See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332016612

CFD ANALYSIS OF A WICKLESS HEAT PIPE

Conference Paper · January 2018

DOI: 10.1615/IHMTC-2017.2260

CITATIONS READS

0 381

3 authors, including:

Sagar Singhal Jeevan Jaidi

Birla Institute of Technology and Science Pilani BITS Pilani, Hyderabad
1 PUBLICATION   0 CITATIONS    19 PUBLICATIONS   66 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

It for lectures, only View project

All content following this page was uploaded by Jeevan Jaidi on 14 July 2019.

The user has requested enhancement of the downloaded file.

 th nd
Proceedings of the 24 National and 2 International ISHMT-ASTFE
Heat and Mass Transfer Conference (IHMTC-2017),
December 27-30, 2017, BITS Pilani, Hyderabad, India.

IHMTC2017-09-0865

CFD ANALYSIS OF A WICKLESS HEAT PIPE

2
Sagar Singhal
1 Abhishek Gaikwad
Student, Department of Mechanical Engineering Student, Department of Mechanical Engineering
BITS-Pilani, Hyderabad Campus,Telangana, India BITS-Pilani, Hyderabad Campus,Telangana, India
[email protected] [email protected]

3*
Jeevan Jaidi
Associate Professor, Dept. of Mechanical Engineering
BITS-Pilani, Hyderabad Campus,Telangana, India
*
[email protected]

