DEVELOPMENT OF CLOSED LOOP

PULSATING HEAT PIPES FOR APPLICATION

AT ELEVATED EVAPORATOR TEMPERATURE

Synopsis Report
Submitted

by
Himanshu Poonia (12ME92R03)

Under the supervision

of

Prof. P.K. Das Dr. A.K. Das

Department of Mechanical Engineering Department of Mechanical and Industrial
IIT Kharagpur Engineering, IIT Roorkee

DEPARTMENT OF MECHANICAL ENGINEERING

INDIAN INSTITUTE OF TECHNOLOGY, KHARAGPUR
1. Introduction

Pulsating heat pipe (PHP) is a closed two-phase heat transfer device in which heat is
transported from the evaporator section to the condenser section due to oscillating motion of
the working fluid (Akachi, 1990). In a typical closed loop PHP (CLPHP), a single capillary
tube is bent in serpentine manner into several interconnected turns between an evaporator and
a condenser. The tube is first evacuated and then filled partially with a working fluid. Due to
vacuum inside the capillary sized tube, the working fluid flashes into vapor and distributes to
produce liquid slug and vapor plug throughout the loop. Upon receiving heat at the
evaporator end, the slug-plug distribution oscillates and transports the heat to the condenser.
Thus the PHP starts its operation and its structure facilitates the oscillating and circulating
nature of different two-phase flow regimes.

Important factors that affect the operation and the performance of a PHP can be broadly
categorized into – design and operational parameters. Important design parameters
considered for the device are – configuration of the PHP, tube diameter (D), working fluid,
number of turns (N), size of evaporator and condenser. At the same time, important
parameters during the operation of the device are – capacity of the source and sink, filling
ratio (FR) and orientation of the device (Zhang and Faghri, 2008; Khandekar, 2002). At high
heat flux, PHP with large number of turns crosses the gravitational barrier and orientation
independent performance is reported (Akachi, 1990, 1993). Additionally, due to light weight
and simple construction, PHP is a suitable candidate for energy transfer where high reliability
is a requirement. Accordingly, PHPs can be utilized in space technology and in nuclear power
systems.

So far PHPs have been developed for a relatively low temperature of the evaporator. The
present work aims to enhance the range of temperature for which the device is operational
keeping in mind its potential application in the field of nuclear engineering.

2. Objectives and Scope of the Present Work

Keeping the target in mind, a thorough literature review has been performed and lacuna has
been observed. Based on the same, the major objectives of this dissertation are decided as:

1. Selection of working fluid for high temperature operation of the PHP.

2. Characterization of fluid flow and thermal behavior in a high temperature withstanding

three-turn CLPHP.

3. Study the performance of the CLPHP to scale up to 14-turn and 30-turn designs.

3. Selection of working fluid for high temperature operation of the PHP

After a thorough literature survey on the CLPHP, it is contemplated that high temperature
application can be achieved by investigating working fluids of higher boiling point and/or by
suitable design modifications. For the purpose, different suitable high temperature heat
transfer fluids are tested considering the thermo-physical properties, experimental constraints

1
and availability. In the feasibility test, at first a preliminary experimental investigation is
carried out to visualize the fluid flow behaviour using high speed camera. Secondly,
temperature signals at three different sections (evaporator, condenser and adiabatic zone) of
the PHP are studied. Finally, the fluid is analysed over the important dimensionless numbers
of the PHP.

Preliminary experiments are conducted in CLPHP with limited number of turns. Two
identical CLPHP set-ups are made, one from quartz tube and another from copper tube of
capillary dimension in accordance with design criteria (White and Beardmore, 1962). Two
standard fluids water and ethanol are tested for benchmarking and novel fluids– KoolGard
(commercial engine coolant), Olive Oil, Therminol-66 (Hydrogenated terphenyl) and
Dowtherm-A (eutectic mixture of biphenyl and biphenyl oxide) are tested thereafter. For
preliminary experiments FR is kept as 60% by volume of the total internal volume of the PHP
and orientation of the PHP is maintained as vertical (evaporator at the bottom and condenser
at the top). Heating at the evaporator section is done by either maintaining isothermal
conditions or constant heat flux conditions. At the condenser section, isothermal conditions
are attempted to maintain. With a certain temperature differential between the two ends, PHP
starts operation after crossing the threshold thermal energy barrier.

