<<< Back to Success Stories and Case Studies Alliance Logo
Link to Printing InstructionsPrinting Instructions
Silica Dust Control Improvement: Grinding of Iron Castings with Portable Tools at Kennedy Valve Foundry in Elmira, New York


The control of airborne silica dust is a constant challenge in a sand casting foundry, where large volumes of sand are used in the molding processes. Kennedy Valve, a sand cast iron foundry in Elmira, New York that manufactures fire hydrants and valves for waterworks applications, developed an innovative approach to the ventilation of its grinding operations that helped overcome some of these challenges. This case study describes the approach and successes achieved by Kennedy Valve.

Kennedy Valve, a division of McWane, Inc., has more than 430 employees. McWayne has implemented a comprehensive environmental, health, and safety management system based on International Organization for Standardization (ISO) 14000, Occupational Health and Safety Assessment Series (OHSAS) 18001, and OSHA’s safety management guidelines. One of the foundations of this management system is the commitment to search for innovative engineering solutions that will reduce hazards in the workplace.

The demanding foundry environment presents unique safety and environmental challenges. To meet these challenges, Kennedy Valve involves all levels of employees in safety and health teams and other facets of the overall safety program. Health and Safety Department staff, along with external experts, coordinate all of Kennedy Valve’s industrial hygiene and occupational health programs.

The Problem: Search for an Exposure Reduction Approach

One challenge that frequently arises in a sand foundry is the control of airborne silica resulting from the chipping and grinding of castings, particularly when portable tools are used. While workers were protected in these jobs through a combination of personal protective gear (primarily respirators) and ventilation, Kennedy Valve searched for an engineering solution that would provide a more consistent and higher degree of protection for workers against overexposure to silica at these work stations. Finding an engineering solution proved especially difficult because of the limitations of current ventilation controls for this process. Ventilated tools have not yet proved feasible for this work, while the protection offered by stationary exhaust hoods, such as downdraft or backdraft benches, is limited if the stream of particles (grinding swarf) emitted from the tools cannot be continuously directed at the exhaust openings.

The Solution: New Approach for Ventilation Controls

Kennedy Valve determined that it needed to devise a new approach for ventilation controls for portable grinding tools on sand castings nearly three feet wide. Kennedy Valve augmented its technical team with a foundry ventilation consultant to investigate the feasibility of silica exposure reduction from the grinding process. While the foundry already used grinding benches with backdraft slots, Kennedy Valve sought more effective controls.

After a broad search of information on available ventilation control methods for grinding with portable tools, the team identified a ventilation approach which had been demonstrated to be effective in controlling emissions from another foundry process, called air carbon-arc gouging, conducted on work benches with steel castings. This method had been identified and documented by the National Institute for Occupational Safety and Health (NIOSH) as Case History #6 in their "Evaluation of Occupational Health Hazard Control Technology for the Foundry Industry" (NIOSH Publication No. 79-114).

The air carbon-arc gouging process is as least as difficult to control as portable grinding. The tabletop booth presented in NIOSH Case History #6 incorporated a wrap-around design, a three-foot diameter turntable for casting repositioning, and a unique way of introducing supply air so that it swept past the worker on both sides of the body (Figure A-1). This design appeared to incorporate the best features seen to date on a ventilated booth. The team decided to determine whether the design could satisfy the ventilation requirements for grinding castings at Kennedy Valve.

There was one design characteristic of the booth that NIOSH evaluated on fume-producing processes that appeared as though it could create a rebounding issue when applied to grinding. That characteristic was the use of spaced exhaust openings along flat collecting surfaces. As shown in Figure A-2, respirable-sized dust follows in the low pressure wake of the large (inertial) particles in the grinding swarf. If the large particles rebound off of a solid wall, the dust will rebound with them and head toward the worker's breathing zone.

An industrial ventilation designer working on the team who is also a firearms instructor offered a way to address this issue. He cited the method of stopping air rifle pellets using an energy-absorbing hanging curtain. In this case, if the grinding swarf impacted a hanging curtain, the large particles would be stopped "in their tracks" and be unable to rebound (Figure A-3). The fine dust particles at that point would be pressed up against the curtain. If vertical dividers were employed to restrict sideways air motion, the fine dust could be readily directed through suction into exhaust plenums both above and below the impact zone for the grinding swarf and be removed from the bench (Figure A-4).

