01 July 2004
Risk or consequences?
New classifying methods may offer options in U.S.
By Ellen Fussell
Say you are classifying your chemical plant as a hazardous location. According to the existing zone method of classification, once you classify your plant as Zone 0 (the most hazardous according to the zone concept), you must use intrinsically safe equipment (equipment ia)—no matter the potential consequences of an explosion. This would be true if you had the same chemical process in the city of Raleigh, N.C., and in the middle of the California desert. However, this wouldn't be the case under a newly proposed method the IEC TC31 standards committee is considering.
With the proposed method, if you had the same process in Raleigh and California, and they were Zone 0 in both cases, you might have future options in classifying your plant—based not only on explosion likelihood and ignition source, but on the consequences of that explosion.
The proposed approach is a decoupling of the prescribed type of protection for equipment from the area classification zone, said Paris Stavrianidis, vice president and general manager at FM Approvals in Norwood, Mass. "Existing standards look at classifying an area in terms of zones," he said. "The end user then installs the type of equipment prescribed for use in a specific zone. If you're an end user and you follow the standards, you come up with Zone 0, and the standard says you must use [intrinsically safe] ia equipment. That's a prescriptive approach," he said.
Under the current system, you would consider two components. The first is the likelihood of an explosive atmosphere—the zone. The higher the likelihood, the lower the zone number. "Zone 0 is the most likely to have a release," Stavrianidis said. The second component to consider is the likelihood of an ignition—the ignition source—present from equipment. "In the existing standards—using the prescriptive approach, if you have Zone 0, you must use equipment that has the least likelihood of creating an ignition source," he said.
The new alternative approach does the same thing, but adds a third component to the risk assessment: consequence of an explosion. "Whereas in the existing approach, you go as far as the likelihood of an explosion, explosion source, and ignition source, the new approach says if you add the consequence of that explosion, then you have a true measure of risk," he said. If the proposal is accepted by TC31 and the method is finally published as an alternative, any end users can use it, provided they employ the zone classification concept, the method is acceptable to the local authority having jurisdiction, and there are clear benefits to the end users and all other stakeholders.
The purpose behind the concept is to give end users options, not dictate the methods to classify their plants. "So the end user is the only one who can make these decisions," said Bill Lawrence, senior engineering specialist for hazardous locations at FM Approvals in Norwood, Mass. End users could include chemical plants, process plants, oil refineries—anyone that uses gas or dust that under the right conditions mixed with air creates an explosive mixture. "Oil and chemical plants are the two biggest players," he said.
With the newly proposed concept, even though hypothetical chemical plants in Raleigh and California might have the same zone classifications, they would have different consequences because one plant is in the middle of the desert. "If there's an explosion in the desert, all you'll lose is equipment," Stavrianidis said. "If there's an explosion in the middle of Raleigh, you'll be a bad neighbor to people around the plant. In this case the new approach says, given the fact you have fewer consequences from the explosion in California, you could use equipment that has a higher likelihood of ignition source, so it gives you more flexibility."
"Although I support the concept of giving end users more options for making their plants safe, I wonder how useful the proposed new method for selecting equipment for hazardous locations would be," said Bob Bailliet, president of Syscon Intl. in Kenner, La. "Many regulatory authorities prefer clear, cookbook rules, which make the regulatory compliance process more straightforward."
"Having options is good," said Paul Gruhn, president of L&M Engineering in Houston. "However, if the proposed methodology doesn't clearly give guidance on how to select likelihoods [of both an explosive atmosphere and ignition] and consequence [of an explosion], then there will be problems with repeatability—different teams will come up with different answers for the same problem—and uniformity—different companies will use completely different assessments," he said. "Also, will the authorities having jurisdiction be familiar with such risk-based approaches, as opposed to simpler prescriptive techniques? Hopefully the zoning community will benefit from some of the lessons learned by the safety system community."
Regulation vs. economics
The new concept started in Europe about ten years ago with the introduction of the ATEX directive, a European legal requirement for equipment in explosive gas atmospheres, Lawrence said. There is also an ATEX workplace directive for end users. The workplace directive requires end users to do a risk assessment. "As they started doing that in Europe, it started to ripple into the rest of the world," Lawrence said. "Most of these companies are multinational companies. They're looking at the same thing everywhere."