(a) Water (b) Dowtherm-A

 Increasing heat flux   Increasing heat flux 

Figure 1:Flow regimes with increasing heat flux for (a) Water (0–150W) and (b) Dowtherm - A
(furnace temperature, FT 50–400°C)

Figure 1(a) shows different flow regimes of fluid water captured at the adiabatic section of
the PHP. Stable initial distribution of the liquid slug and vapour plug after charging of the
PHP with fluid is observed (first image of Figure 1a). Further in line are shown the bubbly
flow, slug flow, transition to annular and annular flow, as observed when PHP is subjected to
varying heat input (0–150 W) at the evaporator section. On the application of heat, during

2
start up, movement of the slug-plug train is observed at the evaporator section. At the
adiabatic section small amplitude oscillation are observed limited to lower half of the PHP.
With an increase in heat input, the pressure of the vapour plug increases and rises above the
saturation conditions; nucleation starts in liquid slug at the evaporator section. In this zone,
due to nucleation and bubble coalescence, large amplitude oscillations comparable to length
of PHP are seen. Dominance of oscillatory flow is seen at this heat flux range. Further at high
heat input, PHP develops enough pressure differentials across its two ends that results in flow
behaviour to bulk flow circulation. Direction of circulation is arbitrary depending upon the
existing conditions inside the system at the time and may change accordingly. At the exit of
the evaporator, dominance of vapour (annular flow) and at the inlet, dominance of liquid
(bubbly flow) is observed. Figure 1(b) shows different flow regimes observed with
Dowtherm-A as the working fluid. Similar to water, liquid slugs and vapour plugs of different
size randomly distribute throughout the PHP loop during charging of Dowtherm-A. Initial
slug-plug distribution is observed to be stable in Dowtherm-A. On the application of heat,
rise in vapour pressure and phase change at the evaporator section leads to low amplitude
oscillations. An increase in heat input above the saturation conditions enhance the pressure
substantially resulting into further high amplitude oscillations. Interplay between the different
driving and restoring forces in the PHP, results in different thermo-physical phenomena
occurring inside the system, before a thrust coming from the evaporator section takes the
Dowtherm-A inventory across the different sections and even to the adjacent channels. A jet
of heated vapours coming out of the evaporator section provides the required thrust and flow
exchange among different sections and adjacent channels of the PHP are observed.
Oscillating flow and circulation with intermittent flow reversal is observed in PHP with
Dowtherm-A. Slug-plug flow and bubbly flow regimes are found dominant in this operating
zone. Further, at high heat input, transition to annular and annular flow with intermittent jet
of superheated vapours is observed at the exit of the evaporator. Here, while working with
Dowtherm-A, dominance of bulk flow circulation is observed. Upon comparison of Figure
1(a) and 1(b), phenomenal difference in behaviour and flow regimes is depicted by different
fluids, which could be attributed to the difference in thermo-physical properties of water and
Dowtherm-A, in transferring heat.

