Post-flashover Fire Research Articles

Experimental justification on thermal empirical equations for post-flashover compartment fires

Many useful correlation equations derived for estimating the heat release rate for a postflashover room fire were applied in performance-based design. A postflashover room fire is very hazardous as demonstrated by several big accidents in the past few years. The authorities are starting to challenge the use of such equations as fire engineering design tools. Additional justification of the results with experiments in fire hazard assessment is now required. Two correlation equations on heat release rates with gas temperature for postflashover fire are commonly used. One was reported by Babrauskas and Williamson and the other by McCaffrey et al., denoted as BW and MQH equations, respectively. Both equations were justified in this article by reported experimental data. Two sets of reported experimental results on postflashover room fires with transient heat release rates measured by oxygen consumption calorimetry were used. One set was reported by Chow et al. on studying flashover in a compartment fire with gasoline pool fires up to 2.8 MW. The other set was the experimental data on cable fires in a long cavity reported by Hietaniemi et al. Direct comparison indicated that heat release rates estimated using the BW equation are lower than measured values of both sets of the experiments. This point should be watched while applying the BW equation in fire hazard assessment. Better agreement was observed in the equation by MQH. A possible reason is that the empirical constants in the MQH equation were deduced by fitting with experimental data.

Assessing levels of hydrogen cyanide in fire experiments using a generalized correlation

A generalized relationship between the normalized yields of carbon monoxide and hydrogen cyanide for nitrogen-containing materials has recently been derived. This correlation is used in the current study to analyze experimentally derived hydrogen cyanide data from three sets of fire tests. For a reduced-scale compartment fire test, the yields of hydrogen cyanide with varied equivalence ratios and the transient hydrogen cyanide concentrations are estimated; for a series of room-corridor sofa fire tests, the extremely high hydrogen cyanide level observed is demonstrated to be a realistic result and a hydrogen cyanide yield value of 0.047 g/g is suggested for this sofa in post-flashover fires for fire safety assessments; and finally, for a series of smoke chamber tests with polyurethane, possible causes for the failure to detect hydrogen cyanide are suggested. Language: en

Possible Air Pumping Action in a Room Fire

The air pumping effect of a fire plume, proposed years ago, to give a higher air intake rate through vertical openings in a compartment fire is further investigated in this paper. Equations for the air intake rate through vertical openings found in the literature are reviewed. As most of the reported correlation expressions were derived empirically from experiments, results might be different if the fire geometry, fuel type and ambient conditions are different. Furthermore, most of the data on heat release rates in earlier experiments were measured from the mass loss rate of fuel, not by the oxygen consumption method. Values used therefore might deviate from the actual heat release rates. Aerodynamic flow through a doorway, induced by a compartment fire, is commonly simulated by Computational Fluid Dynamics (CFD). Factors which are difficult to control in experiments but affecting the doorway airflow can be studied more easily with CFD.In the study presented in this paper a two-room structure with a fire room connected to an adjacent room through a door was selected. Different door widths linking the two rooms were set to give different ventilation factors. A total number of 16 simulations using the popular CFD software, Fire Dynamics Simulator, based on large eddy simulations, were carried out with different heat release rates of a fire in the fire room. Predicted horizontal velocity components were used to estimate the air intake rates through the door to justify the air pumping action. It appears that there is an air pumping effect of the fire plume for cases with lower ventilation factors, even for pre-flashover fires. Results further confirmed earlier experimental data on post-flashover fires. Therefore, further study on possible air pump effects in a room fire is strongly recommended.

