Why the Grinding LEV Stopped Working - and Why N95s Are Not the Fix

There are at least three separate failures feeding each other.

• Transport failure. 1,850 fpm is well under the 4,000 fpm needed to keep aluminum grinding dust suspended. The 2 inches of dust in the basement is the direct evidence: material is depositing in the duct, reducing cross-section, raising static pressure, and further reducing flow. That settled dust is also a re-entrainment source and, for aluminum, a deflagration hazard.
• Capture failure. 55–80 fpm at the hood face is inadequate for a high-velocity, high-momentum operation like grinding, and a hood without enclosure at that distance will lose the plume to crossdrafts.
• Building pressure. With a 6,000 cfm make-up deficit, the building runs under negative pressure - the whistling hatch and hard-to-open doors are the tell. Exhaust air has to come from somewhere, and it arrives as uncontrolled crossdrafts across the hood faces, further degrading capture.

Next measurements: fan static pressure and motor amperage compared with the fan curve; a pitot-tube traverse for total system flow; hood static pressure at each branch; capture velocity by smoke visualization; building pressure differential by manometer; duct velocity at each branch; and a combustible-dust characterization (Kst, MIE, layer ignition) plus personal particulate sampling.

The N95 proposal fails on several grounds.

• Feasibility. Engineering controls are clearly feasible here - this is a maintenance and balancing problem, not a technology limit. Where feasible engineering controls exist, respirators may not be used in place of them.
• It does not address the actual hazards. The settled duct dust is an explosion and re-entrainment hazard that no respirator touches. The eye irritation requires eye protection, not a filtering facepiece. The basement accumulation will continue to grow.
• Performance in this setting. A filtering facepiece on a hot, physically demanding grinding task, with facial hair and poor fit, may deliver a workplace protection factor of 5–10 at best, and only if a full written program (medical evaluation, fit testing, training, change-out, supervision) is in place - which it is not.
• It misdirects capital. Every dollar spent on respirators is deferred from the fix that would actually reduce exposure for all 14 stations.

Acceptable use: as an interim measure while the system is corrected, under a compliant program and with a written schedule for the engineering fix.

Corrective actions (in sequence):

1. Restore transport velocity - clean the basement runs, replace undersized branch ducts, and re-balance all branches with balancing dampers and static pressure taps.
2. Add make-up air to bring the building to roughly neutral (slightly positive), sized to the exhaust total, with tempering for winter conditions.
3. Replace or upsize the cyclone with a cartridge collector rated for the corrected flow; a cyclone alone is a poor choice for fine metal dust.
4. Redesign the three worst hoods as partial enclosures or downdraft tables to reduce the required capture velocity.
5. Institute a housekeeping program using vacuum (never compressed air) and a combustible-dust hazard review.

Verification/commissioning: re-measure hood face velocity, branch duct velocity, hood and fan static pressure, building pressure differential, and total flow against design. Then repeat personal sampling at the same SEGs and compare pre- and post-modification results. Document the as-built balance report; without it, the system will drift again within a year. Write the verification requirement into the purchase order.