### Velocity and Temperature Natural Dissimilarity in a Turbulent Channel Flow

#### Abstract

Natural dissimilarity or de correlation of axial velocity and

temperature fluctuations, in a tur bulent channel flow, is studied

using direct numerical simulation, DNS. Buoyancy effects were

neglected, thus the temperature was considered as a passive scalar. A

uniform energy source case for the thermal field has been

used. Results for molecular Pr or Sc numbers equal to 1.0 and 0.71 are

presented. More evidences of the strong correlation of axial velocity

and temperature in the wall layer are shown, like as the similar

patter of the skin friction and streamwise vorticity correlation, with

that between wall heat flux and streamwise vorticity correlation. The

importance of the most energetic events on the dissimi larity between

the axial velocity and temperature fluctuations is examined using

conditional probability. It is shown that although the most energetic

events are responsible of the strongest instantaneous dis

similarities, their contribution to the mean dissimilarity is less

than a half in the whole channel. As a complement to many previous

results in the literature analyzing fluctuations of longitudinal

velocity and temperature in frequency domain, spectral density

functions is used in order to study dissimilarity. The results

presented here include new variables, as the spectra of the

fluctuations of axial velocity and temperature difference, and the

spectra of the fluctuations of the pressure field. Spectral density

functions at different distances from the wall show, that the main

cause of dissimilarity between axial velocity and temperature

fluctuations is the shift toward higher frequencies of temperature in

comparison to any velocity components, and specially to axial

velocity, in the viscous, buffer, and beginning of the logarithmic

region. However, in contrast with this situation next to the wall,

there is a general tendency to spectral convergence at the center of

the channel. Based on the spectra of the fluctuations of the pressure

field, it appears that one can conclude that such actions next to the

wall and at the center region are driven by the pressure field. It is

speculated, however, that the commented convergence at the center

region can be greater for higher Reynolds numbers than that used in

the present work.

temperature fluctuations, in a tur bulent channel flow, is studied

using direct numerical simulation, DNS. Buoyancy effects were

neglected, thus the temperature was considered as a passive scalar. A

uniform energy source case for the thermal field has been

used. Results for molecular Pr or Sc numbers equal to 1.0 and 0.71 are

presented. More evidences of the strong correlation of axial velocity

and temperature in the wall layer are shown, like as the similar

patter of the skin friction and streamwise vorticity correlation, with

that between wall heat flux and streamwise vorticity correlation. The

importance of the most energetic events on the dissimi larity between

the axial velocity and temperature fluctuations is examined using

conditional probability. It is shown that although the most energetic

events are responsible of the strongest instantaneous dis

similarities, their contribution to the mean dissimilarity is less

than a half in the whole channel. As a complement to many previous

results in the literature analyzing fluctuations of longitudinal

velocity and temperature in frequency domain, spectral density

functions is used in order to study dissimilarity. The results

presented here include new variables, as the spectra of the

fluctuations of axial velocity and temperature difference, and the

spectra of the fluctuations of the pressure field. Spectral density

functions at different distances from the wall show, that the main

cause of dissimilarity between axial velocity and temperature

fluctuations is the shift toward higher frequencies of temperature in

comparison to any velocity components, and specially to axial

velocity, in the viscous, buffer, and beginning of the logarithmic

region. However, in contrast with this situation next to the wall,

there is a general tendency to spectral convergence at the center of

the channel. Based on the spectra of the fluctuations of the pressure

field, it appears that one can conclude that such actions next to the

wall and at the center region are driven by the pressure field. It is

speculated, however, that the commented convergence at the center

region can be greater for higher Reynolds numbers than that used in

the present work.

#### Full Text:

PDF

Güemes 3450

S3000GLN Santa Fe, Argentina

Phone: 54-342-4511594 / 4511595 Int. 1006

Fax: 54-342-4511169

E-mail: amca(at)santafe-conicet.gov.ar

**Asociación Argentina de Mecánica Computacional**Güemes 3450

S3000GLN Santa Fe, Argentina

Phone: 54-342-4511594 / 4511595 Int. 1006

Fax: 54-342-4511169

E-mail: amca(at)santafe-conicet.gov.ar

**ISSN 2591-3522**