Errors in anisokinetic sampling for aerosol concentration measurements are affected by the shapes, wall thickness of the nozzles and size distribution of particles. Moreover, it is expected that the particle concentration distribution in the nozzle is not uniform. The numerical analysis were carried out to obtain the flow field around the nozzle using the finite element method in a potential fluid, and particle motion with the Runge-Kutta-Gill method. For superisokinetic sampling (
Vo/
Vs<1), the shape and thickness of nozzles had little affect on the sampling efficiency, whereas the effect was significant for subisokinetic sampling (
Vo/
Vs>1). The analytical results of the sampling efficiencies agreed fairly well with the experimental data for a Stokes number of
Sk(D)<0.5. The calculated particle concentration distribution showed a maximum at a dimensionless radial distance of around 0.7 for superisokinetic sampling, while it flctuated markedly near the nozzle axis for subisokinetic sampling.
View full abstract