ABSTRACT
Heat pipes are the devices capable of transporting heat NOMENCLATURE
efficiently without the use of any external power supply and ά Volume fraction of liquid
are used in a variety of thermal management applications, 𝐹𝑐𝑠𝑓 Continuum Surface Force (CSF)
such as satellite cooling systems, electronic cooling systems, 𝜌𝑙 Density of liquid
concentrated solar power plants, etc. A conventional heat pipe 𝜌𝑣 Density of vapor
uses wick structure to circulate the working fluid within it. C Surface curvature
Wickless heat pipe, also referred as thermosyphon, uses ⃗
𝑉 Velocity vector
gravity to circulate the working fluid within it. The μ Dynamic viscosity
advancements in this technology have led to the cost efficient p Pressure of the fluid
thermal systems and hence a number of experimental studies 𝑚21 Mass transfer from phase 2 to phase 1
have been done in this area to further improve these systems. 𝑚12 Mass transfer from phase 1 to phase 2
The present study investigates the thermal characteristics E Energy of the fluid
of a copper thermosyphon using Computational Fluid 𝑘𝑒𝑓𝑓 Effective thermal conductivity
Dynamics (CFD). Numerical simulations are performed for 𝑇𝑠𝑎𝑡 Saturated temperature
transient, two-phase flow and heat transfer using ANSYS 𝑇𝑙 Temperature of the liquid phase
Fluent CFD solver. Volume-of-Fluid (VOF) method is used to 𝑇𝑣 Temperature of the vapor phase
track the interfaces between the water vapor and liquid water. LH Latent heat of vaporization of the fluid
In a heat pipe, heat is supplied to an evaporator section T Average temperature of the fluid
resulting in vaporization of water. The water vapor moves 𝜎 Surface tension coefficient
towards the condensor section, where it condenses due to heat
removal from the heat pipe and finally the liquid water
circulates back to evaporator due to capillary action. In a 1. INTRODUCTION
thermosyphon, this capillary action is provided by gravity. Heat pipe is a passive device which possesses the ability to
The evaporation-condensation phenomenon in thermosyphon transfer heat efficiently from one point to another with
has been modelled using Lee model, available in ANSYS minimum drop in temperature [1]. In many cases, their
Fluent CFD software version 16.2. The temperature effective thermal conductivity would be about 100 times that
dependent density and surface tension coefficient of water are of copper indicating their effectiveness. Such favorable
given as inputs through user defined functions. The thermal characteristics are the primary reasons for extensive
simulations are performed for about 60 seconds with a time investigations in this area over the past few decades. Heat
step of 10-4 seconds. The predicted temperatures across the pipes have been used in a variety of thermal management
thermosyphon for a fill ratio (portion of evaporator section applications like aerospace systems, power plants, waste heat
filled with water) of 0.5 are validated with the available units, electronic cooling systems, etc. [2–4]. The heat pipe
experimental data and found a good match. Susbequently, the works on the principle of capillary action [5]. It consists of
effect of fill ratios (0.5-0.8) on the thermal characteristics of three sections; evaporator, condenser and an adiabatic section
the thermosyphon has been studied. An optimum fill ratio for a with a working fluid filled inside the pipe with caps at the
given heat flux is determined by comparing the thermal ends. The functioning of a simple heat pipe is as follows; input
resistance values and are discussed in detail. heat flux given to the evaporator heats the working fluid
leading to its vaporization. Fluid condenses as it moves
Keywords: Thermosyphon, Heat Pipe, Fill Ratio, Volume-of- towards the condenser and is circulated back via a wick
Fluid, Evaporation, Condensation. structure, which creates the capillary pressure necessary for
circulation. A special type of a heat pipe used in many and adiabatic lengths. The authors also noted that unsteady
thermal management applications is the thermosyphon - a violent boiling can occur at higher fill ratios, which can be
wickless heat pipe which works on the same principle as analyzed numerically.
followed by the conventional heat pipe. The only difference in Sukchana et al., [10] studied the influence of adiabatic
functioning is that the role of the wick structure is substituted lengths and fill ratios on the thermal efficiency of a heat pipe.
by gravity, and hence a relatively simple system. In many Experiments were conducted for different heat loads and tilt
cases, the cost can be reduced by relying on a wickless heat angles. Their results indicated that the thermal efficiency
pipe rather than a conventional one. Several input parameters increased with heat fluxes and fill ratios up to a certain point
relevant to the thermosyphon, namely the fill ratio, heat input beyond which a drop in thermal efficiency was found. A
and lengths of the three sections have been investigated by similar trend was found when the adiabatic length was varied.
many researchers. Fill ratio, defined as the ratio of initial The authors also noted that the fill ratio has a significant effect
volume of fluid inside the evaporator section to the volume of on the performance of a heat pipe whereas the influence of
evaporator, is a vital input parameter governs the performance adiabatic length is negligible.
of a thermosyphon. In overall, a number of researchers have From the above discussion, it can be noted that a good
investigated the influence of these parameters on the number of experimental works on thermosyphons are
performance of a thermosyphon and are discussed briefly available in open literature but numerical simulations are in
below. the developmental stages. CFD modelling provides the
Mozumder et al., [6] tested a miniature heat pipe with detailed information in contrast to experimental works and
three different working fluids namely acetone, water and hence many studies are being now conducted to understand
methanol. A number of experiments were conducted for the phase-change phenomenon.
different fill ratios (35%-85%) and at different heat inputs (2- Fadhl et al., [11] examined a copper-water thermosyphon
10 W). The experimental results indicated that the average using ANSYS Fluent. The Volume-of-fluid method was used
temperature of the heat pipe increased with load. Based on the along with the evaporation condensation equations in order to
