Machine Design

Don't get burned by flame standards

Developers must understand how flame, smoke, and toxicity standards vary around the globe to ensure designs pass muster when they reach their final destination.

Edited by

Emily Witthaus
Product Specialist

Rick Hopf
Materials Development Engineer
Rogers Corp.
Carol Stream, Ill.

In a typical passenger-train fire test, technicians measure flame, smoke and toxicity levels then compare these results with those predicted by computer models. This information is used to develop materials and designs that will extend the time passengers have to escape from a railcar in the event of a fire.


The current U.S. standards for materials used in rolling stock are dated at best. When the Federal Railway Association (FRA) developed them in 1989, there were few materials that could meet them. But this is no longer the case as new technologies produce materials that far exceed the current standards. Therefore, it's not surprising that several industry groups are pushing to make the standards more stringent. This is especially true in the area of optical smoke density.

According to the Environmental Protection Agency, similar concerns were raised in a petition from an independent consulting firm that recommended reconsideration of the standards. "Products such as seat foam, elastomers, thermal and acoustic insulation, vacuum forming and wall lining materials have been reformulated to exceed the 1989 FRA guidelines," says Bay State Marketing Consultants. The firm concludes that the final rule ignores the improved materials and products on the market today, and reflects an essential unfamiliarity with both the relevance of the test methods and the operating environment encountered by the majority of passenger railcars, such as those operating in the New York City tunnel system.

Specifically, the petitioner contests that the rule should be continually revised until all products used in railcar construction comply with a smoke or specific optical density limit (Ds) of 100 at 4 min using the ASTM E 662 test procedure. "An acceptance level of 200 provides little protection," says Bay State. "Smoke emitted from one fully combusted window mask complying with a Dsof 200 will completely obscure human vision beyond a distance of 2 feet, disabling people and preventing them from locating emergency exits. Such a standard would not be tolerated by anyone who actually stood in a room with such a smoke density."

Materials suppliers and industry leaders need to work together to ensure the safety of railcar passengers. A better understanding of global standards for flame, smoke, and toxicity of materials will bring industry one-step closer to streamlined railcar development, increased safety, and peace of mind.

It's common for designers of passenger railcars, automobiles, airplanes, and associated equipment to specify components manufactured outside the U.S. Material in these parts must meet flame, smoke, and toxicity (FST) standards. Yet such standards often differ from country to country. They are largely determined by where the equipment is built, and not always by where the end product will be rolling down the tracks, highway, or tarmac.

Comparing passenger railcar FST standards, for example, from four nations — U.S., U.K., France, and Germany — reveals differences in testing requirements, with some more stringent than others. This disparity poses a potential danger to public safety, particularly in the case of components produced in countries where FST standards are less strict or nonexistent. Countries have unique ways of testing materials and determining acceptable limits. However, an understanding of these different standards becomes increasingly important for companies trying to become truly global.

At the Rolling Stock Equipment Technical Forum in 2001, the American Public Transportation Association addressed the need for designers to better understand FST standards. The Association promotes the idea of taking time to fully understand the most stringent standard. It can serve as a benchmark for future material designs. This view will streamline the design-and-build process and help globalize the railcar economy.

At the time most specifications were written, few materials could meet them. But this has changed over the past two decades as new technologies produce materials that far exceed the standards.

Small-use applications, such as elastomeric gaskets, seals, insulators, and gap fillers, are vital parts of all equipment for passenger rolling stock. Equipment from under-car propulsion to roof-mounted HVAC units relies on gaskets to seal out water and reduce vibration. The car body itself requires insulation, door and window gasketing, conduit wrapping, duct insulation, and gap filling materials.

To help evaluate materials used in these components, three key criteria are examined: flame spread and ignition; optical smoke density; and toxicity of emitted smoke.

The devastation caused by fire and toxic fumes in crowded and confined spaces is evidenced by recent tragedies around the globe. They include the nightclub fire in Rhode Island that killed 100 people after a band's pyrotechnics set the nightclub's improperly installed insulation ablaze in only a few minutes. Another is the Austrian cable-car fire that took 155 lives after hot oil ignited the train's plastic flooring as it was passing through a mountain tunnel. The recent suicide bomber attack on a Russian commuter train is an additional example. The ensuing fire forced hundreds of terrified passengers to escape down a smokefilled, 2-km-long tunnel. While not limited to rail, the vulnerability factor for railcars is high because elevated tracks and tunnels make evacuation difficult and dangerous.

