Assessment of airflow measurement uncertainty at terminal devices

Existing protocols for the inspection of mechanical residential systems poorly address both the assessment of uncertainties and recommendations or specifications for measurement methods and devices to be used to guarantee low measurement uncertainties. This paper gives the major elements of a new protocol developed within the Promevent project to overcome this problem. We have analyzed results from 180 airflow measurements performed in laboratory conditions in accordance with this protocol. The methodology developed to analyze uncertainties addresses errors due to repeatability, reproducibility and measurement method consistently with standard error propagation methods. Our analyses of laboratory results show that the measurement method often dominates the overall uncertainty. To contain the overall uncertainty within certain limits, several tables give the Maximum Permissible Error (MPE) of the measurement device as a function of the measurement technique and the geometry of the air terminal devices. These results show that in 6 out of 15 tested configurations, the measurement uncertainty cannot be contained within 15%. For the other tested configurations, the MPE must stay below 9-13% to contain the overall uncertainty below 15%.