skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Large-eddy simulation of a bluff-body stabilized nonpremixed flame

Abstract

Large-eddy simulations have been performed for a turbulent nonpremixed bluff-body stabilized CH{sub 4}:H{sub 2} (50:50 vol.) flame at a Reynolds number of 15,800. The corresponding isothermal flow has also been computed. The Sydney bluff-body burner under consideration has been investigated experimentally by Masri and co-workers, who obtained velocity and scalar statistics. The focus of the current work is on flow and mixing effects with the thermochemistry evaluated using a steady-state laminar flamelet approach. The incompressible (low-Mach-number) governing equations for mass, momentum, and mixture-fraction have been solved on a structured cylindrical grid and resolution effects investigated using up to 3.643x10{sup 6} nodes. The corresponding nonreactive case was resolved by 5.76x10{sup 5} nodes, resulting in a resolution of more than 80% of the turbulence kinetic energy. The reacting case yields a resolution in excess of 75% on the finest grid--arguably sufficient to permit conclusions regarding the accuracy of submodels. Comparisons with experimental data show that for high resolutions comparatively good agreement is obtained for the flow field and for species other than nitric oxide. However, resolution effects are important and results obtained with 4.51x10{sup 5} nodes show that a resolution of less than 70% of the turbulent kinetic energy is insufficient inmore » the context of the Smagorinsky subgrid model combined with the dynamic procedure of Germano. The latter result is consistent with the analysis of Pope.« less

Authors:
;  [1];  [2]
  1. Thermofluids Division, Department of Mechanical Engineering, Imperial College London, London SW7 2AZ (United Kingdom)
  2. Chair for Energy and Powerplant Technology, Department of Mechanical Engineering, Darmstadt Technical University, Petersenstr. 30, 64287 Darmstadt (Germany)
Publication Date:
OSTI Identifier:
20685985
Resource Type:
Journal Article
Resource Relation:
Journal Name: Combustion and Flame; Journal Volume: 144; Journal Issue: 1-2; Other Information: Elsevier Ltd. All rights reserved
Country of Publication:
United States
Language:
English
Subject:
42 ENGINEERING; FLAMES; BURNERS; COMPUTERIZED SIMULATION; TURBULENCE; METHANE; HYDROGEN; GAS FLOW; COMBUSTION KINETICS; CALCULATION METHODS; ACCURACY

Citation Formats

Kempf, A., Lindstedt, R.P., and Janicka, J. Large-eddy simulation of a bluff-body stabilized nonpremixed flame. United States: N. p., 2006. Web. doi:10.1016/j.combustflame.2005.07.006.
Kempf, A., Lindstedt, R.P., & Janicka, J. Large-eddy simulation of a bluff-body stabilized nonpremixed flame. United States. doi:10.1016/j.combustflame.2005.07.006.
Kempf, A., Lindstedt, R.P., and Janicka, J. Sun . "Large-eddy simulation of a bluff-body stabilized nonpremixed flame". United States. doi:10.1016/j.combustflame.2005.07.006.
@article{osti_20685985,
title = {Large-eddy simulation of a bluff-body stabilized nonpremixed flame},
author = {Kempf, A. and Lindstedt, R.P. and Janicka, J.},
abstractNote = {Large-eddy simulations have been performed for a turbulent nonpremixed bluff-body stabilized CH{sub 4}:H{sub 2} (50:50 vol.) flame at a Reynolds number of 15,800. The corresponding isothermal flow has also been computed. The Sydney bluff-body burner under consideration has been investigated experimentally by Masri and co-workers, who obtained velocity and scalar statistics. The focus of the current work is on flow and mixing effects with the thermochemistry evaluated using a steady-state laminar flamelet approach. The incompressible (low-Mach-number) governing equations for mass, momentum, and mixture-fraction have been solved on a structured cylindrical grid and resolution effects investigated using up to 3.643x10{sup 6} nodes. The corresponding nonreactive case was resolved by 5.76x10{sup 5} nodes, resulting in a resolution of more than 80% of the turbulence kinetic energy. The reacting case yields a resolution in excess of 75% on the finest grid--arguably sufficient to permit conclusions regarding the accuracy of submodels. Comparisons with experimental data show that for high resolutions comparatively good agreement is obtained for the flow field and for species other than nitric oxide. However, resolution effects are important and results obtained with 4.51x10{sup 5} nodes show that a resolution of less than 70% of the turbulent kinetic energy is insufficient in the context of the Smagorinsky subgrid model combined with the dynamic procedure of Germano. The latter result is consistent with the analysis of Pope.},
doi = {10.1016/j.combustflame.2005.07.006},
journal = {Combustion and Flame},
number = 1-2,
volume = 144,
place = {United States},
year = {Sun Jan 01 00:00:00 EST 2006},
month = {Sun Jan 01 00:00:00 EST 2006}
}