Search

Concurrent flow flame spread experiments are conducted over thermally thin solid fuels in microgravity aboard the International Space Station (ISS) under varying levels of confinement. Samples of cotton fiberglass blended textile fabric are burned in air flows in a small flow duct. Baffles are placed parallel to the sample sheet, one on each side symmetrically. The distance between the baffles is varied to change the confinement of the burning event. Three different materials of baffles are used to alter the radiative boundary conditions of the space that the flame resides: transparent polycarbonate, black anodized aluminum, and polished aluminum. In all tests, samples are ignited at the upstream leading edge and allowed to burn to completion. The results show that at low flow speeds (

Microgravity experiments are performed to study concurrent-flow flame spread over an array of thin cellulose-based fuel samples, using NASA Glenn Research Center's 5.18 s drop tower. Sample segments are distributed uniformly, separated by air gaps, on a sample holder. The exposed width of each sample segment is 5 cm. Two segment lengths, 0.5 cm and 1 cm, are tested. The gap sizes are varied in different tests, ranging from 0.5 to 5 cm. In all tests, a low-speed air flow (30 cm/s) is imposed and the upstream-most fuel segment is ignited by an electrical ignition wire. Upon ignition, the flame spread is recorded by two video cameras from the front and side-view angles. Spread rates, flame lengths, and burning durations are extracted using a custom video processing code. Similar to continuous fuels, flame spread over discrete fuels is a continual process of ignition. A burning discrete fuel segment, before it is consumed, needs to ignite the subsequent segment in order to have flame propagation across the gap. During this process, larger gaps between samples reduce the effective fuel load, increasing the apparent flame spread rate. However, larger gaps also reduce the heat transfer between adjacent samples, decreasing the sample burning rate. As a result, as the gap size increases, the flame spread rate increases but the burning rate decreases. At the same gap size, the flame spread rate is higher for the shorter tested sample segments. When considering sample configurations of the same fuel ratio (fuel length over the summation of the fuel and gap lengths), the spread rates are similar. This trend remains until a critical gap size is reached and flame fails to propagate across the entire array of samples. The critical gap sizes are similar for the two tested sample segment lengths and are suspected to be determined by the flame length.

International audience; Novel, high-fidelity results related to soot from microgravity flames were obtained by an international topical team on fire safety in space. More specifically, embedded optical techniques for evaluation of the soot-related radiative feedback to the base material from a spreading non-premixed flame in microgravity were developed. The configuration used a non-buoyant axisymmetric flame propagating in an opposed laminar stream overa Low Density PolyEthylene coating of an electrical wire. Within this context, both the standard Broadband Two Color Pyrometry (B2CP) and its recent extension Broadband Modulated Absorption/Emission (BMAE) technique can be deployed to measure the spatial distribution of soot temperature and volume fraction within the flame. Both fields are then processed to establish the field of local radiative balance attributed to soot within the flame, and ultimately the soot contribution to the radiative flux to the wire. The present study first assesses the consistency of the methodology contrasting an experimental frame and a synthetic one, the latter being produced by a signal modeling that processes fields delivered by a numerical simulation of the configuration as inputs. Using the synthetic signals obtained, the fields of local radiative balance within the flame are then computed and significant discrepancies were disclosed locally between the fields originating from the synthetic BMAE and B2CP inputs. Nevertheless, the subsequent evaluation of the soot-related radiative heat feedback to the wire shows that a weak deviation among the techniques implemented is expected. This finding is corroborated by similar evaluations conducted with experimental BMAE and B2CP measurements obtained in parabolic flights. As BMAE is implemented in an ISS configuration within the SCEM rig, BMAE and B2CP will soon provide long-duration soot observations in microgravity. In order to contrast the upcoming results, this current study quantifies discrepancies originating from the post-processing regarding soot temperature and volume fraction, and shows that the radiative feedback evaluation from both methods should be consistent.

