Electrical fires account for about 20% of all structure fires in the United States.1 It may well be the second-largest cause after cooking, which is the cause of about 50% of the structure fires.2 According to National Fire Protection Association (NFPA) statistics,2 heating equipment is the second largest cause at 14%. This is because in its statistics, the NFPA fails to put all electrical fires together in a single category. Instead, fixed wiring and various types of fires involving user appliances or devices are reported separately. Thus, the 20% statistic I cite above is based on my own analysis of the fire data and is intended to encompass all electrical fires, wherever they occur.
A Collaborative Approach to Electrical Vault Fires
The National Fire Incident Reporting System (NFIRS) statistics on intentional fires are unacceptably poor. An in-depth study by Icove and Hargrove3 showed that, in fact, there seems to be little relationship between actual occurrence of arson vs. the NFIRS statistics. Thus, official statistics that intentional fires account for about 7% of the total structure fires2 have to be viewed with significant skepticism. Arson may well be the second largest cause of structure fires but, regardless, if electrical fires constitute the second or third largest category, clearly they need to be of concern to anyone involved with fire safety.
(1) A laboratory-created bad connection at a screw terminal; note the glowing of the wire. (Photo by Yasuaki Hagimoto.)
Some Reporting Problems
A few years ago, the National Association of State Fire Marshals (NASFM) issued a white paper titled: “Conquering the ‘Unknowns’: Research and Recommendations on the Chronic Problem of Undetermined and Missing Data in the Causal Factors Sections of the National Fire Incident Reporting System.”4 The state fire marshals were seriously concerned about what they saw as an emerging trend—the tendency of fire departments to report the cause of fires as “unknown,” whereas evaluating the pertinent data would show that the proper cause is known, to a reasonable, if not 100%, degree of certainty.
Basically, the NASFM found that way too many fire departments were abandoning their previously held commitment to providing full and adequate data for inputting into the NFIRS system. NASFM found several reasons for this unfortunate trend, but foremost was a desire to avoid having to participate in litigation. This is unavoidable, since if a legal case is filed in connection with a fire incident, the personnel involved in the investigation will most likely be called to testify, whether they did a thorough job or shirked their responsibility.
It can be argued that testifying will become less pleasant for the officers involved if they have to justify that they documented circumstances clearly suggesting a known cause yet concluded the cause of the fire was “undetermined.” The NASFM also found evidence to suggest that some departments have been concerned with a potential for lawsuits against them if they identified a faulty item as the cause of the fire.
Some fire department staff have gone down this path for another reason. They ended up concluding that they should not report electrical causes of fires since they are not experts in electricity. This concept is not in accord with how NFPA 921, Fire Investigations Guide,5 teaches investigators to investigate fires, however. The three primary tasks of the fire investigator are performed in the following order:
- Determine the origin of the fire.
- Determine the cause of the fire. The cause must comprise the circumstances that led a source of heat to ignite a fuel, and this has to occur within the area of origin.
- Determine the responsibility for the fire, if appropriate.
Determining the area of origin of a fire does not require engineering expertise on how technical devices work and how they fail. Any properly trained investigator should have the skills needed to use witness information, burn pattern analysis, and possibly some additional data to identify the area of fire origin, assuming that for the particular fire involved this is possible. If this is done successfully, the area of origin has been identified.
Now, the investigator must tally the potential heat sources located in the identified area of origin. If all except one can be successfully excluded, then the cause of the fire has been identified.
For example, if the investigator has successfully excluded all potential sources of ignition within the area of origin except for a computer, then the cause has been identified as being a computer and should be so reported on the NFIRS form, not as “undetermined.” It does not require that the investigator understand or determine why the computer overheated and ignited. In fire incidents leading to litigation, most likely engineers will end up making such a determination, but fire investigators should not be concerned that they are not qualified to make this determination.
Note that to properly assign the cause of the fire does not require that responsibility be assigned. In general, public service investigators are only expected to determine if the incident was because of incendiarism or not. In our computer example, there are two distinct reasons the computer might have ignited. If it was a design or manufacturing defect, the manufacturer will likely be responsible. If the equipment was abused or misused, the person involved is responsible. To determine which occurred is of interest to private-sector engineers but should not concern the fire department investigator, unless there is some suspicion that an individual deliberately abused the device to create a fire.
Electrical fires generally do not involve criminal intent; arson using electricity is rare although not unheard of. However, it is important to point out the nefarious practice of negative corpus, which sometimes occurs in badly done investigations. It is never appropriate to conclude, “All accidental causes have been ruled out; therefore the cause of the fire had to have been intentional.” For a fire to be properly declared as being of intentional cause, there has to exist some positive evidence for this.
