Home

Personnel

Publications

Newsletters

Glossary

Bibliography

Calendar

Services

Products

Instructions,
Descriptions,
Manuals

Links

Design and Evaluation of an Aerosol Wind Tunnel for Applied Industrial Hygiene and Health Physics Studies

Mark D. Hoover, Edward B. Barr, George J. Newton, Carlos Ortiz*, N.K. Anand*, and Andrew R. McFarland*

A new aerosol wind tunnel has been designed, fabricated, and assembled for applied industrial hygiene and health physics studies at the ITRI.  This project is part of a cooperative program with Texas A&M University that takes advantage of their significant experience with the design and operation of aerosol wind tunnels.  This report describes the wind tunnel and presents results for its initial performance.

Figure 1:  Schematic diagram of the Lovelace aerosol wind tunnel.  A Stairmand Disk and a turbulence breaker provide conditioning of the aerosol flow, and clear polycarbonate windows allow access to and observation of the experiments.

A schematic of the new aerosol wind tunnel is shown in Figure 1. The tunnel is constructed from stainless steel to facilitate cleaning and prevent corrosion of its internal surfaces. Each section of the tunnel is mounted on its own support frame with adjustable leveling feet to correctly align the tunnel over its length of 12.5 m. Clear polycarbonate windows allow observation of experiments, visualization of smokes and other flow-tracer media, still or video photography, and remote sensing of particle size or velocity. The windows also provide physical access for installing and removing experiments, and are exchangeable so that different configurations of access ports and penetrations can be used to mount external sampling equipment and insert pitot tubes, hot wire anemometer probes, or aerosol sampling probes.

The flow arrangement for the tunnel is an open-loop design in which incoming air is drawn from the laboratory, and filtered exhaust air is returned to the laboratory. This minimizes the facility requirements for heating, ventilating, and air conditioning that would be necessary for an open-loop, once-through design in which incoming air is drawn from the laboratory, and filtered exhaust air is expelled to the outside environment. It also mitigates problems of heat buildup or humidity changes in a closed-loop, recirculating design. However, some care is needed to make proper background corrections, or to wait a suitable interval between tests with sulfur hexafluoride or other tracer gases that are not removed by filtration.

Air enters the tunnel through a 1.4-m wide by 1.4-m high section of high-efficiency particulate air (HEPA) filters. The filters provide particle-free room air for the experiments and prevent inadvertent release of test aerosols into the room. The HEPA filter module can be removed to accommodate studies that do not require particle-free air or do not involve the injection of special aerosols.

A 1.5-m long HEPA filter house with a 1.6-m by 1.6-m cross section is connected to the test section by a 0.8-m long transition duct. The filter house is serviced through a 0.55-m wide by 1.3-m tall stainless steel door and can accommodate two banks of HEPA filters in series, with a 1-m separation distance between banks. The tunnel is currently configured without the first HEPA bank, which provides adequate space for placement of a full-torso, anthropomorphic manikin in the filter house. The manikin can be equipped with personal samplers, surrogate oral and nasal airways, or a combination of both.

The clean exhaust air from the filter house is passed through a 1-m long silencer to isolate the experiment section from any noise produced by the blower section. Flow through the tunnel is provided by a belt-criven, 45-horse power (hp) blower, connected to a 15-hp, variable speed motor with digital frequency control. A second silencer is located on the exhaust side of the blower to minimize propagation of noise into the laboratory.

The filter module at the entrance of the tunnel is followed by an aerosol generation chamber (1.6-m high by 1.6-m wide by 1.4-m long) in which an aerosol generator can be placed, or into which aerosols can be injected. Replaceable, 0.4-m by 0.9-m windows at the top, bottom, and sides of the generation chamber provide access for inserting and operating the generator or injecting the aerosols. Initial studies are being conducted with a vibrating orifice monodisperse aerosol generator (VOMAG, Model 3053, TSI Inc, St.Paul, MN).

After leaving the generation chamber, the aerosol travels through three flow-conditioning chambers before entering the experiment test section. Each conditioning chamber is 1.2 m long and has a 0.6-m x 0.6-m cross section. The aerosol generation chamber is connected to the first conditioning chamber by a short (0.35-m long) transition section that includes a bulk-head plate with a 30-cm diameter opening. Aerosol entering the first conditioning chamber is dispersed by large-scale turbulence when the airflow contacts a 35-cm diameter disk mounted 13-cm downstream from the opening. This type of mixing plate (known as a "Stairmand disk") improves the spatial uniformity of aerosol leaving the generation chamber. Further mixing and straightening of the flow occurs when the air passes through a small-scale turbulent mixer as it enters the second conditioning chamber. This breaks up the large scale turbulence and creates a relatively flat velocity profile across the entire cross section of the duct. The small-scale turbulent mixer consists of a bulk-head plate with 64 equally spaced, 5-cm holes. The flow is then allowed to establish a uniform profile as it proceeds, without further disturbance, through the second and third conditioning chambers. Particle behavior can be studied in the conditioning chambers, depending on the turbulent conditions being modeled. The Stairmand disk assembly and the small-scale turbulence plate can be reconfigured or removed through the access windows to provide different mixing conditions or to match the mixing conditions to a range of flow rates.

