Evaluation of a new ramping technique for duct leakage testing

Evaluation of a new ramping technique for duct leakage testing

TitleEvaluation of a new ramping technique for duct leakage testing
Publication TypeReport
Year of Publication2006
AuthorsIain S Walker, Darryl J Dickerhoff
Date Published09/2006
InstitutionLawrence Berkeley National Laboratory
Abstract

The DeltaQ duct leakage test has been developed over the past several years as an alternative to duct pressurization testing. Details of the development of the DeltaQ test can be found in Dickerhoff et al. (2004), Walker et al. (2004), and Walker et al. (2002). The DeltaQ test is one of the test methods included in ASTM E1554 "Determining External Air Leakage of Air Distribution Systems by Fan Pressurization" (ASTM 2003).

The DeltaQ test estimates duct leakage to outside under normal operating conditions, and separates supply and return leakage. The DeltaQ test also aims to reduce the time and effort required to leak test ducts. The first time saving is that it does not require all the registers and grilles to be covered. This is a big advantage in homes that have high wall-mounted grilles in two-story rooms that are difficult to access. In occupied houses, access is also limited by furnishings that also hide grilles from view such that they are not noticed by the test crew and therefore are not covered during the test. These uncovered grilles lead to overestimates of leakage. Secondly, if the supply and return sides of the system are to be measured separately, duct pressurization requires separating the return from the supply using internal blocking inside blower cabinetry that is difficult to install and monitor (in case the seal is lost during testing). Thirdly, to determine leakage to outside (this is the value required for energy loss calculations), duct pressurization requires the use of two fans — one to pressurize the ducts and one to pressurize the house, and these fans require synchronization. Finally, the DeltaQ test utilizes a blower door and simultaneously measures envelope and duct leakage. For weatherization programs and other building diagnosticians already measuring envelope leakage, the use of a single fan means that the additional effort to acquire duct leakage information is minimized.

As experience was gained with the DeltaQ test, we looked for ways to make the test faster, simpler and more robust. A couple of key issues have arisen as experience with DeltaQ testing was accumulated. Firstly, the use of distinct pressure stations limited test resolution in the pressure domain and led to the possibility of instability in the multivariate fitting required for the DeltaQ calculations. Secondly, adjusting blower door speeds to achieve the individual pressure stations made the test take longer than desired by potential users such as weatherization crews.

The purpose of this study was to examine an alternative DeltaQ test procedure and several data analysis techniques that would address these issues. The new test procedure does away with specific pressure stations. Instead, the blower door speed is gradually increased and the envelope pressure differences and airflows are continuously recorded. This continuous changing of pressure differences and airflows is referred to as "ramping". The ramping technique was evaluated using both laboratory and field testing. The laboratory tests were carried out under controlled conditions where the duct leakage was precisely known and there was no influence from wind and thermal pressures. These tests allowed us to separate the modeling errors in DeltaQ from errors arising in field measurements. The controlled laboratory tests also allowed evaluation of different data analysis approaches without the variability introduced by field testing. The field tests where the true duct leakage was unknown were used to examine the reduction in precision due to changing wind and thermal pressures on the envelope as well as experimental errors such as poor pressure tubing placement. The field precision estimates were developed based on repeatability testing in several houses.

LBNL Report Number

LBNL-61743