Author: Anton TenWolde

Publisher:

Published: 1997

Total Pages: 48

ISBN-13:

This paper describes a mathematical model developed by the Forest Products Laboratory to predict attic temperatures, relative humidities, and roof sheathing moisture content. Comparison of data from model simulation and measured data provided limited validation of the model and led to the following conclusions: (1) the model can provide reasonably accurate estimates for temperatures of roof sheathing and attic air, although heat storage effects often cause delay of 1 to 2 h in attic air temperatures; (2) the model can accurately predict the frequency of occurrence of high roof sheathing temperatures (> 120ÃF (49ÃC)) during summer, but accuracy is highly dependent on solar absorptance and emissivity values of the roof shingles; (3) the model consistently overpredicts the extent of night-time cooling from sky radiation losses, leading to predicted temperatures that are too low; (4) treatment of the effect of snow cover is too simplistic, but no better alternatives are apparent for simulating this very complex behavior; (5) the model apparently can predict average moisture conditions in the sheathing with reasonable accuracy, generally within 1% moisture content, when moisture content is not excessively high or low; and (6) hourly moisture behavior is not represented as well as is daily or seasonal behavior, especially for north-facing sheathing. The model would benefit from verification with data that include measured emissivity and solar absorptance of the shingles, addition of thermal mass in attic and roof, better algorithms to calculate direct and diffuse solar radiation, and verification for roof with east--west orientation.