The experiment test section is 1.4 m long and has a 0.76-m by 0.76-m cross section. These dimensions are suitable for testing personal samplers, anthropomorphic manikins of the nose and head areas, extractive stack-sampling probes, and most workplace aerosol collection devices.  The conditioning chambers provide a uniformly mixed aerosol with a relatively uniform velocity profile over the cross section of the test chamber.  A very uniform velocity profile can be achieved across the central region of the test chamber by placing a 0.3-m-long, bell-shaped, reducing inlet at the entrance to the test section.  The bell can be inserted or removed through the access windows.  The entrance diameter of the bell is 0.6 m, and the exit diameter of the bell is 0.3 m.

Performance tests are currently underway to meet the U.S. Environmental Protection Agency guidelines for aerosol wind tunnel performance (Ambient Air Monitoring Reference and Equivalence Methods, 40 CFR 53, Subpart D, 1995). Uniformity of aerosol concentration entering the first conditioning chamber is strongly dependent on proper positioning of the aerosol generator outlet on the centerline of the inlet to the Stairmand disk. The blower motor frequency has been tested between a minimum frequency of 10 Hz and a maximum line frequency of 60 Hz, resulting in nominal flow velocities at the outlet of the 0.3-m diameter flow reduction bell of 3.3 m/s to 24 m/s, respectively, with velocity coefficients of variation (COVs) less than 2%. For the 0.76-m x 0.76-m test section, the velocity is 0.7 m/s to 3.6 m/s for frequencies up to 43 Hz. Tracer gas concentration profiles had COVs less than 10% for the entire test section. For particle concentration profiles, the COV is less than 16% over a 20-cm x 20-cm cross sectional area in the test section just after the flow reduction bell and less than 11% over a 30-cm x 30-cm cross sectional area in the center of the 0.76-m x 0.76-m test section without the flow reduction bell. This performance is suitable for workplace aerosol studies and meets the EPA requirements. Velocity profiles and particle and tracer gas concentrations are also being evaluated at other locations in the tunnel, such as in the conditioning chambers, to provide a more complete description of tunnel performance.

As an initial application of the tunnel to a practical problem, inlet efficiency tests were conducted at a wind speed of 1 m/s with 10 um aerodynamic diameter particles for the inline sampling head of the Eberline Alpha-6 continuous air monitor to qualify its use as a room area monitor.  At a flowrate of 2 cfm, aerosol collection efficiencies (mean +/- standard deviation for four tests at each orientation) were 84% +/- 2% with the inlet tube facing toward the direction of flow, 84% +/- 2% with the inlet tube perpendicular too the direction of flow in a horizontal direction, and 84% +/- 6% with the inlet tube facing downstream of the direction of flow.  As Expected, particle collection efficiencies were slightly better at lower flowrates: 87% +/- 3% at 1.5 cfm, and 89% +/- 2% at 1 cfm, for the inlet tube facing toward the direction of flow.  Note that previous work (1990-91 Annual Report, p.9) found the internal delivery efficiency for the inline head (compared to an inline filter) to be essentially 100%, which indicates that particle collection losses for use of the inline head as a room area monitor are associated with aspiration of particles into the sampling inlet.  These results far exceed the generally accepted minimum performance criteria of 50% efficiency for particles with 120 um aerodynamic diameter.  In addition, the performance of the inline sampling head as a room area monitor is very similar to the performance of the radial-inlet version of the Alpha-6 CAM, which is in the range of 83% to 90% collection efficiency for more orientations (A.R. McFarland et al. Health Physics J. 61(1): 97, 1991).

The new aerosol wind tunnel provides a state-of-the-art capability for studying aerosol collection efficiencies and particle deposition patterns in room aerosol monitors, personal samplers, extractive sampling probes, and geometrical configurations representative of human oral and nasal breathing.  It can also be used to study resuspension rates for particles attached to surfaces and patterns for aerosol mixing and movement around subscale models of workplace objects such as glovebox enclosures.  Results from such studies are needed to improve the technical basis for measuring, modeling, and mitigating toxic aerosols in the workplace and environment.

Reference:  Inhalation Toxicology Research Institute Annual Report 1995-1996, ITRI-148, prepared for the U.S. Department of Energy under Contract DE AC04 76EV01013, Albuquerque, NM, December 1996.