thermal resistance calculations, they observed that the heat model the thermosyphon. Experiments were conducted for
pipe without working fluid exhibited higher thermal different heat inputs. The predicted temperatures were
resistance. Also, the heat transfer coefficient calculations compared with the experimental data and a found a good. The
indicated an increment with load for acetone and methanol as boiling behavior indicated formation of large bubbles in the
working fluids but a nearly constant value for water as the evaporator section. The thermal resistance calculations for
working fluid. The fill ratio had a negligible effect with these different heat inputs at a constant fill ratio of 0.5
methanol and water, whereas a temperature drop was observed indicated a decrement with heat input. The study was extended
with increasing fill ratio for acetone. Finally, they concluded by Fadhl et al., [12] with two refrigerants namely R-134a and
that the heat pipe with acetone out-performed the other two R-404a instead of water. The steady-state performance of the
working fluids in terms of the thermal characteristics. thermosyphon with these working fluids was examined and
The heat transfer characteristics of a thermosyphon were compared the predicted temperatures with the available
investigated experimentally by Noie [7]. Fill ratios (30%- experimental data. The results indicated that the pool boiling
90%), heat input (100-900W) and evaporator length were behavior of the refrigerants is very different from that of
varied for analyzing the performance of a copper-water water. The bubbles in the case of refrigerants were very small
thermosyphon. An isothermal temperature distribution was in size in contrast to the large bubbles sizes obtained with
obtained in the evaporator section for all the test cases water.
whereas low temperatures were observed in the condenser Another CFD study on a thermosyphon was conducted by
section due to the resistances between the two sections. He Alizadehdakhel et al. [13]. They investigated the influence of
found that the maximum heat transfer for each aspect ratio heat load and fill ratio on the performance of a thermosyphon.
took place at different fill ratios. A fill ratio of 90% was The 2-D simulation results indicated that increasing the heat
optimum for an aspect ratio of 7.45 whereas the optimum fill flux increased the thermosyphon performance up to a certain
ratio for an aspect ratio of 9.8 was 30%. extent after which a dip in performance was obtained. The
Another experimental investigation on thermosyphon was results also indicated an optimum fill ratio for each input
conducted by Jouhara et al., [8] . Four different working fluids value. The agreement obtained measured predicted is
were considered in their experimental analysis, three of them indicative of the fact that CFD can be used to model
being dielectric fluids while the fourth one was water. They thermosyphons.
analyzed the performance for two different fill ratios with Alammar et al., [14] numerically evaluated the influence
water whereas the performance of the dielectric fluids was of fill ratio and inclination angle on the performance of a
analyzed with a single fill ratio. The water charged thermosyphon. The results indicated that at low inclination
thermosyphon outperformed the thermosyphons with angles and low fill ratios, a significant rise was obtained in the
dielectric fluids for heat loads above 40 W but one of the evaporator temperature. They also noted that the influence of
dielectric fluids, FC-84 performed better for heat inputs less fill ratio and inclination angle was significant at higher heat
than 40 W. Although, water outperformed the dielectric fluids fluxes. The lowest thermal resistance was obtained at a fill
in most cases, they are better suited for electronic cooling ratio of 0.65 and an inclination angle of 90º.
applications with adequate thermal performance up to 40 W. Fadhl et al., [15] recently examined the geyser boiling
Jiao et al., [9] experimentally investigated the steady-state phenomenon inside a thermosyphon. 3-D simulations were
performance of a thermosyphon and proposed a theoretical performed with water and R-134a as the working fluids.
model to analyze the same. Their results indicated that fill Geyser boiling occurs due to the gradual heating of the
ratio must be bounded between CFR and EFR so that the working fluid in the evaporator section, which leads to bubble
thermosyphon functions effectively. Also, they noted the formation. The authors noted that the geyser boiling
geometrical parameters which influenced the values of CFR phenomenon did not occur for higher heat inputs.
and EFR. The parameters are directly proportional to the tube
diameter whereas they were inversely related to the condenser
The objective of this study is to simulate a copper-water flow problems. The advantage of VOF method is that it uses a
thermosyphon using ANSYS FLUENT CFD solver and study single set of momentum equations and tracks the interfaces
the effect of fill ratios on its performance. The temperature between the fluid phases. In VOF method, each cell should be
distribution for a fill ratio of 0.5 has been compared with filled with at least one fluid and accordingly three possible
available experimental data and found a good match. The fill conditions arise; cell woth ά=0 is occupied by vapor, cell with
ratio varied from 0.5 to 0.8. The overall thermal characteristics ά=1 is full of liquid and cell with 0<ά<1 consists of an
including the thermal resistance, temperature distribution, interface between water and vapor.
boiling behavior, etc., have been analysed at each of these fill
ratios. 2.1 Governing Equations
The governing equations involved with the VOF method
2. Mathematical and Numerical Modeling have been described below. Continuum Surface Force (CSF)
A schematic of the thermosyphon investigated in the model has been adopted to incorporate the effects of surface
current study is shown in Fig (1). The outer diameter is 22 mm tension and added as a source term in the momentum equation
with a thickness of 0.9 mm. All the geometric parameters [16].
including the lengths of different sections are given in Table
𝝆𝒍 ×𝜶́𝒍 ×𝑪𝒗 ×𝛁𝜶́𝒗 +𝝆𝒗 ×𝜶́𝒗 ×𝑪𝒍 ×𝛁𝜶́𝒍
(1). This particular setup has been tested rigorously in 𝑭𝒄𝒔𝒇 = 𝟐𝝈𝒍𝒗 (1)
𝝆𝒍 +𝝆𝒗
previous studies and hence used for the current analysis. The
working fluid considered is water and with copper being the
The vector momentum equation after considering the source
pipe material.
term takes the following form;

