FIRE INVESTIGATION TEXTS How Good Are They?
FIRE INVESTIGATION
Some theories considered to be basic truths are questioned in light of the modern fire environment.
A QUESTION commonly asked of fire experts during their depositions is “What fire investigation books do you find particularly authoritative?” It’s a valid question, but as this article hopes to show, the answer should probably be “None.”
As a fire protection engineer and fire expert, I don’t consider any fire investigation books to be particularly authoritative. In fact, I don’t even own a singlefire investigation book. The reason? I have yet to find one that is even relatively free of error or that provides adequate proof of the theories presented.
Why are there so many fire investigation books in print that suffer from so many technical inaccuracies? This article calls the problem to the attention of those who may become engaged in litigation involving fire or explosion, and attempts to provide some solutions to the far-reaching problems enkindled by this thought.
Over the past several years, I have had the opportunity to review the fire investigation books in the libraries of the National Fire Protection Association, the National Institute of Standards and Technology-Center for Fire Research (formerly the National Bureau of Standards), the University of Maryland, and Worcester Polytechnic Institute. The NFPA and NIST need no introduction; the latter two are the only institutions in the country that offer accredited programs in fire protection engineering. If it’s true that these four libraries contain a wide selection of the best fire investigation texts available to the fire service-and no doubt it is-then it follows that there’s no need to reference specific investigation books. Logically, the inaccuracies and errors covered herein are common to most texts available.
CHAR DEPTH AND PATTERN
Char depth and pattern is a subject frequently presented in fire investigation books. Char depth is used as an indicator of origin, time of burn, and the speed at which the fire developed. Many books state that fire will burn through wood at a rate of 1/8-inch per 45 minutes, and that this can be used to determine fire exposure time and the location of the origin by measuring the depth of char. The deeper the char, the longer the exposure; the longest exposure is the location of the origin. Some books further state that the fire growth rate, and even the presence of an accelerant, can be indicated by the characteristics of the char pattern.
These charring rates are based on tests subject to the standard time-temperature curve of ASTM E-l 19, a standardized fire test in the United States.1* This test apparatus is used to determine the fire-resistance rating of assemblies such as fire walls. The test furnace is heated such that the temperature within it is defined according to the expression:
T = T0 + 345 Log (0.133t + 1)
where T (°C) is the temperature within the test chamber at time = t, T (°C) is the initial ambient temperature, and t is the time in seconds.
Although still in wide use, the standard time-temperature curve is based on fire tests conducted over 60 years ago. The fuels typically present in most buildings today, especially residential occupancies, arc vastly different than those used in the derivation of the standardized fire exposure test, rendering’ the lire severity of the test questionable.2,3 The fuels present today are no longer predominantly natural polymers (such as wood, cotton, and wool), but rather, more and more plastics are being introduced into the home. This can drastically increase the heat release rate of the fires occurring in today’s “space age” occupancies.
*For references, see poge 83.
FIRE INVESTIGATION TEXTS
Even if the fuels are the same as those used to determine the standard time-temperature curve, fires can develop at different rates, depending on the conditions present within the fire compartment at the time of ignition. Furthermore, all wood products are not the same, thus they will not char the same way. It has been reported that char rates vary with incident heat fluxes according to:
Cr= 2.2 X l0-2(I)
where Cris the char rate in mm/min., and I is the radiative heat flux incident on the wood (kW/m2)4 Although this equation is based on limited test data itself, it nevertheless indicates the strong dependence of char rate on temperature. During fire development, the hot layer can radiatively heat fuels such that the burning rates (and charring rates) are substantially increased. Moisture and the nonisotropic nature of wood also cause variance in measurements; that is, measurable physical and chemical values found at one point will not necessarily be the same at every other point.4,5
Even if charring rates were constant, substantial error would still be introduced via the measuring techniques and tools used. Many books recommend using a sharpened tire tread gauge or similar tool. Obviously, different results will be attained by different people using different tools and inserting the tool into the char with varying forces. Also, in many cases it is impossible to ascertain just where the char originally began, that is, what the true measuring point is.
Some investigation books discuss using the char characteristics as an indicator of fire growth rate. Some even go so far as to state that certain patterns can be used to identify the presence of an accelerant. However, to date there has been no qualified investigation of char patterns, thus any reference to such use is questionable. Some of the rule-ofthumb guidelines found in fire investigation texts regarding charring can be contradicted by scientific laboratory experiments.