In order to understand the heat transfer characteristics, temperature variations against the heat
input are considered to study at the three sections (evaporator, adiabatic and condenser) of the
PHP. Temperature histories of the thermocouples attached to the PHP charged with ethanol,
another benchmarking fluid like water, are plotted in Figure 2(a). With heat input, initially
temperature of evaporator section rises following the trend of furnace temperature (FT).
Crossing the threshold barrier, operation of the PHP starts when FT is around 150°C. At this
point evaporator temperature starts fluctuating but maintains a lower mean temperature
(75°C). At this stage, temperature signal oscillations observed at all the three sections of the
PHP charged with ethanol signifies fluid movement and heat transfer. With the increase in
heat input (FT 225°C) to PHP charged with ethanol, no marginal rise in the average
evaporator temperature is observed but increase infrequency of oscillation is observed. On
the contrary, Figure 2(b) shows the variation of temperature of the three sections of the PHP
with fluid Dowtherm-A. As it is a suitable candidate for high temperature withstanding PHP

3
fluid, here, the heat input is given in steps, till FT reached 560°C. It is seen that the
temperature of all the sections follows the same trend as that of the FT, until FT crosses by
300°C and the temperature of evaporator reaches 180°C, approximately. At this point
fluctuations in temperature signal at the evaporator section are observed and thereafter high
peaks and low trough are seen throughout the PHP charged with Dowtherm-A with no
substantial increase in the temperature of the evaporator section, though the heat input is
provided till FT reaches 560°C.

(a) Ethanol (b) Dowtherm-A

Figure 2:Time variant behaviour of the three sections of the PHP against heat input (FT) for (a)
Ethanol and (b) Dowtherm-A

Considering non-dimensional analysis, important dimensionless numbers ( Bo , Ja , Pr and

Ku ), their range of variation and inter-relation is obtained by considering independent
variables as diameter and operating temperature. The obtained range of the dimensionless
numbers for different working fluids provides information of the characteristic dominating
forces and their effect on the performance of the device. For the fluids water, ethanol and
Dowtherm-A, saturated conditions at the evaporator section are considered over the
temperatures range of 50-110oC, 50-90oC and 180-300oC, respectively.

Out of other non-dimensional numbers, Kutateladge number ( Ku ) is the most important as it

represents a balance between the dynamic head, surface tension, and gravitational force. It
can be defined as:
Ku  V g 0.5 ( g ( l   g ))0.25 (1)

On the other hand, Charoensawan and Terdtoon (2008) used the Ku to non-dimensionalize
the heat transfer in context of boiling, and represented it as the ratio between output heat flux
to critical heat flux as:
qc
Ku  (2)
h fg  g ( g ( l   g ))0.25
0.5

As it is important to operate the PHP device below the critical heat flux point to avoid dry out
phenomenon, input heat flux should be less than the critical heat flux. To maintain that, by
equalizing equations (1) and (2), one can get the maximum heat flux transfer criterion.

4
Following the same, Figure 3(a), shows the variation of Ku over the temperature range for
the three fluids. High value of Ku represents that the PHP device is transferring high heat
fluxes, and presents high performance towards the critical limit. Figure 3(b), shows the heat
transfer variation over the operated temperature range for different fluids. For a fixed loop
diameter, corresponding heat flux shows the safe operational range of the PHP for the fluid
considered i.e. water, ethanol and Dowtherm-A. This also establishes Dowtherm-A as an
efficient high temperature withstanding PHP fluid.
(a) (b)

Figure 3: Plot of variation of (a) Ku and (b) heat flux against temperature for the three fluids over the
respective operational range

4. Characterization of fluid flow and thermal behavior in a three-turn CLPHP

After selecting Dowtherm-A as the possible high temperature PHP fluid, detailed
experiments are aimed to study the flow behavior and heat transfer characteristic of water
(benchmark fluid) and Dowtherm-A in a three-turn CLPHP. Variation of important
operational parameters, mainly heat input, filling ratio and inclination angle are considered
and their effect on the thermal performance of the device is investigated. Effect of inclination
is investigated for two different axes of rotation (θ and β; shown in a later figure). For the
study, a three-turn CLPHP structure is made of copper tube of ID 2mm. K-type
thermocouples are used for temperature measurement. Heating in steps at evaporator section
is provided by a furnace and constant conditions at inlet to the condenser are maintained by
circulating water from a cooling bath. Heat transferred by the PHP at the condenser section is
calculated by calorimetric method.

