Thermal boundary layers on a rough surface downstream of steps in wall temperature

Abstract

Stanton numbers, mean temperature profiles, and turbulent heat flux profiles are presented for thermal boundary layers which develop over a uniformly rough surface downstream of steps in wall temperature. Artificially thickened hydrodynamic layers were employed in order to create large unheated starting lengths compared to boundary-layer thickness, where the unheated starting length is the distance between the temperature step and hydrodynamic boundary-layer origin. Kernel functions for Stanton number distributions with downstream distance are presented which agree with measurements even when the hydrodynamic layers are significantly thicker than thermal layers. Other effects of unheated starting lengths on rough-wall thermal boundary-layer behavior are described, including significant alterations in profiles of mean temperature and turbulent heat flux. Both change shape as the magnitude of the unheated starting length increases, even though logarithmic regions of temperature profiles in wall coordinates and turbulent Prandtl number distributions are largely unaffected by unheated starting-length effects.

Aerodynamic and heat transfer effects of distributed hemispherical roughness elements inducing step changes in a turbulent boundary layer

“…dicating similarity between these profiles. A logarithmic region is observed over a significant spatial extent, with a slope estimated using a logarithmic regression over that region that compares favorably with the data reported by Pimenta et al [57] and Ligrani and Moffat [16]. Some discrepancies nonetheless exist, these two reference studies having focused on different configurations.…”

“…Pimenta et al [57] studied a rough surface made of closely-packed spheres and such that the entire length of the rough flat plate was heated, providing almost matched dynamic Table 1. (a) Temperature profiles made dimensionless using the temperature difference across the TBL (T w − T e ) and the enthalpy thickness ∆ 2 , compared with the experimental results reported by Pimenta et al [57] and Ligrani and Moffat [16] without and with unheated starting-length respectively. The dash red line ( ) indicates a logarithmic fit of the present data over z/∆ 2 ∈ [1,10].…”

“…A second important aspect of the present work deals with heat transfer modifications in thermal boundary layers induced by increasingly large hemispherical roughness elements. To the knowledge of the authors, the most detailed sets of experiments investigating the effects of distributed geometrical roughness elements on thermal boundary layers were mainly conducted on two roughness geometries: packed spheres [15,16] and staggered hemispheres [17,18] with an arrangement similar to the one used in the present work. These two configurations were however analyzed with different objectives in mind.…”

“…To this end, boundary layer velocity and temperature profiles as well as Reynolds stresses and turbulent heat fluxes were measured. Building on a similar experimental setup, Ligrani and Moffat [16] explored the effects of steps in wall temperature on the characteristics of a thermal boundary layer. Here again, a detailed analysis of the structure of the TBL was conducted providing insights on the structure and development of a thermal layer inside a TBL over packed spheres.…”

Flumes for benthic ecologists: theory, construction and practice

“…The appropriate size and spacing will depend on flow speed and depth and must be determined by trying different configurations and measuring velocity profiles downstream. A more complex device that shows promise is the 'elliptic wedge generator' of Ligrani et al (1979Ligrani et al ( , 1983Ligrani et al ( , 1985. It consists of a set of spires and a small barrier placed in the flow.…”