Some books also state that deep char found at low (floor) levels indicates the presence of an accelerant. The theory is that the deep char is the result of intensive heating; since the hot layer rarely descends to floor level, there must have been an accelerant. Unfortunately, these theories fail to account for radiative heating from the hot, smoky layer that forms during a fire.
We have all felt the warmth of the sun, or a camp fire -this is radiative heating. Radiative heating from the hot layer can cause the phenomenon known as “flashover.” Flashover refers to the transition from localized burning to full room involvement.4 The radiative output of the hot layer can become substantial and cause the remote ignition of fuels within the fire compartment, thus resulting in flashover. Although difficult to quantify, fire researchers typically associate flashover with an average upper layer temperature of 1,112° F or greater.6 At these temperatures, wood flooring can easily be charred.4 Simply stated, deep char at floor level is not necessarily indicative of an accelerant.
FIRE INVESTIGATION TEXTS
Therefore, the use of charring so widely cited in fire investigation literature is scientifically unfounded and inaccurate.
EXPLOSIVE (FLAMMABLE) LIMITS
Explosive (flammable) limits are one of the most abused subjects offered in fire investigation books. Many fire investigators are under the impression that the lower and upper explosive limits arc “written in stone.” What the books fail to do is explain how these values were derived and what their limitations are. Flammability limits are derived through the use of standardized tabletop tests that provide what is more closely a relative ranking than an unchangeable quantitative description. Indeed, the flammable limits are known to change with temperature, pressure, and vessel size.4,5 Ignition source energy also has a profound effect on the limits of flammability.4
Another inaccurate use of the explosive limits is in calculating how much gas is required within an enclosure to initiate an explosion. The common misconception is that an entire volume of the enclosure or structure must be filled to within the explosive limits to result in an explosion. In the first place, ignition can easily occur without an entire area being filled with an explosive mixture, especially if an ignition source is near the fuel supply; secondly, only a small amount of fuel gas is required for sufficient overpressures resulting in structural damage.”
Closely aligned with the investigation of gaseous explosions is the concept that a relationship exists between the vapor density and the location of damages associated with fuel-gas explosions. Some fire investigation texts state that if an explosion involving fuel gases occurs, the type of gas can be determined from the degree and location of damages to the enclosure. It is stated, for example, that if displacement of the upper portion of a structure is greater than that of the lower section, an explosion involving a lighter-than-air gas must be responsible. Apparently, the reasoning is that if more damage is present at the upper sections of the structure or room, then the explosion occurred at the upper section.
The problem here is that these books fail to account for pressure development theory. Basically, deflagrating gases will affect each side of an enclosure equally; this makes sense, as pressure cannot occur until a boundary or restricting membrane (wall) is encountered, thus pressure will not develop until all bounding walls have been affected. On the other hand, detonations, which are characterized by flame propagation speeds greater than the speed of sound, can cause nonuniform pressures.” Fuel-gas explosions in dwellings or similar structures will typically be deflagrations, not detonations. Therefore, pressure must be uniformly applied before resultant structural damage, and the location is not dependent on vapor density alone.
This use of vapor density theory in postincident investigations also fails to account for the construction of the enclosure; that is, if a structure is weak at the foundation, regardless of the location of an explosion, failure of the structure may occur at the foundation. This suggests that this use of vapor density is unfounded; unfortunately, it is commonly referenced.
SPALLING
Spalling, which usually refers to the separation of a portion of a concrete slab or assembly, is commonly listed as an indicator that an accelerant was present. The typical reasoning offered is that high temperatures are required for spalling, thus an accelerant must have burned above the damaged portion of concrete. Some books state that a concrete temperature of around 2,000° F is required for spalling to occur. However, spalling is sometimes the result of trapped moisture that’s converted to steam due to the application of heat. Water normally boils (that is, changes to steam) at 212°F, a value far below the 2,000° F spalling temperature cited in some books.
What makes the above theory even more questionable is the fact that the concrete below an accelerant must remain at temperatures below the boiling point of the accelerant. The liquid cannot be present above its boiling point, and the liquid -not the fire -is contacting the concrete. The area directly around the base of the fire will be heated radiatively and possibly convectively. However, the heat will be dissipated throughout the cool concrete. Thus, it is apparent that attributing spalling to flammable-liquid pool fires above concrete is questionable. Spalling has occurred when hot asphalt has fallen from the roof assembly onto concrete floors below. The molten asphalt rapidly heats a small portion of the concrete, which can cause spalling.
