Quantifying Dynamic Factors Influencing LEV Capture Efficiency

1705 

Student Poster 

P Moshele¹, P Raynor¹, S Arnold²

¹University of Minnesota, Minneapolis, MN, ²University of Minnesota, Minnetonka, MN

Ms Puleng Moshele, MS  
University of Minnesota

Peter Raynor  
University of Minnesota

Susan Arnold  
University of Minnesota

Local exhaust ventilation is one of the most widely used engineering controls for reducing worker exposure to airborne contaminants. However, the effectiveness of these systems depends strongly on operational conditions such as airflow velocity, source proximity, and surrounding air movement. This presentation reports controlled chamber experiments that quantify how ventilation parameters influence capture efficiency for a circular flanged hood. The findings provide empirical evidence that can improve how industrial hygienists evaluate ventilation performance and interpret engineering control effectiveness in occupational settings.

Airborne contaminants remain a major cause of preventable cancer and chronic respiratory disease, particularly in construction and manufacturing where workers rely on LEV for exposure control. Yet even "well‑designed" LEV systems frequently underperform in the field: capture efficiency can drop sharply when cross‑drafts, source distance, or hood geometry change, leaving exposures above recommended limits despite apparently adequate controls. Existing design guidance and most exposure models still assume ideal capture or rely on simple capture‑velocity tables, offering little quantitative support for how far a source can be from the hood, how much cross‑draft is tolerable, or how to represent imperfect LEV in one‑box and near‑field/far‑field models. New modeling approaches, such as extended one‑box models that include an explicit LEV capture efficiency term, require empirical inputs that are rarely available. This work directly targets that gap by generating controlled experimental data on capture efficiency across key dynamic factors, linking LEV performance to both exposure risk and the parameters needed for modern exposure models.

Experiments were conducted in an 11.76 m³ exposure chamber operated at six air changes per hour to approximate a well‑mixed room suitable for one‑box exposure modeling. A 6‑inch circular flanged hood was mounted above an evaporating acetone source delivered at a constant rate by syringe pump. Acetone concentrations were measured at the hood duct and general ventilation exhaust using calibrated ppbRAE 3000+ photoionization detectors, and three related outcomes were derived from a mass‑balance framework: capture efficiency (CE; fraction captured), a capture ratio (R; captured vs. escaped mass), and a relative capture efficiency metric (CE_rel) scaled for modeling applications.
Hood face velocity was characterized and calibrated using a TSI VelociCalc 9565 hot‑wire anemometer and an ACGIH log‑linear annular traverse for circular hoods. Face‑velocity traverses at multiple fan/damper settings were paired with duct static pressure readings from a DG‑700 manometer to generate a regression relating static pressure to face velocity and volumetric flow, enabling later control of target capture velocities from duct pressure alone.

Source distance, face velocity, cross‑draft velocity, and hood angle were then varied one factor at a time while other conditions were held constant, with three replicate trials per level when feasible. For each trial, paired hood‑duct and room‑exhaust concentrations were used to compute CE, R, and CE_rel, which were averaged to obtain trial‑level steady‑state values for each factor level. Distributions were inspected and transformed as needed, and one‑way ANOVA with Tukey post‑hoc tests was applied to each metric to compare levels and identify operational thresholds where capture performance changed meaningfully; these thresholds are now being used to select factor levels for a completed multifactorial experiment evaluating interactions among the same variables.

Face velocity showed a strong positive effect on all three metrics (CE, R, CE_rel), with capture performance generally increasing from low to high velocities and reaching the highest values at the upper test range. Source distance displayed a clear threshold: CE and CE_rel were very high at the closest distance (3 diameters), dropped sharply by about one hood diameter, and changed little at greater separations; R showed the same near‑versus‑far contrast with even stronger separation between levels. Cross‑drafts degraded capture across all metrics, whereas changes in hood angle from 0–60° produced only minor differences compared with the other factors.

These findings explain why LEV that satisfies nominal design criteria can still fail to control exposures when sources are placed too far from the hood or when cross‑drafts are not managed. Practical take‑aways are to prioritize maintaining adequate face velocity at the hood, keep sources within roughly one hood diameter, and minimize disruptive cross‑drafts before fine‑tuning hood angle. Because CE, R, and CE_rel are defined from a common mass‑balance framework, the results can be non‑dimensionalized (for example, by distance/diameter and cross‑draft/face‑velocity ratios) and used to develop predictive models that link these dimensionless groups to capture efficiency for extended one‑box and related exposure models.

Engineering Controls and Ventilation

Exposure Assessment

Work Environments, Occupations, and Industrial Processes