Industry update on arc flash studies and proposed revisions to safety standards
On April 16, Steve Woodring, chief plant operator at the Harrisonville Wastewater Treatment Plant in Harrisonville, Mo., was working in the plant's main lift station, which contains a split wet well with two submersible dry water pumps. Normally, these pumps take care of the flows to the wet well, but when they get too high, the overflow is directed into a larger pump gallery. On this day, the split wet well was flooded, and one of the two submersible dry pumps was reading inoperable — the monitoring system was keeping the 30-hp variable-frequency drive (VFD) from starting it. While Woodring waited for the pump repair people to arrive, he decided to try to bypass the VFD and engage the pump, running it for a few minutes so he could hook the bail, which had come detached from the rope used to track it, to the boom truck to draw the water down. “The controls on these pumps are extremely touchy,” says Woodring. “The slightest bit of moisture or variance in ohm readings will show a problem. It's not uncommon for this system to cry wolf.”
Standing right in front of the closed circuit breaker enclosure, Woodring bent down to reach the bypass switch and started to operate it when there was a loud noise in the cabinet, and the power to the plant shut off. “Just as soon as it began to make contact, there was a humongous bang in that cabinet like a shotgun went off,” says Woodring. “This particular station doesn't have skylights, so when the power goes off, it's dark as a dungeon. It took me a couple of seconds to realize I was all right.”
There was no flash, no smoke, and no smell. The generator started up, restoring the lights, and the working pumps began to operate. After a few minutes, Woodring performed an assessment of the facilities. Initially, no damages were found. He used a screwdriver to open the VFD, which was cool to the touch. There weren't any dropped fuses. After a few minutes, he opened the cabinet. No signs of odor or smoke there either. “I didn't even lose a light bulb,” he says.
A closer inspection by the city's electrical crew uncovered that the main 800A breaker was tripped. According to specifications for this breaker, at 480V, it holds an interrupting rating of 65,000A. “The station should have been destroyed, and I should have been killed,” says Woodring.
He credits the bank of eight arresters primarily installed as protection against lightning and ground-fault incidents at this lift station with saving the plant as well as his life. “None of us planned on giving the arresters this ultimate test,” he says. “It didn't just lessen the arc flash, it prevented it.”
In subsequent investigations of the incident, damage has been found. On the day of the incident, the pump repair people eventually arrived and pulled the inoperable pump. Essentially, the pump was welded to itself. “It was dead,” says Woodring. “It was shorted on all three legs.”
In addition, an electrician checked out the VFD, wiring, and circuit breakers. The contacts were welded shut. “Basically, the fault came backward,” says Woodring. “It went through the input side, and the arresters cut it off.”
On average, approximately five to 10 severe arc flash explosions occur in electric equipment every day in the United States, sending their victims to a hospital or special burn units, according to statistics compiled by CapSchell, Inc., a Chicago-based research and consulting firm that specializes in preventing workplace injuries and deaths. Each year, more than 2,000 people are treated in burn centers with severe arc flash injuries. This number doesn't include cases in which the victim is sent to an outpatient clinic for medical treatment or is unharmed, as was the case with Woodring.
An arc flash is an electric current that passes through air when insulation or isolation between electrified conductors is no longer sufficient to withstand the applied voltage, according to the National Fire Protection Association (NFPA), Quincy, Mass. Arc temperatures can reach 35,000°F, and, as a result, can cause severe burns. An arc flash also generates an enormous explosive force. Even a relatively small 10,000A arc at 480V can create an explosion equivalent to 8MW of power. Unlike electrical shock incidents, a victim of an arc flash does not have to touch live components to sustain an injury. Other injuries from arc flash include blindness, hearing loss, nerve damage, and cardiac arrest.
According to a study conducted by the U.S. Bureau of Labor Statistics, 2,287 U.S. workers died and 32,807 sustained lost-time injuries due to electrical shock or burn injuries over a 7-yr period starting in 1992. The study showed that of the 32,807 non-fatal injuries involving lost time, 38% were classified as electrical burns. The report concluded that: “To decrease the number and severity of non-fatal electrical burn injuries, direct worker exposure to electrical arc energy must be reduced.” In other words, mitigation strategies are key to reducing the number of these incidents.
Ideally, the best way to prevent an arc flash event is to work only on or near de-energized equipment, but sometimes this type of work can't be avoided. On the day of the incident, Woodring was aware of the dangers of arc flash. It's the plant's policy to de-energize equipment before working on it when possible. However, there have been times when actual practice goes against this policy. “I've checked on simple things, like a relay or a fuse, before calling an electrician,” confesses Woodring, who, on those occasions, would open up the cabinet and re-energize to check if power was going to the fuse block, for instance. “Fuses can be checked when power is down, of course,” admits Woodring. “But there are just some instances where there are things to check when it's energized.”