Toxicity is probably the most important factor to address during the design of passenger-carrying trains. For components made in the U.S., France, and the U.K., toxicity is evaluated for different materials by referencing the specified emissions limit for toxic fumes. Sample size and volume of the chamber are irrelevant because all limits are specified in parts per million (ppm) or mg/m3. Critical areas to study in any toxicity specification are the gases being examined, the allowable limits of those gases, and how those limits vary on different types of railcars.

The French NF F 16-101 and U.K. BS 6853 specifications are identical because both require the same test method for elastomers (NF X 70-100). The only difference is that the U.K.'s specification requires an addendum for nitrous oxides. Toxicity limits for French and the U.K. standards are developed from the IDLH values published in the National Institute for Occupational Safety and Health (NIOSH) Guide. IDLH, or Immediately Dangerous to Life or Health, values are calculated based on levels of gas in a particular atmosphere for 30 min that would pose an immediate risk. Converting the mg/m 3 values of the U.K. standard to ppm demonstrates how the standard compares to its U.S. SMP 800C counterpart. With respect to the allowable percentages of toxic fumes, the U.K.'s BS 6853 specification is the most stringent, closely followed by the French, and finally the U.S. It's important to note that Germany does not have a toxicity requirement.

Toxicity component maximums by specification

French (mg/m3)
NF X 70-100

NF X 70-100

U.K (ppm)
NF X 70-100

U.S. (ppm)
SMP 800C








































A true comparison of the U.K.'s standard to the American standard vis-a-vis toxicity limits can't be drawn until the calculation methods have also been studied. That's because the American SMP 800C standard stipulates that designs meet each individual toxic gas emission requirement, while the U.K. and French standards use their results to develop a constant. Based on the constant, materials can be used in different areas of a railcar. BS 6853, ( Appendix B.5), uses a constant value of "R." The R value is calculated using the following equation:

R takes into account each toxic component, but if one is high and another is low, they average to an acceptable limit. The French standard uses a similar calculation to develop their constant:

An example of these constants can be seen if the American standard tested a material at the maximum allowance of all gases. Its constant would have an R-value of 34 for the BS 6853 standard and a CIT value of 108 m3/mg for the French. Under BS 6853, an R-value of 34 would reject all materials deemed acceptable for use in U.S. passenger railcars. Fortunately, most materials never attain a maximum for all component levels, and thus don't reach an R-value of 34 when tested. This example serves as the worst-case scenario for the SMP 800C Standard as it translates to the BS 6853 Standard.

A comparison of French Standard NF X 70-100 to the U.S. SMP 800C can't take place until after studying the rest of the components involved in evaluating materials for flame specifications. This is because the French specification for elastomers, NF F 16-101, uses a constant variable in determining the suitability of a material for an application. Unlike the R-value, the French CIT constant allows both smoke density and toxicity.

Smoke density comparison chart U.K. versus U.S.



U.K. BS 6853 Appendix D.8.3


A0 Class 1/Interior surfaces

2.6 0.01

A0 Class 1/Interior horizontal prone surfaces

2.6 0.01

A0 Class 1/Exterior vertical surfaces

4.4 0.01

U.S. Standard


Ds@ 1.5 min

100 14

Ds @ 4 min

175 55

For the French standard, a CIT value of 108 m3/mg can't be readily compared to SMP 800C without also examining smoke density requirements. The comparison uses values from actual testing of materials manufactured by Rogers Corp. along with the assumption that maximum smoke density values are low. Using the following equation:

With Is= 58 Candela (cd) when data taken from Rogers Bisco Silicone give a Dmvalue of 112 and a VS4 value of 86. In the French standard, an Isvalue of 58 cd will attain a class of F3. This is an average rating that would eliminate materials approved for use in U.S. railcars from all but one type of French railcar. These materials could only be used in applications that travel infrequently through tunnels, called a Class B rolling stock.

To evaluate toxicity based on the strengths of SMP 800C, one could examine the R and CIT values that could still be attained using high levels of one component. If other components in the material are low, the levels of a particular component such as hydrogen bromide (HBr) could be extremely high, thus failing the American standard but still passing its respective U.K. or French standards. This is possible when considering many of the existing fire retardants used in today's elastomers.

Standards for U.S., U.K., France, Germany
American ASTM E 162 SMP 800C ASTM E 662
U.K. BS 6853 Appendix A9 & BS 2782
(material dependent)
BS 6853 Appendix B.5 & NF X 70-100 BS 6853 Appendix D.8.3
France NF T 51-071
& NF C 20-455
NF X 70-100 NF X 10-702
Germany DIN 53438
Parts 1 to 3
& E DIN 54 837

Although the French and U.K.'s specifications may not cover these hazardous materials as stringently as the American, most European countries have their own limits (outside this specification) on HBr and other halogens that are common in flame retardant packages. Halogens and other hazardous materials are strictly monitored by governments and not permitted for use in these products. This eliminates any concerns of toxic fumes from those components.

