Measured and simulated thermal behaviour in rammed earth houses in a hot-arid climate. Part A: Structural behaviour

Abstract

Heating and cooling of residential buildings consumes around ten percent of the world's energy. One approach for reducing these costs is to exploit the high thermal mass of sustainable building materials, for example rammed earth (RE), for intelligent solar passive design. However, there is a lack of scientific evidence about the thermal performance of RE houses in real-world settings.This research investigated to what extent thermal performance in unconditioned RE structures in rural Australia can be captured by current accreditation software. Two custom-designed houses were built in the hot-arid city of Kalgoorlie-Boulder, Western Australia: one comprising traditional solid cement-stabilised rammed earth walls (RE) and the other walls with an insulating polystyrene core (iRE). Otherwise the houses were identical in orientation and design. The houses were instrumented to monitor indoor temperature and humidity conditions prior to and during occupancy. Results were compared to those simulated using cutting-edge assessment software BERS Pro (v4.3) as an example of that used for energy efficiency accreditation in Australia. This first paper in this series discusses the houses' construction and instrumentation and results obtained during the unoccupied period, i.e. those purely demonstrative of the structure's thermal performance. A second paper in the series presents data gathered during occupancy, to contrast occupant thermal comfort with that predicted numerically.Measured data showed that both houses performed nominally-identically: the houses did not receive any relative benefit from including iRE. Simulated data was also similar per house. However, measured performance did not match that simulated: simulated rooms had poorer thermal stability and lag and, consequently, exaggerated internal temperature variations. Collected data has been made publicly available for future analyses.