Numerical investigation of effect of different parameter on heat transfer for a crossflow heat excha

Abstract

Keywords: Computational Fluid Dynamics; Heat exchanger; Nanofluid; Unmanned Air Vehicles*

Introduction

Unmanned aerial vehicles (UAVs) are frequently used to collect data in the fields of real estate photography, mapping & surveying, agriculture, atmospheric studies and so on. Due to compact size of UAVs, thermal management holds a unique and important position UAV design. The

procedure of thermal management generally involves circulation of coolant which would collect heat from distributed parts in the UAV and transfer it either to the fuel or to a heat exchanger. The rugged environment and altitude of UAV along with high speed makes the heat transfer process

*Corresponding author. *E-mail address: mkilic@atu.edu.tr This paper was recommended for publication in revised frm by Regional Editor Tolga Taner Published by Yıldız Technical University Press, İstanbul, Turkey Copyright 2021, Yıldız Technical University. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

complicated which in turn has an effect on the temperature control of electronic equipment [1]. Thermal management of electronics in a UAV is generally achieved by forced convection of external air. However, with increased complexity, it becomes important that thermal management systems are carefully designed system based on air-moving devices [2]. A major bottleneck which poses problems in efficient heat transfer is low thermal conductivity of process fluids [3]. This constraint limits the compactness and effectiveness of heat exchangers. One way to enhance properties of process fluids is to prepare slurries of suspended particles. It is expected that the thermal conductivities of fluids having metallic, non-metallic or polymeric suspended particles will be higher than those of ordinary process fluids. A nanofluid is defined as a suspension of solid particles which have 1-100 nm size in a base fluid. These nano sized particles have much larger surface areas and thus have great potential for heat transfer enhancement. Interactions and collisions between particles cause to increase turbulence intensity in hydrodynamic boundary region. Turbulence intensity and large surface area enables more heat transfer. Nanoparticles carry 20% of their electrons at the surface that makes them ready to heat transfer. Another advantage of using nanofluids is the particle agitation which cause microvortexes to enlarge hydrodynamic boundary layer and decrease thickness of the thermal boundary layer to increase micro-convection between layers of fluid and heat transfer. There are some studies about investigation of thermal characteristics of nanofluids in the literature. Teamah et al. [4] studied on heat transfer and flow structure caused by the impact of Al2O3 nanofluid on a flat plate numerically and experimentally in different Reynolds numbers (Re = 3000-32000) and nanofluid volume ratios (φ=0-10%). As the nanoparticles increased in the fluid, it was observed that the heat transfer from the surface increased according to the situation where only water was used as a fluid, a 62% increase in the heat transfer coefficient could be achieved, and CuO was used as the fluid; It was observed that an increase of 8.9% in the case of using Al2O3 nanofluid and 12% in the case of using TiO2 nanofluid were observed. Manca et al. [5] investigated the effect of limited impinging jets on heat transfer from a flat plate with constant heat flux when pure water and water /Al2O3 nanofluid were used. Jet Reynolds number (Re= 100-400) and dimensionless channel height (H/W= 4-10) are the parameters used in the study. It was stated that as Reynolds number and particle concentration in fluid increased, the local heat transfer coefficient and Nusselt number increased. Chien et al [6] studied on the application of nanofluid in the flat plate heat pipe experimentally. It has been determined that the use of nanofluid can reduce thermal resistance by 40% compared to the use of pure water. Sun et al. [7] examined the effect of a single impinging jet on heat transfer using CuO nanofluid. It was determined that there was a significant increase in the heat transfer according to the use of water only when

