of air flow on flame propagation could also prove troublesome in such systems if good design is not achieved. These latter factors detract from this system's ability to produce a complete design parameter for the material, and could affect reproducibility between systems. This test method offers the most promise of obtaining a meaningful design parameter which relates to the hazard and can be related to the building system with increasing predictive accuracy as more knowledge of building fires is developed. As our understanding has developed on the potential hazard of flame spread over flooring materials, it has become clear that radiation from an external source is a very important factor in this type of fire spread. In particular, the application intended from this program has been the limitation of fire spread through corridors and exitways. Hence, the determination of the minimum radiant flux level necessary to sustain flame spread on flooring materials is a parameter directly related to the hazard and potentially useful in a fire safety design analysis in buildings. Hopefully, a flooring radiant panel system can be designed and operated satisfactorily to measure, with a suitable degree of precision, this "property" referred to as the "critical irradiance to support flame spread." In addition, a theoretical development of flame spread under an external radiant source, as proposed by eq (3), would enhance the understanding of this test method, and support the "property" interpretation of "critical flame spread irradiance." Thus, some additional work needs to be done to support the design and interpretation of this method. R A final consideration is the application of the performance measured by such a flooring radiant panel test method. Initially, judgement might be applied in setting performance acceptance levels for building corridors and exitways in various building occupancies. Guidance from recent full-scale experimental fires could be used to set performance levels. In addition, an empirical design approach could be used to determine acceptable performance and guidance in extrapolation to full-scale data. For example, calculations can be made based on energy release from a fire within a corridor spreading products in two directions. The radiation flux levels to the floor q" can be estimated for different fire energy release rates (E) and different corridor heights (H) and widths (W). Some results of this kind of analysis are shown in figure 15. In this calculation it was assumed that all of the radiation comes from the ceiling and upper walls which are heated to the temperature of the fire plume intersecting the ceiling. A simple line fire plume theory was used and based on an analysis by Alpert [31] the gas temperature of the ceiling jet does not greatly vary with corridor length. The corridor was assumed infinitely long and no accounting was made for direct flame radiation. The result for floor radiant flux is Thus, this equation could be used to calculate the potential radiative floor flux in a building corridor. The "critical irradiance" measured by the flooring radiant panel test could then be used as a means of evaluating the material for the building corridor considered. Figure 15. Estimated radiant flux to floors in building corridor fires. 3. CONCLUSIONS The flammability of flooring materials has been examined through accident analysis and full-scale fire experiments. These results indicate that (except for carpets failing the Pill Test) a large ignition source is necessary to initiate flame spread on flooring material. Radiant heating of the flooring material due to a large ignition fire, along with building interior geometry, can lead to rapid flame spread over the flooring. This is recognized as the potential hazard of flooring materials in building corridors. Flame spread through corridors would block escape routes and transmit the fire to other areas of the building. Available and proposed test methods for flame spread of flooring materials have been examined. Based on this study and on the dynamics of corridor fires a radiant panel test is recommended for flooring materials. Since this test can measure the flame spread behavior of flooring material under a controlled external radiant flux, it can yield data of quantitative value. Moreover, by estimating potential radiant flux levels in particular building occupancy type, a direct means of linking the test result to the application of the product can be achieved. [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] 4. REFERENCES Yuill, C. H., The Flammability of Floor Coverings, J. Fire & Flammability, Carpet and Rug (Pill Test) Standard for the Surface Flammability of Robertson, A. F., Private Communications, National Bureau of Standards, Chandler, S. E., Fires Due to the Ignition of Carpets and Floor Coverings, 1955-62, Fire Res. Note No. 623, Fire Res. Station, Borehamwood, Herts, U.K. Coverings FRO/80/018, Fire