SAMPLING ERRORS IN CYLINDRICAL NOZZLES

This paper reviews publications on aerosol aspiration by axisymmetric tubes, a widely used form of practical sampler. Axisymmetric tubes are widely used, as a rule, in stack sampling and sometimes in other areas of aerosol sampling as well (e.g., workplaces, ambient atmosphere). Numerous reports on aspiration coefficients for particles sampled from disperse flows contain two contradictory viewpoints on the sampling efficiency at suction velocities exceeding that of wind: Although some authors claim that the sample representativeness worsens, others maintain that it is improved. Aerosol aspiration from calm or weakly turbulent air has not been investigated fully, despite the fact that the problem of determining sampling errors under such conditions is important in relation to occupational hygiene and environmentalmonitoring. Along with the analysis of the results published by other investigators (Davies et al., Vincent et al., etc.), this paper contains the axisymmetric sampler aspiration data obtained by us during the last 5-year period.Experimental evidence is given for the secondary aspiration of particles after their bounce or blow-off, not only from the front face of the sampling tube but also from its external side surface. This effect is responsible for the qualitative discrepancy between the aspiration coefficient values obtained by different methods. The sampling conditions, for which aspiration distortions can be compensated for by using the inertial aspiration coefficient calculated from conventional theory, have been determined for axisymmetric samplers. The aspiration coefficient dependences on the anisokinetic coefficient, Stokes number, sampler wall thickness, and yaw angle have been analyzed for the aerosol sampling from steady-state flows. Possibilities of using these dependences to estimate errors in sampling aerosols from flows with the wind vector fluctuating in direction and magnitude are discussed. The poorly predictable secondary aspiration and flow turbulence effects observed with thick-walled samplers are shown to invariably influence the aspiration coefficient, making correction for sampling errors extremely difficult.The inertial aspiration coefficient values measured for low-velocity wind and calm air have been analyzed. These results point to the not-so-obvious dependence of this coefficient on the sampling conditions. Experimental data are included, which make it possible to determine aspiration distortions at the orifices of samplers used with commercial aerosol analyzers. © 1990 Elsevier Science Publishing Co., Inc.