This paper presents new evidence from a nationwide cross-project meta-study investigating the magnitude and extent of the difference between designed and measured thermal performance of the building fabric of 188 low energy dwellings in the UK. The dataset was drawn from the UK Government’s national Building Performance Evaluation programme, and comprises 50 Passivhaus (PH) and 138 non-Passivhaus (NPH) dwellings, covering different built forms and construction systems. The difference between designed and measured values of air permeability (AP), external wall/roof thermal transmittance (U-value) and whole house heat loss were statistically analysed, along with a review of thermal imaging data to explain any discrepancies. The results showed that fabric thermal performance gap was widespread especially in terms of AP, although the magnitude of underperformance was much less in PH dwellings. While measured AP had good correlation with measured space heating energy for PH dwellings, there was no relationship between the two for NPH dwellings. The regression analysis indicated that for every 1m3/h/m2 reduction in designed air permeability, the gap increased by 0.8 m3/h/m2@50Pa. Monte Carlo analysis showed that likelihood of AP gap was 78% in NPH dwellings designed to 5m3/h/m2@50Pa or lower. The study provides useful evidence for improving the fabric thermal performance of new housing through in-situ testing.
Gupta, RajatKotopouleas, Alkis
Faculty of Technology, Design and Environment\School of Architecture
Year of publication: 2018Date of RADAR deposit: 2018-03-27