0
TECHNICAL PAPERS

# Effects of Wall Roughness on Particle Velocities in a Turbulent Channel Flow

[+] Author and Article Information
Michael Benson, Tomohiko Tanaka, John K. Eaton

Department of Mechanical Engineering,Stanford University

J. Fluids Eng 127(2), 250-256 (Dec 13, 2004) (7 pages) doi:10.1115/1.1891149 History: Received May 09, 2003; Revised December 13, 2004

## Abstract

Experimental measurements using a laser Doppler anemometer (LDA) system have been performed on $150μm$ dense glass particles in a fully developed downward channel flow in air. Tests were conducted in smooth, rough development, and fully rough wall conditions with a channel Reynolds number of 13,800, corresponding to a centerline gas phase velocity of $10.5m∕s$ with a dilute loading of particles of 15% by mass fraction. Velocities were measured and statistics compared to see the nature of the effects of the wall roughness in a rebuilt channel facility originally used for important works including Kulick, Fessler, and Eaton, (1994, “Particle Response and Turbulence Modification in Fully-Developed Channel flow  ,” J. Fluid Mech., 277, pp. 109–134) and Paris (2001, “Turbulence Attenuation in a Particle-Laden Channel Flow  ,” Ph.D. thesis, Stanford University, Stanford, CA). Wall roughness has a substantial impact on gas phase mean velocities across most of the channel width, except very near the wall. The turbulence intensity of the gas phase is enhanced across the entire channel in the presence of fully rough walls. The rough walls have an even greater impact on the particle phase. Streamwise particle velocities are reduced up to 40%, and become quite uniform across the channel. Particle fluctuating velocities are nearly doubled near the channel centerplane. Profiles appear uniform, due in large part to strong transverse mixing induced by particle-wall collisions. Much of the data of Kulick and Paris is shown here to be strongly influenced by wall conditions with poorly defined roughness in the development region, followed by rapid flow recovery in a relatively smooth test section.

<>

## Figures

Figure 1

Schematic of the experimental setup

Figure 2

Streamwise single phase mean velocity profile, normalized by the mean gas phase centerline velocity. Smooth wall and rough development conditions are presented for comparison with Kulick, Fessler, and Eaton (1) and Paris (2) data.

Figure 3

Streamwise single phase rms velocity profile normalized by the mean gas phase centerline velocity. Smooth wall and rough development conditions are presented for comparison with Kulick, Fessler, and Eaton (1) and Paris (2) data.

Figure 4

Streamwise particulate phase mean velocity profile normalized by the mean gas phase centerline velocity. Smooth wall and rough development conditions are presented for comparison with Paris (2) data.

Figure 5

Streamwise particulate phase rms velocity profile normalized by the mean gas phase centerline velocity. Smooth wall and rough development conditions are presented for comparison with Paris (2) data.

Figure 6

Gas phase mean streamwise velocities for the smooth wall versus fully rough (5cm) and rough development (30cm) wall conditions. Velocities are normalized by the mean gas phase centerline velocity.

Figure 7

Gas phase streamwise rms velocities for the smooth wall versus fully rough (5cm) and rough development (30cm) wall conditions. Velocities are normalized by the mean gas phase centerline velocity.

Figure 8

Channel centerline PDF distribution for the streamwise gas phase velocity for smooth and fully rough walls

Figure 9

Comparison of streamwise particulate phase mean velocity profiles among smooth, rough development and fully rough walls. Velocities are normalized by the mean gas phase centerline velocity.

Figure 10

Comparison of mean streamwise gas and particle phase velocity profiles for smooth walls, with particle centerline terminal velocity. Velocities are normalized by the mean gas phase centerline velocity.

Figure 11

Comparison of mean streamwise gas and particle phase velocity profiles for fully rough walls. Velocities are normalized by the mean gas phase centerline velocity.

Figure 12

Comparison of streamwise particulate phase rms velocity profiles among smooth and fully rough walls. Velocities are normalized by the mean gas phase centerline velocity.

Figure 13

Comparison of rms streamwise gas and particle phase velocity profiles for fully rough walls. Velocities are normalized by the mean gas phase centerline velocity.

Figure 14

Channel centerline PDF distribution for the streamwise particle phase velocity for smooth and fully rough walls

Figure 15

Near-wall (x2∕h=0.0315) PDF distribution for the streamwise particle phase velocity for smooth and fully rough walls

Figure 16

Wall-normal particulate phase rms velocity for the smooth and fully rough wall conditions. Velocities are normalized by the mean gas phase centerline velocity.

## 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 Proceedings Articles
Related eBook Content
Topic Collections