An investigation of factors affecting the performance of laboratory fume hoods

A `user tracer gas test` was performed on laboratory hoods, with a human subject standing in front of the hood, to assess hood containment ability. The relationship of face velocity and cross draft variables to hood containment ability is investigated. The ability of these variables and other tests, such as smoke challenges or tracer gas tests performed with a manikin at the hood, to predict the results of the user tracer gas test is evaluated. All of the laboratory hoods tested in this study were identical bench top bypass hoods with horizontally sliding sashes. A face velocity traverse, cross draft measurements, a pitot traverse to measure exhaust flow, a smoke test, a manikin tracer gas test, and a user tracer gas test were performed on each hood in several different sash positions. Based on the data collected, face velocity, its distribution and variability, and the magnitude of cross drafts relative to face velocity are important variables in determining hood leakage. `Unblocked` vortices, formed such that no physical barrier exists between the vortex and room air or a person in front of the hood, are identified as important sites of leakage. For the hoods evaluated in this study, unblocked vortices were observed along the beveled side edges. The data support the hypothesis that in the presence of a person standing in front of the hood, leakage is more likely to occur if unblocked vortices are formed than if all vortices are blocked. Evidence suggests that cross drafts are more likely to cause leakage when flowing in a direction that may cause separated flow along a beveled edge of the hood and thereby augment the unblocked vortices along the edge. Results indicate that smoke tests, manikin tracer gas tests, and average face velocity all serve as useful monitoring techniques. Face velocity measurements and smoke tests, which are easy and inexpensive, may provide information which is as valuable as traditional manikin tracer gas tests