| The wind tunnel method is commonly used for sampling odour from area emission sources. The performance of a wind tunnel in odour sampling is influenced by its design and aerodynamic characteristics. A laboratory study was conducted to evaluate the performance of a wind tunnel in odour sampling from a liquid emission surface. The wind tunnel evaluated consisted of an inlet bucket, a carbon filter, a fan, a PVC duct, an expansion section, a flux hood, a contraction section, a mixing chamber and two gas sampling ports. A perforated baffle was installed between the hood and the contraction section to facilitate creating uniform air flow in the hood. N-butanol solutions of various concentrations were used to simulate odour emission from liquid surfaces. An identical tunnel hood, which is 800 mm long, 400 mm wide and 250 mm high, was built with Plexiglas in order to perform velocity measurement and visual smoke testing. A total of 24 sampling ports were drilled for velocity measurement at the top surface of the hood at 200 mm, 400 mm and 600 mm of the hood length, 8 ports each length. The air velocity profiles were measured at bulk velocities of 0.20, 0.24, 0.30 and 0.38 m/s in the wind tunnel by inserting a hot wire anemometer at five different depth of each sampling port. The velocity profiles were deduced from the average velocities in the vertical and horizontal direction at 0.20, 0.24, 0.30 and 0.38 m/s bulk velocities. Smoke tests were conducted to visualize the air flow patterns in the tunnel. The flow pattern was recorded by a high speed motion camera and reviewed frame by frame. Computational Fluid Dynamics (CFD) modeling was performed to verify the flow patterns. The odour pick-up (recovery) rate was calculated from the mass balance of n-butanol in the solution and in the gas. N-butanol solution concentration was measured by using a spectrophotometer together with an ethanol UV testing kit and the gaseous concentration at the exit of the wind tunnel was measured by using a photo ionization detector.; The measured velocity profiles exhibited a substantial degree of non-uniformity in the hood while the velocity at the central hood was relatively uniform. The lower the bulk velocity, the less the velocity variation. The air flow inside the hood was turbulent, confirmed by the calculated Reynolds Number of 4820 at the lowest bulk velocity of 0.20 m/s in the experiment. The pick-up rate ranged from 37% to 56% using the current sampling system. Therefore, caution should be exercised when interpreting and using odour emission data obtained with wind tunnels. Even similar wind tunnel designs could yield different performance results. The odour emission rate measured by the wind tunnel increased with the bulk air velocity in the tunnel and the relationship could be described by the power function.; Keywords. Odour emission, Wind tunnel, Velocity profile, Pick-up rate... |
