Comprehensive Modeling of Turbulent Flames With the Coherent Flame-Sheet Model—Part I: Buoyant Diffusion Flames

1996 ◽  
Vol 118 (1) ◽  
pp. 65-71 ◽  
Author(s):  
C. A. Blunsdon ◽  
Z. Beeri ◽  
W. M. G. Malalasekera ◽  
J. C. Dent

A modified version of the computational fluid dynamics code KIVA-II was used to model the transient behavior of buoyant turbulent diffusion flames burning in still air. Besides extensions to the range of permitted boundary conditions and the addition of buoyancy terms to the turbulence model, KIVA-II was augmented by a version of the coherent flame-sheet model, Tesner’s soot generation model, Magnussen’s soot oxidation model, and an implementation of the discrete transfer radiation model that included both banded and continuum radiation. The model captured many of the features of buoyant turbulent flames. Its predictions supported experimental observations regarding the presence and frequency of large-scale pulsations, and regarding axial distributions of temperature, velocity, and chemical species concentrations. The radial structure of the flame was less well represented. The axial radiative heat flux distribution from the flame highlighted deficiencies in the soot generation model, suggesting that a model of soot particle growth was required.

1996 ◽  
Vol 118 (1) ◽  
pp. 72-76 ◽  
Author(s):  
Z. Beeri ◽  
C. A. Blunsdon ◽  
W. M. G. Malalasekera ◽  
J. C. Dent

This paper describes the application of computational fluid dynamics (CFD) to the prediction of the characteristics of high-momentum vertical and horizontal flames in ambient air flows. The KIVA-II code has been modified by extending the range of boundary conditions and by the addition of the following: a version of the coherent flame-sheet model, Tesner’s soot generation and Magnussen’s soot oxidation model, and an implementation of the discrete transfer radiation model. To assess the accuracy of the complete model for prediction purposes, results are compared with experimental data. Predictions of temperature and flame profiles are in good agreement with data while predictions of radiative heat transfer are not entirely satisfactory.


Following a demonstration that hot gas pockets coalesce in plasma jet vortex cores, various burner systems are designed to induce solid body rotation such as either to promote or to impede the transport into reactants of any islands of hot gases. Promotion results in large increases in the burning velocity and in the stability of premixed turbulent hydrocarbon-air flames, and vice versa. Planar imaging by laser-induced fluorescence of OH at high magnifications reveals numerous small islands of hydroxyl in small turbulent flames, especially near the tips and close to blow-out. Comparison with schlieren photographs and a review of other work suggests that these are sectioned inner cores of vortex filaments or of cusps on the flame front. In rotating conical flames these tend to drift towards the axis. OH concentrations within islands suggest that only a few – generally of the larger ones – are expanding centres of reaction; many of the small ones appear to be diffusing remnants of flame. A rough estimate of the centrifugally induced increase in diffusivity is deduced from the shortening, with rate of rotation, of turbulent diffusion flames. Comparison with the changes in burning velocity of premixed flames of similar geometry and rotation rate suggests that promoting the drift of hot gas and radicals into the reactants, in addition to increasing diffusivity, may also produce a slight augmentation of the reaction rate.


1996 ◽  
Vol 118 (2) ◽  
pp. 128-133 ◽  
Author(s):  
X. Li

Length of jet diffusion flames is of direct importance in many industrial processes and is analyzed by applying scaling method directly to the governing partial differential equations. It is shown that for jet-momentum-dominated diffusion flames, when the buoyancy effects are neglected, the flame length normalized by the burner exit diameter increases linearly with the Reynolds number at the burner exit in the laminar burning regime and decreases in inverse proportion to the Reynolds number in the transitional regime. For turbulent diffusion flames, the normalized flame lengths are independent of the burner exit flow conditions. It is further found that for vertical upward flames, the buoyancy effect increases the flame length in the laminar and transitional regime and reduces the length in the turbulent regime; while for vertical downward flames, the buoyancy effect decreases the flame length in the laminar and transitional regime and increases the length in the turbulent regime, provided that jet momentum is dominated, and there is no flame spreading out and then burning upward like a downward-facing pool fire. Hence, for turbulent flames the flame lengths depend on the Froude number, Fr, and increase (or decrease) slightly as Fr increases for upward (or downward) flames. By comparison, it is found that the foregoing theoretical results are in good agreement with the experimental observations reported in literature.


Author(s):  
Takeshi Ochi ◽  
Norio Arai ◽  
Tomohiko Furuhata ◽  
Naoki Kishi

In this study, the difference of temperature and the component of chemical species on soot formation in CH4/air fuel-rich diffusion flames were investigated. Furthermore, for decreasing soot formation in fuel-rich diffusion flames, we added H2 in CH4, investigated the property of combustion, and compared with methane/air fuel-rich flames. We have paid much attention to the influence of the equivalence ratio of methane (+H2)/air, the swirl strength of combustion air and the concentration of C2H2 on the soot formation. The experimental combustor for CH4(+H2)/air combustion was designed, and the soot resulting from the exhaust gas collected with a silica filter and its weight was measured. The microstructure of the soot particles were analyzed with a Scanning Electron Microscopy (SEM). The temperature profiles in the combustor were measured by thermocouples, and the concentrations of the species O2, CO2, H2, CH4 and C2H2 were determined by a TCD (thermal conductivity detector) gas chromatograph (GC) and FID (flame ionization detector) GC. The soot yields diminished with increasing swirl strength and the C2H2 concentration. When H2 was added to the fuel, combustion was promoted and C2H2 concentration in the exhaust gas was diminished. But using strong swirl, fuel and air were mixed quickly, the effect of H2 addition was decreased.


Sign in / Sign up

Export Citation Format

Share Document