COLOR OF SMOKE AND FLAMES
The color of smoke and flames are frequently referenced indicators in nearly all of the fire investigation books. The investigator is told to determine the color of the smoke and flames from the fire (from witnesses), as these can be indicative of the fuels on fire. Some books go so far as to state that the presence of an accelerant within a building fire can be indicated by the color of smoke issuing from the building. Many books even provide the reader with a table of smoke and flame colors versus fuel type. The books state that dark, black smoke indicates the presence of petroleum-based fuels and, possibly, of an accelerant.
There are fundamental flaws in using this theory for fire investigation. Fuels are rarely homogeneous. Many are comprised of several different fuels combined into one; all these fuel mixtures may burn quite differently.
FIRE INVESTIGATION TEXTS
Secondly, as mentioned previously, the fuels used today are quite different from the fuels used years ago. Many synthetic polymers and fibers are present in large quantities in today’s homes and commercial occupancies, and many of them produce large amounts of thick, black, acrid smoke. Some of the foams, like polyurethane, which is widely used in upholstered furnishings, melt and flow in a similar way to that of a flammable liquid.
Once again, the fire investigation books have failed to account for our changing environment. Flame and smoke colors are seldom, if ever, good indicators of the type(s) of fuel(s) involved, and dark, black smoke should not be relied on as an indicator of a petroleum-based accelerant.
SPRINGS IN UPHOLSTERED FURNITURE
It’s been stated that springs in upholstered furniture are suitable for testing when determining if a fire originated in a piece of furniture. It’s believed that if a furniture spring (or similar item) is compressed (or extended) and springs back into its original form when released, then the fire could not have originated within that particular piece of furniture. If the spring fails to spring back to its original form, then the fire is likely to have originated in that piece of furniture.
Unfortunately, an explanation of this “test” is not provided in most of the books that reference it. A spring does not have the intelligence to know where the fire originated and act accordingly. Although there are exceptions, a fire involving a piece of furniture will burn the same if the fire originated on it or near it. The damages it causes will be the same, suggesting the “spring test” is scientifically unjustified.
OTHER ISSUES
Several other reoccurring investigation indicators are cited in the investigation literature. Melting temperatures are commonly discussed and in some cases relied on to indicate the presence of an accelerant. Most of this information is not supported and generally unfounded. For example, copper has a reported melting temperature of 1,983° F; aluminum, 1,220° F.10 Some fire investigation authors have said that an accelerant must be present for the temperature required to melt copper. The problem is that what appears to be pure copper may in fact be an alloy of copper-aluminum, which has a melting temperature lower than either copper or aluminum. For example, a eutectic alloy of 33 percent copper and 67 percent aluminum has a melting temperature of only 1,018° F.10Also, surface effects on the metal, especially those associated with products of combustion, can alter the melting temperature of the copper. The same goes for other metals and nonmetals.
Hydrocarbon detectors (sniffers) are commonly used by fire investigators to detect hydrocarbon accelerants. Most of these detectors were not designed for this use-they were designed primarily for sampling atmospheres for the presence of combustible gases. Woodbased fuels are comprised mainly of hydrogen, carbon, and oxygen atoms; thus, the possibility that a detector may falsely indicate the presence of a hydrocarbon accelerant on an uncontaminated piece of wood is very real. Indeed, I have witnessed this on numerous occasions. The limitations of these detectors is generally not presented in the literature available.
Another concern with fire investigation books is that they fail to discuss a total fire scene investigation. For most, the fire scene investigation ends upon the identification of the origin and cause of the fire. However, a truly complete investigation should begin at that point. Most fire investigators are not concerned with such topics as building construction, fire and building code analysis, fire detection and suppression systems response to the fire, the burning characteristics of the fuels, and the like. These are all extremely important aspects of the fire. They are also extremely important with respect to litigation.
FIRE INVESTIGATION TEXTS
It is my opinion that fire investigation books are self-perpetuating. Over the years, fire investigators have relied on each other’s books for writing their own. This creates a situation in which no independent technical research is conducted or theories supported. The books become reprints of the same old rules of thumb. Many of these rules of thumb can be challenged, quite easily at times, by simple analysis.
