Study of Swirling Air Flow Characteristics in a Lean Direct Injection Combustor

Dewanji, D.; Rao, Arvind Gangoli; Pourquié, Mathieu; Buijtenen, Jos P. van

doi:10.1115/gt2010-22602

Cited by 1 publication

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Past numerical studies for helical axial swirler type systems were limited to the study of one or two cases of non-reacting/ reacting flows and very limited parametric studies of single swirler (Im et al, 2001;Ren et al, 2016;Heath, 2016;Fu et al, 2007) on aerodynamics and emissions were performed. To capture the rapid fuel-air mixing and recirculation zones, unsteady numerical methods like LES are suitable for resolving all the turbulent eddies (Dewanji et al, 2011(Dewanji et al, , 2010, and a detailed flow investigation of confinement effect on a single helical swirler using LES is studied. The preliminary part of this manuscript presents a non-reacting LES simulation of a single element swirler describing the meshing requirements for LES to capture the required energy spectrum for more reasonable accurate results in the core region of interest.…”

Section: Introductionmentioning

confidence: 99%

“…Two critical geometrical parameters that affect mostly aerodynamics are swirl (tip vane angle) and confinement (combustion cross-section). Swirling flow enables flame stabilization and improved mixing in the combustion zone (Cai et al , 2005; Patel et al , 2007; Dewanji et al , 2010; Patel and Menon, 2008). The term confinement ratio is defined as the cross-sectional area of the combustion chamber to the swirler exit cross-sectional area.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

LES predictions of confinement effect on swirling flow generated by single helical axial swirler

Rajesh

Sivakumar

2022

AEAT

View full text Add to dashboard Cite

Purpose For higher swirling flows (swirl > 0.5), flow confinement significantly impacts fluid flow, flame stability, flame length and heat transfer, especially when the confinement ratio is less than 9. Past numerical studies on helical axial swirler type systems are limited to non-reacting or reacting flows type Reynolds averaged Navier Stokes closure models, mostly are non-parametric studies. Effects of parametric studies like swirl angle and confinement on the unsteady flow field, either numerical or experimental, are very minimal. The purpose of this paper is to document modeling practices for a large eddy simulation (LES) type grid, predict the confinement effects of a single swirler lean direct injection (LDI) system and validate with literature data. Design/methodology/approach The first part of the paper discusses the approach followed for numerical modelling of LES with the minimum number of cells required across critical sections to capture the spectrum of turbulent energy with good accuracy. The numerical model includes all flow developing sections of the LDI swirler, right from the axial setting chamber to the exit of the flame tube, and its length is effectively modelled to match the experimental data. The computational model predicts unsteady features like vortex breakdown bubble, represented by a strong recirculation zone anchored downstream of the fuel nozzle. It is interesting to note that the LES is effective in predicting the secondary recirculation zones in the divergent section as well as at the corners of the tube wall. Findings The predictions of a single helical axial swirler with a vane tip angle of 60°, with a duct size of 2 × 2 square inches, are compared with the experimental data at several axial locations as well as with centerline data. Both mean and unsteady turbulent quantities obtained through the numerical simulations are validated with the experimental data (Cai et al., 2005). The methodology is extended to the confinements effect on mean flow characteristics. The time scale and length scale are useful parameters to get the desired results. The results show that with an increase in the confinement ratio, the recirculation length increases proportionally. A sample of three cases has been documented in this paper. Originality/value The novelty of the paper is the modelling practices (grid/unsteady models) for a parametric study of LDI are established, and the mean confinement effects are validated with experimental data. The spectrum of turbulent energies is well captured by LES, and trends are aligned with experimental data. The methodology can be extended to reacting flows also to study the effect of swirl angle, fuel injection on aerodynamics, droplet characteristics and emissions.

show abstract