The accuracy of three computer programs, DOE-2.1b, TARP84, and EMPS 2.1, for predicting the hourly energy use of passive solar residential buildings was investigated. Measured energy use and interior space temperatures from passive solar test buildings were compared with the results obtained from computer analyses of those buildings. The data base was composed of available data from unoccupied buildings that had passive solar components and had detailed, accurate measurements over a well-defined test period. The results indicate that TARP84 and EMPS-2.1 can accurately model most passive solar features, whereas DOE-2.1b is not as accurate for some high mass structures. Structures with partial shading of the high mass component were more accurately modeled by TARP84 and by EMPS-2.1 than by DOE-2.1b.

Units: Dual