i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8

Available online at www.sciencedirect.com

ScienceDirect
w w w . i i fi i r . o r g

journal homepage: www.elsevier.com/locate/ijrefrig

Experimental analysis of a novel cooling system

driven by liquid refrigerant pump and vapor
compressor

Gang Yan a,*, Yongbin Feng b, Leqin Peng a

a
Department of Refrigeration & Cryogenic Engineering, School of Energy and Power Engineering, Xi'an Jiaotong
University, Xi'an 710049, China
b
Hanjia Design Group CO., LTD, Xi'an Branch, Xi'an 710075, China

article info abstract

Article history: This study introduced a novel energy saving cooling system, i.e. a combined cycle coupled
Received 1 August 2014 with a traditional vapor compression cycle with a pumped liquid two-phase cooling cycle.
Received in revised form The system has two operation modes, i.e. the compression cycle mode driven by
18 September 2014 compressor and the pump cycle mode driven by refrigerant pump. A multi-purpose test
Accepted 28 September 2014 bench was constructed to experimentally evaluate the performance of the integrated cycle
Available online 7 October 2014 system under various operation conditions. The effects of cycle working condition and the
shift temperature between the two operation modes on the overall cycle performance were
Keywords: investigated in detail. It is found that the novel cycle system has a higher EER compared to
Data center the traditional compressor system when the ambient temperature is relatively low. The
Liquid refrigerant pump further experimental results and comparative annual energy saving analysis also indicated
Cooling that the proper shift temperature is about
5  C from the system EER and cooling capacity
EER point of view.
Shift temperature © 2014 Elsevier Ltd and IIR. All rights reserved.
R22

Analyse expe
rimentale d'un nouveau syste  me de

refroidissement fonctionnant gra^ ce a
 une pompe a frigorige
 ne
 un compresseur de vapeur
liquide et a
Mots cl
es : Centre de donne

es ; Pompe a
 frigorige
ne liquide ; Refroidissement ; Coefficient d'efficacite

e
nerge


tique ; Ecart de temperature ;
R22

* Corresponding author. Tel.: þ86 29 82668738; fax: þ86 29 82668725.

E-mail address: gyan@mail.xjtu.edu.cn (G. Yan).
http://dx.doi.org/10.1016/j.ijrefrig.2014.09.017
0140-7007/© 2014 Elsevier Ltd and IIR. All rights reserved.
12 i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8

Nomenclature EER energy efficiency rate [-]

ho,ref refrigerant-side outlet enthalpy [kJ$kg
1]
Qe,air air-side cooling capacity [kW]
hi,ref refrigerant-side inlet enthalpy [kJ$kg
1]
Qe,ref refrigerant-side cooling capacity [kW]
ho,air air-side outlet enthalpy [kJ$kg
1]
Qsen sensible cooling capacity [kW]
hi,air air-side inlet enthalpy [kJ$kg
1]
Qcomp cooling capacity of the compression cycle mode
E energy consumption [kW$h]
[kW]
Qpump cooling capacity of the pump cycle mode [kW] Subscripts
T Temperature [ C] ref refrigerant
Ta ambient air temperature [ C] sen sensible
m_ ref refrigerant mass flow rate [kg$h
1] comp compressor
m_ air air mass flow rate [kg$h
1] e evaporator
Wevap evaporator fan input power [kW] a ambient air
Wcond condenser fan input power [kW] i inlet
Wpump pump input power [kW] o outlet
Wcomp compressor input power [kW] evap evaporator
SHR sensible heat rate [-] cond condenser
xe,o outlet vapor quality of evaporator [-]
Dp pressure drop [kPa]

