Gabriel Remion, Bassam Moujalled, Mohamed El Mankibi
Languages: English | Pages: 8 pp
Bibliographic info:
40th AIVC - 8th TightVent - 6th venticool Conference - Ghent, Belgium - 15-16 October 2019

The measurement of natural airflows is practically challenging. Driving forces that induce natural airflows are characterized by low pressure differences. Conventional airflow-meters would introduce pressure drops, which can significantly affect the flow pattern. Besides, the measurement of the flow crossing a window is difficult to implement and poorly reliable. Thus, indirect methods called tracer gas methods are widely used to bypass these difficulties, as they do not interfere with the flow pattern. They rely on the analysis of the evolution of the concentration of a tracer gas, injected before or during the measurement.  
However, tracer gas methods are subject to several uncertainty sources. To reduce the uncertainty due to concentration measurements, least squares regressions are often realised, which allow to smooth the measurement noise. If the regression is realised, airflows have to be stationary during the measurement, which is particularly questionable for natural airflows. Actually, for lack of better methods, these techniques are often used in natural conditions, assuming that the bias induced by the regression is inferior than the bias due to the measurement noise.  The aim of the present paper is to experimentally assess a dynamic airflow rate with the commonly used multi-points tracer decay method, which theoretically assumes a constant ACH. The variation of airflows is realised thanks to a mechanical controllable fan, which allows a direct measurement of the airflow in the extract duct to test the accuracy of the tracer gas-dynamic ACH measurement.