nanofluid was used, there was no significant change in the pressure drop, and a higher heat transfer coefficient was obtained when the circular nozzle was used, and the highest heat transfer was obtained when the jet angle was 90°. Kang et al. [8] using silver nanoparticles and pure water in their experimental work with nanofluids; When using 10 nm nanoparticles according to the use of pure water, it was determined that the thermal resistance was reduced by 50% and when using nanoparticles with 35 nm diameter 80%. Shang et al. [9] examined the heat transfer properties of a closed-circuit vibrating heat pipe with Cu-water nanofluid. Compared to pure water, it was found that the heat transfer capacity of the system could be increased by 83% when this nanofluid was used. Umer et al. [10] in the study; Using CuO-H2O nanofluid, studied the heat transfer from a constant flowing surface under laminar flow conditions in different volumetric ratios. As a result, as the particle volume ratio increases and the Reynolds number increases, the heat transfer coefficient increases, and the highest increase in the heat transfer coefficient (61%) occurs when the particle volume ratio is 4% and the Reynolds number is Re = 605. Qu et al. [11] examined the thermal performance of the closed-circuit vibrating heat pipe in their experimental studies using Al2O3-water as a nanofluid. As a result, they found that the thermal resistance of the system decreased by 32.5% compared to the use of pure water. Kilic and Abdulvahitoglu [12] studied on numerical investigation of heat transfer with nanofluids and swirling jets in a vehicle radiator. As a base coolant Al2O3-H2O nanofluid was chosen for all parameters. It was found that increasing Reynolds number (Re) from 12000 to 21000 results in an increase of 51.3% in average Nusselt number (Nu). Using 1-jet causes an increase of 91.6% and 29.8% on average Nu number according to the channel flow and 2-jet. Using Cu-H2O nanofluid causes an increase of 3.6%, 7.6%, and 8.5% on the average Nu as compared to TiO2-H2O, Al2O3-H2O and pure water, respectively. Yan et al. [13] studied heat transfer for the case where the channel flow and jet flow were applied together. The study was experimental in nature and covered Re = 10 × 103 to 40 × 103 for channel flow and Re 50 x 103 to 20 × 103 for jet flow. Kilic and Ali [14] studied on numerical investigation of heat enhancement and fluid-flow from a heated surface by using nanofluids with three impinging jets. It was found that that increasing volume ratio of suspended particles from φ = 2% to 8% caused an increase of 10.4% on average Nu. Using Cu-water nanofluid resulted in an increase of 2.2%, 5.1%, 4.6%, and 9.6% in average Nu with respect to CuO-water, TiO2-water, Al2O3-water, and pure water respectively. The corresponding convective heat transfer coefficient and friction factor of nanofluids for nanoparticle weight concentrations of 0.025, 0.075, and 0.1% were evaluated in the study of Sadri et al [15]. Akdag et al. [16-18] investigated effect of change of flow characteristics as pulsating Flow conditions on heat transfer by using nanofluids. They obtained that heat transfer performance

considerably increased with increasing pulsating amplitude at low frequencies compared with that in steady flow. Nidal et al. [19] investigated; the effect of the amount of surfactant added to the base fluid on the thermal properties of the base fluid and nanofluid numerically. They obtained that the addition of surfactant molecules to the nanofluid increases the thermal conductivity of the structure from 0.741 W/m K to 0.783 W/m K. Vinot and Sachuthananthan [20] investigated effect of CuO/Water (0.3%), Al2O3/Water (0.3%) nanofluids and Al2O3-CuO/Water (0.3%) Hybrid nanofluid on heat transfer. They obtained that by using hybrid nanofluid, the heat transfer rate enhanced by 4.2% and 5.5% with pentagonal microchannel heat sink when compared to Cuo/ Water and Al2O3/Water nanofluids respectively. Tekir et al. [21] investigated forced convection heat transfer of Fe3O4/ water nanofluidflow in a straight pipe under constant and alternating magnetic field experimentally. They obtained that the constant magnetic field offers 13% convective heat transfer enhancement compared to the absence of a magnetic field. On the other hand, the alternating magnetic field increases the convective heat transfer in the pipe up to 35%. A brief summary of above-mentioned studies, shows the potential of nanofluids. Further, these studies can be classified into two distinct categories. One category focuses on preparation and characterization of nanofluids while the other on application of nanofluids for heat transfer enhancement. The present work can be considered as a contribution to the second category. Different form the literature in this study; cooling performance of a cross flow heat exchanger by using nanofluids in an unmanned air vehicle (UAV) was investigated for different parameters according to the four different scenarios to design new types of heat exchangers. For this purpose, internal flow of a crossflow heat exchangers was simulated for different parameters such as Re, different inlet velocity (Vairinlet) of cooling air, different inlet temperature for different attitude (Tairinlet) of cooling air and different types of nanofluids. So this manuscript is focused on to enhance heat transfer from inner tube by using nanofluids for different parameters. To obtain the effect of nanofluids on effectiveness of a cross flow heat exchanger barely, flow of the inner tube was analyzed. So no fin was used for the air side to enhance heat transfer. The effect of these parameters on heat transfer and heat exchanger effectiveness was investigated computationally.

