Arc Flash and Molten Metal — the Hidden Hazard

Hazard analysis and incident energy modeling are excellent tools for determining what level of arc rating should be worn for a given situation, but they are not as reliable for determining whether FR is needed.

• By Scott Margolin
• May 01, 2010

Since the inception of the NFPA 70E 2000 Edition 10 years ago, there has been a major evolution in our understanding of the electric arc flash hazard. NFPA 70E has been revised twice and is now in revision cycle again; NESC (National Electrical Safety Code) has dramatically expanded its arc flash rules; and there are a number of hazard analysis tools, tables, calculators, and consultants, as well as seemingly countless new flame-resistant (FR) fabric options. All of this progress has driven the industry toward safer work practices and better PPE for those unfortunate occasions when an arc flash does occur. However, new research has dramatically demonstrated a dangerous and underappreciated component to arc flash: molten metal. The molten metal created by an arc flash presents a significant ignition hazard in addition to the arc itself, which to date has been largely overlooked by incident energy and hazard analysis models.

Through use of super-slow-motion video and innovative research, we are beginning to understand just how large a role molten metal plays in arc flashes in industrial equipment. As you may know, the primary risk to a worker in an arc flash is ignition of nonfl ame-resistant (non-FR) clothing; the vast majority of catastrophic injuries and fatalities in arc flashes are a result of sustained clothing fires. Once flammable clothing ignites, it will spread rapidly and continue to burn long aft er the arc flash itself is over, increasing both the severity and the extent of body burn injury. Within a few seconds, a clothing fire will burn areas of the body the arc never touched.

The primary risk factor for fatality is body burn percentage, that is, the higher the percentage of the body which receives second- or third-degree burn, the greater the likelihood of death. This is especially true for burns covering more than 50 percent of the skin. The body area from the belt up of the average American worker is usually 50 to 60 percent, so it is easy to see how shirt ignition can lead within seconds to a fatal level of body burn.

Why is percent body burn, rather than severity, linked to fatality? Because the significant majority of deaths among burn victims are a result of infection; second-and third-degree burn both break the skin, and the larger the area of the skin surface broken open, the greater the likelihood of infection.

Flame-resistant clothing typically has two primary purposes: 1) by not igniting and sustaining combustion aft er the arc flash (or other thermal event), the percentage of the skin surface receiving second- or third-degree burn is limited to, at most, areas directly impacted by the arc flash; and 2) in addition to not spreading injury by ignition, FR clothing ideally also should insulate the wearer from as much second- or third-degree burn as possible. This level of insulation is reported as the arc rating (ATPV, Arc Thermal Performance Value, or Ebt, Energy Breakopen Threshold). The higher the arc rating, the greater the level of incident energy necessary to cause onset of seconddegree burn through the fabric. Most standards today require the arc rating of the FR clothing to meet or exceed the arc energy predicted by hazard analysis of the system, in the hope of significantly reducing or eliminating anything worse than sunburn (first degree) underneath the garments.

Hazard analysis and incident energy modeling are excellent tools for determining what level of arc rating should be worn for a given situation. However, they are not as reliable for determining whether FR is needed because molten metal can cause fabric ignition even if the arc is below ignition level or does not impact the garment at all.

We can use these tools to quantify the probable energy released in an arc flash, and we can match that to the protection afforded by a given FR fabric so we don't wear something with an arc rating of 4 cals into a job where the arc flash energy is predicted to be 8 cals. For instance, a windbreaker might be a fine choice for a day when the temperature is 55 degrees F, but it would not provide adequate insulation if the temperature were 5 degrees F. However, when these same models are used to predict ignition of non-FR fabrics, they fall short because they quantify only the arc flash energy, not the magnitude of the molten metal hazard that accompanies that energy.

A Hazard Seen in Slow Motion

Electrical equipment primarily utilizes copper to carry the current. When an arc occurs, most or all of the copper is essentially instantly melted and thrown a significant distance. Copper melts at about 1,900 degrees F, so this large volume of molten metal is at least that hot and often hotter. Non-FR-cotton clothing and other non-FR fabrics typically ignite at about 800 degrees F or less, and an arc flash creates large volumes of molten metal at 1,900 degrees F+, traveling significant distances at a high rate of speed. When this molten metal lands on flammable fabric at more than twice the ignition temperature of the fabric, it can easily cause fabric of any weight to ignite, causing a sustained fire. That is why workers in the steel industry and even welders have worn FR clothing since the 1970s.

Two or three drops of molten metal are unlikely to cause a clothing fire, and an individual "spot fire" can be easily patted out if it occurs. However, real-world arc flashes create thousands or tens of thousands of molten metal projectiles, meaning a worker is likely to be hit with 2,000 or 3,000 drops, not two or three. That quantity of molten metal can easily cause a large, aggressive clothing fire the victim cannot pat out.

New HD video and super-slow-motion video equipment recently used to research arcs definitively and graphically illustrate this hazard. Arcs ranging from 0.6 cals to more than 45 cals were created in disconnects, switchgear, and MCCs and analyzed at 3,000 frames per second. (For perspective, standard video cameras generally record at 30 frames per second. Using a camera that is 100 times faster allowed for unprecedented analysis of the arcs on a millisecond-by-millisecond basis.) Very large volumes of molten metal are projected hemispherically out from the arc gap, routinely traveling 5-10 feet, and significant quantities are usually thrown 20-30 feet. In one notable instance, packing blankets used to protect our camera and computer cables from foot traffic and which were thought to be well outside the hazardous zone were ignited by molten metal at a distance of nearly 80 feet.

The potentially dangerous conflict with hazard analysis and non-FR clothing arises when arc flash incident energy predictions are below published ignition thresholds for non-FR fabrics. The arc laboratory originally used to determine ignition thresholds of non-FR fabrics in the late 1990s produces almost no molten metal; it uses stainless steel electrodes a foot apart and a very thin wire strung between them to initiate an arc. This is an excellent test method for reliably and reproducibly rating FR fabrics, but it does not (and was never intended to) reproduce "real world" arcs in industrial electrical equipment. Thus, the ignition threshold data accurately quantifi ed the arc energy necessary to ignite a given fabric (when new and clean) but did not capture the ignition eff ect of molten metal. Arcs in working equipment create hundreds or thousands of times more molten metal than this lab, much of which is projected directly at the worker by the arc-in-a-box eff ect. This molten metal is present at sufficient temperatures and in more-than-sufficient volume to cause ignition of non-FR clothing independent of the arc flash itself.

Molten metal also raises an issue around which FR fabric brand to select for an FR clothing program. The criteria for this decision typically include all of the standard considerations you would expect, including arc rating and long-term durability of FR to laundering, shrinkage control, proven performance in the market, comfort, availability, and the overall value equation. An additional criterion that has been overlooked until now is whether the fabric has the ability to shed molten metal. Although there is currently not a standard that compares competitive products against molten metal in an arc flash, certain fabrics are recommended for molten metal exposure in general, while others are not. As we understand more about the large role molten metal plays in an arc flash, it's reasonable to select fabrics that have been proven to perform in this area over products that are not recommended for molten metal applications.

Hazard analysis and arc flash incident energy calculations are very valuable tools that help to save lives and reduce the frequency and severity of injuries every day. However, they are not absolutes and, as we are learning, do not always account for the entire thermal hazard. They absolutely should be used to determine what FR to wear, but not whether to wear FR. This new research is helping the industry and the standards committees recognize that every arc flash, regardless of incident energy, can ignite non-FR clothing with molten metal.

This article originally appeared in the May 2010 issue of Occupational Health & Safety.