Reducing Uncertainty for the DeltaQ Duct Leakage Test
The thermal distribution system couples the HVAC components to the building envelope, and shares many properties of the buildings envelope including moisture, conduction and most especially air leakage performance. Duct leakage has a strong influence on air flow rates through building envelopes (usually resulting in much greater flows than those due to natural infiltration) because unbalanced duct air flows and leaks result in building pressurization and depressurization. As a tool to estimate this effect, the DeltaQ duct leakage test has been developed over the past several years as an improvement to existing duct pressurization tests. It focuses on measuring the air leakage flows to outside at operating conditions that are required for envelope infiltration impacts and energy loss calculations for duct systems. The DeltaQ test builds on the standard envelope tightness blower door measurement techniques by repeating the tests with the system air handler off and on. The DeltaQ test requires several assumptions to be made about duct leakage and its interaction with the duct system and building envelope in order to convert the blower door results into duct leakage at system operating conditions. This study examined improvements to the DeltaQ test that account for some of these assumptions using a duct system and building envelope in a test laboratory. The laboratory measurements used a purpose-built test chamber coupled to a duct system typical of forced air systems in US homes. Special duct leaks with controlled air-flow were designed and installed into an airtight duct system. This test apparatus allowed the systematic variation of the duct and envelope leakage and accurate measurement of the duct leakage flows for comparison to DeltaQ test results. This paper will discuss the laboratory test apparatus design, construction and operation, the various analysis techniques applied to the calculation procedure and present estimates of uncertainty in measured duct leakage.