Velocity fields of fuel lean premixed turbulent opposed jet flames

Velocity measurements in the turbulent reacting flow created by an opposed nozzle configuration are reported with emphasis on the axis and stagnation plane. The instrumentation comprised particle image velocimetry (PIV) with Al 2O 3 particles and silicon oil droplets added upstream of both nozzles a...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings of the Combustion Institute 2007, Vol.31 (1), p.1459-1466
Main Authors: Lindstedt, R.P., Luff, D.S., Whitelaw, J.H.
Format: Article
Language:English
Subjects:
Combustion
Computational fluid dynamics
Droplets
Flames
Fluid flow
Nozzles
Opposed jets
PIV
Planes
Stagnation
Turbulence
Turbulent
Turbulent flow
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Velocity measurements in the turbulent reacting flow created by an opposed nozzle configuration are reported with emphasis on the axis and stagnation plane. The instrumentation comprised particle image velocimetry (PIV) with Al 2O 3 particles and silicon oil droplets added upstream of both nozzles and with the laser sheet passing through the axis between the nozzles. Results are reported for a set of fuel lean (0.7 ⩽ ϕ ⩽ 0.9) turbulent premixed counterflow CH 4, C 2H 4 and C 3H 8 flames covering a range of Reynolds numbers (4900 ⩽ Re ⩽ 13,700). The study extends past work featuring single point measurements to include comparisons of mean and fluctuating velocity components along the stagnation point streamline and stagnation plane. The study permits a more detailed assessment of flame structures for a comparatively wide range of conditions and experimental results are further compared with computations using eddy viscosity and comprehensive second moment closures for velocity and scalar fields. The effects of closure approximations and boundary conditions were explored and it is shown that a primary sensitivity is associated with the prescribed initial length scale of turbulence. The study also shows that modifications, based on the ratio of the turbulent to mean strain timescale, do not have a significant impact in the current flow. However, sensitivities to closure approximations relating to mean and fluctuating pressure gradient terms are confirmed. Comparisons with data obtained for the corresponding isothermal flows show that the recorded levels of anisotropy are strongly affected by heat release.
ISSN:1540-7489
1873-2704
DOI:10.1016/j.proci.2006.08.085