Influence of Radial Rotation on Heat Transfer in a Rectangular Channel With Two Opposite Walls Roughened by Hemispherical Protrusions at High Rotation Numbers

2011 ◽  
Vol 134 (1) ◽  
Author(s):  
Shyy Woei Chang ◽  
Tong-Miin. Liou ◽  
Wei-Chun Chen

Detailed heat transfer distributions over two opposite leading and trailing walls roughened by hemispherical protrusions were measured from a rotating rectangular channel at rotation number up to 0.6 to examine the effects of Reynolds (Re), rotation (Ro), and buoyancy (Bu) numbers on local and area-averaged Nusselt numbers (Nu and Nu¯) using the infrared thermography. A set of selected heat transfer data illustrates the Coriolis and rotating buoyancy effects on the detailed Nu distributions and the area-averaged heat transfer performances of the rotating channel. The Nu¯ for the developed flow region on the leading and trailing walls are parametrically analyzed to devise the empirical heat transfer correlations that permit the evaluation of the interdependent and individual Re, Ro, and Bu effect on Nu¯.

Author(s):  
S. W. Chang ◽  
T.-M. Liou ◽  
W. C. Chen

Detailed heat transfer distributions over two opposite leading and trailing walls roughened by spherical protrusions were measured from a rotating rectangular channel at rotation number up to 0.6 to examine the effects of Reynolds (Re), rotation (Ro) and buoyancy (Bu) numbers on local and area averaged Nusselt numbers (Nu and Nu) using the infrared thermography. A set of selected heat transfer data illustrates the Coriolis and rotating-buoyancy effects on the detailed Nu distributions and the area-averaged heat transfer performances of the rotating channel. The Nu for the developed flow region on the leading and trailing walls are parametrically analyzed to devise the empirical heat transfer correlations that permit the evaluation of the interdependent and individual Re, Ro and Bu effect on Nu.


2010 ◽  
Vol 132 (2) ◽  
Author(s):  
Shyy Woei Chang ◽  
Tong-Miin Liou ◽  
Tsun Lirng Yang ◽  
Guo Fang Hong

Endwall heat transfer measurements for a radially rotating rectangular pin-fin channel with the width-to-height ratio (aspect ratio) of 8 are performed at the parametric conditions of 5000≤Re≤20,000, 0≤Ro≤1.4, and 0.1≤Δρ/ρ≤0.21. Centerline heat transfer levels along the leading and trailing endwalls of the rotating pin-fin channel are, respectively, raised to 1.77–3.72 and 3.06–5.2 times of the Dittus–Boelter values. No previous attempt has examined the heat transfer performances for the pin-fin channel at such high rotation numbers. A selection of experimental data illustrates the individual and interactive Re, Ro, and buoyancy number (Bu) effects on heat transfer. Spanwise heat transfer variations between two adjoining pin rows are detected with the averaged Nusselt numbers (Nu) determined. A set of empirical equations that calculates Nu values over leading and trailing endwalls in the developed flow region is derived to correlate all the heat transfer data generated by this study and permits the evaluation of interactive and individual effects of Re, Ro, and Bu on Nu. With the aid of the Nu correlations derived, the operating conditions with the worst heat transfer scenarios for this rotating pin-fin channel are identified.


Author(s):  
S. W. Chang ◽  
T.-M. Liou ◽  
T.-H. Lee

This experimental study examines the pressure drop coefficients (f) and the detailed Nusselt numbers (Nu) distributions over two opposite leading and trailing walls roughened by 45° ribs for a rotating parallelogram channel with radially outward flow. For the first time the isolated effects of Reynolds (Re), rotation (Ro) and buoyancy (Bu) numbers on local and area averaged Nusselt numbers (Nu and Nu) measured from the infrared thermography method were successfully examined at the parametric conditions of 5000≤Re≤15000, 0≤Ro≤0.3 and 0.001≤Bu≤0.23 for the single-pass parallelogram channel. A set of selected heat transfer data illustrates the Coriolis and rotating-buoyancy effects on the detailed Nu distributions and the area-averaged heat transfer performances of the rotating parallelogram channel. With the consideration of the f data generated at the isothermal conditions, the thermal performance factors (η) for this radially rotating channel were evaluated. The Nusselt numbers obtained from the leading and trailing walls of the rotating test channel fall in the respective ranges of 0.78–1.34 and 1.09–1.38 times of the stationary levels; while the η factors are in the range of 0.979–1.575 for the present test conditions.


