Turbine blade entropy generation rate part I: The boundary layer defined

M. R.D. Davies, F. K. O'Donnell, A. J. Niven

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The profile loss of a gas turbine blade is normally associated with the entropy increase due to the boundary layer phenomena of viscous shear, Reynolds stress generation and heat transfer. To establish the relative contributions of laminar, transitional and turbulent adiabatic boundary layer flow, to the overall entropy generation (as described in part two of this paper), detailed hot film and hot wire measurements have been made over the suction surface of a turbine blade mounted within a subsonic linear cascade. At a Reynolds number of 185 × 103, a natural transition region was found between 53 and 70% suction surface length, followed by a slowly relaxing turbulent boundary layer. The wall shear stress distribution indicated a peak in the leading edge region, dropping to a minimum value prior to transition, with a sharp rise over the transitional length before decreasing within the turbulent portion. The measurements were compared with predictions obtained from a commercial computational fluid dynamics code, which utilised the renormalisation group theory (RNG) turbulence model.

Original languageEnglish
Title of host publicationHeat Transfer; Electric Power; Industrial and Cogeneration
PublisherAmerican Society of Mechanical Engineers (ASME)
ISBN (Electronic)9780791878569, 9780791878569
DOIs
Publication statusPublished - 2000
EventASME Turbo Expo 2000: Power for Land, Sea, and Air, GT 2000 - Munich, Germany
Duration: 8 May 200011 May 2000

Publication series

NameProceedings of the ASME Turbo Expo
Volume3

Conference

ConferenceASME Turbo Expo 2000: Power for Land, Sea, and Air, GT 2000
Country/TerritoryGermany
CityMunich
Period8/05/0011/05/00

Fingerprint

Dive into the research topics of 'Turbine blade entropy generation rate part I: The boundary layer defined'. Together they form a unique fingerprint.

Cite this