Better understanding the burning behavior of solid fuels is necessary in order to be able to estimate more accurately the fire risk that they might represent. The burning rate is an essential component when assessing the growth of a fire, since it determines the heat released during the fuel burning. The objective of the present work is to study the effect of ambient pressure and oxygen concentration on the burning of cylindrical samples of polymethyl-methacrylate (PMMA) in opposed flow flame spread conditions. Experiments are conducted using pressures ranging between 100 and 30 kPa and oxygen concentrations between 19% and 23%, with a forced flow velocity of 100 mm/s. The experimental conditions reflect environments potentially encountered in future spacecraft cabins. The data collected show that during downward opposed flame spread the surface regression rate, and consequently the mass burning rate, decreases with decreasing ambient pressure and increases with decreasing oxygen concentration. The data presented is correlated with a simplified model to estimate the surface regression rate based on the mixed heat convection coefficient and the B number. Model predictions are made based on fuel properties as well as the relevant environmental conditions. The predictions of the model agree well capturing the effect of ambient pressure and oxygen concentration. The correlations provide information about the variations in the burning rate in environments with variable pressure and oxygen concentration, providing guidance for potential for fire safety testing and design in different ambient conditions.

Low-pressure blowoff experiments were conducted with a stagnation flame stabilized on the forward tip of cast PMMA rods in a vertical wind tunnel. Pressure, forced flow velocity, gravity, and ambient oxygen concentration were varied. Both normal gravity and microgravity tests were conducted to determine the influence of buoyant flow velocity on the blowoff limits. A linear shift of a buoyant flow of ~ 25 cm/s was adequate to match the microgravity data. The flame standoff distance before blowoff varies with the inverse square root of pressure and stretch rate via the gas phase length scale δ = (α/a)1/2, where the thermal diffusivity varies inversely with pressure. For a constant critical Damkohler number, the reaction rate must increase in tandem with the flow rate. The flame standoff distance responds to environment changes to maintain both the critical Damkohler number and the molar flux of oxygen along the blowoff boundary. Via Fick's law, there is a ubiquitous linear relationship between the partial pressure of oxygen at blowoff and the total pressure regardless of forced flow velocity and gravity level. At very low pressures, reactant leakage appears to cause a two-stage reaction zone. For very dilute oxidizer mixtures, an apparent high pressure blowoff limit is found, believed to be caused by chain-terminating three body reactions. The overall reaction order becomes negative along the high pressure limit. A normal gravity materials flammability screening methodology is suggested for spacecraft exploration atmospheres where the blowoff partial pressure of oxygen versus total pressure blowoff limits are measured for the expected exploration atmospheres (oxygen concentrations). As long as the blowoff boundary is above the Earth-normal partial pressure of oxygen (21 kPa) threshold, the material is safe for use in exploration spacecraft.

Normal gravity flame blowoff limits in an axisymmetric PMMA rod geometry in upward axial stagnation flow are compared with microgravity Burning and Suppression of Solids-II (BASS-II) results recently obtained aboard the International Space Station. This testing utilized the same BASS-II concurrent rod geometry, but was done in normal gravity mixed convective flow. Cast polymethylmethacrylate (PMMA) rods of diameters ranging from 0.635 cm to 3.81 cm were burned at oxygen concentrations ranging from 15% to 18% by volume. The forced flow velocity where blowoff occurs was determined for each rod size and oxygen concentration. These blowoff limits compare favorably with the BASS-II results when the buoyant stretch rate is included and the flow is corrected by considering the flow blockage factor of the fuel. From these results, the normal gravity blowoff boundary for this axisymmetric rod geometry is determined to be linear, with oxygen concentration and flow velocity as coordinates. We describe a new normal gravity upward blowoff test method which extrapolates this linear blowoff boundary to the zero mixed stretch rate limit to resolve microgravity flammability limits—something current methods cannot do. This new test method can improve spacecraft fire safety for future exploration missions by providing a tractable way to obtain accurate but conservative estimates of material flammability in low gravity based on tests performed on Earth.

