Structure of igniting ethanol and n-heptane spray flames with and without swirl

52nd Aerospace Sciences Meeting

et al. 2014

Self Cite

Laser Induced Breakdown Spectroscopy (LIBS) in uniform flowing methane-air mixtures of compositions ranging from only air to only fuel was investigated in order to explore calibration schemes for further measurements in turbulent non-premixed and premixed flames. A portable low-cost spectrometer was used to record plasma emission spectra. The differences among the various spectra are discussed and the intensity ratios H α (656.3nm)/O(777.3nm) and C 2 (516.5nm)/CN(388.3nm) were found to depend monotonically to the mole fraction of methane in the ranges 0-0.3 and of 0.3-1.0 respectively, therefore providing a scheme for measurement in non-premixed systems spanning a wide range of equivalence ratios. Based on this and using laser excitation conditions such that the escaping plasma emission contains spectral signatures from reactants, flame intermediates, and combustion products, the measurement can be thought to give an estimate of the local and instantaneous mixture fraction. The mean estimated mixture fraction in turbulent axisymmetric non-reacting and reacting jets was in reasonable agreement with empirical correlations. The technique was also applied to a swirling recirculating premixed flame and it is shown that the equivalence ratio is successfully measured.

Section: A Flow Conditionsmentioning

confidence: 99%

Laser-induced breakdown spectroscopy measurements in turbulent methane flames

Kotzagianni

Yuan

52nd Aerospace Sciences Meeting

et al. 2014

Self Cite

“…Nevertheless, their presence in the kernel or its vicinity has been shown to be detrimental to ignition in cases of large or slowly-evaporating droplets at the flame front [8], and beneficial to ignition in experiments with small droplets [9,6]. Experiments focusing on ignition probability have only looked at the burner-scale aspects of the problem [10][11][12], and although some experiments have evaluated MIE in terms of global parameters of the flow [13][14][15], ignition probabilities have not been investigated in canonical configurations in order to understand the effects of global flow parameters and their fluctuations controlling the early-phase ignition of sprays.…”

Section: Introductionmentioning

confidence: 99%

Ignition of uniform droplet-laden weakly turbulent flows following a laser spark

Oliveira

Allison

2019

Combustion and Flame

Self Cite

The forced ignition process has a stochastic nature, which can be intensified due to turbulence and mixture fluctuations. Although fuel droplets represent strong inhomogeneities which are generally detrimental to ignition, the presence of small droplets has been found to enhance flame speeds, decrease minimum ignition energy, and improve the ignitability of overall lean mixtures. In order to understand which factors are conducive to ignition of sprays, a spherically expanding flame is investigated, which is produced by a laser spark in a uniform dispersion of ethanol droplets in turbulent air. The flame is visualised by schlieren and OH*-chemiluminescence for overall equivalence ratios of 0.8 to 2, Sauter mean diameter of approximately 25 µm, and u /S L ranging from 0.9 to 1.3, where u and S L denote the rms axial velocity and laminar burning velocity, respectively. The timescales of the spark's effects on the flame are measured, as well as quenching timescales and initial kernel sizes conditional on ignition or failure. Small kernels quenched faster than approximately 0.6 ms, that is, the duration of the flame overdrive, and a minimum kernel radius for ignition of 1 mm was observed. The short-mode of ignition failure was suppressed by increasing the laser energy and, consequently, the initial kernel size. Nevertheless, the ignitability of lean mixtures was only effectively improved through high-energy sparks and partial prevaporisation of the fuel. Virtually all kernels ignited once prevaporisation was increased, and the gas-phase equivalence ratio was approximately 75% of the lower flammability limit, with ignition being limited only by laser breakdown.

“…The temperature of the droplets Fig. 2 The computational domain with the boundary conditions (left f igure) and sketch showing detailed dimensions of the bluff-body and combustion chamber [13] was 300 K, their velocity were estimated based on the mass flow measurement and was 9.9 m/s, and the droplets were injected in a deterministic manner without any velocity fluctuations. We assumed ten classes for the droplet diameters, which were computed from a modified Rossin-Ramler distribution with the Sauter mean diameter equal to 30 μm and with the exponential factor q = 3 in the Rossin-Ramler formula [39].…”

Section: Flow Configuration and Initial And Boundary Conditionsmentioning

confidence: 99%

“…Recently, an experimental analysis of spark ignition and blow-off processes of an n-heptane spray flame in a bluff-body swirl burner was done [12][13][14] and the corresponding LES-CMC of spark ignition [15] showed that the LES-CMC can capture the spatial distribution of the ignition probability. Both experiment and LES have shown a very rich and variable behaviour of the flame following spark ignition and, to a large extent, ignition probability becoming lower than unity has been attributed to localised quenching, a physical phenomenon that also lies at the heart of the global blow-off of flames.…”

mentioning

confidence: 99%

LES/CMC of Blow-off in a Liquid Fueled Swirl Burner

Tyliszczak

Cavaliere

Flow Turbulence Combust

2013

Self Cite

Large Eddy Simulations of two-phase flames with the Conditional Moment Closure combustion model have been performed for flow conditions corresponding to stable and blow-off regimes in a swirl n-heptane spray burner. In the case of stable flame (i.e. low air velocity), the predicted mean and r.m.s. velocities and the location and shape of the flame agree reasonably well with experiment. In particular, the presence of localised extinctions is captured in agreement with experiment. Using model constants previously calibrated against piloted jet methane flames (Sandia F) with localised extinction, we obtain that at the experimentally determined blow-off velocity of the swirling spray flame, the predicted flame also blows off, demonstrating that the LES-CMC approach can capture the global extinction point in a realistic configuration.