Demonstration of Dust Control Effectiveness. Figure A-5 shows the control method selected for the demonstration phase at Kennedy Valve. A prototype grinding booth was constructed for the demonstration and tested in an isolated part of the facility to eliminate the potential for cross-contamination from any other silica-producing process. Figure A-6 shows the grinding booth following the prototype demonstration, in its current form for production grinding. Castings are moved on and off of the work surface by overhead hoist. After the castings are set down on the workbench, they may be rotated via a turntable for better access to surfaces to be ground, to assist the ergonomic aspects of the work, and to allow the grinding swarf to be directed as far as possible into the capture zones. The worker can initiate the turntable using a "bump" switch which does not require hand use and thus does not slow down the grinding operation.

In the process of evaluating the ventilated grinding booth, respirable dust was measured in the breathing zone of the grinding operator and in the general background air next to the grinding booth. For this purpose, two real-time particle sensors were used. Each sampling inlet was fitted with a cyclone to remove the non-respirable portion of dust. These instruments simultaneously logged respirable dust during grinding operations that were also video recorded. The particle sensor data was downloaded to a computer and graphed.

Since the particle sensors did not provide a real-time measure of silica in the dust, the data was used to evaluate general dust control from the process. Follow-up testing was therefore conducted to determine the time-weighted-average (TWA) of respirable silica dust throughout a workshift in which both respirable dust levels and silica content of that dust were measured and averaged.

The first testing was done on a single type of casting (i.e., large fitting with four-inch flanges) selected to eliminate variability of the work in the initial assessment. From a visual standpoint, the only obvious exposure was associated with the manner in which dust was emitted from the inside of the casting in a "chimney effect" during internal grinding. This dust was discharged very close to the breathing zone before it was withdrawn by the push of the supply air and the pull of the hood exhaust. The dust produced by one particular task was directed in such a way that the exhaust hood could not directly capture it.

The task in question involved grinding a portion of the casting which was overhanging the front edge of the bench. The grinding swarf was directed downward toward the floor. This type of grinding did not appear to affect the dust exposure to the same extent that the chimney effect did, but the dust induced into the grinding swarf was definitely fugitive to the ventilated grinding bench. For this reason and also because of the anticipated housekeeping issue associated with directing grinding swarf at the floor, the decision was made to alter the basic prototype bench to add a small hood in front of the grinding bench to capture the grinding swarf and the dust induced with it when grinding was conducted on the overhanging portions of castings.

The tests also included measurements made without the supply air operating. The absence of supply air almost doubled the average dust exposure level. It is postulated that supply air improved the speed of dust removal by pushing the air which was just in front of the operator into the capture hood, thus reducing the residence time of this potentially dust-laden air in the breathing zone.

After it was confirmed that the supply air was important to the protection offered by the ventilated bench, it was decided to try to optimize the interworking of the exhaust and supply airstreams to enhance the benefit received. It had already been established that testing would be conducted over a range of exhausts from roughly 2,000 to 6,000 cubic feet per minute (CFM). Dampers had been installed to adjust airflow rates and a flow meter was used to measure these rates. Three exhaust rates were tested and at each of these exhaust rates baby powder was used to set supply air at what appeared to be the most effective, non-turbulent pattern.

The time-weighted-average dust concentrations measured during grinding of the test casting at three different ventilation rates are presented in Figure A-7. The lowest personal dust exposure occurred at the middle ventilation condition. This finding was consistent with expectations for, and observations of, the capture efficiency of the grinding bench. At the lowest ventilation rate, evacuation of dust appeared to be complete but not rapid. The dust seemed to dwell above the bench briefly before being extracted. At the other end of the spectrum, the highest ventilation rate produced a near-turbulent condition which tended to spread the dust. Area samples taken simultaneously with the personal samples showed consistent readings for the lower two ventilation rates, with a sharp increase at the highest ventilation rate. This finding is consistent with the observations summarized above, which suggested that turbulence could lead to dust loss from the capture zone of the bench.

