perimeter. The first step in acquiring the whole-wall rating label was to construct and test an 8 ft x 8 ft clear wall section in a guarded hot box. A guarded hot box is a test apparatus that measures the thermal conductivity of full-size walls according to ASTM C 1363-97 (ASTM 1997). The results from the hot-box test are compared with a three-dimensional finite difference model, HEATING 7. This calibration check is a quality control exercise. If the modeling results and the hot box test results do not agree, then one or both sets of results must be inaccurate. Sometimes the assumptions made about the material properties of a wall being tested prove to be incorrect. In one case, we found that the metal in a steel frame assembly was made from a very high content of recycled materials, which gave the frame a lower thermal conductivity than we had assumed it would have. Once acceptable (within +5%) agreement is attained between the modeling and the test results, the interface details are modeled using the calibrated HEATING model for that wall. A database of validated thermal conductivities is generated for the clear wall and for all of the interface details for each wall system. The Whole-Wall Thermal Performance Calculator accesses this database.

Home Energy can be reached at: contact@homeenergy.org
Home Energy magazine -- Please read our Copyright Notice