The most significant safety issue in the electrical industry is the threat of electrical workers being injured or killed by arcs and blasts. The principal hazards associated with such arcs and blasts include thermal burn injuries and physical trauma from the blast concussion and flying projectiles caused by partially melted components being propelled by the force of the blast. Therefore, it’s important to understand that the focus of arc flash hazard analysis (AFHA) is to mitigate hazards — not merely to select flame-resistant (FR) clothing.
Many electrical engineers have used the methodologies discussed in this article to calculate the “heat” associated with electric arcs, but then used that information only to recommend electrical workers wear higher levels of FR clothing when performing energized work. The appropriate approach would be to first exhaust reasonable attempts to make engineering changes to the system to reduce the “heat,” and then select the appropriate FR clothing for the risk that cannot be adequately controlled through engineering interventions. Let’s take a closer look at the AFHA process and examine what factors electrical engineers should consider when attempting to protect electrical workers in the workplace.
AFHA integrates both electrical engineering and safety engineering. Unless the electrical engineer performing the AFHA is fluent with proper methods of hazard analysis and systems safety principles, he will be at a disadvantage when determining appropriate methods to address safety issues in an AFHA (see Avoiding Traps in the AFHA Process). Furthermore, the person conducting an AFHA must have a great depth of experience in power systems design. This is because there are many occasions where the arc flash calculations will provide erroneous results, and the engineer must have enough knowledge/experience to identify when reasonable results are produced — or when investigation is needed to identify why an erroneous value was produced during calculations.
An AFHA consists of four distinct engineering functions, including:
- System modeling
- Data entry and validation
- System analysis
- Reporting and recommendations.
Let’s take a closer look at each of these phases for additional clarification.
System modeling — Given that all subsequent analysis of the project hinges on the accuracy of the front-end information, it’s critically important to accurately capture the electrical system in a commercially available AFHA software program. This step involves physically gathering data relative to the system components and settings on electronic system protective devices, such as power circuit breakers, protective relays, and fuses. The initial evaluation captures the facility electrical system in its normal operating state, which includes the normal position of bus ties, generator operation, and feeder contribution from the electric utility.
Data entry and validation — This stage includes populating the AFHA software with the needed information to predict system function in both normal operation and during faulted conditions. An added benefit of this stage is that an accurate schematic diagram of at least the main feeders of the facility is created as a natural output of the study.
From a practical standpoint, these first two stages constitute about two-thirds of the total time involved in an AFHA study. In addition, the graphical representation of the facility (schematic diagram) must be verified before proceeding to the analysis process because the software program uses the diagrams in running the engineering calculations.
System analysis — The analysis stage of the study includes evaluation of the system from several perspectives, including performing a short circuit analysis, protective device duty analysis, and coordination analysis. Brushing up on the technical vocabulary involved in such analysis is also useful (see Terms and Definitions).
You must first determine the amount of short circuit current (SCC) generated by the system during faulted conditions at each “node” (i.e., location) in the facility. This information is valuable for ensuring protective devices are properly rated to interrupt the available SCC and properly sizing grounding cables. SCC is mostly a function of the mega-volt-amperes (MVA) ratings of the source generators and transformers from the electric utility; however, many modern industrial customers have extremely large internal generation systems (>50MVA). The AFHA must take these internal generation capabilities into account when performing the study.
One key output of the SCC analysis is a report known as the “protective device duty analysis,” which compares the capability of protective devices (fuses, circuit breakers) to interrupt SCC to which it is subjected. In cases where the SCC exceeds the interrupting rating of the protective device, a “through-fault” results, which means the protective device operates but is unable to interrupt the flow of SCC. Because the result is the same effect as not having a protective device in the circuit, the SCC must then be interrupted by the next protective device in series with the system “upstream” toward the source. This results in much slower arc clearing times, which, in turn, translates into far greater incident energy exposure levels for electrical workers.
Coordination analysis involves evaluating the time current curves (TCC) of the protective devices to ensure that the electrical system will clear faults in an orderly or “coordinated” manner. A TCC refers to the speed at which a device will “clear” SCC as a function of the amount of SCC to which it is exposed. In general, the higher the SCC, the faster the protective devices will operate. This reality often explains why systems with low SCC can actually have more incident energy — because the time an arc continues to “burn” determines how much heat eventually develops. The coordination study evaluates two scenarios that will later appear in reports. The first evaluates the coordination of the current configuration (i.e., the “as-built” case) of the system. The second looks at the system once the recommended engineering changes have been implemented (i.e., the “revised” case). The recommended engineering changes that come out of this step can involve any combination of the following:
- Reducing trip times on adjustable circuit breakers.
- Using current-limiting fuses.
- Reducing fuse sizes of non-current-limiting fuses.
