Numerical and Experimental Analysis of Turbulent Flow in Corrugated Pipes

2010 ◽  
Vol 132 (7) ◽  
Author(s):  
Henrique Stel ◽  
Rigoberto E. M. Morales ◽  
Admilson T. Franco ◽  
Silvio L. M. Junqueira ◽  
Raul H. Erthal ◽  
...  

This article describes a numerical and experimental investigation of turbulent flow in pipes with periodic “d-type” corrugations. Four geometric configurations of d-type corrugated surfaces with different groove heights and lengths are evaluated, and calculations for Reynolds numbers ranging from 5000 to 100,000 are performed. The numerical analysis is carried out using computational fluid dynamics, and two turbulence models are considered: the two-equation, low-Reynolds-number Chen–Kim k-ε turbulence model, for which several flow properties such as friction factor, Reynolds stress, and turbulence kinetic energy are computed, and the algebraic LVEL model, used only to compute the friction factors and a velocity magnitude profile for comparison. An experimental loop is designed to perform pressure-drop measurements of turbulent water flow in corrugated pipes for the different geometric configurations. Pressure-drop values are correlated with the friction factor to validate the numerical results. These show that, in general, the magnitudes of all the flow quantities analyzed increase near the corrugated wall and that this increase tends to be more significant for higher Reynolds numbers as well as for larger grooves. According to previous studies, these results may be related to enhanced momentum transfer between the groove and core flow as the Reynolds number and groove length increase. Numerical friction factors for both the Chen–Kim k-ε and LVEL turbulence models show good agreement with the experimental measurements.

Author(s):  
Sam Ghazi-Hesami ◽  
Dylan Wise ◽  
Keith Taylor ◽  
Peter Ireland ◽  
Étienne Robert

Abstract Turbulators are a promising avenue to enhance heat transfer in a wide variety of applications. An experimental and numerical investigation of heat transfer and pressure drop of a broken V (chevron) turbulator is presented at Reynolds numbers ranging from approximately 300,000 to 900,000 in a rectangular channel with an aspect ratio (width/height) of 1.29. The rib height is 3% of the channel hydraulic diameter while the rib spacing to rib height ratio is fixed at 10. Heat transfer measurements are performed on the flat surface between ribs using transient liquid crystal thermography. The experimental results reveal a significant increase of the heat transfer and friction factor of the ribbed surface compared to a smooth channel. Both parameters increase with Reynolds number, with a heat transfer enhancement ratio of up to 2.15 (relative to a smooth channel) and a friction factor ratio of up to 6.32 over the investigated Reynolds number range. Complementary CFD RANS (Reynolds-Averaged Navier-Stokes) simulations are performed with the κ-ω SST turbulence model in ANSYS Fluent® 17.1, and the numerical estimates are compared against the experimental data. The results reveal that the discrepancy between the experimentally measured area averaged Nusselt number and the numerical estimates increases from approximately 3% to 13% with increasing Reynolds number from 339,000 to 917,000. The numerical estimates indicate turbulators enhance heat transfer by interrupting the boundary layer as well as increasing near surface turbulent kinetic energy and mixing.


2000 ◽  
Author(s):  
Stephen E. Turner ◽  
Hongwei Sun ◽  
Mohammad Faghri ◽  
Otto J. Gregory

Abstract This paper presents an experimental investigation on nitrogen and helium flow through microchannels etched in silicon with hydraulic diameters between 10 and 40 microns, and Reynolds numbers ranging from 0.3 to 600. The objectives of this research are (1) to fabricate microchannels with uniform surface roughness and local pressure measurement; (2) to determine the friction factor within the locally fully developed region of the microchannel; and (3) to evaluate the effect of surface roughness on momentum transfer by comparison with smooth microchannels. The friction factor results are presented as the product of friction factor and Reynolds number plotted against Reynolds number. The following conclusions have been reached in the present investigation: (1) microchannels with uniform corrugated surfaces can be fabricated using standard photolithographic processes; and (2) surface features with low aspect ratios of height to width have little effect on the friction factor for laminar flow in microchannels.


