Local ventilation effectiveness of single-zone naturally ventilated spaces

The greenhouse gas (GHG) emissions and energy demand from the building sector have been rapidly increasing since the 1970s. In many developed countries, energy use in buildings has risen above the industrial and transport sectors. As building envelopes become more efficient at reducing fabric transmission losses due to better thermal insulation, it is expected that ventilation heat losses will become the dominant factor in low energy demand building design. In order to prevent excessive ventilation heat losses and ensure good indoor air quality (IAQ), the supplied fresh air should be distributed effectively throughout the space.

The IAQ related local ventilation effectiveness (VE) relates to how effectively the fresh air is distributed in the space. Two types of metrics are defined in the literature to investigate the IAQ related local VE. These are the contaminant concentration and age of air. Whilst the contaminant related metrics are sensitive to the contaminant strength and distribution in the space, the age of air related metrics are only sensitive to the local air flow-field vectors, making them ideal for use at the initial design stage when the detailed information regarding the contaminants are often unknown. The only age of air related metric investigating the VE locally (i.e. specific location in the room) is the local air change index (LACI). In the LACI metric, the local mean age of air (LMAA) at the exhaust is compared with the LMAA at a point in the space. In the literature, specific LACI values are associated with distinct air flow patterns. For instance, a 100 % LACI value indicates a perfectly mixed environment and 200 % LACI corresponds to an ideal piston air flow scenario. There is a consensus in the literature regarding the importance of the local VE on IAQ and energy efficiency, which is consolidated further by design guidelines (CIBSE Guide A (2016) and ASHRAE 62.1 (2016)) where the local VE is used as a factor in minimum fresh air requirement calculations.

There is a relatively large body of literature investigating the local VE of MV systems, whilst the equivalent research on local VE in NV buildings is lacking. Globally, NV is one of the most commonly used and effective low energy ventilation design solutions; therefore, it is important to investigate the local VE of common NV applications and propose robust new design suggestions to improve existing standards. Consequently, this study aims to quantify the local VE of typical NV solutions, identify the influencing parameters of local VE and provide design suggestions for ensuring a high local VE and consequently good IAQ and energy efficiency in NV spaces.

This research employed computational fluid dynamics (CFD) parametric modelling supported by full-scale experimental data. The experiments took place in a fully controlled environmental chamber at the University of Sydney, where the NV was achieved by top-hung and side-hung windows in a single-zone space, making the test configuration relevant for many spaces found in domestic and non-domestic building sectors. As a result of modelling 120 variations of particular single-sided and cross-flow NV configurations tested in the experimental facility, the local VE of such spaces was quantified and influencing parameters were identified.

This research has contributed to the existing body of knowledge by providing three main findings: (i) measuring the tracer gas concentration at all open windows in NV local VE experiments is essential to calculate the LMAA hence LACI values accurately; (ii) the local VE (i.e. median LACI value) of the tested single-sided NV scenario was 103 % (with standard deviation (SD) of 7.5 %) and 101 % (with SD of 8 %) at breathing heights for seated and standing occupants respectively; and similarly, the local VE of the tested cross-flow NV scenario was 102 % (with SD of 45 %) and 99 % (with SD of 41 %); and finally, (iii) the local VE of cross-flow NV is significantly more sensitive to the tested parameter compared with the single-sided NV.