What Is an Electrical Fire?
Perhaps surprisingly, the fire safety profession has not yet agreed on the definition of an electrical fire. I offer this definition1: “Electrical fire: A fire occurring due to a static electricity discharge or due to an electrical fault or failure.” Electricity can be divided into two categories: static electricity and dynamic electricity. The latter term has seen very little usage, so we can replace it with ”current-carrying electricity.” Only a very small fraction of electrical fires result from static electricity, but the definition needs to be broad enough to properly encompass them.
It may also be noted that lightning is a form of static electricity. Mostly for historical reasons, however, fire statistics agencies, including the Federal Emergency Management Agency (FEMA) and the NFPA, assign them to a specific category of lightning-caused fires rather than subsuming them under the broader category of static electricity.
One problem seen in some foreign countries is that they create too broad a definition of electrical fires. If a person puts a paper grocery bag on an electric stove, which has a hot element, the bag and its contents might ignite. Some countries actually tally this as an electrical fire. In the United States, however, such incidents are properly categorized as “a combustible placed too close to heat,” not as an electrical fire. This is because there was no fault or failure of any electrical device.
Common Electrical Fires Causes
Neither FEMA nor NFPA statistics are good at categorizing the reasons for electrical fires. Notably, NFIRS does not contain a useful listing of the dominant physical mechanisms responsible for electrical fires. The NFIRS forms are filled out by fire departments without assistance from private-sector engineers. Thus, more accurate statistics would not be expected, even if this aspect of NFIRS were upgraded.
Because of this situation, we depend on forensic experience for guidance as to what are the dominant mechanisms leading to electrical fires. A few years ago, I published Table 1 based on experience and judgment.6 Apart from “miscellaneous phenomena,” which may entail a wide variety of mechanisms (all of low incidence), seven distinct mechanisms emerge. There is a relatively broad consensus among fire investigators and forensic electrical engineers that such ranking is appropriate.
Note that “short circuits” does not appear in this table, since it does not comprise a unique mechanism. When a short circuit occurs, the ignition mechanism may be arcing (of either type), overload, or ejection of hot particles, with some rarer mechanisms also potentially being possible. By contrast, the layman commonly envisions only two mechanisms for electrical fires—short circuits and overloads. However, the reality is this: Overload is important but not one of the top causes, and “short circuits” does not correspond to a unique ignition mode.
It is important to emphasize that forensic experience indicates poor connections are the top cause of electrical fires (Table 1). To gather reliable statistics on this issue requires electrical engineering experts. As a result, such a study has not been undertaken in the United States, but in Ontario, Canada, an engineering study was done a few years back,7 and the results also showed that the top mechanism for electrical fires was poor connections.
Poor connections usually involve exactly what the term would suggest—a connection that is mechanically poor, inadequate, or failed. Screw terminals in 120-volt alternating current (AC) outlets are a common example. The simplest explanation might be that whoever did the wiring forgot to tighten a screw connection. Firefighters may occasionally encounter houses from the 1960s or 1970s that were wired with aluminum wire. They were notorious for failing. The technical explanation is lengthy and involved,1 but basically it was impossible to make good connections between the aluminum wires and the outlet screw terminals of domestic outlets.
The second most important cause of electrical fires is arcing across a carbonized path. Electricians refer to this as arc tracking, sometimes distinguishing “wet tracking” vs. “dry tracking.” Tracking is an interesting phenomenon. The earliest research goes back nearly a century, but our understanding is still limited and mostly empirical. The basic situation is that a charred material is not an electrical insulator but a conductor, albeit not a great one. Materials susceptible to charring include all types of wood, wood products, and plastics. Plastics vary in their propensity to charring, but there is no plastic that is impossible to char.
The easiest way to create tracking is to have a wet environment in the vicinity of an electrical installation. If a path can be established from one conductor to the other, wet tracking may ultimately result. This can result in overt flaming and fire, although not every incident will end up with a fire.
Dry tracking occurs because an insulator material is subjected to excessive temperatures. Photo 2 shows an electrical plug that started undergoing dry tracking because it was subjected to excessive temperatures. Other laboratory research has shown that the process can start at temperatures as low as 110°C (230°F), which can readily be reached if current flows are excessive and the wiring is overheated.
(2) Degradation of a PVC plug after oven-aging at 165°C (329°F) for 250 hours. (Photo by Kiyomi Ashizawa.)
