The calculation of design heat transmission coefficients is typically based on thermal performance data generated by laboratory instrumentation on samples that have been artificially conditioned. To verify the validity of this practice, an in-situ thermal performance experiment has been performed on an insulated low-slope roof system for 23 months, and laboratory analyses on laboratory-conditioned insulation samples of the product used in the construction of the roof system have been concurrently performed. Laboratory tests were also undertaken on samples of the insulation product removed from the roof system after the termination of the in-situ experiment. Comparisons of these results indicate that changes in the thermal performance of insulation products may not be adequately described by laboratory conditioning. Details of the instrumentation and data analysis for the in-situ experiment are described.

Units: I-P