Observable characteristics of flashover

Definitions of flashover are multifarious but may be categorized as three types: mathematical, visual and those relying on diction. There are at least two arguments to be had with regard to the definition of flashover:(i) Is rapidity of transition to a state of total surface involvement an essential criterion for flashover?(ii) Is instability in the heat balance an essential criterion of flashover?This article postulates a one-dimensional scale with three major types of compartment fire (that evolve beyond smoldering but rely on unpressurized contained fuel); and two of those fire types can be split into two sub-categories. In all, five sub-categories exist ranging from fuel-restricted pre-flashover fires at one extreme, to vent-restricted pre-flashover fires at the other, via post-flashover fires in the middle where there is neither fuel nor vent restriction sufficient to prevent flashover; and with borderline (or weak flashover) fires lying between them on the scale.Each sub-category has a tendency to produce its own particular pattern of fire growth (in the form of hot gas temperature rise), which is different from one another and for which some explanation can be offered. By examining the temperature–time plots of 53 fixed-vent, single-source scaled laboratory fires it is concluded that rapidity of fire growth is a common but not an essential characteristic of flashover. Furthermore, conditions considered to be post-flashover can be attained without much/any evidence of thermal instability.It is suggested that the most convenient definition of ‘flashover’ for experimentalists is a visual one, based on observation of flame projecting from the vent; and that any dictionary style definition should be based on this observation.

Experimental review of the homogeneous temperature assumption in post-flashover compartment fires

Traditional methods for quantifying and modelling compartment fires for structural engineering analysis assume spatially homogeneous temperature conditions. The accuracy and range of validity of this assumption is examined here using the previously conducted fire tests of Cardington (1999) and Dalmarnock (2006). Statistical analyses of the test measurements provide insights into the temperature field in the compartments. The temperature distributions are statistically examined in terms of dispersion from the spatial compartment average. The results clearly show that uniform temperature conditions are not present and variation from the compartment average exists. Peak local temperatures range from 23% to 75% higher than the compartment average, with a mean peak increase of 38%. Local minimum temperatures range from 29% to 99% below the spatial average, with a mean local minimum temperature of 49%. The experimental data are then applied to typical structural elements as a case study to examine the potential impact of the gas temperature dispersion above the compartment average on the element heating. Compared to calculations using the compartment average, this analysis results in increased element temperature rises of up to 25% and reductions of the time to attain a pre-defined critical temperature of up to 31% for the 80th percentile temperature increase. The results show that the homogeneous temperature assumption does not hold well in post-flashover compartment fires. Instead, a rational statistical approach to fire behaviour could be used in fire safety and structural engineering applications.

N-Class Fire Resistant Divisions in U.S. Naval Ships

Abstract The U.S. Navy has now developed an N-Class division system to classify fire resistant boundaries in accordance with MIL-STD-3020, fire resistance on U.S. naval surface ships. The N-Class division is analogous to the commercial International Maritime Organization (IMO) system (e.g., A-Class) division. The key difference is that N-Class divisions are designed to have satisfactory fire resistance when exposed to hydrocarbon pool fire (UL 1709 fire exposure) following a shock event. In addition, N-Class is applicable to fire resistant divisions constructed from polymer composites structures when they meet the requirements of DDS-078-1. MIL-STD-3020 requires a hydrocarbon pool fire exposure (UL-1709 fire exposure) which provides a minimum average total heat flux of 188 kW per square meter (kW/m2) (60,000 British thermal unit per square foot/hour (Btu/ft2∣h)) within the first 5 min of the test exposure. At all times after 5 min, the average temperature within the furnace shall be maintained at a minimum of 1038 °C (1900 °F) and a maximum of 1149 °C (2100 °F) for the duration of the test. These fire conditions simulate a post-flashover fire scenario for shipboard applications. Performance is defined as the time period during which assemblies will continue to perform their intended function when subjected to fire exposure. The minimum duration of fire exposure is 30 min. Shock testing prior to fire resistance test is performed to ensure that fire resistance of divisions, with fire insulation, penetrations, and associated attachment methods, is not degraded in a combat environment. Shock test specimen size of 1219 by 3048 mm (4 by 10 ft) was selected to reduce the cost of shipping a full size fire test specimen of 3048 by 3048 mm (10 by 10 ft) or greater to the fire test laboratory. A full scale fire resistance test is still required for qualification purposes after successful completion of the shock test prior to fire resistance test. MIL-STD-3020 also permits AN-Class fire resistant divisions when fire threat is from common combustibles (Class A) and where liquid flammable (Class B) fire threat is non-existent. AN-Class divisions are those divisions formed by bulkheads and decks that are designed to protect against structural failure and prevent the passage of flame or hot gases when exposed to IMO Resolution A.754(18) fire exposure after shock testing. For all AN-Class divisions, including those with penetrations, the fire test duration is a minimum of 60 min. This paper will discuss fire resistance test results conducted using steel, aluminum, and composite substrates.