4.1 Observations with water as working fluid

4.1.1 Effect of heat input on fluid flow behavior and heat transfer characteristics

Effect of heat input is first explained from the assessment of fluid flow behavior by making
conclusions from temperature signals. Following this, the heat transfer characteristics are
derived from experimental data of average temperatures and heat flux. Considering a
representative experimental run for water with FR 50%, 75° inclination (in θ-plane), the
characteristics of PHP is explained. The PHP starts its operation when FT 100°C. At this

5
condition, temperature at evaporator section shows fluctuations and drops to maintain low
temperature compared to FT. Increase in FT to 150°C, crossing the threshold range brings the
PHP in safe operational domain, and the pressure difference between evaporator and
condenser is enough to transfer the fluid across different sections of the PHP. Increase in
frequency and decrease in amplitude of temperature oscillations is observed. With the
increase in heat input (FT 200 – 250°C), a slight increase in evaporator temperature is
observed. Increase in amplitude of oscillations and tendency to bulk flow circulation is also
seen. In Figure 4(b), plot of heat transfer offered by the device for the operating condition is
shown. During start-up conditions (FT 100°C) low heat transfer is observed. With the
increase in heat input (FT 150–250°C) device enters the stable operational domain and a
significant increase in heat transfer with a corresponding decrease in thermal resistance is
observed.
(a) (b)

Figure 4: (a) Temperature history and (b) Heat Transfer plot during the experimental run (Water- Fr
50% & 75° inclinations (θ-plane)).

θ β

Z X

Figure 5: Effect of filling ratio on the performance of Figure 6:Schematic of CLPHP and the
PHP operated at 90° inclinations orientation position

6
4.1.2 Effect of FR on heat transfer characteristics

Thermal resistance offered by the PHP system for the three FRs –40%, 50% and 60% at 90⁰
inclinations, are plotted in Figure 5. For all the FRs PHP is observed to be fully operational
with water. When the furnace maintains 100⁰C, unstable behavior of the PHP is observed due
to low pressure differential which resulted in high thermal resistances. With the increase in
heat input (FT 150–250⁰C), decrease in thermal resistances is observed for the three FRs.
Lowest thermal resistance of 0.18⁰C/W is observed when FT is 250⁰C for the FR 40 % at 90⁰
inclination.

4.1.3 Effect of inclination on fluid flow behavior and heat transfer characteristics

The thermal resistance of the PHP for FR 50% at inclinations 30⁰- 90⁰ (θ-plane) is shown in
Figure 7(a). In Figure 7(b), temperature history is plotted for FR 50% at 30⁰ inclination. At
0⁰ inclination, high evaporator temperatures following the trend of FT are observed and PHP
does not initiate operation. With the increase in inclination to 15⁰, operation of the device is
observed; yet behavior is somewhat erratic. With the increase in inclination towards 30⁰,
proper start-up is observed when FT is 100⁰C. Thermocouple at the adiabatic section
indicates the oscillating flow at low heat input and tendency shifts to circulatory flow with
the increase in heat input (FT 150–200⁰C). At high heat input (FT 250⁰C), substantial
increase in evaporator temperature is observed. Temperature oscillations of adiabatic
channels suggest oscillatory flow behavior. Further rise in inclination to 45⁰- 90⁰, stable
characteristic operation leading to improvement in thermal performance is observed.
Dominance of oscillatory flow is seen at low heat input and the behavior shifts to bulk
circulation at high heat input for these inclinations.