𝝏(𝝆𝑽⃗)
⃗ ⃗𝑽) = −𝛁𝒑 + 𝛁. [𝝁(𝛁𝑽
+ 𝛁. (𝝆𝑽 ⃗ + 𝛁𝑽
⃗ 𝑻 )] +
𝝏𝒕
⃗⃗ + ⃗𝑭𝒄𝒔𝒇
𝝆𝒈 (2)

A volume fraction equation used to track the interface is given

by,

́ )
𝟏 𝝏(𝜶𝝆
́ ⃗𝑽) = 𝑺 + ∑𝒏𝟏(𝒎𝟐𝟏 − 𝒎𝟏𝟐 )
+ 𝛁. (𝜶𝝆 (3)
𝝆 𝝏𝒕

Similar to the momentum equation, the energy equation is also

shared among the phases and is given by,

𝝏(𝝆𝑬)
⃗ (𝝆𝑬 + 𝒑)) = 𝛁. (𝒌𝒆𝒇𝒇 × ∆𝑻) + 𝑺𝒆
+ 𝛁. (𝑽 (4)
𝝏𝒕

where, 𝑺𝒆 is the source term associated with the evaporation

and condensation phenomenon. The mass transfer and energy
transfer during the evaporation and condensation phenomenon
are added in the form of source terms to the continuity and
energy equations, respectively. The equations used by
Schepper et al., [17] have been utilized to model this complex
phase-change phenomenon.

The mass transfer phenomenon is given by the following

vapour transport equation:

Figure 1. SCHEMATIC OF THERMOSYPHON. 𝝏(𝜶́𝒗 ×𝝆𝒗 )

+ 𝛁. (𝜶́𝝆𝒗 ⃗⃗⃗⃗
𝑽𝒗 ) = 𝒎̇𝒍𝒗 − 𝒎̇𝒗𝒍 (5)
𝝏𝒕

Table (1). DIMENSIONS OF THERMOSYPHON.

Depending on the cell temperature, the mass transfer is
Part Value (mm) defined as follows;
Pipe outer
22 𝑻𝒍 −𝑻𝒔𝒂𝒕
diameter ̇ = 𝟎. 𝟏 × 𝜶́𝒍 × 𝝆𝒍 ×
If 𝑻𝒍 > 𝑻𝒔𝒂𝒕 ; 𝒎𝒍𝒗 (6)
𝑻𝒔𝒂𝒕
Pipe thickness 0.9
Length of 𝑻𝒔𝒂𝒕 −𝑻𝒗
evaporator section
200 If 𝑻𝒗 < 𝑻𝒔𝒂𝒕 ; ̇ = 𝟎. 𝟏 × 𝜶́𝒗 × 𝝆𝒗 ×
𝒎𝒗𝒍 (7)
𝑻𝒔𝒂𝒕
Length of
100
adiabatic section The source term in the energy equation is obtained by
Length of multiplying the source term used in the mass transfer equation
200
condenser section with the latent heat of vaporization of the working fluid.

𝑻𝒍 −𝑻𝒔𝒂𝒕
ANSYS Fluent CFD solver coupled with the Volume-of-Fluid If 𝑻𝒍 > 𝑻𝒔𝒂𝒕 ; 𝑺𝒆 = 𝑳𝑯 × 𝟎. 𝟏 × 𝜶́𝒍 × 𝝆𝒍 × (8)
𝑻𝒔𝒂𝒕
(VOF) technique was used for modeling the two-phase flow
phenomenon. The VOF method is applicable for an Eulerian 𝑻𝒔𝒂𝒕 −𝑻𝒗
mesh (fixed mesh) and is widely used for solving multi-phase If 𝑻𝒗 < 𝑻𝒔𝒂𝒕 ; 𝑺𝒆 = 𝑳𝑯 × 𝟎. 𝟏 × 𝜶́𝒗 × 𝝆𝒗 × (9)
𝑻𝒔𝒂𝒕
 vaporization
Figure (2) shows a schematic of the mesh used. The mesh near Saturation temperature 373.15 K
the walls are fine, whereas away from the walls and towards Convergence criteria 10-3
the center are coarser. A total of 75,000 quad elements were Pressure velocity
used in the analysis. A mesh independence study indicated the SIMPLE
coupling
similar trends in temperature distribution beyond 75,000 Discretization schemes First Order Upwind
elements. Hence, the given mesh was chosen for all the test
cases. The simulations were done for a period of 60 seconds after
which a quasi steady-state solution was achieved. The
2.2 Boundary conditions performance of the thermosyphon was quantified in terms of
In order to simulate this complex two-phase flow thermal resistance, which is a measure of the overall heat
phenomenon, appropriate conditions have been assigned to the transfer between the evaporator and condenser section and is
boundaries. A constant heat flux of 12512.30 W/m2 was inversely related to the heat exchange. The mathematical
assigned to the evaporator section, whereas the condenser expression for the thermal resistance is given by the following
section was assigned with a convective heat transfer [11];
coefficient of 707.6 W/m2-K and a free stream temperature of
301.45 K. The adiabatic section and the end walls were Thermal resistance=
assigned a zero heat flux value. These particular boundary
conditions were assigned based on the experimental data 𝑨𝒗𝒈. 𝒕𝒆𝒎𝒑. 𝒐𝒇 𝒆𝒗𝒂𝒑. 𝒔𝒆𝒄𝒕𝒊𝒐𝒏 − 𝑨𝒗𝒈. 𝒕𝒆𝒎𝒑. 𝒐𝒇 𝒄𝒐𝒏𝒅𝒆𝒏𝒔𝒆𝒓 𝒔𝒆𝒄𝒕𝒊𝒐𝒏
provided by [11]. 𝑯𝒆𝒂𝒕 𝒊𝒏𝒑𝒖𝒕