2003 ◽  
Vol 125 (4) ◽  
pp. 726-733 ◽  
Author(s):  
Peeyush Agarwal ◽  
Sumanta Acharya ◽  
D. E. Nikitopoulos

The paper presents an experimental study of heat/mass transfer coefficient in 1:4 rectangular channel with smooth or ribbed walls for Reynolds number in the range of 5000–40,000 and rotation numbers in the range of 0–0.12. Such passages are encountered close to the mid-chord sections of the turbine blade. Normal ribs (e/Dh=0.3125 and P/e=8) are placed on the leading and the trailing sides only. The experiments are conducted in a rotating two-pass coolant channel facility using the naphthalene sublimation technique. For purposes of comparison, selected measurements are also performed in a 1:1 cross section. The local mass-transfer data in the fully developed region is averaged to study the effect of the Reynolds and the rotation numbers. The spanwise mass transfer distributions in the smooth and the ribbed cases are also examined.


Author(s):  
S. W. Chang ◽  
T. L. Yang ◽  
G. F. Hong ◽  
Tong-Miin Liou

Endwall heat transfer measurements for a radially rotating rectangular pin-fin channel with the width-to-height ratio (aspect ratio) of 8 are performed at the parametric conditions of 5000≤Re≤20000, 0≤Ro≤1.4 and 0.1≤Δρ/ρ≤0.21. Centerline heat transfer levels along the leading and trailing endwalls of the rotating pin-fin channel are respectively raised to 1.77–3.72 and 3.06–5.2 times of the Dittus-Boelter values. No previous attempt has examined the heat transfer performances for the pin-fin channel at such high rotation numbers. A selection of experimental data illustrates the individual and interactive Re, Ro and buoyancy number (Bu) effects on heat transfer. Spanwise heat transfer variations between two adjoining pin rows are detected with the averaged Nusselt numbers (Nu) determined. A set of empirical equations that calculates Nu values over leading and trailing endwalls in the developed flow region is derived to correlate all the heat transfer data generated by this study and permits the evaluation of interactive and individual effects of Re, Ro and Bu on Nu. With the aid of the Nu correlations derived, the operating conditions with the worst heat transfer scenarios for this rotating pin-fin channel are identified.


Author(s):  
Ruquan You ◽  
Haiwang Li ◽  
Zhi Tao ◽  
Kuan Wei

An experiment investigation was conducted in a smooth cooling channel under rotating conditions on a new rotating facility. The heat transfer phenomenon of rotating channel with constant heat flux boundary conditions was measured using PIV and TLCs (Thermography Liquid Crystal). The square cross-section channel is manufactured from Plexiglas. Four 1 mm thick transparent ITO (Indium Tin Oxide) heater glasses taped on four Plexiglas walls independently to provide the same constant heat flux boundary conditions. The size of the heated channel is 600 mm*80 mm*80 mm. TLCs was used for measuring the two-dimensional detail distribution of the temperature on the leading surface and trailing surface of the channel. The pixel-level calibration method was used for reducing the error of the TLCs. In the experiments, the Reynolds number, based on the channel hydraulic diameter (D) of 80 mm and the bulk mean velocity, is 20000, and the rotation number are 0, 0.068, 0.135, 0.205, 0.273, respectively. The obtained result shows that rotation has an important different effects on the heat transfer of leading and trailing side. On the trailing side, the heat transfer increases with the rotation number monotonically along the stream wise (X direction), and rotation enhances the local Nu/Nus up to 20%. On the leading side, rotation weakens the Nu/Nus around 25% with the rotation number of 0.273. However, on the leading side, with the increase of rotation number, the Nu/Nus is not decreasing monotonically along the stream wise, there is a slight enhancement (turning point) along the stream wise with different rotation numbers. The position of turning point on the leading side is different with various rotation numbers. With higher rotation number, the turning point moves closer to the inlet of the channel. This phenomenon is an important reference to investigate the flow field in our future work with PIV. More details of two-dimensional distribution Nu on the leading and trailing side will be shown in this work.