An arc flash analysis must be performed prior to allowing personnel to work on energized equipment or even when performing a voltage check to confirm de-energization. The analysis defines the flash protection boundary distance and the type of personal protective equipment (PPE) required. On the day of the incident in Harrisonville, Woodring was not wearing any PPE. “That should not have been attempted without proper protective gear on,” he says.
Protective relays that detect arc flash conditions and send a trip signal to a circuit breaker to interrupt current can be installed on electrical systems. They are often installed in facilities that must remain running 24/7, such as power plants and industrial plants. Some of these arc flash detection devices use the arc's bright flash against it. When the devices' fiber-optic sensor detects the flash of light at the same time as the overcurrent, the relay issues a trip signal to the circuit breaker. That removes the voltage and current from the circuit, which, in turn, reduces the amount of interim energy from the arc flash.
These devices don't eliminate the use of PPE, but they reduce the category required by NFPA 70E, “Standard for Electrical Safety in the Workplace,” which is widely cited by OSHA and covers safety-related work practices, including employee training, hazard/risk evaluation, establishing electrically safe work conditions, establishing approach boundaries, and selecting and using PPE and flame-resistant (FR) clothing. A Hazard Category 2 requires lighter weight PPE than a Hazard Category 4.
Woodring is going to begin requiring members of his crew to wear the proper PPE when they're checking equipment. “Even if it's just one set, we need to have it,” he says. “That policy was just not in place before, and that's unfortunate. PPE should be used no matter how well you think this is going to work or how secure you think the system is. If the worst thing that happens is somebody gets hot and sweaty from having the gear on, that's okay.”
In January 2005, the Institute of Electrical and Electronics Engineers (IEEE), Piscataway, N.J., and the NFPA formed the Arc Flash Collaborative Research Project to gain deeper insight into arc flash phenomena and the hazards they pose for those working on or near electrical equipment operating at or above 50V. This collaborative effort brings together the two organizations that have been at the forefront of electrical safety issues and, more importantly, that will apply the knowledge gained to create practices and standards that enhance workplace safety.
The $6- to $7-million initiative supports research and additional testing to increase the understanding of a variety of issues related, such as heat and thermal effects, blast pressure, sound, and light intensity, which need further research and testing validation to provide relevant information that can be used to develop safety strategies that will protect workers, according to the program's project manager, Dr. Wei-jen Lee, University of Texas, Arlington, and Ben Johnson, co-chair of the program's Steering Committee, IEEE Life Fellow, and VP of Thermon Industries, San Marcos, Texas.
In 2008, four high-power test/laboratory facilities — three independent test laboratories and one corporate sponsor of the project — were selected. The project management team identified these third-party-certified, independent test laboratories for quality assurance, qualified personnel, previous experience, and adequate facilities for arc flash testing. In February 2010, a fifth independent test laboratory was pre-qualified through a similar process. These test laboratories were invited to submit technical and cost proposals for RFPs issued for the arc flash collaborative research project.
The first round of scouting tests, otherwise known as Phase 1, was designed to check the instrumentation functionality, sensitivity, overall measurement accuracy, repeatability of experiments, and consistency of test results with identical test protocols and arrangements in different environment and independent test facilities. Phase 1 test results on heat flux and thermal energy measurements have been verified for accuracy and conformity and justified (or compared) with the various theoretical models. Next, an expanded set of similar tests for Phase II will focus on quantifying the significance of many of the variables that influence the energy of an arc fault event, such as operating voltage, 3-phase bolted fault-current level, air-gap length, type of enclosures, and electrode orientation.
In order to maintain consistency, the project managers worked with the test facilities personnel to verify the test setup and witness the calibration procedure and the actual tests. It was the intention of the proposed testing to discuss, when appropriate, with the test laboratories in advance and ensure that all the testing is done under exactly similar test setup.
Analyses and validation of test results were performed by the IEEE/NFPA research team to verify and compare with the basic circuit-theory model, input-output characteristics, conservation of energy, and energy-transfer mechanism; compare with available theoretical and/or empirical models published in the literature for similar operating and testing conditions; and compare the results from different laboratories for identical tests. After preliminary data analysis, the project team has observed that different labs have yielded relative consistent results when configurations are identical.
The goal of the Phase II testing is to provide new information regarding typical operating ranges among the industry. Due to budget and time constraints, it is difficult — if not impossible — to perform tests for all possible operating conditions. Therefore, the project team has established a balanced approach to the design of the Phase II test plan. Additional spot tests may be performed later to verify/clarify/confirm the results/conclusion of the data analysis. The tests will include different levels of bolted fault current, fault duration, and measurement distances. For instance, there are 270 tests scheduled for 4,160V alone. Since the distance to the back panel is another variable that can affect the incident energy, the project management team plans to perform selective tests to extract the correlation between the incident energy and the distance to the back panel.
