Modelling and Measurements of Combustor Cooling Tile Flows

2003 ◽  
Author(s):  
Paul A. Denman ◽  
Ashley G. Barker ◽  
Charith W. Jayatunga ◽  
James J. McGuirk
Keyword(s):  
Heat Transfer ◽  
Large Scale ◽  
Experimental Testing ◽  
Reynolds Numbers ◽  
Scale Model ◽  
Pressure Drops ◽  
Flow And Heat Transfer ◽  
Experimental Side ◽  
Cooling Technology ◽  
Cooling Air

Pressure to reduce available cooling air in modern combustors has driven recent interest in cooling technology based on double-skinned combustor liners, i.e. tiles containing multiple pin-type pedestals to enhance heat transfer. The design of such systems is, however, hampered by the multiplicity of parameters needing optimisation: feedhole configuration, pedestal configuration, tile configuration (e.g. tile overlap). Much experimental testing is currently needed. In addition, the simulation of flow and heat transfer in cooling tile geometries using RANS-based CFD is made particularly difficult by the impossibility of resolving every individual pin in the pedestal matrix whilst retaining an overall CFD problem of reasonable size. The present paper describes a mixture of experimental and computational work undertaken to explore cooling tile flows. On the experimental side, a large-scale Perspex aerodynamic rig of a cooling tile was constructed. Measurements at representative Reynolds numbers were possible and delivered information on discharge coefficients, pressure drops and flow splits for various tile configurations. The same tile geometries were subsequently modeled using a RANS-based CFD approach. The novelty in these simulations was the use of a ‘sub-grid-scale’ model for the pedestal flow and heat transfer. This approach has previously been used in combustor heatshield predictions; it is demonstrated in the present work how it may also be applied to cooling tiles.

Heat Transfer in a “Cover-Plate” Preswirl Rotating-Disk System

1999 ◽  
Vol 121 (2) ◽  
pp. 249-256 ◽  
Author(s):  
R. Pilbrow ◽  
H. Karabay ◽  
M. Wilson ◽  
J. M. Owen
Keyword(s):  
Heat Transfer ◽  
Rotating Disk ◽  
Reynolds Numbers ◽  
Turbine Disk ◽  
Cover Plate ◽  
Blade Cooling ◽  
Nusselt Numbers ◽  
Flow And Heat Transfer ◽  
Cooling Air ◽  
Plate System

In most gas turbines, blade-cooling air is supplied from stationary preswirl nozzles that swirl the air in the direction of rotation of the turbine disk. In the “cover-plate” system, the preswirl nozzles are located radially inward of the blade-cooling holes in the disk, and the swirling airflows radially outward in the cavity between the disk and a cover-plate attached to it. In this combined computational and experimental paper, an axisymmetric elliptic solver, incorporating the Launder–Sharma and the Morse low-Reynolds-number k–ε turbulence models, is used to compute the flow and heat transfer. The computed Nusselt numbers for the heated “turbine disk” are compared with measured values obtained from a rotating-disk rig. Comparisons are presented, for a wide range of coolant flow rates, for rotational Reynolds numbers in the range 0.5 X 106 to 1.5 X 106, and for 0.9 < βp < 3.1, where βp is the preswirl ratio (or ratio of the tangential component of velocity of the cooling air at inlet to the system to that of the disk). Agreement between the computed and measured Nusselt numbers is reasonably good, particularly at the larger Reynolds numbers. A simplified numerical simulation is also conducted to show the effect of the swirl ratio and the other flow parameters on the flow and heat transfer in the cover-plate system.

Heat Transfer in a “Cover-Plate” Pre-Swirl Rotating-Disc System

1998 ◽  
Author(s):  
Robert Pilbrow ◽  
Hasan Karabay ◽  
Michael Wilson ◽  
J. Michael Owen
Keyword(s):  
Heat Transfer ◽  
Reynolds Numbers ◽  
Cover Plate ◽  
Swirl Ratio ◽  
Rotating Disc ◽  
Blade Cooling ◽  
Nusselt Numbers ◽  
Flow And Heat Transfer ◽  
Turbine Disc ◽  
Cooling Air

In most gas turbines, blade-cooling air is supplied from stationary pre-swirl nozzles that swirl the air in the direction of rotation of the turbine disc. In the “cover-plate” system, the pre-swirl nozzles are located radially inward of the blade-cooling holes in the disc, and the swirling air flows radially outwards in the cavity between the disc and a cover-plate attached to it. In this combined computational and experimental paper, an axisymmetric elliptic solver, incorporating the Launder-Sharma and the Morse low-Reynolds-number k-ε turbulence models, is used to compute the flow and heat transfer. The computed Nusselt numbers for the heated “turbine disc” are compared with measured values obtained from a rotating-disc rig. Comparisons are presented, for a wide range of coolant flow rates, for rotational Reynolds numbers in the range 0.5 × 106 to 1.5 × 106, and for 0.9 < βp < 3.1, where βp is the pre-swirl ratio (or ratio of the tangential component of velocity of the cooling air at inlet to the system to that of the disc). Agreement between the computed and measured Nusselt numbers is reasonably good, particularly at the larger Reynolds numbers. A simplified numerical simulation is also conducted to show the effect of the swirl ratio and the other flow parameters on the flow and heat transfer in the cover-plate system.

