Evaluation of mixing downstream of tees in duct systems with respect to single point representative air sampling.

Air duct systems in nuclear facilities must be monitored with continuous sampling in case of an accidental release of airborne radionuclides. The purpose of this work is to identify the air sampling locations where the velocity and contaminant concentrations fall below the 20% coefficient of variation required by the American National Standards Institute/Health Physics Society N13.1-1999. Experiments of velocity and tracer gas concentration were conducted on a generic "T" mixing system which included combinations of three sub ducts, one main duct, and air velocities from 0.5 to 2 m s (100 to 400 fpm). The experimental results suggest that turbulent mixing provides the accepted velocity coefficients of variation after 6 hydraulic diameters downstream of the T-junction. About 95% of the cases achieved coefficients of variation below 10% by 6 hydraulic diameters. However, above a velocity ratio (velocity in the sub duct/velocity in the main duct) of 2, velocity profiles were uniform in a shorter distance downstream of the T-junction as the velocity ratio went up. For the tracer gas concentration, the distance needed for the coefficients of variation to drop 20% decreased with increasing velocity ratio due to the sub duct airflow momentum. The results may apply to other duct systems with similar geometries and, ultimately, be a basis for selecting a proper sampling location under the requirements of single point representative sampling.