Before completing a dust hazard analysis, conduct a sampling plan to improve prevention and mitigation strategies
By Timothy L Cullina, P.E., Fauske & Associates, LLC
As Published in the Chemical Processing Magazine Hazardous Dust eHandbook
If you were creating a feature film on combustible dust, which would be the prequel — the sampling plan or the dust hazard analysis (DHA)?
Combustible dusts are fine particulates that present a deflagration hazard when suspended in air. You’ve seen that movie and know the characters, plot and staging from all the reruns. Oxygen plays itself on the enclosed stage when the Fugitive Shape Shifter, interpreted by a local Particulate Solid, reprises the villain’s role, eluding good housekeeping and dispersing onto the scene where the hot Ignition Source and Friends spark about.
How will this episode end? Will local kid, Pea Solid, and the beautiful Eye Source make explosive fireworks together? Or will Pea Solid steer clear and go straight to complete the process? That depends. How well did the sampling plan and the DHA inform the facility?
Industry experts, a.k.a., the people who write the National Fire Protection Association (NFPA) Standards, recognized that a DHA needs direction, and a written sampling plan would provide a start when they included this requirement in Chapter 5 of the NFPA 652 Standard:
A sampling plan shall be developed and documented to provide data as needed to comply with the requirements of this chapter n(Hazard Identification) …shall include the following:
- Identification of locations where fine particulates and dust are present
- Identification of representative samples
- Collection of representative samples
- Preservation of sample integrity
- Communication with the test laboratory regarding sample handling
- Documentation of samples taken
- Safe sample collection practices
These writers must have thought of the sampling plan as the prequel. The sampling plan is like a movie screenplay and identifies the steps to ultimately managing the risks associated with combustible dusts. Ideally, the sampling plan should be devised and documented before valid samples are taken and tested; otherwise, it would be called a sampling history.
But, for most of us, life and work and dust are not that clean and simple. The problem is where and how to begin knowing that sampling mistakes and omissions can lead to lost time and wasted money — or worse, incorrect risk assessment and an unhappy ending.
Unless you’re in “the business,” you rarely see the screenplays to the movies you pay to see.
Is the same true of the sampling plan and the DHA? Screenplays describe the scenery, provide comments on expressions and give actors cues to movements and messages. Let’s imagine a sampling plan as screenplay.
THE SCREENPLAY
Here you’re the writer, the director and maybe even a minor character or an extra in the background, but, please, do not be a tragic hero.
Imagine the shot. You’re filming on location. The screen play identifies the locations and describes the scene laid out before the cameras:
- It takes place on the factory floor.
- Hot lights with illuminated dust halos hang above.
- White over grey two-tone ventilation ducts cross overhead, disappearing off stage.
- Silver sprinkler heads sprout from grey over black, water pipe blending into the dark ceiling above the out-of-focus small black tress-work that supports the roof.
- A pedestal fan whines off to the side.
- Movable three-step stairs provide access to the top of a round mixing vessel, hatch opened, faint vapors visible, gone briefly and then back again.
- Parallel trail marks reveal the wheeled path of the stairway across a well-dusted floor.
- Extras move about wearing dust masks and look away while pouring bags into other mixing vessels.
- Noise fills the air with a steady mechanical din.
The antagonist waits in the wings.
Quiet on the set… and … “Action!”
From stage left enters an orange-trimmed forklift sporting a beer keg-like fuel tank mounted behind the driver. It’s used to deliver a well-stacked pallet of 50-lb. bags held in place by a partially torn wrap of grimy, stretched plastic. Bags have been removed, and others of a different color have been added. The side bar instructs the camera to zoom in on one of the added bags neatly stenciled, WHEAT FLOUR. The bag below has a script type spelling out Maltodextrin.
“CUT!!”
Then the screenplay flops unforgivingly into the producer’s round file as he mutters, “Rubbish,” and then adds “No one, not even Harry Potter, is going to believe in evil flour.”
ASSESS A POTENTIALLY DANGEROUS SETTING
So, forget the analogy, but keep the vision and imagination. Light halos tell us dust is in the air. Wheel tracks from the mobile stairs indicate dust on the floor. Light dust clouds rising from the manholes during powder loading suggest failed or absent dust collection at the mixing vessels.
The descriptions in the screenplay (sampling plan) identify where the potential enclosures, ignition sources, fuels, dispersers and oxidizers are. Alfred Hitchcock called them “MacGuffins.” We call them sides of the explosion pentagon.
To know what and how to sample, identify and write down the combustible dust hazards that each process component makes possible, even if you think they are improbable. The components include ducts, conveyors, silos, bunkers, vessels, lights, fans, forklifts, the inside of the process equipment and the fugitive emissions that get out. This doesn’t mean testing everything, unless you really need to test everything. This is simply identification. Vetting comes later. Remember: no dust, no hazard.
Each facility compartment, whether building, room or box containing combustible particulate solids, should be evaluated. Attention also should be given to identifying hidden areas where accumulations may go unnoticed. This approach applies to the out-of-sight tops of tall equipment, ducts, ledges and pipes. Remember: dust hides.
The same can be said of oxygen and the ignition source. Because getting rid of the oxygen takes more than just a little effort, unless you’re applying a lot of effort on oxygen displacement, consider that it is present. Ignition sources are not always controlled as easily as we want and should almost always be considered a potential.