Concurrent-flow flame spread over a thin charring material was studied in an unmanned spacecraft. Two sample sizes were used: 5 cm by 29 cm and 41 cm by 94 cm, the largest ever samples burned in controlled experiments in microgravity. A low-speed ambient airflow of 20 cm/s was used. The samples were ignited from their upstream ends and were allowed to burn for several minutes. Video recorded during the burning process and readings from thermocouples on the sample surfaces were examined. The results showed that flames reached a quasi-steady state with nearly constant flame length for both sample sizes. However, the pyrolysis length exhibited an initial overshoot before reaching steady state for the large sample. This phenomenon has not been reported or observed until now. While steady state pyrolysis lengths were similar, the 5 cm wide sample had a slightly larger spread rate (by ∼13%) and a shorter burnout time (pyrolysis length over spread rate) compared to the 41 cm wide sample. A previously developed three-dimensional transient model was used to conduct the numerical study. Detailed profiles of the gas and solid phases, including flow patterns, species concentrations, temperature, solid burning rate, and heat flux distributions are examined. The modeling results reveal that flow reduction in a growing boundary layer along the sample surface accounts for the overshoot of the flame length observed for the large sample. Compared to a wide sample, a narrow sample has more effective oxygen transport across the width of the sample. This results in a stronger flame, a shorter flame standoff distance from the sample surface, and larger heat feedback to the sample. This accounts for the shorter burnout time for the narrow sample compared to the large sample.

The spread of flames over the surface of solid combustible material in an opposed flow is different from the mass burning (or fuel regression) in a pool fire. However, the progress of a flame front over a solid fuel includes both flame spread and fuel regression, but the difference between these two processes has not been well clarified. In this work, experiments using cylindrical PMMA samples were conducted in normal gravity and in microgravity. We aim to identify the transition from opposed flame spread to fuel regression under varying conditions, including sample size, opposed flow velocity, pressure, oxygen concentration, external radiation, and gravity level. For a thick rod in normal gravity, as the opposed flow increases to 50–100 cm/s, the flame can no longer spread over the fuel surface but stay in the recirculation zone downstream of the cylinder end surface, like a pool fire flame. The flame spread first transitions to fuel regression at a critical leading-edge regression angle of α ≈ 45°, and then, flame blow-off occurs. Under large opposed flow velocity, a stable flat blue flame is formed floating above the rod end surface, because of vortex shedding. In microgravity at a low opposed flow (

During blowoff extinction of clear cast PMMA rods in concurrent axial flow for microgravity BASS-II experiments, a dynamic flame oscillation was observed after the flame was blown off of the stagnation point but briefly stabilized on the periphery of the rod. Complementary normal gravity experiments were conducted and flame oscillations were tracked using a high-speed color camera at 240 frames per second. The side-stabilized flame oscillated up and down the rod with increasing amplitude until the entire flame extinguished. In none of the BASS-II or normal gravity tests could the side-stabilized flames persist (Hopf subcritical bifurcation). Since the oscillations occurred even in microgravity, the mechanism does not depend on gravity. For the larger fuel radius tests, the flame developed asymmetric oscillation (pitchfork bifurcation). The oscillation time and the number of oscillations scale with the inverse square of the rod radius (∼ Fourier no.) for the preheated microgravity rods. The average flame oscillation frequency is found to be linearly dependent on the mixed convective stretch rate (inverse of the flow time). The flame intensity varied in concert with its direction, either increasing or decreasing as the flame moved upstream or downstream, respectively. The oscillation frequency decreased as the amplitude increased and the flame slipped slightly farther down the rod with each oscillation. The flame speed increased with each subsequent oscillation, both flashing forward upstream and retreating downstream. The oscillations were found to closely follow a power law log-periodic dependence similar to those that describe systems approaching a critical point, such as diffusion-limited aggregation clusters, earthquakes, ruptures, and even stock market crashes. The net flame speeds varied linearly with ambient oxygen concentration, and linearly with the mixed convective stretch rate. Based on these observations, a mechanistic theory of the oscillations is described, and is consistent with the thermodiffusive instability.