- Replacing fuses with other styles of fuses that have different TCC characteristics.
- Changing protective relay settings on systems where an electronic relay actuates a separate circuit breaker. Note: These systems are far more expensive but provide maximum flexibility for engineering interventions because many different relays can be connected to a single circuit breaker. This means the protective systems can be “smarter” than simply sensing magnetism or heat, as is the case in a simple thermal-magnetic circuit breaker found in a home.
- Inserting additional protective devices in series with existing devices. Note: Often, the use of motor overloads in series with fuses can result in much lower values of incident energy because fuses can be set to interrupt only SCC while relying on the overload sensors to interrupt overloaded conditions.
The normal approach to making incident energy calculations is to use commercially available software programs that have essentially automated the use of the IEEE Standard 1584 Arc Calculation spreadsheet. The most impressive use of this software is in the area of adjusting system coordination. The engineer can adjust an individual protective device in the system, which, in turn, simultaneously updates the values of any other related system elements. This allows the engineer to test different scenarios and receive instantaneous feedback. Needless to say, the investment in AFHA software is worth the price in terms of saved man-hours of engineering time.
Reporting and recommendations — This stage of an AFHA typically includes five major steps:
- Tabular data from the study: It is important to provide tabular data for each section of the report, because doing so allows critical review by other engineers and enables others to catch data entry mistakes in equipment labeling, etc.
- Protective device duty analysis: Identifies devices at or near their interrupting duty ratings. Some software programs produce an “equipment duty report,” which is synonymous with the protective device duty analysis.
- Incident energy calculations: Highlights areas where incident energy levels exceed 10 cal/cm2. We recommend using a 10 cal/cm2 threshold because studies have shown that third-degree burns result from exposures to 10.7 cal/cm2 (unprotected skin) or more.
- Recommended engineering interventions: Provides revised breaker/relay settings when those changes will result in satisfactory outcomes. This section also includes a cost-benefit section for recommended interventions that necessitate either equipment replacement or significant retrofitting of equipment to lower incident energy exposure levels.
- Equipment labeling: The NEC (110.16) requires that all equipment with arc flash hazard potential (i.e., >1.2 cal/cm2) be field marked to warn electrical workers of the hazardous condition. This label normally includes the calculated incident energy value and other important safety information.
The hazards represented by electrical arc blasts have been identified as a significant hazard for many years. Previously, the preferred method for protecting electrical workers from arc flash hazards was through the use of personal protective equipment (PPE), such as FR clothing, face shields, etc. However, many accident reports revealed that incident energy levels could easily reach values for which there was no PPE capable of providing adequate protection. As such, it was necessary to reduce incident energy to manageable levels through the use of engineering controls.
The use of formal engineering studies, such as AFHA, represents a significant improvement for protecting electrical workers from arc flash hazards in the workplace. The IEEE 1584 methodology has emerged as the standard for AFHA in the United States since its publication in 2002. Although the methodology presented in this standard is very powerful, there are inherent weaknesses within it that qualified engineers must consider when determining the best methods to protect electrical workers.
A number of commercially available AFHA software programs make AFHA significantly easier — programs that allow an engineer to easily test different coordination scenarios and result in better systems analysis and improved recommendations. Automating the many calculations needed to perform AFHA reduces mathematical errors and improves the end product.
In the final analysis, safety professionals and engineers alike must remember that AFHA is about people, not engineering calculations or OSHA regulations. A miscalculation by an engineer or errors in data gathering can result in death of a human being. Thus, all parties must maintain great diligence in ensuring that every facet of AFHA is performed to the highest standards possible.
Kolak is president of Praxis Corp., a company specializing in electrical engineering and electrical safety training based in Granbury, Texas. He can be reached at email@example.com.
Sidebar: Terms & Definitions
Arc clearing time: The time from the onset of the arcing current to the moment the arc is extinguished. The clearing time is comprised of three separate variables: the time it takes for the protective device to “sense” the fault, the mechanical operating time of the protective device (circuit breakers or fuses), and the time it takes for the protective device to extinguish the arc.
Arcing fault current: A fault current flowing through an electrical arc plasma, also called arc fault current and arc current.
Arc-in-a-box: The estimated incident energy for an arc in a cubic enclosure with sides of 20 in.
Arc rating: The maximum incident energy resistance demonstrated by a material (or a layered system of materials) prior to break-open or at the onset of a second-degree skin burn. Arc rating is normally expressed in calories per square centimeter.
Available fault current: The electrical current that can be provided by a serving utility and facility-owned generation devices and large electrical motors, considering the amount of impedance in the current path.
Bolted fault current: A short circuit or electrical contact between two conductors at different potentials in which impedance between the conductors is essentially zero.