What I find most amazing is that the fire investigation books are written by investigators with 20, 30, even 40 years of experience, but many suffer from technical errors. Most of this can be traced back to their training, which rarely includes formal technical education beyond fire investigation seminars and schools. To add to the problem, these investigators are frequently seminar and course lecturers, based on their years of experience. This explains, in part, the year-to-year continuity of errors present in the field of fire investigation.
This does not mean that there aren’t any good books or papers on the subject, and it certainly doesn’t mean that all fire investigators are undertrained – indeed there are many fine fire investigators. One problem is that many accurate, informative articles that could be of benefit to fire investigators never reach the journals read by most of them. The articles most often see their way into the well-respected technical journals that aren’t as well-known or widely read. However, with a little research these generally are available.
Another frustrating characteristic of fire investigation books, besides their constant failure to provide an explanation or proof for their theories, is that most never offer a single reference. I find this extremely questionable and indicative of the lack of credibility of the books.
The main point of this article is to aid those who investigate fires and who are involved in litigating fire cases. First and foremost, any fire expert should provide their understanding of the fundamentals of fire. Only when these are mastered can one hope to be an effective investigator. However, everyone is a fire expert these days. Fire is an extremely complex phenomenon, and understanding it requires more than occasional study. Every day, technical information is released by the research community. Fire investigators have an obligation to seek out such information and learn from it.
Readers should never assume that what is written in fire investigation books is accurate. Hopefully, enough examples have been presented to call attention to this. Challenge every concept presented by an author or expert, and if need be, consult a qualified expert.
The amount of excellent technical references on fire science is considerable and continues to grow rapidly. Many of these can be understood by laymen; others require the translation assistance of a technically oriented fire expert. In particular, one source of information that appears to go unnoticed by most fire investigators is the National Institute of Standards and Technology-Center for Fire Research in Gaithersburg, Maryland. In addition to its library and other resources, the NIST-CFR maintains an on-line computer reference data base, called “FIREDOC,”11that can be accessed by anyone. The cost is only that of your telephone line fee-there are no other charges. FIREDOC is an excellent source for those seeking information on any subject related to fire or fire protection.
Of course, there are several universities across the country conducting research in the area of fire science. Two were cited at the beginning of this article. Also, the National Fire Protection Association, located in Quincy, Massachusetts, is a good source. In particular, they offer two excellent films, “FIREPOWER” and “Fire: Countdown to Disaster,” that graphically demonstrate the speed at which fire can grow in modern settings, showing actual footage of developing fires. These films are highly recommended and considered to be an excellent means of conveying the overwhelming danger of fire to jurors and others unfamiliar with the speed and intensity at which fires can develop.
It is time we all work together to move fire investigation literature out of the dark ages and into the present day by utilizing the information provided by modern scientific understanding and research.
REFERENCES
1“Standard Methods of Fire Tests of Building Construction and Materials,” ASTM E-l19, American Society for Testing and Materials, Philadelphia, PA, (1983).
2Fang, J.B. and Breese, J.N., “Fire Development in Residential Basement Rooms,” NBSIR 80-2120, National Bureau of Standards (U.S.), (1980).
3“Fire Safety in Multi-Family Housing,” U.S. Congressional Hearing, Subcommittee on Science, Research, and Technology, July 28, 1988.
4Drysdale, D., An Introduction to Fire Dynamics, John Wiley and Sons, New York, (1985).
5The SFPE Handbook of Fire Protection Engineering, National Fire Protection Association, Quincy, MA, (1988).
6Peacock, R.D. and Breese, J.N., “Computer Fire Modeling for the Prediction of Flashover,” NBSIR 82-2516, National Bureau of Standards (U.S.), (1980).
7Monakhou, V.T., Methods for Studying the Flammability of Substances, Amerind Publishing Co., New Delhi, India, (1985).
“Zukoski, E.E., “Development of a Stratified Ceiling Layer in the Early Stages of a Closed-Room Fire,” Fire and Materials, Vol. 2, No. 2, (1978).
9Fire Protection Handbook, 16th Edition, National Fire Protection Association, Quincy, MA, (1986).
10Van Vlack, L.H., Element of Material Sciences and Engineering, ” AddisonWesley, Reading, MA, (1980).
11Jason, N.H., “FIREDOC Users Manual,” NBSIR 87-3562, National Bureau of Standards (U.S.), (1987).”
