Abstract
This work attempts to enlighten the influence of core flow disturbance and boundary layer disturbance on momentum and energy transport. Local temperature distributions over entire heated surfaces with the associated pressure drops are examined using Infrared Thermometry (IT) and pressure transducers with Reynolds number (Re) ranging from 5,000 to 20,000. The detailed turbulent flow features are measured with Particle Image velocimetry (PIV) to provide the flow physics attributed to the heat transfer performances at a fixed Re of 10,000. The first part of the present study explores the detailed flow fields, heat transfer distributions, and pressure drop penalties in a stationary two-pass smooth parallelogram channel with 180-deg sharp bend as the baseline case. The particular asymmetrically and suddenly contracted inlet condition (hereinafter referred to as “ASI”) emulates the abrupt flow entrance of the real gas turbine blade cooling conditions. The secondary-flow patterns as well as the near-wall streamwise mean velocity components and turbulent kinetic energy are analyzed to correlate the relationship between flow characteristics and heat transfer distributions. The most distinct finding of this study is the asymmetric thermal and fluid flow features on the top and bottom wall side, in contrast to the symmetric ones in the corresponding square and rectangular channels. Compared with the previous fully developed inlet condition (hereinafter referred to as “FDI”) case, it is found that the thermal performance factors (TPF) of the ASI case are respectively 43.5% and 53.7% higher than those of the FDI case at Re=5,000 and 20,000 because of the entrance-induced secondary flow and disturbance. The correlations of Nu and fanning friction factor (f) are obtained to compare with those generated from the corresponding square channels. The second part of the present study explores the detailed flow patterns and turbulence parameters, including streamwise and spanwise mean flow field, secondary-flow mean velocity map, turbulent intensity, Reynolds stress and turbulent kinetic energy as well as the heat transfer performances for the two-pass ribbed parallelogram channels with three different rib orientations (45-deg, -45-deg, and 90-deg). It is found that the combined effect of inlet condition and rib orientation extends to the entire parallelogram channel rather than the first passage only. Among the rib orientations investigated, the 45-deg ribs enhance the effects of parallelogram slant sidewalls and entrance to extend the Nu differences between the top and bottom walls. In contrast, the other two rib configurations, 90-deg ribs especially, weaken these effects and narrow the corresponding differences. Based on the ability to elevate the surface-averaged Nu/Nu∞ ratios, the respective orders of (45-deg, −45-deg, 90-deg) and (90-deg, 45-deg, −45-deg) for the present top and bottom walls are significantly different from those in square and rectangular channels. The variations of channel-averaged Nu/Nu∞ and f/f∞ with the associated TPF values against Re are compared with the previous results obtained from literature. In general, the three tested angled ribs all provide the TPF values above than unity. Only the TPF values in 45-deg case are always higher than those in corresponding smooth parallelogram channel (ASI case). Divided by the TPF values in ASI case, the normalized TPF values are respectively 1.09-1.04, 1.07-0.87, and 0.83-0.99 for 45, 90, −45-deg ribbed cases. It is thus important to take the entrance effect into account when applying the angled ribs in a two-pass parallelogram channel. The third part of the present study aims to propose innovative louver-type turbulators to enhance the heat transfer rate via three flow mechanics, namely core flow disturbance, jet impingement, and extended heat transfer surface. These louvers are installed in the twin-pass square channel with a fully developed inlet condition. The parameters are examined to disclose optimal design in terms of the pitch ratio (Pi/DH=1, 2, 3, 4, and ∞) and the number of slat per half louver (1≤Ns≤4). The pressure drop measurements are also performed to estimate the Fanning friction factors (f) and the thermal performance factors (TPF). The results show that both (Nu) ̅/Nu∞ and f ̅/f∞ ratios rise with descending Pi/DH and ascending Ns under the present test conditions. Among all the tested cases, the case with Pi/DH=1 and Ns=4 provides the highest Nu/Nu∞, almost twice the value of smooth reference. Nevertheless, it suffers from high f/f∞ penalty. The TPF level is a relatively weak function of Pi/DH. There exists a critical slat number of Ns=3 above which the TPF value is a weak function of Ns. Below the critical Ns, the TPF value increases with decreasing Ns. From the viewpoint of heat transfer enhancement, one could apply the louvered channel as a heat exchanger with small Pi/DH and large Ns. The boundary layer disturbance, on the other hand, is more cost-effective than core flow disturbance as a mechanism to augment heat transfer from the viewpoint of thermal performance.