This report is the first of three in a multi–year study comparing thermal performance of steel and wood framed houses conducted for the U.S. Department of Housing and Urban Development (HUD), the North American Steel Framing Alliance (NASFA), and the National Association of Home Builders (NAHB). This study is conducted by the NAHB Research Center, Inc.

Steel framing has been used for many years for interior non-load bearing and curtain walls in commercial construction. However, cold-formed steel members are only recently attracting attention for use in load bearing wall, floor, and roof framing applications in residential construction. Steel stud framing for residential building is gaining popularity due to consistently low material cost, simplicity of construction and similarity to wood frame assembly. Despite the availability of cold-formed steel framing, there are still basic barriers that impede its adoption in the residential market. The largest barrier is generally believed to be that the building industry is generally reluctant to adopt alternative building methods and materials unless they exhibit clear cost or quality advantages. A second barrier is the question of how the higher thermal conductivity of steel affects energy use in homes.

When building with steel framing members, it is necessary to compensate for the thermal bridging inherent in steel. If a structurally equivalent steel stud were to replace wood without consideration of thermal performance, the overall clear wall R-value of a wall can be reduced by 25 percent1 with a typical wall section.

The approach taken in Valparaiso was to build a wood house to local standard practices. A nearly identical steel house was also designed using the builders’ standard practices that required ¾” exterior foam insulation and 24” on center stud spacing (in lieu of 16” o.c.). The long-term (1-year) monitoring was designed to determine how these two houses perform thermally in a northern Indiana climate. Monitoring various temperatures and heating and cooling energy use during the test period in unoccupied houses are the basis of the evaluation. Appendix B and C contain graphs reflecting monitoring results for four seasonal months (July, September, January, April).