West Palm Beach, Fl (PRWEB) January 13, 2007
Dust explosions are extraordinarily complex. They do not happen frequently enough to develop a meaningful statistical data base. Accordingly, prediction of occurrence and consequence is not always possible. For the owner/operator of a manufacturing plant handling combustible particulate solids, the best chance for success is to start with an explosion risk analysis performed by an expert in the field.
Bill Stevenson, VP Engineering for CV Technology, a company solely focused on explosion protection and mitigation, illustrates this point by describing an event where proper risk analysis could have prevented or significantly have reduced the damage caused by dust explosion.
The system in question was a central dust extraction system comprised of a large, filter-element dust collector with a smaller cyclone air-separator on the clean air return. The main dust collector housing was built to withstand the maximum explosion pressure of the dust and was fitted with isolation valves as follows: Between the main dust collector and the cyclone, a self-actuating, float type explosion-isolation valve; on the inlet side of the dust collector, a gate type explosion-isolation valve; and on the discharge cone, double gate-valves interlocked so that only one could be opened at a time. Below the double gate-valves there was a collection bin, and below the cyclone there was a rotary valve. On the discharge side of the cyclone there was a fan.
Shortly after startup there was a primary explosion in the cyclone. It spread to the large filter-element dust collector because the float type explosion isolation valve is a one way valve, in essence it is a check valve, and it was specified on the assumption that an explosion would initiate in the large dust collector, not in the cyclone. All of the gate type isolation valves allowed flame spread past them, which resulted in burning dust and flame through the ducting and into the work area and into the collection bin which was heavily damaged. The cyclone was destroyed.
The investigation of this incident revealed the following observations: Filter bags had been removed and not replaced. This led to a buildup of dust in the clean air parts of the dust extraction system. The cyclone fan is believed to have ignited the dust that became suspended in high concentration in the clean air side as a result of startup. All of the gate valves had dust deposits in the valve bodies and in the slide mechanisms that prevented them from fully closing. In the case of the gate type explosion-isolation valve, it was installed improperly which allowed it to harbor dust deposits. The end-stop of this valve was configured to signal closure of the double gate valves, but since it didn't fully close the needed signal was not available.
Clearly there were a lot of things that went wrong in this incident and I don't think any reasonable analysis could have foreseen all of them. Particularly in the case of human error (removing bags and not replacing them, and improperly installing a valve critical for safety) there is a clear disconnect between the intentions of the designer and the execution of his vision. On the other hand, there are several things that could have altered the outcome if a qualified risk analysis had formed the basis of safety for this system. First, allowing a dust explosion to reach maximum pressure is never as safe as taking steps to reduce pressure to the maximum extent possible. The dust in question here was an ordinary organic so there can be no argument that there were overriding safety considerations such as toxicity, or very low minimum ignition energy to justify specifying a containment strategy. Second, the choice of a float type valve between the main dust collector and the cyclone neglected to take into account that ignition could occur from either end. Third, the decision to cause both double valves closed only upon signal from the gate type explosion isolation valve rather than from separate controls, undoubtedly motivated by cost, was not wise
"A few simple changes can make a big difference in terms of the consequences in this event." States Stevenson, "Valves that can isolate two connected vessels from either end are readily available. Explosion isolation valves that cannot harbor dust deposits are available to ensure valve closure when needed irrespective of orientation. Separate controls for each valve location should be specified in most cases because it is not possible to predict the exact spot in the process train where ignition will occur. Most importantly, measures such as explosion venting or suppression are available to significantly reduce the maximum pressure in a vessel and doing so would have reduced the consequences of this event." He concludes, "Clearly having a strong vessel is only as good as the weakest component in the system. "
A qualified risk analysis can reveal more options to the owner/operator and could have formed a basis of safety that prevented much of the consequential damage. CV Technology Inc., of West Palm Beach, Florida is exclusively involved in the explosion prevention and protection business and is a strong proponent of risk analysis as the necessary first step to developing a sound basis of safety
CV TECHNOLOGY, INC
2580 Metrocentre Boulevard
West Palm Beach, FL 33407
Phone: (561) 683 - 1200