The test results of the project will provide information that is expected to help more accurately predict the hazards associated with high-energy arcing faults and the accompanying arc blasts, thereby improving electrical safety standards and providing practical safeguards for employees in the workplace. The proposed research and testing plan will focus on the development of physics- and engineering-based modeling and testing to validate the theory related to heat transfer and thermal effects, arc blast pressure, sound, and light intensity. This effort will include both the open-air arc hazard and the effect of enclosures commonly found in electrical systems, such as switchgears, motor control centers, and power panels.
The data and information generated by this project will strengthen electrical safety standards, IEEE 1584, “Guide for Performing Arc-Flash Hazard Calculations” and NFPA 70E. The IEEE 1584 Committee will be able to use the new information and data to expand the current guidelines. “This will help IEEE provide guidance for re-engineering safer electrical equipment and systems that reduce the potential for arc flash,” says Johnson.
In addition, the NFPA 70E Technical Committee will use the new information and data to enhance work practice requirements outlined in NFPA 70E (NFPA 70E Proposed Revisions on page 26). “This program should have a huge payoff in preventing injuries,” says Johnson. “It will yield standards for the industrial, commercial, and utility electric power industries that more closely reflect arc flash and arc blast experience in the workplace. The information provided by these tests and used in revising these standards will allow the industry to take steps to prevent or mitigate hazards and help workers protect themselves against the possibility of injury.”
Up for adoption
Unfortunately, there are still some jurisdictions in the United States that have not adopted codes that include reference to arc flash mitigation (Map on page 24). “Unless the authority having jurisdiction (AHJ) is using the 2005 or later edition of the NEC, you cannot push for arc flash protection,” says Laurel C. Talbot, electrical inspector and owner, Lightning Surge Protection Co., Appleton City, Mo. “In versions of the Code prior to 2005, arc flash is not considered a problem, so unless the local municipality accepts the 2005 or 2008 NEC as its operating code, facilities don't have to do a thing about arc flash.”
For instance, Missouri does not adopt an electrical code statewide. Instead, it allows local AHJs to decide on it. In Harrisonville, where Woodring works, the 2005 Code is standard. However, for construction at the wastewater treatment plant, the facility has chosen to adopt the 2008 Code. “We're using the 2008 Code, even though the city has adopted the 2005 Code,” Woodring explains.
The plant has a head works building, two aeration basins, two clarifiers, and a digester. Recently, Harrisonville voters approved a $9.7-million bond, issued to upgrade the Harrisonville sewer system and a 5-yr program of improvements, which is substantially complete.
“There's been quite a bit of discussion about arc flash,” says Woodring. “A couple of months ago, even our electric superintendant said we're going to have to start using PPE. So even though it's not in full force at this time, we're very much in the know — and that's the direction we're going.”
Sidebar: NFPA 70E Proposed Revisions
Currently, proposals for revision to NFPA 70E, “Standard for Electrical Safety in the Workplace,” are under consideration. Proposed changes accepted in the Report on Proposals (ROP) include adding columns to the hazard/risk category classifications table to provide the maximum available short circuit current and overcurrent protective device clearing time parameters for every task associated with all equipment currently covered, including equipment rated over 600V. A column with the arc flash boundary for every task has also been added.
In addition, a revision to the equipment labeling requirement has been accepted. Under the proposal, equipment that may require service or maintenance while energized would be field-marked with a label, including available incident energy or the minimum arc rating of protective clothing (but not both) as well as the date of the most recent arc flash hazard analysis, the nominal system voltage, and the arc flash boundary.
Currently, NFPA70E does not specify that it only applies to alternating current (AC) systems. Because of its predominant use, AC systems were used as the basis of many of the standard's requirements, but proposed changes to the 2012 edition of NFPA 70E will include the hazards associated with direct current (DC) systems. Included in the accepted committee actions on DC system safety requirements are actions to add a DC approach boundary table similar to current Table 130.2(C) and a DC task table similar to current Table 130.7(C)(9).
Revisions to the definition of a “qualified person” in the National Electrical Code (NEC) will include DC systems, particularly renewable energy sources and fuel cells. A “qualified person,” according to the NEC, is one who has “skills and knowledge related to the construction and operation of the electrical equipment and installations and has received safety training to recognize and avoid the hazards involved.” Because the articles covering the renewable energy sources have many requirements related to the installation of DC systems, workers will have to be trained in safety-related work practices associated with these systems.
The first step in processing the 2012 edition of NFPA 70E, “Standard for Electrical Safety in the Workplace,” has been completed. The ROP and a draft of the proposed revisions will be available at http://www.nfpa.org/70E on or around June 25. Public comments on the committee's proposed actions are due September 3. The report on comments (ROC) will be posted on or around Feb. 25, 2011, and ROC meeting will take place from October 25-29 in Savannah, Ga. The Notice of Intent to Make a Motion (NITMAM) closing date is April 8, 2011, and the posting date is May 6, 2011.