Effect of an Oscillating Flow Direction on Leading Edge Heat Transfer

1984 ◽  
Vol 106 (1) ◽  
pp. 222-228 ◽  
Author(s):  
M. L. Marziale ◽  
R. E. Mayle
Keyword(s):  
Heat Transfer ◽  
Strouhal Number ◽  
Large Scale ◽  
Flow Direction ◽  
Leading Edge ◽  
Periodic Variation ◽  
Scale Model ◽  
Transfer Rates ◽  
Large Scale Model ◽  
Turbulence Length

An experimental investigation was conducted to examine the effect of a periodic variation in the angle of attack on heat transfer at the leading edge of a gas turbine blade. A circular cylinder was used as a large-scale model of the leading edge region. The cylinder was placed in a wind tunnel and was oscillated rotationally about its axis. The incident flow Reynolds number and the Strouhal number of oscillation were chosen to model an actual turbine condition. Incident turbulence levels up to 4.9 percent were produced by grids placed upstream of the cylinder. The transfer rate was measured using a mass transfer technique and heat transfer rates inferred from the results. A direct comparison of the unsteady and steady results indicate that the effect is dependent on the Strouhal number, turbulence level, and the turbulence length scale, but that the largest observed effect was only a 10 percent augmentation at the nominal stagnation position.

A Numerical Study of Heat Transfer and Flow Structure in Channels With Miniature V Rib-Dimple Hybrid Structure on One Wall

2018 ◽  
Author(s):  
Peng Zhang ◽  
Yu Rao ◽  
Yanlin Li
Keyword(s):  
Heat Transfer ◽  
Turbulent Flow ◽  
Numerical Simulations ◽  
Numerical Study ◽  
Reynolds Numbers ◽  
Hybrid Structure ◽  
Transfer Enhancement ◽  
Nusselt Numbers ◽  
Flow And Heat Transfer ◽  
Flow Mixing

This paper presents a numerical study on turbulent flow and heat transfer in the channels with a novel hybrid cooling structure with miniature V-shaped ribs and dimples on one wall. The heat transfer characteristics, pressure loss and turbulent flow structures in the channels with the rib-dimples with three different rib heights of 0.6 mm, 1.0 mm and 1.5 mm are obtained for the Reynolds numbers ranging from 18,700 to 60,000 by numerical simulations, which are also compared with counterpart of a pure dimpled and pure V ribbed channel. The results show that the overall Nusselt numbers of the V rib-dimple channel with the rib height of 1.5 mm is up to 70% higher than that of the channels with pure dimples. The numerical simulations show that the arrangement of the miniature V rib upstream each dimple induces complex secondary flow near the wall and generates downwashing vortices, which intensifies the flow mixing and turbulent kinetic energy in the dimple, resulting in significant improvement in heat transfer enhancement and uniformness.

Heat Transfer and Pressure Drop in a Converging Pedestal Array With Exit Area Variation

2018 ◽  
Author(s):  
L. W. Soma ◽  
F. E. Ames ◽  
S. Acharya
Keyword(s):  
Heat Transfer ◽  
Pressure Drop ◽  
Flow Rate ◽  
Film Cooling ◽  
Reynolds Numbers ◽  
Flow Path ◽  
Internal Cooling ◽  
Channel Measurements ◽  
Trailing Edge ◽  
Cooling Air

The trailing edge of a vane is one of the most difficult areas to cool due to a narrowing flow path, high external heat transfer rates, and deteriorating external film cooling protection. Converging pedestal arrays are often used as a means to provide internal cooling in this region. The thermally induced stresses in the trailing edge region of these converging arrays have been known to cause failure in the pedestals of conventional solidity arrays. The present paper documents the heat transfer and pressure drop through two high solidity converging rounded diamond pedestal arrays. These arrays have a 45 percent pedestal solidity. One array which was tested has nine rows of pedestals with an exit area in the last row consistent with the convergence. The other array has eight rows with an expanded exit in the last row to enable a higher cooling air flow rate. The expanded exit of the eight row array allows a 30% increase in the coolant flow rate compared with the nine row array for the same pressure drop. Heat transfer levels correlate well based on local Reynolds numbers but fall slightly below non converging arrays. The pressure drop across the array naturally increases toward the trailing edge with the convergence of the flow passage. A portion of the cooling air pressure drop can be attributed to acceleration while a portion can be attributed to flow path losses. Detailed array static pressure measurements provide a means to develop a correlation for the prediction of pressure drop across the cooling channel. Measurements have been acquired over Reynolds numbers based on exit flow conditions and the characteristic pedestal length scale ranging from 5000 to over 70,000.