Materials And Methods

In this study, a crossflow heat exchanger is modeled having dimensions of 300 × 37.5 × 22.5 mm (length × width × height respectively). The heat exchanger is shown in Figure 1. a. Hot fluid tube is made up of copper which has higher thermal conductivity for high performance of heat exchanger. The hydrodynamic diameter of the channels through which hot fluid passes is 7.5 mm. The k-ε

Figure 1.(c) Computational domain. Table 1. Fluid properties for different altitude Altitude

turbulence model available in ANSYS FLUENT® was used for present analysis. A part (two inner tubes) was chosen as a computational domain. Figure 1.b shows mesh structure of chosen part of heat exchanger. In the present study, 1.2 million cells were used in mesh structure. Mesh quality based on maximum skewness was 0.66. Number of inflation layers used were

10. Computational domain is shown in Figure 1c. Figure

2 shows different flight scenarios from the altitude point of view. This altitudes were chosen according to the tactical task scenarios. Fluid properties for different altitudes are shown in Table 1.

Theory

Major governing equations used during the simulation are presented (Cengel and Ghajar [22]) below. ∂ρ + div ( ρu) = 0 ∂t

The most dominant heat transfer mechanism in the current situation is convection. Convective heat transfer is governed by (Cengel and Ghajar [22]); Qconvention = h. A.∆T

Where h is the convective heat transfer coefficient, A is the surface area, ΔT (ΔT = Ts–Tbulk) is the difference between the measured surface temperature and the bulk fluid mean temperature. Nu is a dimensionless parameter indicating the ratio of heat transfer with convection to heat transfer with conduction (Cengel and Ghajar [22]),

Where Ts is the measured surface temperature, Dh is the hydraulic diameter of the channel, and knf is the coefficient of thermal conductivity of the nanofluid. Re number is used to determine for forced convection whether the flow is laminar or turbulent. Reynolds number based on turbulent flow (Cengel and Ghajar [22]); Re =

Where ρnf is the nanofluid density, Vjet is the jet velocity, and μnf is the nanofluid dynamic viscosity. The heat exchanger effectiveness ε is defined (Cengel and Ghajar [22]) as; Actual HeatTransfer Rate Q = ε =  Qmax MaximumPossible HeatTransfer Rate

In Eq. 4, the actual heat transfer rate Q̇ is calculated from an energy balance on the hot or cold fluids and is expressed (Cengel and Ghajar [22]) as;  pc . (Tcold ,out − Tcold ,in ) = mC  ph (Thot ,in − Thot ,out ) Q = mC

Where Cpc and Cph are heat capacity of cold and hot fluids. The maximum possible heat transfer rate (Cengel and Ghajar [22]) is;

Where Cmin is the smaller one of Cc and Ch. Thermal properties of nanofluids are calculated according to the equations which are given below. The density of nanofluids is calculated (Pak and Cho [23]) as;

Where ρbf is the base fluid (water) density, φ is the volume fraction of the solid particles in nanofluid, and ρp is the density of the solid particles in the nanofluid. The volumetric ratio of nanoparticles is defined (Pak and Cho [23]) as

Where ω is the density difference between the fluid and the main fluid (water). The nanofluid specific heat is calculated (Wang et al. [24]) as; C pnf =

Where Cp(p) is a specific heat of particle Cp(f) is the specific heat of the base fluid. The effective thermal conductivity of nanofluid (Corcione [25]) is; 10

 Tnf   k p  = 1 + 4.4 Re 0.4 Pr 0.66     kf  T fr   k f  keff

suspended nanoparticles, Tnf in the nanofluid temperature (K), Tfr is the freezing point of the base liquid. Nanoparticle Reynolds number is defined (Corcione [25]) as; Re =

Kb is the Boltzmann’s constant. The effective dynamic viscosity of nanofluids defined (Batchelor [26]) as;

Boundary conditions applied in numerical model were shown in Table 2. This study focused on enhancing heat transfer by using nanofluids for a cross flow heat exchanger of an Unmanned Air Vehicle. It was not mentioned the power need like pressure losses. It is assumed that cooling the electronic system and maintaining them is more important than power needs. Additionally, pressure losses of any fluid depends on viscosity of the fluid and friction factor caused by surface roughness. So viscosity of the fluid depends on volume ratio and particle diameter of solid particles for nanofluids. In this study, it is assumed that particle diameter and volume ratio of nanofluids were Dp=20 nm and φ=1%. So effects of solid particles on pressure losses of nanofluids can be assumed negligible and it can be assumed as one phase.