3. Results and Discussions

In this section, the thermal characteristics obtained from
the CFD simulations are discussed. The temperature
distribution, boiling phenomenon and thermal resistance have
been studied at the different fill ratios. In order to validate the
CFD simulations, the geometry and boundary conditions were
chosen similar to that considered by Fadhl et al. [11]. The
steady-state temperature obtained after a time period of 60
seconds at a fill ratio of 0.5 has been compared (see Fig. 3)
with the variation obtained by Fadhl et al., [11] and the match
found to be good with a maximum variation of about 3% in
average temperature for the adiabatic section for FR=0.5.
Figure 2. MESH OF THERMOSYPHON
(COMPUTATIONAL DOMAIN).
390 Present simulation
Fadhl (2013)
375
The density of liquid is assumed to be a function of
Temperature (°C)

temperature and accordingly an equation based on the steam 360

table, has been used to model the same;
345
𝝆𝒍 = 𝟖𝟓𝟗. 𝟎𝟎𝟖𝟑 + 𝟏. 𝟐𝟓𝟐𝟐𝟎𝟗 × 𝑻 − 330
𝟎. 𝟎𝟎𝟐𝟔𝟒𝟐𝟗 × 𝑻𝟐 (10)
315
where, 𝑻 is the absolute temperature shared among the two 300
phases of water. 0 100 200 300 400 500
Length (mm)
In order to capture the effect of surface tension, the surface Figure 3. TEMPERATURE VARIATION IN
tension coefficient is developed again based on the steam table THERMOSYPHON FOR 0.5 FILL RATIO.
and is given by the following:

𝝈 = 𝟎. 𝟎𝟗𝟖𝟎𝟓𝟖𝟓𝟔 − 𝟏. 𝟖𝟒𝟓 × 𝟏𝟎−𝟓 × 𝑻 − Figure (4) shows the distribution of temperature vs. time for a
fill ratio of 0.5. The solution is initialized with saturation
𝟐. 𝟑 × 𝟏𝟎−𝟕 × 𝑻𝟐 (11)
conditions and hence the entire fluid present inside the
thermosyphon is found to possess a temperature close to
The fill ratio was varied from 0.5 to 0.8 in increments of 0.1.
saturation conditions during initial phases.
The phase-change phenomenon is a time dependent process
and hence a transient simulation was performed. The time step
size, convergence criterion and other relevant parameters used
in the analysis are given in Table (2).

Table (2). PARAMETERS USED IN THE NUMERICAL ANALYSIS.

Parameter Description
Flow type Laminar
Time step 0.0001 sec
Latent heat of 2455 KJ/kg
 Time (s) 0.1 0.5 1 5 10 60 Fill ratio 0.5 0.6 0.7 0.8

Figure 6. STEADY-STATE TEMPERATURE

Figure 4. TEMPERATURE DISTRIBUTION WITH
DISTRIBUTION FOR DIFFERENT FILL RATIOS.
TIME FOR 0.5 FILL RATIO.