Microgravity experiments are performed to study the effects of confinement on the burning behavior of polymeric solid materials. Flat, 100 × 22 × 1 mm PMMA samples are burned in concurrent air flow in a small flow duct aboard the International Space Station. Three different burning scenarios are examined, double-sided, single-sided, and parallel samples. In the first two scenarios, single samples are burned on both sides and on one side, respectively. Flat baffles are placed parallel to the sample to confine the available space for combustion. The distance between the baffle and the sample (H) is varied in different tests. In each test, imposed flow is reduced in steps and steady flame spread is achieved at each flow speed until the flame quenches. The results show that at the same confined condition, steady state flame length and spread rate are proportional to flow speed over the range tested. When confinement increases (or H decreases), the flame spread rate and flame length increase first and then decrease. In addition, the quenching flow speed decreases and then increases with decreasing H. These results suggest that the confinement can increase or decrease solid fuel flammability depending on conditions. In the third burning scenario, two PMMA samples are placed parallel to each other separated by a distance H. Twin flames are observed and combustion is confined between the two samples. Among the three tested burning scenarios, twin flames have the largest flame length and burning rate at the same confinement level (H). This is because the thermal interaction between the twin flames enhances the heat feedback to the solid fuel and reduces the relative heat loss to the surrounding flow duct. Comparing single- and double-sided flames with the same baffle-sample distance, the spread rate of a single-sided flame is slightly less than half of that of a double-sided flame. This is due to the halved pyrolysis area exposed to the flame and heat loss on the back side of the sample. Optimal transport of oxygen to the flames also plays a role.

For the first time, a series of concurrent-flow rod flammability tests were conducted in microgravity aboard the International Space Station. A small flow duct was used to create 0 to 55 cm/s flows past three sizes of clear and black PMMA rods. The ambient oxygen concentration in the Microgravity Science Glovebox was varied from 13.6% to 22.2%. Oxygen, carbon dioxide, and carbon monoxide gas sensors provide initial and final readings for each test and indicate that the flames are globally stoichiometric at higher oxygen concentrations, but become more globally fuel rich as the minimum oxygen concentration is approached due to excess pyrolyzate leakage out of the open tail of the hemispherical flames. Quenching extinction occurs at very low forced flows, where the flame shrinks to a hemispherical blue flame and oscillates with increasing amplitude just before going out. Blowoff extinction is initiated by the formation of a hole in the flame sheet in the stagnation region of the flame. A critical Damkohler number formulation is applied across the flammability boundary, and the critical flame temperatures are derived. These critical flame temperatures are then used in a Nusselt number correlation to estimate the convective heat flux to the stagnation region of the rod. A model of surface energy balance is formulated that uses the critical flame temperature and convective heat flux to derive the mass burning rate along the boundary. The rod regression rates calculated from this model compare favorably with the experimental measurements. The surface energy balance reveals that along the blowoff branch, heat losses are negligible whereas in the quenching region, surface radiative loss dominates. At the bottom of the flammability map, the transition from blowoff to quenching occurs when the convective flows become the same order of magnitude as diffusive flows, shifting the critical Damkohler number from residence time limitations to diffusive time limitations.

The three regimes of opposed-flow flame spread – radiative, thermal, and kinetic regimes – are well known. For thermally thin fuels, the spread rate is independent of opposing flow velocity in the thermal regime. It decreases with an increase in the flow velocity in the kinetic regime, leading to blow off extinction. In the radiative regime which occurs mostly in a buoyancy-free environment of microgravity, the spread rate decreases with a decrease in flow velocity leading to radiative extinction unless the oxygen level is very high. In a recent experiment aboard the International Space Station, thin sheets of PMMA were ignited in a flow tunnel with the opposing flow varying over a wide range. All three regimes of flame spread were captured in a single set of experiments for the first time. Instantaneous spread rates were obtained from digital video processing and compared with a computational model in all three regimes along with the evolution of flame shapes. Spread rates in the radiative and thermal regimes are also compared with existing theories of flame spread in the thermal and the radiative regime producing remarkable qualitative agreement.

