The Modification, Development, and Characterization of an Aerosol Dispersion Wind Tunnel for the Use of Candidate Air Sampler Testing

Before a candidate ambient air monitoring sensor can be used confidently in the field, it first must undergo characterization in a tightly controlled environment. To create those predictable testing conditions, the Aerosol Dispersion Wind Tunnel at the West Texas A&M University Environmental Quality Research and Technology Laboratory was physically modified to ensure sufficient mixing and fully developed flow at the testing section, then verified for acceptable velocity and aerosol concentration profile standards as specified by 40 CFR 53 Subpart D (USEPA, 1987). Using a particle image velocimetry system, the applicable cross section inside the test chamber was non-intrusively interrogated with planar sheets created by the Class IV Nd:YAG laser. By imaging the displacement of seed aerosols illuminated by the laser light sheet, nearly instantaneous velocity flow fields were calculated and time-averaged at 5 levels across the Y-axis of the test section, resulting in a high resolution velocity profile. These tests showed that the percent coefficient of variation (C(v%)) for the target velocities of 2 kilometers per hour and 8 kilometers per hour were 3.8 percent and 3.4 percent, respectively. To characterize the concentration profile within the applicable cross section, a vibrating orifice aerosol generator was used to introduce highly mono-disperse aerosols tagged with a fluorescent tracer into the wind tunnel at the same target speeds mentioned above. An adjustable isokinetic rake fitted with four isokinetic sampling cones was used to collect the aerosols of varying sizes upon glass fiber filters once isokinetic conditions had been established. The filters were then soaked in an aliquot, and a fluorometer was used to calculate the concentration of the fluorescent tag in each sample. From this the concentration profile within the tunnel was established. To be considered acceptably uniform, the C(v%) of the concentration values for each run could not exceed 10 percent. The aerosol concentration distributions within the test section proved more uniform when tested with middle range droplet sizes (7, 10, and 13 micrometers), while concentration values at the high and low ends of the droplet size range (3 and 17 micrometers) did not meet the 10% C(v%) standard.