1. Introduction experimentally evaluated a pumped liquid multiphase cool-

Data centers are computer infrastructure facilities in which a
large number of enterprise servers, server communication
equipment, and cooling power equipment are contained.
Along with the rapid development of electronics and the
advent of cloud computing, the number of data centers is
increased dramatically, which requires a huge amount of
electrical power for their normal operation. However, this has
a noticeable impact on the energy consumption and carbon
footprint due to the size and number of data centers. Ac-
cording to an Environmental Protection Agency (EPA) report
on energy efficiency in data centers, the energy consumption
of data centers in the United States was more than 100 billion
kWh by 2011, which represents an annual energy cost of
approximately $7.4 billion (Brown, 2008). The report empha-
sized that there are opportunities for data centers to improve
their efficiencies - both on the facility and server sides of the
data center infrastructure.
The most widely used cooling strategies in computer room
air conditioners (CRAC) is refrigerated air cooling due to its
high reliability, and low initial and maintenance costs. How-
ever, it has several major disadvantages. One is the by-pass air
flow in the racks of servers in data centers and it cannot
effectively participate in the cooling process of servers.
Furthermore, the working liquid driven by a compressor has
to be lifted to a higher condensing temperature to ensure the
safe and stable operation of the compressor at a lower outside
ambient temperature, which means a significantly amount of
energy will be consumed. The poor energetic performance of
traditional cooling strategies motivates thermal designers to
search for green thermal solutions for higher performance
servers while providing the possibility to revise the tradition
cooling cycle to improve the overall performance.
Due to the performance limits of conventional cooling
technologies in both military, commercial electronics and
electro-optical systems, Hannemann et al. (2004)
ing system (PLMC) to cool microprocessors (such as com-
puters, telecommunications, and phased array radar systems)
and Rack-based system in data centers. They emphasized the
significant benefits of performance, cost, size, weight and
reliability that can be provided with the PLMC solution as
compared with existing active air cooling or simple liquid-
loop systems. They also recommended that the PLMC can be
applied to military, commercial and industrial systems due to
its excellent thermal management capabilities. Chanwoo
et al. (Park et al., 2007; Crepinsek and Park, 2012) presented an
advanced Hybrid Two-Phase Loop (HTPL) technology which
integrated active mechanical pumping with passive capillary
pumping for electronics thermal management. The experi-
mental results indicated that the hybrid loops with multi-
evaporators were able to remove high heat fluxes and the
total heat inputs govern the system temperature and pres-
sures. A shortcoming of the multi-evaporators loop is that
each evaporator line needs to be identical or controlled by a
valve. Although there are many limitations about the tradi-
tional cooling strategies in CRAC as mentioned above, it has
still attracted more attention (Mongia et al., 2006;
Trutassanawin et al., 2006; Zhou et al., 2010). All studies
selected lower pressure refrigerant R134a or R600a as the
working fluids and mini-compressor to drive the working fluid
for the small-scale refrigeration system. Active two-phase
cooling system such as evaporative spray cooling (Glassman
et al., 2004; Lin et al., 2006; Silk et al., 2008; Mudawar et al.,
2009) has been extensively investigated over the past several
years due to advantages, such as high heat flux performance
and robust operation. However, it requires complex fluid
control due to two-phase mixed flow and flooded boiling in
the system. Ohadi et al. (2012) compared the cooling perfor-
mance of electronics in data centers by using air, liquid and
two-phase on-chip cooling. Advantages and disadvantages of
various methods were also discussed. It was concluded that
air cooling may remain a very attractive solution for locations
 i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8
where a cool air climate is available during most of the year,
and two-phase cooling reduced thermal resistance as much as
