0
Gas Turbines: Combustion, Fuels, and Emissions

Experimental and Numerical Investigation of Combustor-Turbine Interaction Using an Isothermal, Nonreacting Tracer

[+] Author and Article Information
Chong M. Cha1 n2

Turbine Systems,  Rolls-Royce PLC, PO Box 31, PCF-1, Derby DE24 8BJ, UKchong.m.cha@gmail.com

Sungkook Hong

 Korea Institute of Energy Research, PO Box 305-343,102 Gajeong-ro, Yuseong-gu,Daejeon, South Korea

Peter T. Ireland

 Turbine Systems, Rolls-Royce PLC, PO Box 31, PCF-1, Derby DE24 8BJ, UK

Paul Denman, Vivek Savarianandam

Department of Aeronautical and Automotive Engineering,  Loughborough University, LE11 3TU, UK

1

Corresponding author.

2

Present address: Turbine Aerodynamics, Rolls-Royce Corp., PO Box 420, Speed Code T-63, Indianapolis, IN 46206.

J. Eng. Gas Turbines Power 134(8), 081501 (Jun 11, 2012) (18 pages) doi:10.1115/1.4005815 History: Received February 19, 2011; Revised October 02, 2011; Published June 11, 2012; Online June 11, 2012

Understanding the interaction between the combustor and turbine subsystems of a gas turbine engine is believed to be key in developing focused strategies for improving turbine performance. Past studies have approached the problem starting with an existing turbine rig with inlet conditions provided by “representative” hardware which attempts to mimic some key flow features exiting the combustor. In this paper, experiments are performed which center around complete engine hardware of the combustor, including engine geometry turbine nozzle guide vanes (NGVs) to solely represent the upstream impact of the complete turbine. This domain ensures that the traditional interface between combustor and turbine is sufficiently encompassed and not compromised by obfuscating or limiting effects due to approximating combustor hardware. The full-annular experimental measurements include all components of the velocity and pressure fields at various planar sections perpendicular to the primary flow direction. These include detailed, two-dimensional measurements both upstream and downstream of the NGVs. The combustor is a classic rich-burn design. Passive scalar (CO2 ) tracing measurements are performed to gain insight into the flow responsible for the temperature fields in the coupled system, including the impact of the NGVs on the upstream flow at the conventional combustor-turbine interface. CFD simulations are used to develop a complete picture of the combustor-turbine interface and the coupling between the two subsystems. The complementary experimental and simulation datasets are together intended to provide a benchmark for future, more traditional turbine rig tests and turbine CFD simulations where inlet conditions are at the exit plane of the combustor.

Copyright © 2012 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Schematic of the Loughborough test rig

Grahic Jump Location
Figure 2

Photograph of the NGVs installed in the test rig. View is looking downstream from the combustor.

Grahic Jump Location
Figure 3

Schematic of the CO2 supply system and photograph of the connections to the six FSNs

Grahic Jump Location
Figure 4

Locations of the three measurement planes

Grahic Jump Location
Figure 5

Sector-to-sector variability as visualized by YCO2* distributions at Plane A

Grahic Jump Location
Figure 6

Sector-to-sector variability quantified using the “RTDF” of YCO2* for each of the two sectors in Fig. 5 (left subplot) and at the other two measurement planes B and C

Grahic Jump Location
Figure 7

The computational domain is a single sector of the combustor annulus. Two simulation cases are performed, one with the NGVs and one without. For the No-NGV case, the domain is identical to that shown apart from the removal of the NGVs.

Grahic Jump Location
Figure 8

Dilution port arrangement for the single-sector computational domain. Subplot (a) shows the pattern on the outer combustor liner and (b) on the inner liner. The annulus surfaces have been removed from Fig. 7.

Grahic Jump Location
Figure 9

Highlights of the computational mesh. Subplot (a) shows the FSN swirler passage and heat-shield inflows, (b) shows the dilution port feeds and relative scale of the RDN holes, and (c) shows the tile film cooling inflow patches on the outer liner.

Grahic Jump Location
Figure 10

CFD predictions of normalized velocity magnitude and mean streamlines for the No-NGV case. Subplot (a) shows the centerline plane which bisects the FSN and primary dilution ports; (b) shows an off-centerline plane which intersects a secondary dilution port.

Grahic Jump Location
Figure 11

Experimental measurements (left column) and CFD predictions (right column) of normalized axial velocities (contours) and radial and circumferential velocity components at each plane (overlayed vector plots). Axial velocity magnitudes have been normalized by the mean, area-averaged velocity of each plane.

Grahic Jump Location
Figure 12

Dilution jet dynamics leading to the characteristic “hot-spot” between burner sectors at the combustor-turbine interface (last downstream plane shown). Figure 8 shows the dilution port arrangement.

Grahic Jump Location
Figure 13

Experimental measurements (left column) and CFD predictions (right column) of YCO2* distributions. No-NGV case.

Grahic Jump Location
Figure 14

Experimental data (symbols) and CFD predictions (bold solid lines) of RTDF calculated using YCO2*. No-NGV case.

Grahic Jump Location
Figure 15

CFD predictions of normalized velocity magnitude and mean streamlines for the NGV case. Subplot (a) shows the centerline plane which bisects the FSN and primary dilution ports; (b) shows an off-centerline plane which intersects one of the two dilution ports for this single sector simulation. Velocities have been normalized by the same reference velocity used for the No-NGV case of Fig. 1.

Grahic Jump Location
Figure 16

CFD predictions of normalized axial velocities (contours) and radial and circumferential velocity components (overlayed vector plots) for the NGV case (left column). Corresponding CFD results for the No-NGV case from Fig. 1 have been copied on the right column.

Grahic Jump Location
Figure 17

CFD predictions of normalized CO2 concentrations for the NGV case (left column). Corresponding CFD results for the No-NGV case from Fig. 1 have been copied on the right column.

Grahic Jump Location
Figure 18

Plane C experimental data and CFD predictions of normalized CO2 concentrations for the NGV case (left column) and No-NGV case (right column)

Grahic Jump Location
Figure 19

CFD predictions of normalized CO2 RTDFs for the NGV case (left column) at Plane C. Corresponding CFD predictions for the No-NGV case from Fig. 1 has been included in the right column. CFD results are shown by the bold solid lines, experimental data by symbols connected by the thin solid lines.

Grahic Jump Location
Figure 20

Normalized turbulent lengthscales (ℓ/H) for the two CFD cases. H is the duct height, which decreases for each respective plane. Note the changing contour scale for each plane.

Grahic Jump Location
Figure 21

Turbulent intensities (u′/⟨⟨U⟩⟩) for the two CFD cases. U is the mean axial velocity, which increases for each respective plane.

Grahic Jump Location
Figure 22

Turbulent viscosity ratio (νt  /ν) for the NGV case at Plane A

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In