Electrical hazard: A dangerous condition in which inadvertent contact or equipment failure can result in shock, arc flash burn, thermal burn, or blast.
Exposed: Capable of being inadvertently touched or approached nearer than a safe distance by a person. It is applied to parts that are not suitably guarded, isolated, or insulated.
Fault current: A current that flows from one conductor to ground or to another conductor through an abnormal connection (including an arc) between the two.
Flame-resistant (FR): The property of a material whereby combustion is prevented, terminated, or inhibited following the application of flaming or non-flaming source of ignition — with or without removal of said flaming source.
Flash hazard analysis: A method to determine the risk of personal injury as a result of exposure to incident energy from an electrical arc flash.
Flash protection boundary: An approach limit at a distance from live parts that is un-insulated with which a person could receive a second-degree burn. This is defined as incident energy levels of 1.2 cal/cm2 or more.
Incident energy: The amount of energy impressed on a surface, a certain distance from the source, generated during an arc event. Incident energy is measured in joules per square centimeter or calories per square centimeter.
Sidebar: Avoiding Traps in the AFHA Process
Inexperienced engineers (or non-engineers) often fall into some of the more common traps when performing arc flash hazard analysis (AFHA), including those outlined below.
Hyper focus on high-voltage (>600V) systems — There is a general belief that high-voltage systems present a much greater threat from arc flash than low-voltage (<600V) systems. The tabular approach in NFPA 70E contributes to this misunderstanding, because the tables in the document require high levels of flame-resistant (FR) clothing for high-voltage systems and relatively little FR clothing for low-voltage systems. However, there are numerous occasions when low-voltage systems represent significantly higher arc flash hazards to workers than do high-voltage systems. One reason for this is that most high-voltage work is performed using insulated sticks, which can range in length from 4 ft to more than 35 ft. The increased working distance when using insulated sticks often renders the actual incident energy exposure levels to values far less than working on low-voltage circuitry.
Another reason to avoid high-voltage hyper focus relates to employee exposure to high-voltage systems. In the commercial/industrial realm, most electricians or maintenance workers only interface with high-voltage systems perhaps once per year. As such, many organizations will use a “phased approach” to AFHA due to costs, almost invariably devoting scarce resources to the high-voltage system first. However, this is the antithesis of what should be done. By far, electrical workers interface with low-voltage systems more often than high-voltage systems. Therefore, the focus of the AFHA should first be to mitigate arc flash hazards on the low-voltage system in most organizations.
Modeling the system based upon voltage levels — A dangerous practice in AFHA is to model the electrical system to only a certain voltage level rather than the actual arc flash hazard. The most common practice is to stop at the 480V level of the system, assuming that lower levels of the system do not present a real hazard. However, experience teaches us that some of the highest incident energy levels in a facility are on the 208V side of 480V/208V dry transformers. There are literally thousands of examples where there are incident energy levels of only 0.1 cal/cm2 on the 480V side of the transformer and levels of as high as 600 cal/cm2 on the 208V side of the same transformer.
Stopping analysis when low incident energy levels are achieved — In many cases, once incident energy has been reduced to acceptable levels at one level of the system, it’s often true that lower levels of the system on the same circuit have the same or lower incident energy level. However, a good practice to follow when performing AFHA is to “sample” downstream circuits, just to ensure they do, in fact, have low incident energy levels.
A recommended way of sampling would be to model each style of circuit breaker or fusible element on lower level circuits. The reason for this approach is that short circuit current (SCC) levels usually decrease as you move to lower levels of the system. The clearing times on protective devices will increase in response to the lower SCC levels. However, in some cases, incident energy levels can actually increase when SCC levels dip below the level where a protective device will operate. This results in arcs that can last for several seconds and often translates into dangerously high incident energy levels.
Blindly accepting computer-generated results — Only competent electrical engineers should perform AFHA, because the software is not perfect and can occasionally produce erroneous results. In these cases, it takes someone with both experience and engineering training to first recognize the error and then hand calculate the correct results. Another difficult step requires the engineer to re-integrate the correct results into the software program for use with the rest of the study.
Here’s a good example. Looped systems can present a challenge for AFHA software because algorithms in the software may “look” for certain mathematical results when determining the “source” for a fault. This can also happen when generators are involved. In these cases, the software can sometimes produce erroneous results, because these situations usually require more than one protective device to operate in a specific order to clear the fault. If the software “picks” the wrong source — or if it doesn’t properly calculate the sequence of events in the fault event — it will produce inaccurate results.
It’s also important to note that AFHA software programs often do not include comprehensive “reasonability check” systems to verify that inputted results are reasonable for the system being modeled. Once again, the responsibility for catching these situations rests with the engineer doing the analysis. Clearly, not having a properly qualified person doing the analysis increases the likelihood of erroneous results.