an order of magnitude less than that of air and significantly
below that of liquid cooling. Additionally, the thermal resis-
tance of the two-phase cooling by the force-fed manifold
microchannels is 20 times lower than that of the liquid cooling
at 5 times lower pumping power consumption. The phase
change cooling systems will have significant competitive
advantage over conventional systems due to the lower costs
and energy savings. Marcinichen et al. (2010, 2012) and Wu
et al. (2013) proposed three two-phase cooling cycles for
cooling data center servers and the cycles using two-phase
boiling in microchannels for removing the heat from the mi-
croprocessors and memories. The results showed that the
pumping power consumption is on the order of 5 times higher
for the water-cooled cycle when considering the potential
saving by implementing on-chip cooling with waste heat re-
covery. The results also showed that, when compared with
the traditional air-cooling systems, the energy consumption
of the data center could be reduced by 50% when using a liquid
pumping cycle and by 41% when using a vapor compression
cycle. Further investigation indicated that a PI structure and
simple SISO and SIMO strategies were sufficient to fulfill the
requirements.
However, previous studies of two-phase cooling in data
centers mainly focused on the application in a microchannels
evaporator for removing the heat of on-chip cooling. In a
pump liquid two-phase cycle, the potential of energy saving
with lower input power feature has been already exploited,
where it is the basis of improved cycle performance. In fact,
the two-phase cooling can be used to modify the existing
tradition air-cooled systems in data centers considering the
energy and cost savings, i.e. obtaining a higher system per-
formance with a lower pump input power during lower
ambient temperatures. For this purpose, this paper presents a
novel hybrid cooling cycle coupled with liquid refrigerant
pump cycle and vapor compressor cycle to be used in data
centers. Based on existing vapor compressor cycle, a liquid
refrigerant pump associated with a receiver is introduced into
Fig. 1 e Schematic diagram of the proposed cooling cycle.
13
the vapor compressor cycle system. Such a novel cycle system
may be more effective for improving energy efficiency and
cost saving of data centers. The performance of the novel
cooling system driven by liquid refrigerant pump and
compressor in data centers is experimentally evaluated.
2. Experimental setup
Fig. 1 illustrates the integrated cooling cycle driven by a liquid
refrigerant pump and a vapor compressor and it can be
operated in two different operation modes by means of the
solenoid valves (SVs) 1 to 4. The choice of the system opera-
tion mode depends on the cooling demand and ambient air
temperature. When the ambient air temperature is low, the
system is shifted to the pump operation mode by means of
opening the SVs 1 and 3, and shutting off SVs 2 and 4. In the
liquid pump operation cycle, a feed custom gear pump with
constant speed compresses the working fluid from the satu-
rated liquid state at the receiver to a subcooled liquid state at
pump outlet (process 5e6). Subsequently, the subcooled fluids
absorb the data center heat in the evaporator and then change
to a saturated or superheat vapor state (process 8e1). The high
pressure and temperature vapor enters the connected pipe
between the evaporator and condenser, and then down to a
lower pressure and temperature (process 1e2e3). Lastly, the
heat will be rejected to the ambient in the condenser where
working fluid changes from the superheat state to the liquid
state at the inlet of the receiver (process 3e4). The major
function of the check valves (CVs) 1 to3 in the loop diagram is
to ensure the proper flow direction of working fluids in
different operation modes. When designing a pumping sys-
tem, it is important to prevent the phenomenon of cavitation.
Cavitation occurs when bubbles form in the liquid at the
pump inlet. In order to make the liquid pump work stable and
avoid cavitation, the 6 L receiver is lifted up to 1.5 m higher
than the liquid pump. The liquid refrigerant pump cycle is
characterized with low condensing temperature and high
14 i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8
Table 1 e Evaporator geometric parameters.
Geometric parameters Values
Number of tube rows in air direction 4 Inner tube diameter/mm
