The structural analysis program that is being developed includes a capability f either computing the material property tables from room temperature values or reading in a full set of material property tables for the full range of tempera cures. The creep analysis in this program is based on the Harmathy-Dorn [10,11] theory and on a modification of the time hardening rule5. [20] The thermal properties of concrete and steel are based mainly on the comprehensive summary in Bizri's thesis [21] and on Harmathy's recommendations. [9,11] 2.2.1. Outline of Creep Theory The Dorn [22] theory states that if the creep curves for metals at the same stress but at different temperatures are plotted against the temperature compensated time, which is defined below, all the curves will be coincident. This immediately leads to a great simplification in analysis, as now only a single curve (and its associated equation) is needed to calculate creep strains at constant stress. This is shown in figure 2. Harmathy [10] extended the temperature range of the Dorn theory and added a term to account for the stress level. The term turns out to be the Zener-Hollomon parameter. [9,11,23] The Zener-Hollomon parameter is the slope of the secondary portion of the creep curve when it is plotted against the temperature compensated time. This is shown in figure 2. Thus, this parameter may be considered as a temperature compensated creep rate and it is a function of stress only. Harmathy [10] proposed the following equations for calculating the creep in reinforcing and prestressing steels: /de/de = coth2 (ɛ/εo) which when integrated and rearranged gives (4) where Z = Zener-Hollomon parameter (function of material and stress; (fig. 2) = temperature compensated time (function of temperature and material) = intercept of the secondary portion of creep curve plotted against the temperature compensated time (function of material and stress; see figure 2). The restriction on eqs (4) and (5) is that either the stress is constant or at most the stress is slowly changing with time. Under the conditions of a fire test, this restriction is satisfied. 5This is in conformity with Harmathy's theory which is discussed below. used. In order to solve for creep using eq (5), an iterative procedure must be Thus, Harmathy's recommendations greatly simplify the computation problems for estimating the creep in the steel, since varying temperature level and moderately varying stress level are included in a single equation. It is assumed that since the creep of the concrete in the compressive zone of the beam is much smaller than the creep in the steel, it may be neglected in the analysis of simply supported beams. The assumption is based on the fact that in most fire tests the average temperature of the concrete compressive zone is generally within the range of elastic or elasto-plastic behavior. In the structural analysis program, the calculations are controlled by two program elements. These are designated MAIN and BEAMB. MAIN controls the reading in of the problem data, the preparation of tables of material properties, the initializing of the problem, controlling the time sequencing of the problem and reading in the temperature data that had previously been prepared by AMG065. BEAMB controls the structural calculations within each time cycle and calculates either the ultimate normal temperature capacity of the beam section, the stresses and strains in the concrete and steel under working loads (required for initializing the creep calculations), or the capacity of the beam section during the process of a fire test exposure. These two program elements also call other subprograms which perform specific calculations during a run, such as setting the temperature of each element, calculating or reading in the temperatures of each bar, creep calculations and so forth. 2.4. Verifications of Structural Analysis The structural analysis computer program will be verified by comparing the analytic results with available test data. In this, some discrimination is necessary as it has been found that test results from different laboratories vary. The computer programs will be used to generate fire performance information for a variety of beam shapes. These data will be used to develop design aids such as graphs, tables, and nomographs for use by designers, architects, building code officials and other interested users. Based on experience with its use, the computer program will be modified. Possible modifications are: inclusion of creep in concrete, calculation of available strength at selected fire exposure times and testing of alternative creep algorithms. This work was sponsored in part by the Public Building Service, General Services Administration. Isotherm data were provided by the Portland Cement Association. [1] Standard Methods of Fire Tests of Building Construction and Materials, American Society for