(a) FR 50% (b)FR 50%, 30⁰ inclination

Figure7 (a) Thermal resistance against heat input for varying inclinations(θ-plane) at the FR 50%, (b)
Temperature history plot at FR 50%, 30⁰ inclinations

4.2 Working fluid Dowtherm-A

4.2.1 Effect of heat input on fluid flow behavior and heat transfer characteristics

Figure 8 (a) shows the representative temperature history for FR 70 %, 75° inclinations (θ-
plane). When FT is 200°C, PHP is not operational. When FT attains 300°C, a sudden drop in

7
evaporator temperature is seen. At this point, the evaporator section maintains a considerable
lower temperature than the FT, and the PHP starts its operation. With few initial oscillations,
the fluid exchange takes place in bulk. Dowtherm-A circulates and takes a particular flow
direction. During stable circulation, fluid exhibits local pulsating behavior; and temperature
oscillations of very low amplitude and high frequency are observed. With the increase in FT
in the range 400–500°C, small rise in evaporator temperature and adiabatic temperature is
observed. Increase in frequency of temperature signal is also observed. The attained
directional bulk circulation is maintained. Figure 8(b) shows the corresponding heat transfer
plot during the experimental run.
(a) (b)

Figure 8 (a) Temperature history and (b) Heat Transfer plot during the experimental run (Dowtherm-
A - Fr 70% & 75° inclinations (θ-plane)).

4.2.2 Effect of FR on heat transfer characteristics

Figure 9 shows the thermal performance of the PHP for the four tested FRs of 50%, 60%,
70% and 80% at 90⁰ inclinations. FR of 50% is observed to offer high thermal resistance, as
proper start-up of the PHP charged with Dowtherm-A is not observed probably due to
insufficient liquid inventory available for the stable operation. Comparatively, low thermal
resistance is observed for the other three FRs i.e. 60%, 70% and 80%. In this context,
decrease in thermal resistance with the increase in heat input is observed. FR 70% is found to
offer least thermal resistance at high heat input (FT 500˚C).

Figure 9 Effect of filling ratio on the PHP operated at 90° inclinations

8
 (a) FR 70% (b) FR 70%, 30⁰ inclination

Figure 10 (a) Thermal resistance against heat input for varying inclinations (θ-plane) at the FR 70%,
(b) Temperature history plot at FR 70%, 30⁰ inclinations

4.2.3 Effect of inclination on fluid flow behavior and heat transfer

The thermal resistance of the PHP charged with Dowtherm-A for FR of 70% at inclinations
15⁰- 90⁰ (θ-plane) is shown in Figure 10(a). The temperature history plot for FR 70 % at 30⁰
inclinations is shown in Figure 10(b). At 0⁰ inclinations no start-up is observed. With
increase in inclination to 15⁰, start-up is seen yet unstable behavior is depicted by the device.
At 30⁰ inclinations, PHP charged with Dowtherm-A starts operation when FT reaches 300⁰C.
With increase in heat input, PHP continues its operation. But at FT of 500⁰C, suddenly a
substantial rise in evaporator temperature is observed. Also, an increase in the amplitude and
frequency of temperature oscillations is observed. Increase in inclination, is observed to bring
more stability in the operation of the PHP charged with Dowtherm-A. At higher inclination
(45⁰ - 90⁰), temperature readings confirm a stable operation.

Effect of gravity over the thermal performance of the PHP charged with Dowtherm-A is
observed at low heat input conditions. With increase in heat inputs the gravity effect over the
performance of device diminishes gradually for higher inclinations (60⁰ - 90⁰), as thermal
resistance is observed to converge. Lower inclinations (30⁰-45⁰) are observed well
operational at low heat input but at high heat input (FT 500⁰C) performance is observed to
degrade.

5. Performance of the CLPHP to scale up to 14-turn and 30-turn

Considering the design constraints of the three-turn CLPHP, two CLPHP with 14-turn and
30-turn are made of long copper tubes. Experiments are performed by varying FR,
inclinations and heat input. In 14-turn PHP, both water and Dowtherm-A, whereas in 30-turn
PHP only Dowtherm-A is experimented. In 14–turn CLPHP, characteristic operation of the
PHP is observed with water. CLPHP is observed operational for the tested FRs 40%, 50%
and 60% at vertical orientation. At higher inclinations proper performance is observed and

9
with increase in heat input, effect of inclination over the thermal performance is observed to
diminish. For lower inclinations, high FRs and medium heat input is observed as favorable.
At horizontal position, start-up at comparatively higher heat input is seen, but performance
deteriorates with increase in heat input.

