Using Measured Exponents and Fractional Leakage Flow to Characterize Air Leakage in Ducts

The issue of leakage in air distribution systems is one of the most underappreciated efficiency improvement opportunities, even though potential improvements are significant. One of the main reasons for this lack of awareness lies in the current ineffective practices for determining and addressing the energy consequences due to poor tightness. The airtightness of ducts is generally only evaluated based on classifications (e.g. leakage class or seal class), which are in turn used to estimate duct leakage flows during normal operation using an assumed operating pressure, and a fixed flow exponent of 0.65. This paper presents technical arguments for why it is worthwhile to characterize duct leakage with measured flow exponents and operating pressures, or with direct measurements of leakage flows during normal operation. The focus is non-residential buildings, for which there are large variations in duct pressures during normal operation, between different duct sections, and between testing/classification values and normal-operation values. Experimental evidence is presented to show that the current methodology is not suitable for describing the real behavior of ductwork. Two alternative methodologies are described, both related to expressing leakage in percentage terms with respect to the nominal flow rate. The difference between the two indexes is that the first one utilizes the estimated leakage at operating pressure, the other indicates the maximum pressure at which the system can operate in order to not exceed a predetermined percentage leakage value.