Flame growth, spread, and quenching extinction over a thin composite cotton-fiberglass fabric blend (referred to as the SIBAL fabric) were studied in low-speed concurrent purely forced flows aboard the International Space Station. The tests were conducted in a small flow duct within the Microgravity Science Glovebox. The fuel samples measured 1.2 and 2.2 cm wide and 10 cm long. Ambient oxygen was varied from 21% down to 16% molar concentration and flow speed from 55 cm/s down to 1 cm/s. A slow purely forced flow resulted in a small flame, enabling us to observe the entire history of flame development including ignition, flame growth, steady spread (in some cases), and decay at the end of the sample. In addition, by decreasing flow velocity during some of the tests, low-speed flame quenching extinction limits were determined as a function of oxygen percentage. The quenching speeds were found to be between 1 and 5 cm/s with higher extinction speeds in lower oxygen atmospheres. The shape of the quenching boundary supports the prediction by earlier theoretical models. These long duration microgravity experiments provide a rare opportunity for solid fuel combustion since microgravity time in ground-based facilities is generally not sufficient. This is the first time that a low-speed quenching boundary in concurrent spread is mapped in a clean and unambiguous manner. A previously developed three-dimensional transient model is modified to compare with the experiment. The modification includes the use of two-step SIBAL fabric pyrolysis kinetics where the rate constants are determined using Thermo-Gravimetric Analysis data. The model yields good quantitative comparison on the quenching boundary, the flame transient development, and the steady flame spread rates.

The thermal regime of flame spread over solid fuels constitutes the reference condition for all other flame spread research. Although the theory of flame spread in the thermal regime is well understood, the well-known closed-form formulas for flame spread do not compare well with available experimental data. The comparison is further complicated by the fact that establishing a thermal regime in a normal-gravity environment is difficult because of the buoyancy induced flow which may usher in finite-rate kinetics effect. As a result, even the transition thickness, when a fuel can be considered a thermally thick fuel, still lacks a widely accepted formula. In this work we present opposed-flow flame spread data over varying thicknesses of poly-methyl methacrylate (PMMA) obtained in the International Space Station where the opposing flow velocity can be reduced arbitrarily without any interference from the gravity induced flow. We also present a larger set of spread rate data for the downward spreading configuration at normal gravity. A comparison between the two data set allows us to establish the thermal limit for thin fuel for which the spread rate is independent of the opposing flow velocity. The classical thin-fuel spread rate formula is shown to fit well with the experimental results provided the adiabatic flame temperature is used in the flame coefficient that appears in the formula. The experimentally determined flame coefficient along with downward flame spread data for thick fuels are used to develop a closed-form expression for the transition thickness between thermally thin and thick fuels for downward spread in the thermal regime.

The effect of opposing flow on flame spread rate over thin solid fuel is investigated with the help of scaling theory, a comprehensive computational model, and experiments conducted aboard the International Space Station. While spread rate over thin fuels is independent of the opposing flow velocity in the thermal regime, in the microgravity regime, where the opposing flow can be very mild or even completely absent in the absence of buoyancy induced flow, the spread rate is known to decrease as the opposed flow is reduced. Under certain conditions, this can lead to flame extinguishment at a low enough flow velocity. This paper combines scaling arguments with computational results to predict a critical flow velocity for such flame extinction. Results from the recently conducted limited number of space based tests, presented in this paper, seem to confirm the prediction validating the closed-form formula for the critical extinction velocity.

The influence of environmental conditions on the flammability of combustible solids is of importance to spacecraft fire safety because of the differences with the conditions encountered on Earth. In a manned spacecraft there is reduced gravity and low velocity flows. Additionally, the environment is maintained at a Normoxic condition, which is the combination of ambient pressure and oxygen concentration that results in a partial pressure of oxygen equal to that of normal atmosphere at sea level ( P O 2 = 21 k P a ) . Future spacecraft will have atmospheres with reduced pressures and oxygen concentrations at Normoxic conditions (Space Exploration Atmospheres - SEA), designed to reduce preparation time for spacewalks. This work studies the effect of ambient pressure and oxygen concentration, on opposed flame spread and mass burning in cylindrical samples of polymethyl methacrylate (PMMA). Experiments in normal gravity are conducted using ambient pressures ranging between 100 and 60 kPa and oxygen concentrations between 21% and 35% by volume, while maintaining Normoxic conditions. Results show that moving to Normoxic environments with reduced pressure and increased oxygen concentration increases the flammability of the PMMA cylinders. The data presented here provide information about the flammability of spacecraft materials in future SEA, yielding insight for future designs when considering fire safety in spacecrafts.