Fuel Load and Peak Heat Release Rate Correlations in Post-Flashover Room Fires

Flashover might occur rapidly as a consequence of non-accidental building fires. Response of the stored combustible items (being ignited and burned) should be studied carefully in such developed fires, which are ventilation-controlled. Air supply rate and heat fluxes acting at the materials would affect burning of the combustibles. A wide range of peak heat release rates with different fire durations results under different heat fluxes. The amount of combustibles denoted by fire or fuel load commonly adopted to quantify fire risks is no longer the key point in generating the heat release rate. A room with a large fire load may not give high heat release rates if the air supply is inadequate, or if the radiant heat flux is low. In this article, correlation of the peak heat release rate in burning the material under flashover with the fire load of the combustibles with adequate ventilation will be investigated. Data compiled from full-scale burning tests on well-developed room fires with steady burning are used to deduce a linear correlation between the possible peak heat release rate and the fire load density under adequate ventilation conditions.

Assessment of the Use of Fire Dynamics Simulator in Performance-Based Design

This paper discusses a procedure for the use of fire modelling in the performance-based design environment to quantify design fires for commercial buildings. This procedure includes building surveys, medium-and full-scale experiments and computer modelling. In this study, a survey of commercial premises was conducted to determine fire loads and types of combustibles present in these buildings. Statistical data from the literature were analysed to determine the frequency of fires, ignition sources, and locations relevant to these premises. Based on the results of the survey and the statistical analyses a number of fuel packages were designed that represent fire loads and combustible materials in commercial buildings. The fuel packages were used to perform medium- and full-scale, post-flashover fire tests to collect data on heat release rates, compartment temperatures and production and concentration of toxic gases. Based on the experimental results, input data files for the computational model, Fire Dynamics Simulator (FDS), were developed to simulate the burning characteristics of the fuel packages observed in the experiments. Comparative analysis between FDS model predictions and experimental data of HRR, carbon monoxide (CO), and carbon dioxide (CO2), indicated that FDS model was able to predict the HRR, temperature profile in the burn room, and the total production of CO and CO2 for medium- and large-scale experiments as well as real size stores.

Design Fire Experiments for Commercial Premises

Two series of experiments were conducted to determine the burning characteristics of different fuel packages, in both pre- and post-flashover fires, to develop data to characterize design fires for commercial premises. These fuel packages represented fuel loads determined from the survey of commercial buildings. Results suggest substantial differences in the burning characteristics of different stores. In this article, the description of the recommended design fires includes: (1) types of combustibles, (2) fire load density (MJ/m 2), (3) fire growth rate, and (4) yields of CO and CO2.

Are Two 2-Hour Fire Rated Shutters Equivalent to a 4-Hour Shutter Using ASTM E119?

Performance-based design is commonly applied to projects that have difficulties in complying with prescriptive fire codes in the Far East. There are procedures for evaluating fire resisting construction and structural elements with fire resistance ratings lower than the values specified in the fire codes. The equal temperature-time exposure concept or t-equivalent rule is commonly applied to support the argument. This paper will discuss whether installing two fire shutters each of a 2-hour fire resistance period is equivalent to one 4-hour shutter. The standard temperature-time curve under ASTM E119 was applied to show they are not equivalent and to illustrate the inadequacy. Further, there are concerns in applying such an approach for bigger fires as the radiative heat flux acting on the construction element is not included. Fire hazard assessment on fire resisting construction with lower ratings under postflashover fires should be evaluated with caution. Full-scale burning tests with bigger fires are suggested as a means to support the argument or demonstrate that the equal temperature-time concept would work as expected.

PREDICTION FOR FIRE BEHAVIOR FOCUSED ON THE CHANGES IN SPECIES CONCENTRATION AND QUANTITY OF HEAT TRANSFER IN A COMPARTMENT

A model for predicting fire behavior in a compartment is formulated that considered the following two points regarding mass loss rate: (i) the heat transfer from flame to fuel is decreased with lack of oxygen concentration; and (ii) the heat transfer from walls, ceilings and smoke layer to fuel with development of compartment fire. In addition a computer simulation model based on zone concept is developed that considers conditions in a compartment from the initial period to the decay period of compartment fire. Namely, this approach applies a two-layer zone model for the growth or pre-flashover stage and a one-layer zone model for the fully developed or post-flashover fire. The compartment fire experiment is performed to validate this model. In general, good agreement between experimental and theoretical results about the mass loss rate and the temperature in the compartment is obtained.