The result here is that the U.K.'s BS 6853 specification is the most difficult to pass, followed by the French, then the American.

Optical smoke density is another critical component that must be considered when designing passenger railcars and associated equipment. While the level of risk to passengers is reduced when a railcar is designed with nontoxic materials, an inability to see the nearest exits could prove hazardous. A thick cloud of smoke can make breathing difficult, even if fumes are nontoxic.

Testing methods for optical smoke density are similar in all three specifications.

The French NF X 10-702 and American ASTM E 662 standards expose material to a mode of combustion inside a closed chamber. A record of the percentage of light transmittance lost is taken and expressed as Specific Optical Density, or Ds. Dstakes into account the percent of transmittance, the chamber volume, the sample size, and other factors. It lets designers compare materials despite chamber and sample size differences.

The American ASTM E 662 standard evaluates Dsat 1.5 min, 4 min, the maximum, and a corrected value due to deposits on chamber windows. The French NF X 10-702 standard not only determines Dsat 1.5 min and the maximum but also uses Dsdata taken at 1-min intervals up to 4 min (i.e., D1, D2, D3, and D4) and employs the following equation for determining VS4:

Because the American specification does not take measurements at D1to D4an extrapolation of a curve from the specified maxima (Dm) in the U.S. to the determined smoke density is used in the comparison.

Using the previous toxicity data, it's now possible to unravel the mysteries surrounding the smoke density of the French specification. The same toxicity equation for Is is employed in the calculation. Here the actual toxicity values attained from Rogers Bisco Cellular Silicone materials, for example, can be substituted making it possible to compare the French and American standards.

With VS4 = 432.5 and a value of 375 for Dm( calculated by assuming a like curve to previously tested materials), the final Isvalue would be equal to 20 cd. This provides a French rating of F1 and gives the materials approval for use in any type of rolling stock, equivalent to the American standard.

Although a comparison has been drawn, keep in mind the assumptions. A high smoke density does not necessarily equate to a low toxicity level and vice versa. In addition, the French specification requires both a SI and a fire index to determine product suitability. Therefore, a product with an F1 rating may still be of little value in a railcar depending on its fire rating.

The current ASTM Dslevel of 200 is being evaluated by both the Federal Railway Association and the Environmental Protection Agency and may be reduced to a level of 100 in the near future making the ASTM E 662 specification more stringent than the French.

Likewise, to compare the U.K.'s BS 6853 (Appendix D.8.3) to the American ASTM E 662 specification also requires an understanding of the difference between each standard's calculation. For BS 6853, the optical density, calculated by the following equations:

Due to the complexity of assigning a value to Dsand A0, it is reasonable to compare the U.K. and American standards through testing of two like materials. Rogers Bisco Cellular Silicones, for example, produce test results well beneath the American ASTME 622 requirements for Dsat both the 1.5 and 4-min marks. With respect to the U.K.'s BS 6853, the materials fall well below the most stringent standards of A0= 2.6. Although this does not indicate which standard is more difficult to pass it does show that a material that passes either specification by a large margin will also pass other specifications.

Flame testing according to NF T 51-071 and NF C 20-455
Class Oxygen Index
(NF T 51-071)
Glow Wire
(NF C 20-455)
I-0 > 70 No ignition at 960°C
I-1 > 45 No ignition at 960°C
I-2 > 32 No ignition at 850°C
I-3 > 28 Ignition does not persist
at 850°C after glow wire is removed
I-4 > 2 -
Not classified < 20 -


The flame, smoke, and toxicity standards of railcar materials vary throughout the world. The American Public Transportation Association (APTA) encourages designers to familiarize themselves with standards used in other developed countries. The APTA maintains that this approach will assist companies in streamlining their design-and-build process and will help to globalize the railcar industry.

The final category necessary for evaluating different specifications is fire tests. The French standard (NF C 20-455 and NF T 51-071) has extreme test conditions where materials must withstand exposure to a hot wire, without ignition. Also, the materials must meet or exceed a specified oxygen index. It is difficult to imagine a material achieving a rating higher than I-3, which states that ignition does not persist after exposure to an 850C glow wire and an atmosphere of >28% oxygen.

Under the U.K.'s Standard BS 2782 and BS 6853 Appendix 9, materials are exposed to different tests based on their end use. Because materials evaluated are typically used in gasketing, the fire test requirements of the minor use materials serve as an example.