2017 ◽  
Vol 139 (6) ◽  
Author(s):  
Ali Mohammadi ◽  
Ali Koşar

This study focuses on microheat sinks with different staggered arrangements of micro pin fins (MPFs). A rectangular microchannel with the dimensions of 5000 × 1500 × 100 μm3 (l′ × w′ × h′) was considered for all the configurations while different MPF diameters, height over diameter ratio (H/D), and longitudinal and transversal pitch ratios (SL/D and ST/D) were considered in different arrangements. Using the ansys fluent 14.5 commercial software, the simulations were done for different Reynolds numbers between 20 and 160. A constant heat flux of 30 W/cm2 was applied through the bottom heating section. The performances of the microheat sinks were evaluated using design parameters, namely pressure drop, friction factor, Nusselt number, and thermal-hydraulic performance index (TPI). The effect of each geometrical parameter as well as wake-pin fin interaction patterns were carefully studied using the streamline patterns and temperature profiles of each configuration. The results reveal a great dependency of trends in pressure drops and Nusselt numbers on the wake region lengths as well as the local velocity and pressure gradients. Moreover, the wake region lengths mostly contribute to the increase in obtained pressure drop and Nusselt number with Reynolds number. Although an increase in the H/D and SL/D ratios results in an increase and a decrease in pressure drop, respectively, the effect on the Nusselt number depends on other geometrical parameters and Reynolds number. A larger ST/D ratio generally results in a decrease in the pressure drop and Nusselt number. Finally, while the friction factor decreases with Reynolds number, two different trends are seen for the TPI values of configurations with the H/D ratio of 1 and 2 (D = 100 and 50 μm). While the trend in the TPIs is increasing for Reynolds numbers between 20 and 40, it reverses for higher Reynolds numbers with a steeper slope in the configurations with the ST/D ratio of 1.5.


2009 ◽  
Author(s):  
Shobeir Aliasghar Zadeh ◽  
Rolf Radespiel

Three-dimensional laminar and turbulent water flows in smooth and rough micro channels with rectangular cross-section were numerically simulated. The hydraulic diameter of the smooth micro channel is 190 μm and 191 μm for the rough one. The roughness inducing surfaces, which were modelled by three rectangular elements placed on the sidewall of the micro channel, are 50 μm high and 50 μm wide. The simulations were conducted for Reynolds numbers between 100 and 4000. The effects on the friction factor and flow characteristics due to the roughness elements, varying Reynolds numbers and low-Reynolds number turbulence models were investigated and compared with the experimental values reported by Hao et al. [1]. Furthermore, the velocity profiles in various Reynolds number and flow regimes have been compared with μPIV measurements. At Reynolds numbers less than 2100 the computed friction factors in the smooth micro channel agree well with the measurements and the values of the conventional theory. For the micro channel with roughness elements, the friction factor approaches the value of measurements and conventional theory, when Re < 900. Transition from laminar to turbulent flow occurs at about Reynolds numbers of 2100 and 900 in smooth and rough micro channel, respectively. Comparison of simulated results using the Spalart-Allmaras and SST K-ω turbulence models with experimental values show good agreement. By contrast, the K-ε model overestimates the pressure loss in micro channels.


Author(s):  
Yasuo Koizumi ◽  
Hiroyasu Ohtake ◽  
Hiroki Takahashi ◽  
Yoshiaki Ohno

The friction characteristics of water in a sub-millimeter scale channel were investigated experimentally. The friction factors and the critical Reynolds number were measured using water flow through circular tubes with diameters of 0.5, 0.25 and 0.17 mm. The experimental results show that the measured friction factor for water agreed well with the conventional Poiseuille (λ = 64/Re) and Blasius (λ = 0.316 Re−0.25) equations in laminar and turbulent flow regime; the laminar-turbulent transition Reynolds number was approximately 2300 for diameter 0.5 mm. For diameter 0.25 mm, the friction factor evaluated by the form pressure drop also agreed well with the Poiseuille equation. For diameter 0.17 mm, the measured total friction factor was close to the Poiseuille prediction.


1986 ◽  
Vol 108 (3) ◽  
pp. 507-512 ◽  
Author(s):  
H. Chegini ◽  
S. K. Chaturvedi

Friction factors for fully developed flow in triangular ducts with fins of various height and width are investigated for Reynolds numbers ranging from 150 to 90,000. Two triangular ducts having apex angles of 60 and 38.8 deg are studied. Results are presented in the form of standard plots of friction factor as a function of Reynolds number. Friction factor values for the smooth triangular duct cases are in good agreement with the existing results. For the finned-duct cases, the fully developed axial velocity profiles in laminar flow are determined by solving the x-momentum equation iteratively by the Gauss–Seidel finite-difference technique. The theoretically determined friction factors for these cases are in good agreement with the experimental values of friction factors based on pressure drop measurements.