An Overview of Design Fires for Building Compartments

The worldwide trend of moving towards performance-based codes has brought about a need for computational tools, most of which rely on suitably-defined design fires, to adequately predict the impact of fires on buildings and their occupants. This literature review was carried out to determine the range of methods used to characterize design fires. The methods currently available were found to be largely empirical in nature and unsophisticated. The two main quantities used to describe design fires are the heat release rate (pre-flashover scenario) and temperature-time profiles (post-flashover). The most commonly used pre-flashover design fires are t 2 fires, whereas a host of empirical correlations are available for post-flashover design fires.

Development of Design Fires for Performance-Based Fire Safety Designs

This paper discusses a procedure for the development of design fires for buildings and presents a study undertaken to quantify design fires for commercial buildings. This procedure includes building surveys, full-scale experiments and computer modelling. In this study, a survey of commercial premises was conducted to determine fire loads and types of combustibles present in these buildings. Based on the results of the survey a number of fuel packages were designed that represent fire loads and combustible materials in commercial buildings. The fuel packages were used to perform full-scale, post-flashover fire tests to collect data on heat release rates, compartment temperatures and production and concentration of toxic gases. Results from these experiments are presented and discussed. The paper also discusses a computer modelling study in which a computational fluid dynamics model was used to simulate the tests and to model a fire in a real-scale commercial store.

BRE large compartment fire tests—Characterising post-flashover fires for model validation

Reliable and comprehensive measurement data from large-scale fire tests are needed for validation of computer fire models, but is subject to various uncertainties, including radiation errors in temperature measurement. Here, a simple method for post-processing thermocouple data is demonstrated, within the scope of a series of large-scale fire tests, in order to establish a well characterised dataset of physical parameter values which can be used with confidence in model validation. Sensitivity analyses reveal the relationship of the correction uncertainty to the assumed optical properties and the thermocouple distribution. The analysis also facilitates the generation of maps of an equivalent radiative flux within the fire compartment, a quantity which usefully characterises the thermal exposures of structural components. Large spatial and temporal variations are found, with regions of most severe exposures not being collocated with the peak gas temperatures; this picture is at variance with the assumption of uniform heating conditions often adopted for post-flashover fires.

Effects of varying Smagorinsky constant on simulating post-flashover fires

Air flow post-flashover room fire is weakly compressible. In this paper, Navier–Stokes equations in an appropriate form will be applied to simulate such motion. Large eddy simulation (LES) is used with the Smagorinsky sub-grid scale model. A pool fire was described by the combustion model based on simple and fast chemical reaction. Three-dimensional simulations were carried out by using different values of Smagorinsky sub-grid scale constant. Predicted results on air temperatures and radiative heat fluxes were compared with the experimental values. Results illustrated that the value of the Smagorinsky constant would affect the simulations for strong buoyant plume. Value of the Smagorinsky constant from 0.20 to 0.21 appears to be reasonable in simulating flashover with a filter width of 0.1 m.

Smoke Toxicity Data for Fire Hazard and Risk Assessment of Cable Products

Abstract This paper presents the methodology for and results from a series of room-scale fire tests to produce data on the yields of toxic products in both preflashover and postflashover fires. Trays of common household electric cable burned in a room with a long adjacent corridor. The yields of CO2, CO, HCl, HCN, and carbonaceous soot were determined. Other toxicants (e.g., NO2, formaldehyde and acrolein) were not found; concentrations below the detection limits were shown to be of limited toxicological importance relative to the detected toxicants. The uncertainties in the post-flashover data are smaller due to the higher species concentrations and the more fully established upper layer from which the fire effluent was sampled. The uncertainty values are comparable to those estimated for the fractional effective dose calculations used to determine the time available for escape from a fire. The uncertainty in the yield data is sufficiently small to determine whether a bench-scale apparatus is producing results that are similar to or different from the real-scale results here. The use of Fourier transform infrared spectroscopy was shown to be a useful tool for obtaining concentration data of toxicants. However, its operation and interpretation is far from routine. The CO yield was slightly lower than the expected value of 0.2, which should be used in hazard and risk analyses. The accuracy of the results is verified, and a hypothesis is offered for the lower CO yield values.