At each point through the process, identify whether a fire (no dispersion necessary), a deflagration (no enclosure required) or an explosion hazard may exist. Remember: know dusts, know hazards. In our screenplay analogy we’d make note of the following “MacGuffins”:
Know what happens to raw materials in the process by following the money. That is, follow the raw material, the products they become and the waste left behind. Does the material start as a powder or larger solids? Does it become a smaller particulate, potentially fugitive dust? Does the material change chemical makeup and change its hazard? Are there different sizes of the same material at different process stages? When is the material wet and when is it dry?
The dust has two environments — inside equipment and outside. Consider each environment on its own and with its own hazards. One process simply may be a series of different spaces with sources of fugitive dusts. The other, equipment process, is where material gets smaller or larger, wetter or drier, conveyed mechanically or pneumatically, heated or cooled, reacted, mixed, sieved, molded, cut or manipulated in other ways that may alter the material. Do process enclosures change volumes, head spaces, freeboard, pipe or duct sizes during the process? What fresh hazard are introduced or eliminated by any of these changes?
The NFPA Standards writers must have been thinking a sequel. Nothing states sampling must be before the DHA. If you have dust, inside equipment or out, it’s probably safe to assume that two and maybe even three of the five sides of the explosion pentagon are in place: the dust (fuel), its enabler, oxygen and a potential ignition source lurking nearby. Looking at each situation this way brings you face to face with your risk much more quickly, and while it may not be the deciding factor for testing or not, it should get you thinking about how to reduce the risk.
FOLLOW YOUR PROCESSES FROM START TO FINISH
Is there rail delivery of particulate materials to silos? Pellets usually are too large to present a flash fire or explosion hazard, but pellet handling creates a potentially ignitable dust. These pellets do not require independent sampling or testing, but if the dust is not contained and becomes a flash fire or an accumulation hazard, this location and material should be identified as a candidate fugitive material sample. This same particulate may be collected upstream in the process and therefore may not be required to be sampled at receiving. If this is the case, document it.
In the absence of fugitive dust, there still may be hazards during receiving. Silos that receive pellets also receive the dusts from pellet damage. Cloud formation will occur in the silo, making this dust a candidate for sampling and testing. Is this handling damage particulate the same or smaller? When the same material exists in several different places, the sample with the smallest size particle can represent them all.
The DHA should acknowledge that this is an explosion risk, and therefore ignition source control is critical to explosion protection. Bonding and grounding are important ignition source controls. Explosion venting can be provided but is not required by NFPA on enclosures with problematic, difficult-to-protect geometries, such as tall, narrow silos. The sampling plan won’t necessary address these protections, but knowing this advises the sample selection and testing.
A straightforward approach is to screen all area and process samples identified before collecting them to narrow the field based on different combinations of raw materials and particle size, before testing. Samples of fugitive dust should be collected and tested to identify the risk level in the areas that require concurrent housekeeping efforts to keep area accumulations less than 1/32 of an inch each day.
MIXING AND SIZE REDUCTION/GRINDING
For any number of unit operations, internal considerations will need to look at the potential for ignition by hot spots, tramp materials, sparks or other means to determine effective protection measures. If the fugitive material collected is not representative of the in-process particulate, then collecting an in-process sample is required, but only if this particulate is smaller.
DUST COLLECTION
Dust collectors typically have the finest dust clinging to filters. Sampling from the dust collector should include, if possible, dust from the filter and from the collection bin.
SAMPLING PROTOCOL CONSIDERATIONS
Careful notes should be taken at each sample collection time and location. Taking a picture makes good documentation reference. Include complete material identification, sample location, date, time, method of collection and any unusual conditions at the time. Material sample temperature and moisture content and ambient temperature and humidity may be useful for in- plant interpretations but is not required for dust explosion testing.
ALTERNATE TESTING STRATEGY
One strategy is to test the smallest particulate from each process material type of concern, but even this can quickly get prohibitively expensive if you process dozens or hundreds of materials. If all your materials are likely to have a Kst less than 200 b m/s, consider using that Kst value as the design parameter. Finally, a dust collector sample should be collected for testing.
RECOMMENDED TESTS
For materials that are of uncertain potential combustibility and explosibility, a go/no-go test series can be performed to make these determinations. When known combustible dusts are involved, this step may be unnecessary unless special circumstances exist such as a predominance of large particulate handled or samples mixed with large quantities of inert materials in which you want to establish that the sample is not combustible.
It must be representative of actual conditions, or there may be more than just regret if an explosion event occurs. Criminal neglect or intent investigations may follow. All samples should receive particle size analysis and moisture content determination.
Other common tests include:
- explosion severity test (KSt and PMAX–ASTM E1226),
- minimum ignition energy (MIE – ASTM E2019), and
- minimum explosible concentration (MEC – ASTME1515).
Additionally, you must consider whether other tests may be of value depending on the potential ignition sources in your facility. They may include layer ignition testing (LIT) if there are hot surfaces where dust may accumulate or the limiting oxygen concentration (LOC) test if this means of ignition control is a practical consideration for the unit operation. For nonconductive materials, static charging and relaxation parameters may be useful.
Do not hesitate to discuss your test results with your testing lab or other consultants and DHA providers to ensure your understanding of the results.
But wait! You might ask, if I figure this all out now, what does the DHA guy do? Good question. Clearly, the sample plan is like a first draft of the DHA. Sample plan development should anticipate the need to evaluate test results with the location, equipment and dust conditions identified such as when unacceptable airborne emissions and accumulations of fugitive dusts or process contained dust clouds and accumulations may be exposed to ignition sources.
From here, prevention and mitigation strategies can be developed. As the dust consideration complexity increases, the number of factors to consider also increases and brings the sampling plan and the DHA closer to a singularity.
If you enjoyed this blog, you might also enjoy reading about FAI's 3 step approach to dust hazards analysis.