2014 ◽  
Vol 18 (4) ◽  
pp. 1145-1158 ◽  
Author(s):  
Kamil Arslan

In this study, steady-state turbulent forced flow and heat transfer in a horizontal smooth semi-circular cross-sectioned duct was numerically investigated. The study was carried out in the turbulent flow condition where Reynolds numbers range from 1?104 to 5.5?104. Flow is hydrodynamically and thermally developing (simultaneously developing flow) under uniform surface heat flux with uniform peripheral wall heat flux (H2) boundary condition on the duct?s wall. A commercial CFD program, Ansys Fluent 12.1, with different turbulent models was used to carry out the numerical study. Different suitable turbulence models for fully turbulent flow (k-? Standard, k-? Realizable, k-? RNG, k-? Standard and k-? SST) were used in this study. The results have shown that as the Reynolds number increases Nusselt number increases but Darcy friction factor decreases. Based on the present numerical solutions, new engineering correlations were presented for the average Nusselt number and average Darcy friction factor. The numerical results for different turbulence models were compared with each other and similar experimental investigations carried out in the literature. It is obtained that, k-? Standard, k-? Realizable and k-? RNG turbulence models are the most suitable turbulence models for this investigation. Isovel contours of velocity magnitude and temperature distribution for different Reynolds numbers, turbulence models and axial stations in the duct were presented graphically. Also, local heat transfer coefficient and local Darcy friction factor as function of dimensionless position along the duct were obtained in this investigation.


Author(s):  
Levi A. Campbell ◽  
Satish Kandlikar

In studying the fluid flow and heat transfer in microchannels and minichannels, various claims have been made regarding transition at Reynolds numbers significantly below 2300. As a first step in identifying the reasons for such reports on early transition, the effect of entrance geometry on the pressure drop and transition to turbulence was studied in a conventional channel of 19 mm inside diameter (Kandlikar and Campbell [1]). As a second step, the effect of entrance condition on pressure drop and transition to turbulence is studied in small channels with diameters of 1.067 mm and 0.457 mm. The two entrance conditions employed for both channels are re-entrant and smooth. The experimental results show the effect of entrance condition on local friction factor, transition Reynolds number, and Hagenbach’s factor.


An apparatus was designed on the counter-flow system to study heat transfer between tube walls and gases at low temperatures in a region in which careful measurements had not previously been made. Oxygen, nitrogen and carbon dioxide were used, covering a temperature range from + 45° to –167° C, pressures up to 11 atm., and Reynolds numbers from 3000 to 60,000. Results were correlated by the use of dimensionless groups and a general equation ob­tained, independent of the nature of the gas and applicable over the whole range of experi­ments. With Reynolds numbers evaluated at mean film temperatures, the coefficient in the equation was found to be 5% lower than that obtained from measurements made at normal and high temperatures. This is regarded as justifying the extension of the ordinary equation to low-temperature regions. Determinations on friction accompanying heat transfer with gases in turbulent flow at low temperatures showed that the effect of heat transfer on the friction factor was small. Nomenclature C constant in Sutherland equation. D diameter of tube; equivalent diameter of annulus, i. e. internal diameter of outer tube minus external diameter of inner tube. F frictional force per lb. of fluid. L length of tube. T absolute temperature, ° K. V linear velocity of gas, as calculated from mass flow per unit time per unit of cross sectional area, divided by the mean density of the fluid. c specific heat of fluid at constant pressure. f friction factor, or coefficient of proportionality in pressure drop equation. g acceleration due to gravity. h coefficient of heat transfer between fluid and surface. k thermal conductivity of fluid. r, s constants (used as exponents). α, β constants. ϕ(x) function of x . μ absolute viscosity of fluid. ρ absolute density of fluid. Δp pressure drop in pipe. Subscripts a refers to annulus. i refers to inner tube. f refers to properties evaluated at film temperatures. Film temperature is taken as the arithmetic mean of the bulk fluid temperature and the tube-wall temperature.


Sign in / Sign up

Export Citation Format

Share Document