Validation

Where Re is the nanoparticle Reynolds number, Pr is the Prandtl number of the base liquid. kp is the nanoparticle thermal conductivity, φ is the volume fraction of the

In order to validate the CFD model, numerical results were compared with experimental results of Sadri et. al. [27] the highest difference between numerical results and experimental results is less than 10%. Figure 3 shows differences between numerical and experimental results. It was obtained that increasing Re number shows a decrease on

Figure 3. Comparison of numerical and experimental results. Figure 4. Average Nusselt number for different Reynolds numbers.

difference between numerical and experimental values. The reason of this change is that numerical model was modeled as one phase and increasing Re number (velocity of the fluid) causes a decrease of effect of micro vortexes which was caused by velocity of the solid particles. Secondly, numerical results were verified by using y+ value. This non-dimensional wall distance is defined (Minkowycz et al.[28] as; y+ =

Where z is the distance to the nearest wall, uT is the friction velocity at the nearest wall and γ is the local kinematic viscosity of the fluid. For verification of CFD model for heat transfer analysis, y+ value was compared. In this investigation, when y+ values were related to mesh elements for course, medium and fine mesh structure. The y+ values used in the current work are shown in Table 3. Fine mesh, which is used in this study, was seen as the higher performance of heat transfer than other mesh structures.

Results And Discussion

In this section, we will present the quantitative results for the effect of Re number, inlet velocity and type of nanofluid on Nu number.

Figure 5. Temperature contours for different Reynolds numbers.

Effect of Re number of hot fluid on Nu number Effect of inlet velocity of the hot fluid as Reynolds number for Cu-H2O nanofluid on heat transfer is investigated. It is obtained that increasing Reynolds number for Re=6000-8000 causes an increase of 24.99%, for Re=600010000 causes an increase of 37.56% and for Re=6000-12000 causes an increase of 44.65% on average Nusselt number. So increasing Reynolds number causes an increase on heat transfer. The reason of this is increasing Re number causes an increase on fluid velocity. This causes an increasing on thickness of hydrodynamic boundary and

decrease on thermal boundary layer. So this condition causes an increase on value of heat convection coefficient and decrease on thermal resistance. As a result heat transfer increases. Figure 4 and Figure 5 show variation of average Nusselt number and temperature contours for different Reynolds number. Increasing Reynolds number causes an increase on thermal boundary layer thickness. This causes a temperature variation (temperature decrease) on internal side of the inner tubes. Effect of Inlet Velocity on Nu number Effect of different inlet velocity (Vair,inlet = 30, 45, 60, and 90 m/s) for cooling air on heat exchanger effectiveness: effect of inlet velocity of cooling air on average Nusselt number of the heat exchanger at Re = 12000 and Thot,inlet = 90°C was numerically investigated. It is obtained that increasing velocity of cooling air three times (from Vair,inlet = 30 m/s to 90 m/s) causes only an increase of 6.96% on average Nusselt number of heat exchangers. The reason of this low increase is that heat capacity of air is too less than heat capacity of nanofluids. So fins should be used on air

side surface of inner tube to increase surface area of outside of inner tube. Figure 6 and Figure 7 variation of Nusselt number of heat exchanger and temperature contours for different inlet velocity of air.

Figure 6. Average Nusselt number for different air inlet velocity.

Figure 7. Variation of temperature contours for different air inlet velocity.

Effect of Altitude on Nu number In this section, effect of cooling air temperature at different on heat transfer is investigated for Re=12000 and Vair,inlet = 90m/s. It is obtained that increasing or decreasing air inlet temperature at different attitude does not cause any significant change on average Nusselt number. The reason of this, decreasing inlet temperature of air causes a decrease on temperature difference in thermal boundary layer. It also causes a decrease on local and average Nusselt number. So not only cooling performance of

crossflow heat exchanger but also performance of UAV do not change for different attitude. Figure 8 shows variation of average Nusselt number for different air inlet temperature.

Figure 8. Average Nusselt number for different air inlet temperature.

Figure 10. Variation of temperature for different nanofluids.

Effect of type of Nanofluid on Nu number Effect of nanofluid on heat transfer is investigated for different nanofluids at Re=12000 and Thot,inlet= 90°C. Particle diameter and volume ratio of nanofluids were assumed as Dp=20 nm and φ=1%. It was obtained that

Cu-H2O nanofluid shows the best heat transfer performance. Using Cu-H2O nanofluid causes an increase of 5.28%, 4.89% and 6.63% on average Nusselt number according to CuO-H2O and TiO2-H2O nanofluids and H2O. So higher thermal conductivity of nanofluid causes higher heat transfer. This result shows that the thermal conductivity of nanofluid is a significant parameter on enhancing heat transfer. Figure 9 and Figure 10 show variation of average Nusselt number and temperature contours for different type of nanofluids.

References

1. The article references can be accessed from the .pdf file.

Share and Cite