Boundary layer loss mechanism and justification of wall functions for turbulence modeling

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations


The mechanism of entropy creation in an adiabatic boundary layer is studied. The local entropy creation rate has two sources: shear stress work and heat flux gradient. These two factors have the same order of magnitude everywhere in the boundary layer. Their balance results in a fairly uniform distribution of entropy creation rate across the boundary layer. In the inner layer region, the large rate of entropy creation due to the velocity gradient and shear stress is offset by the gradient of the heat flux. The heat flux gradient is so large that a negative entropy creation region is generated in the inner layer. With the new theory of the entropy creation mechanism, the wall function methods for the internal turbulent flow loss prediction are justified. Detailed theoretical derivation and numerical proof are given in the paper. The previous theory of local entropy creation mechanism in a boundary layer given by Denton ignored the important heat flux gradient factor. Denton's theory will lead to a large false error when it is used to study the applicability of wall function methods to loss prediction. The numerical solutions of a transonic cascade using wall functions and integrating to the wall are compared. The difference for the loss prediction using these two methods is small and is consistent with the error range given from the loss mechanism theory of this paper.

Original languageEnglish
Title of host publicationAIAA Paper
Number of pages14
StatePublished - Jul 1 2004
Event42nd AIAA Aerospace Sciences Meeting and Exhibit - Reno, NV, United States
Duration: Jan 5 2004Jan 8 2004


Other42nd AIAA Aerospace Sciences Meeting and Exhibit
Country/TerritoryUnited States
CityReno, NV

ASJC Scopus subject areas

  • Engineering(all)


Dive into the research topics of 'Boundary layer loss mechanism and justification of wall functions for turbulence modeling'. Together they form a unique fingerprint.

Cite this