Upward flame spread tests were conducted on thin fuels in a sealed chamber capable of accommodating large-scale samples (1 m length). The primary objective of these tests was to measure flame spread and pressure rise in a large sealed chamber during and after flame spread and to characterize that data as a function of sample material, initial pressure, and sample size. The flame spread rate as a function of initial pressure has been measured for a given fuel and found to vary as ∼ P 2 in agreement with Grashof number scaling. The burning rate per unit area for a fixed pressure has been shown to be a constant independent of fuel area density or quantity of fuel burned. A steady upward flame spread was observed only at low pressure. The pressure rise in a sealed chamber has been shown to scale with the quantity of fuel burned, and the peak pressure has been shown to scale inversely with initial pressure, in agreement with the pressure dependence of the characteristic time associated with a simple analytical solution of an energy balance. The pressure rise per mass of fuel burned exhibits an exponential decay with burn-time, also in agreement with the analytical solution.

Upward flame spread experiments were conducted on long thin composite fabric fuels made of 75% cotton and 25% fiberglass of various widths between 2 and 8.8 cm and lengths greater than 1.5 m. Symmetric ignition at the bottom edge of the fuel resulted in two sided upward flame growth initially. As flame grew to a critical length (15–30 cm depending on sample width) fluctuation or instability of the flame base was observed. For samples 5 cm or less in width, this instability lead to flame blow off on one side of the sample (can be either side in repeated tests). The remaining flame on the other side would quickly shrink in length and spread all the way to the end of the sample with a constant limiting length and steady spread rate. Flame blow off from the increased buoyancy induced air velocity (at the flame base) with increasing flame length is proposed as the mechanism for this interesting phenomenon. Experimental details and the proposed explanation, including sample width effect, are offered in the paper.

Experimental observations of anomalous combustion of n-heptane droplets burning in microgravity are reported. Following ignition, a relatively large n-heptane droplet first undergoes radiative extinction, that is, the visible flame ceases to exist because of radiant energy loss. But the droplet continues to experience vigorous vaporization for an extended period according to a quasi-steady droplet-burning law, ending in a secondary extinction at a finite droplet diameter, after which a vapor cloud rapidly appears surrounding the droplet. We hypothesize that the second-stage vaporization is sustained by low-temperature, soot-free, “cool-flame” chemical heat release. Measured droplet burning rates and extinction diameters are used to extract an effective heat release, overall activation energy, and pre-exponential factor for this low-temperature chemistry, and the values of the resulting parameters are found to be closer to those of “cool-flame” overall reaction-rate parameters, found in the literature, than to corresponding hot-flame parameters.

Pressure–gravity modeling is proposed as a means to simulate upward flame spread and burning rates over vertical solid samples in partial gravity environments, such as on the Moon and on Mars. Based on experimental results in reduced gravity, the upward flame spread rate data over thin solids can be correlated by the expression p 1.8 g (where p is the ambient pressure and g is the gravity level). This is close to the theoretical p 2 g factor in preserving the Grashof number and is also supported by detailed numerical simulations. Since the flame size, shape and standoff distance are preserved in this simulation, it is expected that combustion properties controlled chiefly by convective heat transfer are properly accounted for by the present technique. This includes upward flame spread rates, growth rates, and burning rates over thin and thick solids in both laminar and turbulent flames. In flames where the heat transfer is dominated by soot emission, more studies are needed to verify the validity of this correlation.

Diffusion flame spread over a thin solid fuel in quiescent and slowly moving atmospheres is studied in microgravity. The flame behavior is observed to depend strongly on the magnitude of the relative velocity between the flame and the atmosphere. In particular, a low velocity quenching limit is found to exist in low oxgen environments. Using both the microgravity results and previously published data at high opposed-flow velocities, the flame spread behavior is examined over a wide velocity range. A flammability map using molar oxygen percentages and characteristic relative velocities as coordinates is constructed. Trends of flame spread rate are determined and mechanisms for flame extinction are discussed.