Europe's changes are driven by the ATEX regulations. "Part of the ATEX directive [EU regulation] requires the EU end users to perform a risk assessment for the processes to evaluate the potential risk," Stavrianidis said. "EU end users need a framework that employs risk assessment and risk reduction principles for hazardous locations." But end users actually drive the process. "In a global economy you look for economies of scale," he said. "The principle of one design used everywhere in the world is cost effective for most global companies across the whole supply and maintenance chain."
However, the focus of the new method in the U.S. is on "economic benefit, because the existing approach in the U.S. is still safe," Stavrianidis said. "The end-user community needs to make choices, but at end of the day, it's fair to say divisions have worked exceptionally well."
Because the National Electrical Code (NEC) included an alternative method of area classification to divisions—the zone system in 1996—the new concept isn't in place of the zone concept. It is actually "an expansion of the existing IEC standards that use the zone concept," he said. "The zone concept remains intact in the new proposed method."
"It only makes sense [to change in the U.S.] if you're building a new plant," Lawrence said. "And how many new plants have we built in this country lately? If you're going to go through the expense of reclassifying an existing plant from division to zones, you'll probably be looking for economic benefit in the end, but if you leave existing equipment in the plant, there's no benefit, so why spend money?"
Time will tell
Yet Lawrence said there's still room for the new concept to make its way into the U.S. in time. "If you're building or modifying a process, this concept might give you the ability to do something you might not be able to do otherwise," he said. "It might be enough impetus to switch from divisions to zones. But it's the end-user community that makes the decision."
"As we speak, most of the industry continues to use divisions in the U.S.," Stavrianidis said. "If end users in the U.S. are building a new plant, and if they want some scale of economies, they may choose to go to the zone concept. If the end users are employing the zone concept, and if they want to use the risk assessment method, it will be an alternative method. But we're not there yet," he said. "It's only a proposal to TC31 to develop this framework of risk assessment as an alternative method. It's agreed it's a good thing to do and someone should modify existing standards."
But it might be a five- to ten-year process, Lawrence said. "A lot of things have to change—like all the standards. Little pieces have to appear in equipment standards, larger parts in the installation standards," he said. "And what TC31 is looking at now is simply making a decision as to whether we should proceed with this, and then developing a road map to establish when, where, and by whom. There are at least 12 standards identified in the first pass that need some sort of changes."
So far in the U.S., the movement away from divisions has been insignificant, Stavrianidis said. "This concept of risk assessment exists in the U.S. as well. It doesn't just exist in Europe. If you have a chemical plant or refinery and you fall under OSHA or EPA rules, you have to perform process safety management [risk assessment]," he said.
But even though the U.S. uses the zone classification, it is still not on a large scale. "The whole thing is a work in progress," Lawrence said. "It'll be gradual. You won't just turn the switch on and be ready to roll."
"The bottom line is the new approach gives the end users additional options," Stavrianidis said. "They can go with the new risk assessment approach—if the standards are revised. The good thing about this approach is nobody is trying to prescribe to the end-user community how to run their plants. But it gives them options."
Risk assessment review
The new concept is a proposed introduction of TC31—the IEC committee on electrical apparatus for explosive atmospheres. The proposal is to introduce a risk assessment method and equipment protection level designation. The topic was discussed at several meetings in the last two or three years. Discussions have been taking place since the 1999 meeting in Sydney, Australia. An ad hoc committee looked at the concept of risk assessment and made a proposal to TC31. The last meeting took place in February of this year in London.
This committee proposal to TC31 was to incorporate the risk assessment method as an alternative method to existing standards. It says the risk assessment approach should be introduced as an option—an alternative method to the current prescriptive approach linking equipment to zones.
The alternative risk assessment method gives the end user more options to identify ways to reduce the potential risk associated with a specific process by adjusting the three major components of risk: (a) the likelihood of an explosive atmosphere (zones), (b) the likelihood of an ignition source present from the equipment—equipment protection level, and (c) the potential consequences of the explosion. An end user can adjust any or all three of these components to reduce the risk of an explosion to a tolerable level (defined by the end user or regulatory authority). In the existing standards the third component (i.e., consequences) is not an available option.
In the existing method, the zones and potential equipment are prescribed. For example, in Zone 0 you can only use equipment ia. In the new proposed risk assessment method, under certain conditions, you may also use equipment ib provided the risk of an explosion is tolerable to the end user and the regulators.