Figures (7, 8) show the volume fraction contours

The influence of fill ratio on the temperatures of the
corresponding to this boiling phenomenon. Figure 7 presents
thermosyphon has been presented below. Figure (5) shows the
the volume fraction for fill ratio 0.5 at different time intervals.
average temperature variation along the length of the
The charged liquid pool is heated by the input heat flux
thermosyphon and for different fill ratios. Figure (6) shows the
resulting in formation of bubbles. The evaporation of the
steady-state temperature distribution for all the considered fill
liquid leads to an increase in the vapour content and hence
ratios. It can be clearly observed from Figs (5-6) that the
volume fraction of vapour increases, as shown in Figs (7, 8).
temperatures in the evaporator region is nearly isothermal and
Figure (8) presents the volume fraction distribution for all the
hardly affected by the change in volume of the charged liquid,
considered fill ratios. The figures indicate that the boiling
whereas the average temperature in the condenser region is
phenomenon is affected by the volume of charged liquid
affected by the fill ratio. In practical applications, the
present initially. Although, the size of the bubbles for all the
thermosyphon is bound to lose temperature in the adiabatic
fill ratios is comparable, the rise is different due to the
region but these losses are difficult to model and hence not
different initial heights and hence, the heat transfer would be
been considered in the current simulations. Thus, a smooth
different for each case. The temperature distribution is a clear
transition was obtained in temperature curves in the adiabatic
indication of the different heat transfer behaviour.
region for all the cases considered cases.

390
380 Fill ratio = 0.5
Fill ratio = 0.6
370 Fill ratio = 0.7
Temperature (°C)

360 Fill ratio = 0.8

350
340
330
320
310
0 100 200 300 400 500 600
Length (mm)

Figure 5. AVERAGE TEMPERATURE

VARIATION FOR DIFFERENT FILL RATIOS.

Time (s) 0.1 0.5 1 5 10 60

Figure 7. VOLUME FRACTION DISTRIBUTION

FOR 0.5 FILL RATIO.
 that the optimum fill ratio is 0.7. The temperature distribution,
bubble formation and velocity vectors were captured at
different time intervals for all the test cases. The simulation
data indicates that CFD can be a useful tool to understand the
complex phase-change phenomenon within the thermosyphon.
Suggestions for further works would be to analyze the
same test cases with 3-D simulations. A 3-D analysis is
expected to provide better outcomes but would be
computationally expensive. Such an analysis would also
indicate whether a 3-D analysis is essential or a 2-D analysis is
able to provide sufficient information about the thermosyphon.

5. REFERENCES
[1] Reay, D.A., Kew, P.A., and McGlen, R.J., 2014.
Introduction-Heat Pipes, 6th edition, in: Butterworth-
Heinemann, Oxford, 2014: pp. xxi–xxvii.
doi:https://doi.org/10.1016/B978-0-08-098266-3.00012-1.
[2] Faghri, A., 2014. “Heat Pipes: Review, Opportunities and
Fill ratio 0.5 0.6 0.7 0.8 Challenges,” Frontiers in Heat Pipes, vol. 5.
doi:10.5098/fhp.5.1.
Figure 8. STEADY-STATE VOLUME FRACTION [3] Sharifi, N., Bergman, T. L., Allen, M. J., and Faghri, A.,
DISTRIBUTION FOR DIFFERENT FILL RATIOS. 2014. “Melting and solidification enhancement using a
combined heat pipe, foil approach,” Int. J. Heat Mass
Transf., vol. 78, pp. 930–941.
The performance of this thermosyphon was quantified using doi:10.1016/j.ijheatmasstransfer.2014.07.054.
the thermal resistance concept, as described above. A [4] Robak, C. W., Bergman, T. L., and Faghri, A., 2011.
thermosyphon would function effectively only if its thermal “Enhancement of latent heat energy storage using
resistance is low. At lower fill ratios, a dry out is possible due embedded heat pipes,” Int. J. Heat Mass Transf., vol. 54,
to the low liquid charge, whereas at very high fill ratios, an pp. 3476–3484.
abrupt burst of bubbles is possible leading to unstable doi:10.1016/j.ijheatmasstransfer.2011.03.038.
functioning of the working fluid [18]. Therefore an optimum [5] Mozumder, K., Akon, A. F., Chowdhury, M. S. H., and
fill ratio is necessary for smooth functioning of the Banik, S. C., 2011. “Performance of Heat Pipe for
thermosyphon. Figure (9) shows the variation of thermal Different Working Fluids and Fill Ratios,” J. Mech. Eng.,
resistance with different fill ratios. vol. 41, pp. 96–102. doi:10.3329/jme.v41i2.7473.
It is clear from Fig (9) that for this particular [6] Noie, S. H., 2005. “Heat transfer characteristics of a two-
thermosyphon, a fill ratio of 0.7 is optimum at the given phase closed thermosyphon,” Appl. Therm. Eng., vol. 25,
loading conditions. Previous results have indicated that pp. 495–506. doi:10.1016/j.applthermaleng.2004.06.019.
thermal resistance is dependent not only on fill ratio but on [7] Jouhara, H., and Robinson, A. J., 2010. “Experimental
heat flux as well. investigation of small diameter two-phase closed
thermosyphons charged with water, FC-84, FC-77 and
FC-3283,” Appl. Therm. Eng., vol. 30, pp. 201–211.
0.3 doi:10.1016/j.applthermaleng.2009.08.007.
Thermal resistance (K/W) [8] Jiao, B., Qiu, L. M., Zhang, X. B., and Zhang, Y., 2008.
Thermal Resistance (K/W)