Full Scale Operations. After successful completion of the prototype test program, 15 production booths were constructed and installed in the renovated finishing area. These benches have consistently controlled silica exposures during grinding to below OSHA’s Permissible Exposure Level (PEL) for Kennedy Valve’s grinding needs when operating at exhaust rates down to 3,000 CFM and supply airflow rates at half of that flow rate. The supply air is provided by a tempered (i.e., heated in winter) makeup air unit with proportional heater control for steady temperature conditions.

The particles which are "stilled" through contact with the enclosure walls readily fall out into collection areas in the workbench. This feature has reduced the loading of abrasive particles on the filter media in the baghouse. Foundries considering this type of approach should perform prototype demonstrations before proceeding with full scale operations.

The expected effectiveness of this exposure reduction approach depends on a number of variables, including:
  • Extent of sand burn-in/burn-on on the castings.
  • Effectiveness of shot blasting in removing sand material from surfaces.
  • Amount of grinding performed on a casting.
  • Proportion of grinding internal to the casting.
  • Ability to perform all grinding on top of the workbench.
The Impact

Kennedy Valve workers performing this grinding have long been protected by ventilation controls, augmented as needed by respiratory protection. The improvement program described here has resulted in consistent levels of silica exposure below the OSHA PEL.

Although the engineering controls have been effective, the worker in Figure A-6 voluntarily continues to wear the air-supplied helmet. This worker and other workers appreciate the eye and face protection and the stream of air circulating around the head. In addition, spikes in exposure are not impossible with any local ventilation method applied to manual processes; the air-supplied helmet provides a safety factor against these spikes. In sum, Kennedy Valve is closer to its goal of state-of-the-art processes, including worker protection, for cleaning a variety of castings.

Source: Arne Feyling, Assistant General Manager; Mike Maziur, Plant Manager; Tom Shaw, Health and Safety Manager, Kennedy Valve, Elmira, New York

The views expressed herein do not necessarily represent the official position or policy of the U.S. Department of Labor (DOL).

-- As of June 2009.
Grinding hood with supply air plenums. NIOSH foundry ventilation assessment.Grinding hood with supply air plenums. NIOSH foundry ventilation assessment.
Figure 1. Grinding hood with supply air plenums. NIOSH foundry ventilation assessment.

Effect of inertial particle rebound from a rigid surface causing respirator particles to escape from the exhaust plenum.
Figure 2. Effect of inertial particle rebound from a rigid surface causing respirator particles to escape from the exhaust plenum.

Effect of inertial particle not rebounding from an energy absorbing surface causing respirable particles to remain in the exhaust plenum.
Figure 3. Effect of inertial particle not rebounding from an energy absorbing surface causing respirable particles to remain in the exhaust plenum.

Exhaust patterns in the grinding hood equipped with impact absorption curtains to trap respirable particles in the exhaust plenums.
Figure 4. Exhaust patterns in the grinding hood equipped with impact absorption curtains to trap respirable particles in the exhaust plenums.

Effect of supply air on worker protection from dust exposure.
Figure 5. Effect of supply air on worker protection from dust exposure.

Production module of grinding booth during grinding of a casting.
Figure 6. Production module of grinding booth during grinding of a casting.

Measurements of respirable particulate matter for three different castings and at different grinding bench ventilation rates.
Figure 7. Measurements of respirable particulate matter for three different castings and at different grinding bench ventilation rates.


Chart Title: Measurements of Respirable Particulate Matter
Chart Type: Vertical Bar
Chart Elements: 6 elements - Two bars each measured by respirable particulate matter (mg/m3) for three different castings at different grinding bench ventilation rates.
  • Exhaust (CFM) = 2,016
  • Supply Rate (CFM) = 1,554
  • Casting = 4" flange
    • Personal Air Sample = 0.754
    • Area Air Sample = 0.598
  • Exhaust (CFM) = 4,032
  • Supply Rate (CFM) = 2,190
  • Casting = 4" flange
    • Personal Air Sample = 0.551
    • Area Air Sample = 0.596
  • Exhaust (CFM) = 6,250
  • Supply Rate (CFM) = 2,684
  • Casting = 4" flange
    • Personal Air Sample = 0.813
    • Area Air Sample = 1.229