Number of tube rows perpendicular to air direction 24
Transverse tube spacing/mm 25.4
Longitudinal tube spacing/mm 22.0
Outer tube diameter/mm 9.52
overall efficiency. This is a good operation option during the
winter season.
When the ambient temperature is high, the system is
shifted to the compressor operation mode by means of
opening SVs 2 and 4 and shutting off SVs 1 and 3. In the
compressor operation cycle, a VR30KM-TFP type compressor
compresses the working fluid from the superheat state at the
outlet of evaporator to a high temperature and pressure state
at the compressor outlet (process 1e2). Subsequently, the heat
will be rejected to the ambient in the condenser where the
working fluid passes from the superheat state to the saturate
state at the inlet of the receiver (process 3e4). Finally, the
saturated or subcooled state liquid refrigerant is expanded
through thermal expansion valve into the evaporator at a
lower evaporation pressure to cool the data center (process 6-
7-8e1). This cycle is characterized by a high condensing
temperature and a medium overall efficiency when compared
with the liquid refrigerant pump cycle. This is a good opera-
tion option when the ambient temperature is high, typically
during the summer season.
During the experimental test, a 7 kW fin-tube evaporator
was designed to transfer the heat generated by the servers to
the refrigerant. The evaporator used is a cross-flow heat
exchanger and is composed of four circuits. Each circuit has a
single inlet and outlet. The coil geometry used is summarized
in Table 1. The condenser used is a cross-flow fin-tube heat
exchanger with two rows and two circuits, and each circuit
has a single inlet and outlet.
Experiments were conducted in a standard psychromet-
ric chambers with two insulated rooms to evaluate the
performance of the designed cycle under a range of oper-
ating conditions. The temperature and humidity were
independently controlled within each room to the desired
operation conditions. The evaporator air stream was con-
nected to an ASHRAE standard air measurement box to yield
a measurement of the air flow rate and the evaporator air
flow rate was altered by a variable speed blower inside the
box. Two pressure transducers are used to measure the
pressure drop across the nozzle and the pressure at the
nozzle inlet, respectively. Three 1000U platinum TEP series
probes RTDs measure the air temperature near the inlet of
the nozzles. All temperature measurements were Type T
copper-constantan thermocouples, which were calibrated
using an ice bath and boiling water to make sure the read-
ings were within 0.3  C of the known values. Low pressure
transducers and high pressure transducers were installed in
the refrigerant flow line to measure the refrigerant pres-
sures of the interested state points with an accuracy of
±0.25% over the full scale range. The refrigerant tempera-
tures and pressures were measured at the locations of the
inlet and outlet of each component throughout the system,
Geometric parameters Values
8.85
Fin thickness/mm 0.12
Fins per inch 14
Fin type Louver
Circuitry number 4
as indicated in Fig. 1. The flow rate of the working fluid was
measured using a turbine-type flow meter with an accuracy
of ±0.5%. One sight glass was installed downstream of the
Micro Motion mass flow meter to identify the case of two-
phase flow. The electric power output of the generator (i.e.
fan, pump and compressor) was measured using a power
analyzer made by WT230, with an accuracy of ±0.02%. The
air-side cooling capacity can be achieved based on data of
the airflow rate and different sensors. A steady state was
maintained for 30 min in psychrometric lab to gather data in
each test. During the operation, the thermodynamic prop-
erties of the working fluids in cycle were calculated by
REFPROP (Lemmon et al., 2007) using measured tempera-
tures and pressures values. All the measurement data was
recorded and stored on a computer by using a HP 75000 se-
ries data acquisition system. For steady state performance
tests, the sampling interval was 10 s.
The refrigerant enthalpy was determined by the local
pressure and temperature measurement except the two-
phase mixture state. The air enthalpy was determined by
the measured dry bulb temperature and wet bulb temperature
at the corresponding location.
The refrigerant-side cooling capacity is calculated as:


Qe;ref ¼ m_ ref ho;ref
hi;ref (1)
where hi,ref is determined by the pressure and temperature
ahead of the expansion device. ho,ref is determined by the
pressure and temperature leaving the evaporator. It should be
noted that the refrigerant-side cooling capacity cannot be
determined when two-phase flow occurs at the exit of an
evaporator.
The air-side total cooling capacity is calculated as:


Qe;air ¼ m_ air ho;air
hi;air (2)
where hi,air and ho,air are determined using the measured inlet
and outlet air dry bulb and wet bulb temperature of the
evaporator, respectively.
The air-side total sensible cooling capacity is calculated as:


Qsen ¼ m_ air ti;air
to;air (3)
where ti,air and to,air are determined using the measured inlet
and outlet air dry bulb temperatures of the evaporator,
respectively.
Energy efficiency ratio is given as.


I For compressor operation mode :
(4)
Qe;air
EER ¼
Wcomp þ Wevap þ Wcond
 i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8
Fig. 2 e Effect of ambient air temperature on overall system
SHR and sensible capacity with the pump cycle mode.

 Qe;air
II For pump operation mode : EER ¼
Wpump þ Wevap þ Wcond
The quality of the experimental results was determined by
estimating the uncertainty of the test results. The un-
certainties were calculated for cooling capacity and EER, given
by Equation (6).
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u  2
uX vY
UY ¼ t U2X
i
vXi
where Xi and Y represent measured and calculated variables,
respectively. UX and UY are the measured and calculated un-
certainties, respectively (Kline and McClintock, 1953).
The uncertainties of the dependent variables can be
determined with the partial derivatives and independent
uncertainties. The uncertainties of the dependent variables
air-side cooling capacity and EER in a steady-state test of the
7 kW R22 unit are 5.38% and 6.49%, respectively. The data of
the other steady-state tests have similar uncertainties.
3. Results and discussion
In data centers, the cooling capacity plays an essential role to
provide a proper working environment for servers. In order to
highlight the general characteristics of the novel cooling sys-
tem driven by liquid refrigerant pump and vapor compressor,
and better understand the effect of the ambient temperature
on the evaporator cooling capacity, a number of capacity tests
were conducted under a typical data center working condi-
tion. The working conditions used in the test are as below.
▪ The psychrometric indoor dry bubble temperature set as
23  C.
▪ Air inlet wet bubble temperature is 17  C, and outdoor
ambient air temperature is from-20  C to 10  C.
▪ Working fluid used is R22.
15
Fig. 3 e Effect of ambient air temperature on overall system
capacity with different cycle modes.
Fig. 2 shows the sensible cooling capacity and sensible heat
ratio (SHR) varied with the ambient air temperature in the
pump cycle mode. The sensible heat ratio is the sensible
cooling capacity divided by the total cooling capacity. The
(5)
sensible heat ratio is given as:
Qsen
SHR ¼ (7)
Q
It can be seen that the sensible cooling capacity increases
rapidly with the decrease of ambient air temperature. How-
ever, the sensible heat ratio of the novel system decreased
(6)
along with the decreased ambient air temperature in the
pump cycle mode. The sensible cooling capacity was 2.74 kW
when the ambient air temperature is 10  C, which is lower 51%
than that of the compression cycle mode as indicated in Fig. 3.
The total cooling capacity and the sensible cooling capacity of
the novel system were 94% and 90% respectively, as compared
with the compression cycle mode when the ambient air
temperature is
5  C. Meanwhile, the system SHR was 76.63%.
For a typical data center room, the cooling load decreased with
the decrease of the ambient air temperature due to larger
Fig. 4 e Effect of ambient air temperature on the
evaporating temperature, condensing temperature and the
outlet quality of evaporator in the pump cycle mode.
16 i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8

temperature difference between the ambient environment

and indoor room. Therefore, the novel system can nearly
reach the same level cooling capacity when compared with
the traditional compressor system.
Fig. 4 shows the effect of ambient air temperature on the
evaporating temperature, condensing temperature and the
outlet quality of evaporator in the pump cycle mode. The
evaporating temperature and condensing temperature are
corresponding to the saturation temperatures of the
measured pressure at the outlet of the evaporator and
condenser. As can be seen clearly, the evaporating tempera-
ture and condensing temperature of the novel system in the
pump operation mode decreased with the decrease of the
ambient air temperature. The evaporating and condensing
temperature were 6.73  C and 0.94  C, respectively, when the
ambient air temperature is
5  C. It can be concluded that the
condensing pressure is more sensitive to ambient air tem-
perature compared with the evaporating pressure. A higher Fig. 6 e Effect of ambient air temperature on the overall
cooling capacity can be obtained with the decrease of the system EER with different cycle modes.
ambient temperature decreased. However, it might happen
that the evaporator temperature will reach the freezing point
if no control strategy is adopted. Therefore, a condenser fan Fig. 5 shows the overall pressure drop of the pump cycle
speed control strategy is needed for the lower temperature mode decreased when the ambient air temperature varied
region but it not considered in the present study. The liquid from
20 Cto 10  C. This is because the decrease of the
refrigerant pump cycle is characterized with low condensing ambient temperature greatly affects the refrigerant thermal
temperature compared with the traditional compressor cycle. dynamic parameter (outlet quality of evaporator as seen in
The outlet vapor quality of the evaporator was calculated Fig. 4) and fluid viscosity, resulting in larger refrigerant pres-
considering equations 8e10: sure drops. For example, the percentage of the pressure drop
  for the evaporator and pipe play a significant role to the total
xe;o ¼ f he;o ; pe;o (8) pressure drop and their contributions are 35% and 45% when
the ambient air temperature is -5 C, respectively. Therefore,
Qe this means a larger connection pipe between the condenser
he;o ¼ þ he;i (9)
m_ and evaporator can decrease the system pressure drop and
  improve the cycle performance for the novel system.
he;i ¼ f pe;i ; Te;i (10) It is known that the performance of the refrigerant cycle
system always plays a significant role in influencing the en-
where xe,o,he,o are determined from Refprop software ergy consumption. As mentioned before, the novel cycle sys-
(Lemmon et al., 2007). In general, the outlet vapor quality of tem has two operation modes dependent on the ambient air
evaporator is increased along with the decreased ambient air temperature and cooling load demand. The cycle will shift to
temperature. the compression cycle mode when the ambient air