0.29 “Investigation on the effect of filling ratio on the steady-

state heat transfer performance of a vertical two-phase
0.28 closed thermosyphon,” Appl. Therm. Eng., vol. 28, pp.
1417–1426. doi:10.1016/j.applthermaleng.2007.09.009.
0.27 [9] Sukchana, T., and Jaiboonma, C., 2013. “Effect of Filling
Ratios and Adiabatic Length on Thermal Efficiency of
0.26 Long Heat Pipe Filled with R-134a,” Energy Procedia,
vol. 34, pp. 298–306. doi:10.1016/j.egypro.2013.06.758.
0.25 [10] Fadhl, B., Wrobel, L. C., and Jouhara, H., 2013.
0.4 0.5 0.6 0.7 0.8 0.9 “Numerical modelling of the temperature distribution in a
Fill Ratio two-phase closed thermosyphon,” Appl. Therm. Eng., vol.
Figure 9. VARIATION OF THERMAL 60, pp. 122–131.
RESISTANCE WITH FILL RATIOS. doi:10.1016/j.applthermaleng.2013.06.044.
[11] Fadhl, B., Wrobel, L. C., and Jouhara, H., 2015. “CFD
modelling of a two-phase closed thermosyphon charged
4. Conclusions and Scope of future Work
with R134a and R404a,” Appl. Therm. Eng., vol. 78, pp.
The effect of four different fill ratios (0.5 to 0.8) on the 482–490. doi:10.1016/j.applthermaleng.2014.12.062.
performance of a thermosyphon was studied numerically using [12] Alizadehdakhel, A., Rahimi, M., and Alsairafi, A. A.,
ANSYS Fluent CFD tool. A good match was found between 2010. “CFD modeling of flow and heat transfer in a
the predicted temperatures and previous experimental data. thermosyphon,” Int. Commun. Heat Mass Transf., vol. 37,
The thermal resistance initially decreased up to a fill ratio of pp. 312–318.
0.7 and then increased with increasing fill ratio, which means doi:10.1016/j.icheatmasstransfer.2009.09.002.
[13] Alammar, A. A., Al-Dadah, R. K., and Mahmoud, S. M.,
2016. “Numerical investigation of effect of fill ratio and
inclination angle on a thermosiphon heat pipe thermal
performance,” Applied Thermal Engineering, vol. 108,
pp. 1055–1065.
doi:10.1016/j.applthermaleng.2016.07.163.
[14] Jouhara, H., Fadhl, B., and Wrobel, L. C., 2015. “Three-
dimensional CFD simulation of geyser boiling in a two-
phase closed thermosyphon,” Int. Journal of Hydrogen
Energy, vol. 44, pp. 1–14.
doi:10.1016/j.ijhydene.2016.02.038.
[15] ANSYS Fluent Theory Guide.
[16] Schepper, S. C. K, Heynderickx, G. J., and Marin, G. B.,
2009. “Modeling the evaporation of a hydrocarbon
feedstock in the convection section of a steam cracker,”
Computers and Chemical Engineering, vol. 33, pp. 122-
132.
[17] Imura, H., Sasaguchi, K. S., and Kozai, H., 1983.
“Critical heat flux in a closed two phase thermosyphon,”
Int. J. Heat Mass Transfer, vol. 26 (8), pp. 1181–1188.

View publication stats