Electrical Arc Explosions
If two wires are shorted together and then they separate, an electrical arc is created, a highly luminous discharge of electric current flowing through air, which just a moment earlier was acting as an electrical insulator. If the current available during this process is high, then you can expect a mighty bang called an electric arc explosion. The phenomenon has been known for a long time but studied only to a limited extent. In fact, I wrote the first-ever review paper on this topic a few years ago.8
The most common types the fire service encounters are fuel/air explosions. In such an incident, some fuel gas or vapor escapes and accumulates in some confined locale. If the gas or vapor reaches a sufficient concentration and an ignition source is present, an explosion can result. This is most commonly encountered with natural gas, liquefied petroleum gas, and gasoline vapors. Such an explosion is a combustion reaction—fuel and oxygen combine and burn; since they can do this very rapidly and in a large volume, an explosion results. Such explosions, of course, can demolish houses and cause very serious damage.
An electrical arc explosion, however, is something quite different. There is no fuel involved or fuel involvement is insignificant, but a severe incident can occur because the electrical arc going through air can increase pressure very rapidly and raise the pressure to a high value in the vicinity.
Electricians and electrical inspectors started focusing on this problem a few decades ago because of the arc flash problem. Arc flash is the heat output that results from the initiation of an electric arc.1 The profession became concerned about arc flash injuries to electricians, and protective equipment requirements started to be required by the Occupational Safety and Health Administration; NFPA 70E, Standard for Electrical Safety in the Workplace®9; and other jurisdictions. Arc flash and electric arc explosion pressure blast are two aspects of the same phenomenon—a high-current arc, discharging through the air, creates both a heat output and a pressure rise.
The focus on arc flash protection is appropriate, since an electrician is more likely to be injured from the heat output than from the blast pressure. Compared to fuel/air explosions, electric arc explosion pressures are more likely to be lower, leading to lower injury potential. Achieving high blast pressures requires confinement in a small volume; thus, the most likely cause of injury is flying shrapnel, such as switchboard doors.
Additional Resources
NFPA 921 offers a brief introduction to electrical fires,5 but I recently published a textbook devoted solely to Electrical Fires and Explosions.1 Similar to the earlier Ignition Handbook,10 the new book covers the needs of engineers, fire investigators, and fire training instructors. One chapter focuses on the principles of science while another focuses on how failures and fires occur. The latter provides practical information to fire service personnel but does not require any engineering background. Readers who successfully go through them will not become qualified engineers but will become comfortable with the topic of electrical fires and explosions. This should enable us to see improvements in report writing and in the data fed into NFIRS. The new book gives supplemental pointers to investigators beyond what they can obtain in NFPA 921 that can serve to improve their understanding of how to examine and report fires that are potentially electrical fires.
References
1. Babrauskas, V., Electrical Fires and Explosions, Fire Science Publishers, New York (2021), http://doctorfire.com.
2. Ahrens, M., and Maheshwari, R., Home Structure Fires, National Fire Protection Association, Quincy MA (2020).
3. Icove, D. J., and Hargrove, T. K., “Project Arson: Uncovering the True Arson Rate in the United States,” pp. 283-292 in Proc. Intl. Symp. on Fire Investigation (ISFI 2014), National Association of Fire Investigators, Sarasota FL (2014).
4. “Conquering the “Unknowns”: Research and Recommendations on the Chronic Problem of Undetermined and Missing Data in the Causal Factors Sections of the National Fire Incident Reporting System,” National Association of State Fire Marshals, Washington DC. (2014).
5. NFPA 921, Guide for Fire and Explosion Investigations, National Fire Protection Association, Quincy MA (2021).
6. Babrauskas, V., “Research on Electrical Fires: The State of the Art,” pp. 3-18 in Fire Safety Science—Proc. 9th Intl. Symp., International Association for Fire Safety Science (2008).
7. Hardy, F., “Electrical Ignition Causes of Fires in Ontario from 2002 to 2007,” Electrical Safety Authority, Mississauga, Ont., Canada (2008).
8. Babrauskas, V., “Electric Arc Explosions—A Review,” Fire Safety J, 89, 7-15 (2017).
9. NFPA 70E, Standard for Electrical Safety in the Workplace, National Fire Protection Association, Quincy MA (2021).
10. Babrauskas, V., Ignition Handbook, Fire Science Publishers/Society of Fire Protection Engineers, Issaquah WA (2003).
Vyto Babrauskas, Ph.D., is known for his research contributing to the field of fire investigation. In addition to his earlier book, Ignition Handbook, he has recently published two new books, Electrical Fires and Explosions and Smoldering Fires. His firm, Fire Science and Technology Inc., is located in Clarkdale, Arizona.