Author(s):  
Jiang Lei ◽  
Je-Chin Han ◽  
Michael Huh

In this paper the effect of rib spacing on heat transfer in a rotating two-passage channel (AR=2:1) at orientation angle of 135° was studied. Parallel ribs were applied on leading and trailing walls of the rotating channel at the flow angle of 45°. The rib-height-to-hydraulic diameter ratio (e/Dh) was 0.098. The rib-pitch-to-rib-height (P/e) ratios studied were 5, 7.5, and 10. For each rib-spacing, tests were taken at five Reynolds numbers from 10,000 to 40,000 and for each Reynolds number, experiments were conducted at four rotational speeds up to 400 rpm. Results show that the heat transfer enhancement increases with decreasing P/e from 10 to 5 under non-rotation condition. However, the effect of rotation on the heat transfer enhancement remains about the same for varying P/e from 10 to 5. Heat transfer enhancement due to rotation can be correlated on all surfaces (leading, trailing, inner and outer walls and tip cap region) in the two-passage 2:1 aspect ratio channel.


Author(s):  
Michael Huh ◽  
Yao-Hsien Liu ◽  
Je-Chin Han ◽  
Sanjay Chopra

The focus of the current study was to determine the effects of rib spacing on heat transfer in rotating 1:4 AR channels. In the current study, heat transfer experiments were performed in a two-pass, 1:4 aspect ratio channel, with a sharp bend entrance. The channel leading and trailing walls in the first pass and second pass utilized angled rib turbulators (45° to the mainstream flow). The rib height-to-hydraulic diameter ratio (e/Dh) was held constant at 0.078. The channel was oriented 90° to the direction of rotation. Three rib pitch-to-rib height ratios (P/e) were studied: P/e = 2.5, 5, and 10. Each ratio was tested at five Reynolds numbers: 10K, 15K, 20K, 30K and 40K. For each Reynolds number, experiments were conducted at five rotational speeds: 0, 100, 200, 300, and 400 rpm. Results showed that the sharp bend entrance has a significant effect on the first pass heat transfer enhancement. In the second pass, the rib spacing and rotation effect are reduced. The P/e = 10 case had the highest heat transfer enhancement based on total area, whereas the P/e = 2.5 had the highest heat transfer enhancement based on the projected area. The current study has extended the range of the rotation number (Ro) and local buoyancy parameter (Box) for a ribbed 1:4 aspect ratio channel up to 0.65 and 1.5, respectively. Correlations for predicting heat transfer enhancement, due to rotation, in the ribbed (P/e = 2.5, 5, and 10) 1:4 aspect ratio channel, based on the extended range of the rotation number and buoyancy parameter, are presented in the paper.


2003 ◽  
Vol 125 (3) ◽  
pp. 575-584 ◽  
Author(s):  
P. M. Ligrani ◽  
G. I. Mahmood

Spatially resolved Nusselt numbers, spatially averaged Nusselt numbers, and friction factors are presented for a stationary channel with an aspect ratio of 4 and angled rib turbulators inclined at 45 deg with perpendicular orientations on two opposite surfaces. Results are given at different Reynolds numbers based on channel height from 10,000 to 83,700. The ratio of rib height to hydraulic diameter is .078, the rib pitch-to-height ratio is 10, and the blockage provided by the ribs is 25% of the channel cross-sectional area. Nusselt numbers are given both with and without three-dimensional conduction considered within the acrylic test surface. In both cases, spatially resolved local Nusselt numbers are highest on tops of the rib turbulators, with lower magnitudes on flat surfaces between the ribs, where regions of flow separation and shear layer reattachment have pronounced influences on local surface heat transfer behavior. The augmented local and spatially averaged Nusselt number ratios (rib turbulator Nusselt numbers normalized by values measured in a smooth channel) vary locally on the rib tops as Reynolds number increases. Nusselt number ratios decrease on the flat regions away from the ribs, especially at locations just downstream of the ribs, as Reynolds number increases. When adjusted to account for conduction along and within the test surface, Nusselt number ratios show different quantitative variations (with location along the test surface), compared to variations when no conduction is included. Changes include: (i) decreased local Nusselt number ratios along the central part of each rib top surface as heat transfer from the sides of each rib becomes larger, and (ii) Nusselt number ratio decreases near corners, where each rib joins the flat part of the test surface, especially on the downstream side of each rib. With no conduction along and within the test surface (and variable heat flux assumed into the air stream), globally-averaged Nusselt number ratios vary from 2.92 to 1.64 as Reynolds number increases from 10,000 to 83,700. Corresponding thermal performance parameters also decrease as Reynolds number increases over this range, with values in approximate agreement with data measured by other investigators in a square channel also with 45 deg oriented ribs.


Sign in / Sign up

Export Citation Format

Share Document