Aerodynamics of a Letterbox Trailing Edge: Effects of Blowing Rate, Reynolds Number, and External Turbulence on Aerodynamic Losses and Pressure Distribution

2010 ◽  
Vol 132 (4) ◽  
Author(s):  
N. J. Fiala ◽  
J. D. Johnson ◽  
F. E. Ames
Keyword(s):  
Heat Transfer ◽  
Reynolds Number ◽  
Film Cooling ◽  
Large Scale ◽  
Reynolds Numbers ◽  
Design Flow ◽  
Trailing Edge ◽  
Film Cooling Effectiveness ◽  
Aerodynamic Loss ◽  
External Turbulence

A letterbox trailing edge configuration is formed by adding flow partitions to a gill slot or pressure side cutback. Letterbox partitions are a common trailing edge configuration for vanes and blades, and the aerodynamics of these configurations are consequently of interest. Exit surveys detailing total pressure loss, turning angle, and secondary velocities have been acquired for a vane with letterbox partitions in a large-scale low speed cascade facility. These measurements are compared with exit surveys of both the base (solid) and gill slot vane configurations. Exit surveys have been taken over a four to one range in chord Reynolds numbers (500,000, 1,000,000, and 2,000,000) based on exit conditions and for low (0.7%), grid (8.5%), and aerocombustor (13.5%) turbulence conditions with varying blowing rate (50%, 100%, 150%, and 200% design flow). Exit loss, angle, and secondary velocity measurements were acquired in the facility using a five-hole cone probe at a measuring station representing an axial chord spacing of 0.25 from the vane trailing edge plane. Differences between losses with the base vane, gill slot vane, and letterbox vane for a given turbulence condition and Reynolds number are compared providing evidence of coolant ejection losses, and losses due to the separation off the exit slot lip and partitions. Additionally, differences in the level of losses, distribution of losses, and secondary flow vectors are presented for the different turbulence conditions at the different Reynolds numbers. The letterbox configuration has been found to have slightly reduced losses at a given flow rate compared with the gill slot. However, the letterbox requires an increased pressure drop for the same ejection flow. The present paper together with a related paper (2008, “Letterbox Trailing Edge Heat Transfer—Effects of Blowing Rate, Reynolds Number, and External Turbulence on Heat Transfer and Film Cooling Effectiveness,” ASME, Paper No. GT2008-50474), which documents letterbox heat transfer, is intended to provide designers with aerodynamic loss and heat transfer information needed for design evaluation and comparison with competing trailing edge designs.

Unsteady RANS Simulation of Turbulent Flow and Heat Transfer in Ribbed Coolant Passages of Different Aspect Ratios

2004 ◽  
Author(s):  
A. K. Saha ◽  
Sumanta Acharya
Keyword(s):  
Heat Transfer ◽  
Secondary Flow ◽  
Rotation Number ◽  
Large Scale ◽  
Density Ratio ◽  
Navier Stokes ◽  
Flow Structures ◽  
Flow And Heat Transfer ◽  
Aspect Ratios ◽  
Unsteady Rans

The flow and heat transfer in ribbed coolant passages of aspect ratios (AR) of 1:1, 4:1, and 1:4 are numerically studied through the solution of the Unsteady Reynolds Averaged Navier-Stokes (URANS) equations. The ribs are oriented normal to the flow and arranged in a staggered configuration on the leading and trailing surfaces. The URANS procedure can resolve large-scale bulk unsteadiness, and utilizes a two equation k-ε model for the turbulent stresses. Both Coriolis and centrifugal buoyancy effects are included in the simulations. The computations are carried out for a fixed Reynolds number of 25000 and density ratio of 0.13 while the Rotation number has been varied between 0.12–0.50. The average duct heat transfer is the highest for the 4:1 AR case. For this case, the secondary flow structures consist of multiple roll cells that direct flow both to the trailing and leading surfaces. The 1:4 AR duct shows flow reversal along the leading surface at high rotation numbers with multiple rolls in the secondary flow structures near the leading wall. For this AR, the potential for conduction-limited heat transfer along the leading surface is identified. At high rotation number, both the 1:1 and 4:1 AR cases exhibit loss of axial periodicity over one inter-rib module. The friction factor reveals an increase with the rotation number for all aspect ratio ducts, and shows a sudden jump in its value at a critical rotation number because of either loss of spatial periodicity or the onset of backflow.

Sign in / Sign up

Export Citation Format

Share Document