0
Research Papers

Three-Dimensional RANS Prediction of Gas-Side Heat Transfer Coefficients on Turbine Blade and Endwall

[+] Author and Article Information
Hee-Koo Moon

Solar Turbines Incorporated,
A Caterpillar Company,
San Diego, CA 92101

Manuscript received November 27 2011; final manuscript received December 15 2011; published online November 1, 2012. Editor: David Wisler.

J. Turbomach 135(2), 021005 (Nov 01, 2012) (11 pages) Paper No: TURBO-11-1249; doi: 10.1115/1.4006642 History: Received November 27, 2011; Revised December 15, 2011

This paper presents a study using 3D computational fluid dynamics (CFD) based on Reynolds-averaged Navier-Stokes (RANS) equations to predict turbine gas-side heat transfer coefficients (HTC) on the entire airfoil and endwall. The CFD results at different spanwise sections and endwall have been compared with the flat-plate turbulent boundary layer correlation and with the data in a NASA turbine rotor passage with strong secondary flows, under three different flow conditions. The enhancement effects of secondary flow vortices on the blade surface and endwall heat transfer rate have been examined in detail. Analyses were conducted for the impact of Reynolds number and exit Mach number on heat transfer. The SST, k-ɛ, V2F, and realizable k-ɛ turbulence models have been assessed. The classical log-law wall-functions have been found to be comparable to the wall-integration methods but with much reduced sensitivity to inlet turbulence conditions. The migration of hot gas was simulated with a radial profile of inlet temperature. CFD results for mid-span HTCs of two other airfoils were also compared with test data. Overall, results are encouraging and indicate improved HTC and temperature predictions from 3D CFD could help optimize the design of turbine cooling schemes.

Copyright © 2013 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Comparison of CFD mesh: wall-function (WF) versus low-Reynolds-number (LRN)

Grahic Jump Location
Fig. 2

CFD-predicted loading versus data at three spanwise sections (50 %, 10 %, 2.5 %, Case #2). (a) k-ɛ with wall-function. (b) SST with wall-function and LRN.

Grahic Jump Location
Fig. 3

Streamlines and secondary vortices (predicted by SST with wall-function) through the turbine passage (wall colored by heat flux). (a) Suction side view (vortex cores shown in white curves). (b) Second view (horseshoe vortex, downwash; passage vortex).

Grahic Jump Location
Fig. 4

CFD-predicted HTC versus data at three spanwise locations: impact of turbulence model and near-wall treatment (SST_wf, SST_lrn versus V2F). (a) 50 % span (S/C < 0 pressure side; > 0 suction side). (b) 25 % span. (c) 10 % span.

Grahic Jump Location
Fig. 5

HTC predicted by k-ɛ_wf and k-ɛ_lrn models versus data at mid-span (Case #2; Tu1 = 8 %): impact of length scale and wall treatment. (a) k-ɛ_lrn model. (b) k-ɛ_wf model.

Grahic Jump Location
Fig. 6

Heat transfer coefficients predicted by the k-ε, realizable k-ε, and SST models with wall-function versus the data and flat-plate correlation at 50 %, 25 %, 10 % span (Case #2). (a) 50 % span. (b) 25 % span. (c) 10 % span.

Grahic Jump Location
Fig. 7

Impact of the definitions of HTC (HTC1 = Q•w/(Tt1-Tw), HTC2 = Q•w/(Taw-Tw)) on the predictions (by the SST model with wall-function; 10 % span of Case #2)

Grahic Jump Location
Fig. 8

Predicted Stanton number (St × 1000) on the endwall (by SST model with wall-function) compared with data

Grahic Jump Location
Fig. 9

Predicted Stanton number (St × 1000) on the blade (by SST model with wall-function) in comparison to data

Grahic Jump Location
Fig. 10

CFD-predicted HTC on the blade and endwall: effects of turbulence modeling. (a) Suction side view. (b) Pressure side view.

Grahic Jump Location
Fig. 11

HTC predicted by k-ε and SST with wall-function versus flat-plate correlation and data at the 50 % span (Case #6)

Grahic Jump Location
Fig. 12

SST-predicted HTC versus flat-plate correlation and data at 25 %, 10 % span: Impact of Reynolds number (Case #2 versus Case #6). (a) 25 % span. (b) 10 % span.

Grahic Jump Location
Fig. 13

HTC predicted by k-ε and SST with wall-function versus flat-plate correlation and data at 25 % span (Case #5)

Grahic Jump Location
Fig. 14

CFD-predicted (by SST_wf) isentropic Mach number versus data at 10 % span (Case #6 versus Case #5)

Grahic Jump Location
Fig. 15

CFD-predicted HTC (by SST with wall-function) versus data at 10 % span (Case #6 versus Case #5)

Grahic Jump Location
Fig. 16

CFD-predicted (by SST with wall-function) HTC versus data at three sections (Case #5)

Grahic Jump Location
Fig. 17

CFD-predicted (with the SST_wf model) contour plots of total temperature on cross-sections and streamlines. (a) Total temperature (normalized). (b) Streamlines and total temperature.

Grahic Jump Location
Fig. 18

CFD-predicted HTC (by the SST model and wall function) on the nozzle [23] versus the flat-plate laminar and turbulent boundary layer correlations and test data. (a) CFD mesh. (b) Predicted HTC versus data.

Grahic Jump Location
Fig. 19

CFD-predicted HTC (by the low-Reynolds-number SST model) on NASA turbine nozzle (MarkII) versus the flat-plate TBL correlation and test data. (a) CFD mesh. (b) HTC (ARC = surface arc length).

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