The U.K.'s specification requires that materials pass both a limited oxygen test and a flammability temperature test. Depending on the rating, materials must require 28 to 34% oxygen/air content to burn. In conjunction, materials must have a flammability temperature of >250 or 300°C. The flammability temperature test is similar to the French specification in procedure, but as one can see, the U.K.'s specification is much less stringent.

In the case of the French standard, minor use materials must have a rating of I-3 or better. Therefore, the French specification requires nonpersisting ignition at 850°C where the U.K.'s specification requires no flammability at 300°C in its worst-case scenario.

The American specification ASTM E 162 evaluates materials based on flame spread as opposed to ignition temperature so it doesn't directly correlate to either the British or French specification. Rogers Bisco Silicones again serve as an example to help make conclusions about how the American specification compares to its French and U.K. counterparts. The Rogers materials are approved for use in all railcar applications. They are limited in the application area by both the French and U.K. specifications. Thus the American standard is less stringent because our products are limited in application area in both the French and U.K.'s.

The only requirement for the German standard DIN 53438 (Parts 1 to 3) for gaskets and other small parts is, quite simply, that flames not exceed 15 cm (5.9 in.) during a 20-sec observation period. Using Rogers materials as a test basis, this condition is even easier to pass than the American specification. Thus, as before, with respect to fire requirements, France is the most stringent, followed by the U.K., then the U.S., and finally Germany.

Rigid polyurethane invented by German chemist Otto Georg Wilhelm Bayer in 1937 quickly found use in aircraft wings and boot soles. And in the 1950s polyurethane foams were developed that had excellent load-bearing properties and resilience. They replaced horsehair, cotton, wool, and feathers in mattresses and upholstered furniture including car and airplane seats.

Unfortunately, the physical properties that give polyurethane foam its outstanding resilience also contribute to its flammability. The open-cell, foam structure gives flames ready access to oxygen after ignition. It's not surprising, therefore, that catastrophic accidents happen when designers don't properly specify materials. For example, in 2000, a South Korean train was engulfed in fierce flames, choking smoke, and toxic fumes after a disgruntled passenger ignited a milk carton full of explosive chemicals. In this case, appropriate material selection in the train's construction, could perhaps have saved 300 lives.

In most cases of fire, smoke asphyxiation is the leading cause of death, not flames. While it's important that materials be fire retardant, it's equally important that the materials emit low to nonexistent levels of smoke and that the smoke contains little that could harm passengers attempting to escape.

Since 1980 (during the aftermath of a San Francisco BART subway fire), the Polyurethane Foam Association, Wayne, N.J., has been educating flexible polyurethane foam specifiers, converters, and end users on the importance of fire prevention. It provides facts on environmental, health and safety issues related to polyurethane foam and urges designers to select the proper materials to meet end-use requirements, including compliance with applicable building codes, occupancy requirements, and flammability standards. It strongly encourages the installation of sprinkler systems and other fire-suppression technologies and advises caution even around products designed to be combustion resistant. Given enough heat and a source of ignition, even combustion-modified products can burn and/or generate dense smoke and toxic gases. More information about safety considerations is available at

According to the Alliance of the Polyurethanes Industry, Arlington, Va., there is no federal governance of building codes in the United States. The API is particularly concerned about deaths and injuries from fires involving residential furniture and mattresses that contain flexible polyurethane foam. The organization supports a combination of approaches to reduce these incidents including:

  • Addressing technically sound national standards.
  • Ensuring that products are properly labeled.
  • Fire safety education.

Information about the Alliance's position on polyurethane combustibility in the regulatory environment can be found on its Website:


A0= The optical density generated across the faces of a 1-m cube when one "unit" of material is tested (m2/unit). The "unit" varies with the method.

Am= The optical density measured in the 3-m cube test chamber, dimensionless.

cx= The emission of the x th species, mg/gm

Ci= Critical concentration, mggas/m3.It is the maximum concentration of gas that a human can withstand for 15 min without irreversible biological effects.

CIT = The Conventional Index of Toxicity constant, m3/gmmaterial

Dm= Maximum smoke density, dimensionless f

x = The reference value for the x th species, mg/gm

I0= The initial luminous intensity, cd

Is= The smoke index rating, cd

It= The transmitted luminous intensity, cd

l = The length of the optical path, m

r = The individual index for a particular species, dimensionless

R = The weighted summation of toxic fume, dimensionless

rx= The individual index for the x th species, dimensionless

ti= Content of gas (i) in a material, mggas/gmmaterial

T = The quantity of toluene in the fire source, as a percentage of volume.

V = The volume of the chamber, m 3

VS4= Value of obscuration due to smoke in the first 4 min of the test, dimensionless.

Rogers Corp., (630) 784-6200,

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.