Quantitative comparison of FDS and parametric fire curves with post-flashover compartment fire test data

This paper presents a comparison of two parametric fire modelling techniques (Eurocode 1, and the BFD curve method) and one field model (fire dynamics simulator) against large-scale post-flashover test data. Using a method of the product moment correlation coefficient, it is shown that the BFD curve predictions are most closely representative of reality. For the computational test data, two grid resolutions are adopted in the FDS field model, the finer of which having comparable results in terms of regression analysis to the BFD method. Both the field model and the BFD curve method were found to give better predictions compared to the Eurocode method over the duration of the test. However, a direct comparison of the maximum gas temperatures shows the field model to be poorer in its predictive capability than the parametric methods, under-predicting the maximum gas temperatures. In addition, a more in-depth analysis of the FDS predictions indicates that by considering simply average compartment temperatures the more inaccurate spatially specific temperature predictions were disguised. This study provides useful quantitative data on the three techniques presented and discusses more general issues concerning fire modelling.

Simulating the effects of fuel type and geometry on post-flashover fire temperatures

This paper discusses the effect of fuel type and geometry on predicted compartment temperatures derived from computer modelling of post-flashover compartment fires. Many previous studies have investigated post-flashover fires with either wood crib or liquid pool fuels, but very few analytical or experimental studies have considered realistic wood-based fuels with different ratios of surface area to volume, combined with plastic-based fuels. A simple single zone fire model was used to calculate the temperatures in post-flashover compartment fires. The program includes a catalogue of furniture items, each with fuel mass loss rate evaluated on the basis of a constant regression rate on all exposed surfaces. The program also includes a pool-burning model and considers wood fuels and thermoplastic fuels burning together inside a compartment. Use of the model shows that the total fuels load alone is not sufficient to characterise a post-flashover fire. The fire temperature is highly dependent on the fuel type and geometry. For given ventilation and total fuel load, the resulting temperature depends greatly on the average thickness of the wood fuels and the presence of thermoplastic fuels. The ratio of the available fuel surface area to the ventilation opening is particularly important. Several fire scenarios involving different fuel types and characteristics are simulated and compared with Eurocode parametric fires.

Species Transport from Post-Flashover Fires

A study was performed to determine the use of an equivalence ratio to predict gas levels (CO, CO2, O2, and unburned hydrocarbons) transported to locations remote from a post-flashover compartment fire. A series of tests were conducted in a reduced-scale facility to measure the evolution of post-flashover compartment fire gases flowing down a hallway. Test variables included air entrainment into gases in the hallway, stoichiometry of the compartment fire gases entering the hallway, mass flow rate of compartment fire gases, and the presence of a vitiated smoke layer accumulated in the hallway. In cases with no layer accumulated in the hallway, species yields in the hallway were found to correlate with a control volume equivalence ratio. The control volume equivalence ratio is the ratio of the mass loss rate of fuel inside the compartment to the air flow into the compartment plus the air entrained into compartment fire gases flowing along the hallway. Layers that accumulate in the hallway were determined to limit oxidation, which in some cases resulted in CO yields transported to remote locations being 20% higher than those inside the compartment. Based on the experimental data, a methodology was developed for predicting species levels transported to remote locations.

Evaluation of Models of Fully Developed Post-flashover Compartment Fires

Closed-form models of fully developed enclosure fires are evaluated by comparing model predictions to temperature and burning rate data from experiments. The experimental data selected represent a wide spectrum of ventilation conditions, including fuel-controlled and ventilation-controlled burning. Most methods are found to underpredict either compartment fire temperatures or duration of burning under some conditions. Based on the comparisons, a recommendation is made as to which method should be used in a design or analysis context to model post-flashover fires in compartments that are nearly cubic in shape.