14-turn PHP 30-turn PHP

(a) FR 70 % (b) FR 70 %

(c) FR 70 %, 75° inclinations (d) FR 70 %, 75° inclinations

(e) FR 70 %, 75° inclinations (f) FR 70 %, 75° inclinations

Figure 11(a, b) thermal resistance against FT for inclination 0º-90 ºat FR-70%, (c, d) temperature
history and (e, f) heat transfer at FR-70%, 75° inclinations of Dowtherm-A for 14 and 30-turn CLPHP

10
Stable operation of 14 and 30-turn CLPHP is observed with fluid Dowtherm-A, offering high
heat transfer and low thermal resistance. Among the tested FRs, FR 60%, 70% is found to
perform better, whereas FR 50% seems insufficient for PHP to operate properly. Tendency of
the fluid behavior to have bulk circulation is seen at higher inclinations. With decrease in
inclination, flow behavior gradually shifts to oscillating flow regime. At high heat input (FT
600 ⁰C), orientation independent operation of 30-turn PHP is observed. Figure 11 shows
variation of thermal resistance with heat input for all inclinations at FR 70%, temperature
history plot and corresponding heat transfer plot during the experimental run at FR 70%, 75°
inclinations for fluid Dowtherm-A in 14 and 30-turn CLPHP side by side.

6. Major Contributions

The major contributions of this study are as follows:

1. Characteristic behavior, depicted by PHP with water as the working fluid over a large
range of parameters, both with 3 and 14-turns is found to be consistent with literature
published by researchers earlier (Saha et al., 2014; Han et al., 2014). After start up, an
increase in heat input enhances the performance of the CLPHP. Tested FRs (40%,
50%, 60%) are observed to be well operational. Higher inclinations are observed to
perform better than lower inclinations. Operation at horizontal orientation is not
observed with 3-turn PHP, but with 14-turn PHP operation is observed at high heat
input.
2. Proper operation of the fluid Dowtherm-A is observed over the operated parameters.
Start-up (at evaporator temperature 180ºC) and stable operation (up to evaporator
temperature 380ºC) is observed. Among the tried FRs, FR 60%, 70% are observed
well operational depicting the characteristic behavior. Effect of gravity is well
observed over 3-turn PHP, whereas over 30-turn PHP at high heat input orientation
independent operation is observed.

References

Akachi, H., United States Patent: 4921041-Structure of a Heat Pipe, 1990

Akachi, H., US Patent, No. 5219020, 1993.
Charoensawan P. and Terdtoon P., Thermal performance of horizontal closed loop oscillating
heat pipes, Applied Thermal Engineering, Vol. 28, pp. 460-466, 2008.
Khandekar, S., Thermo-hydrodynamics of closed loop pulsating heat pipes (Doctoral
dissertation), IKE, University Stuttgart, Germany, 2004.
Zhang Y. and Faghri A., Advances and unsolved issues in pulsating heat pipes, Heat Transfer
Engineering, Vol. 29, No. 1, pp. 20-44, 2008.
Han, H., Cui, X., Zhu, Y. and Shun, S., The comparative study of the behavior of working
fluids and their properties on the performance of pulsating heat pipes, International Journal of
Thermal Sciences, Vol. 82, pp. 138-147, 2014.
Saha, N., Das, P. K. and Sharma, P. K., Influence of process variables on the hydrodynamics
and performance of a single loop pulsating heat pipe, International Journal of Heat and Mass
Transfer, Vol. 74, pp. 238-250, 2014.

11