Testing and Materials Designation E119, American Society for Testing and Materials, Philadelphia, Pa. (1973). [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] Issen, L. A., Scaled Models in Fire Research, Journal of the PCA Hamming, R. W., Numerical Methods for Scientists and Engineers (McGraw- Zienkiewicz, O. C., The Finite Element Method in Engineering Science (McGraw-Hill Book Co., New York, 1971). Issen, L. A., Gustaferro, A. H. and Carlson, C. C., Fire Tests of Concrete Members: An Improved Method for Estimating Thermal Restraint Forces, Fire Test Performance ASTM STP 464, American Society for Testing and Materials, Philadelphia, Pa. (1970). Abrams, M. S., Gustaferro, A. H. and Salse, E. A. B., Fire Tests of Gustaferro, A. H., Abrams, M. S. and Salse, E. A. B., Fire Resistance Ehm, Herbert, A Contribution to the Design of Fire Endurance of Rein- Harmathy, T. Z., Thermal Performance of Concrete Masonry Walls in Fire, Harmathy, T. Z., A Comprehensive Creep Model, Transactions A.S.M.E. Harmathy, T. Z. and Stanzak, W. W., Elevated-Temperature Tensile and Gustaferro, A. H., Fire Endurance of Concrete Slabs and Beams, unpublished proposals to the American Concrete Institute Committee, No. 216 (1973). Edwards, A. L., TRUMP: A Computer Program for Transient and SteadyState Temperature Distribution on Multi-dimensional Systems (available from the NTIS, Springfield, Va., Report No. UCRL-14754, Rev. II). Schauer, D. F., FED: A Computer Program to Generate Geometric Input for the Heat-Transfer Code TRUMP (available from the NTIS, Springfield, Va., Report No. UCRL-50816). Edwards, A. L., TRUMP/XB2: A Computer Program for Processing the TRUMP [16] [17] [18] [19] [20] [21] [22] [23] Brisbane, J. J., Heat Conduction and Stress Analysis of Solid Propellant Perry, Robert H., et al, John H. Perry's Chemical Engineers Handbook, 4th Edition (McGraw-Hill, New York, 1963). Myers, G. E., Analytic Methods in Conduction Heat Transfer McGraw-Hill Hansen, N., BEAMBUSTER: Program Solves for the Moment-Curvature Leckie, F. H., Some Structural Theorems of Creep and Their Implications, Bizri, H., Structural Capacity of Reinforced Concrete Columns Subjected Zener, C. and Hollomon, J. H., Effects of Strain Rate on the Plastic 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) SMOKE AND CARBON MONOXIDE GENERATION FROM BURNING SELECTED PLASTICS AND RED OAK Thomas Y. King1 Armstrong Cork Company, Lancaster, Pennsylvania This paper presents preliminary results of simultaneous smoke Key words: Carbon monoxide; electrostatic precipitation; particu- A material's flammability (i.e., flame spread, ease of ignition, etc.) has in the past been the primary consideration in fire safety evaluation. Now with a greater understanding of those factors which contribute to fire injuries and deaths, smoke and toxic products are becoming of increasing concern. Even a relatively good flame retardant or thermally stable material, which by itself would not ordinarily propagate a flame, should be considered in the total combustible system. In most actual fire situations other more flammable material will be available which could result in partial or full involvement of materials of low flammability. The deleterious effects of smoke in a fire result from decreased visibility, both for the escaping occupant and the rescuing firefighter; inhalation of particulates which restrict respiration and can induce panic; and the toxic gases that accompany the smoke and account for a large fraction of fire deaths. A study of the physical and chemical parameters which affect the smoke generating potential of various materials is important not only in assessing possible hazardous materials, but also in determining the means by which the hazard can be reduced. This paper presents preliminary results of measurements of smoke and gaseous combustion products from selected materials in a smoke density chamber. Since the research described has recently been started, this paper is designed primarily to introduce the overall program and indicate its likely direction. 2. SMOKE AND CARBON MONOXIDE GENERATION Smoke can be defined as the solid particulate and/or liquid aerosol generated from the incomplete oxidation of organic materials and the release of certain inorganic materials during the burning process. The gaseous combustion products are not included in the above definition of smoke. Factors which can affect the smoking tendency of materials include: composition, density and thickness, heat flux, air flow, oxygen availability, pressure, surface characteristics, and geometry. Carbon monoxide is formed from the incomplete oxidation of the decomposition products, and the same factors which affect smoke production might be expected to influence carbon monoxide generation. 1The author was a Research Associate at the National Bureau of Standards, Washington, D.C. when the work reported here was performed. |