Res. Station, Borehamwood, Herts, U.K., Fung, F. C. W., Suchomel, M. R. and Oglesby, P. L., The NBS Program on Huggett, C., Carpet Flammability and the NBS Corridor Fire Program, Fung, F. C. W., Suchomel, M. R. and Oglesby, P. L., NBS Corridor Fire Christian, W. J. and Waterman, T. E., Fire Behavior of Interior Finish Waterman, T. E., Private Communications, IIT Research Institute, Segall, W. M., Private Communications, The Carpet and Rug Institute, McGuire, J. H., The Spread of Fires in Corridors, Fire Technology, Vol. 4, Steiner, A. J., Method of Fire-Hazard Classification Building Materials, Standard Method of Test for Surface Burning Characteristics of Building Lee, T. G. and Huggett, C., Interlaboratory Evaluation of the Tunnel Test (ASTM E-84) Applied to Floor Covering, Nat. Bur. Stand. (U.S.), NBSIR 73-125 (Mar. 1973). Surface Flammability of Materials Using a Radiant Heat Source, ASTM Robertson, A. F., Gross, D. and Loftus, J., A Method for Measuring Surface Flammability of Materials Using a Radiant Energy Source, Proceedings, Am. Soc. Testing and Materials, Vol. 56, 1956, 1437-1453. Gross, D., and Loftus, J. J., Flame Spread Properties of Building [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] Lee, T. G., Loftus, J. J. and Gross, D., Effect of Moisture on Engermann, H. S., Floor Covering System A Test for Flame Propagation Denyes, W. and Raines, J., A Model Corridor for the Study of Flamma- Denyes, W. and Quintiere, J., Experimental and Analytical Studies of Hartzel, L. G., Development of a Radiant Panel Test for Flooring Roberts, A. F. and Clough, G., The Propagation of Fires in Passages de Ris, Duct Fires, Combustion and Science Technology, Vol. 2, 239- Rockett, J., Mathematical Modeling of Radiant Panel Test Methods, Rhodes, A. C. and Smith, P. B., Experiments with Model Mine Fires, The Kashiwagi, T., Private Communications, National Bureau of Standards, [30] Huggett, C. and Lee, T. G., Investigation of Carpet Flammability Test Methods, Nat. Bur. Stand., NBS Report 10585 (Sept. 24, 1971). [31] Alpert, R. L., Fire Induced Turbulend Ceiling-Jet, Factory Mutual NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 411, Fire Safety Research, Proceedings of a Symposium Held at NBS, Gaithersburg, Md., August 22, 1973, (Issued November 1974) MATHEMATICAL MODELING OF RADIANT PANEL TEST METHODS J. A. Rockett National Bureau of Standards, Washington, D.C. Standard flame spread tests characterize complex physical phe- Key words: Fire modeling; fire test methods; flame spread; test method. There are several recognized fire test methods in which a sample is exposed to a radiant energy flux and, after either self or piloted ignition, burns [1,2]. The standard measurement of material performance is either the extent or rate of burning. Commonly, an index is computed from the measured quantities and is the only test result usually reported. An analytic model of a test method can provide information on the material properties and sample characteristics controlling the test outcome. This information, much more than the contrived test figure of merit, tells something about the burning characteristics to be expected from materials in actual fire situations. The modeling process yields material properties and sample characteristics important to the specific burning situations modeled and illuminates what the test does (and does not) measure. The model also allows some inferences on the extent to which test results fully characterize the material and sample variables needed to predict performance in fire situations. 2. FLAME SPREAD AGAINST THE WIND A decade ago Gross and Loftus presented a simple model for radiant panel tests [3]. Their model suggested some similarity relations which were well verified by their experiments. Recently, J. Quintiere has developed a more detailed model for flame spread under a situation similar to that found in radiant panel tests. His model, derived for the case of propagation of flames against the ambient wind, appears to represent quite well the principal features of the test methods for sufficiently thin samples. This is appropriate because, although there is no forced convection with most radiant panel test methods, a certain amount of natural convection is always present, and in some test configurations quite a bit. Usually, the experimental configuration is arranged so that the flame propagates against such wind as is present. Because of the importance of test modeling, Quintiere's model has been rederived in a somewhat more general way and extended to more general sample configurations. The present model is for thermally thick and laminated fuel beds. The model assumes there is a heat flux, Q, acting on the sample which is composed of two parts: the heat flux, QR, supplied to the sample from the radiant panel, which is normally a function of distance along the sample length, and the reflexive heat flux, Qp, from the flame back to the sample surface. |