Fig. 5 e Effect of ambient air temperature on the refrigerant

side pressure drop of the overall components in the pump Fig. 7 e Effect of ambient air temperature on the overall
cycle mode. input power with different cycle modes.
 i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8
temperature is higher due to the lower cooling capacity of the
pump cycle mode as indicated in Fig. 3. This compression
cycle mode commonly happens in the summer season. Along
with the decreased ambient temperature, however, the cool-
ing capacity of the pump cycle mode is increased. Thus, the
novel cycle system should be shifted to the pump cycle mode,
which commonly happens in the winter season. In this case,
there is as proper shift temperature while the two cycle modes
are shifted to each other. Figs. 6 and 7 show the overall EER
and input power of the novel system operated in the pump
cycle mode and compression cycle mode, respectively. It can
be clearly seen that the novel cycle system has a higher EER
compared with the compression cycle system when the
ambient temperature is below 7 C, which is due to the lower
input power of the pump cycle mode compared with the
compression cycle mode. The overall EER for the novel system
and compression cycle is 6.84 and 3.46, respectively, when the
ambient air temperature is
5  C. Meanwhile, the total input
power for the pump cycle mode is just about 50% of the
compression cycle mode. Therefore, the novel system has
better performance when compared with the traditional
compression cycle during the lower ambient conditions. This
means that the novel system can provide significant energy
savings and it also provides an alternative method to improve
the existed traditional compressor system. The experimental
results also indicated that the proper shift temperature is
about
5  C in terms of the system EER and cooling capacity.
The calculation of energy savings is an important step in
the validation of the benefits of the novel system. The annual
energy saving was calculated based on the measured data
using the bin method. Three typical places were selected
during the energy saving analysis process and the bin hour
distribution under different ambient temperatures is dis-
played in Fig. 8, which were generated by using EnergyPlus. It
can be seen clearly that the annual Bin hours in the low
temperature region of Minneapolis is higher than that of
Washington and Atlantic. In the following analysis, the sen-
sible cooling capacity is considered for both the novel hybrid
system and the traditional compressor system. The ambient
air temperature kept in a range of
30  C to 10  C when
Fig. 8 e Percentage of Bin hours' distribution under
different ambient temperatures.
17
Fig. 9 e Effect of shift temperature on the overall system
energy saving with the novel hybrid cycle.
considering the pump cycle mode of the novel system in the
winter season. Using the Bin method the total annual energy
saving of the novel system can be determined using equation
(11) (Liu, 1993).
Xn Xm  
Esave;tot ¼ i¼1 j¼1
Eprei;j
Eposti;j (11)
where Eprei,j is the bin mode predicted hourly electricity use
during hour (j) of day (i) in the novel hybrid cooling cycle. i is
the distinct day type varying from i ¼ 1 to i ¼ 365 and j is 1e24 h
in each day. Eposti,j is the bin mode predicted hourly electricity
use during hour (j) of day (i) in the traditional compression
cycle.
Fig. 9 shows the results of the overall system energy saving
of the novel system compared with the traditional compres-
sion cycle. It can be concluded that higher percentage of the
lower ambient temperature is, the more energy can be saved.
There is a maximum energy saving point when the ambient
air temperature is about 5  C. After the optimal shift ambient
temperature, the novel system in the pump cycle mode loses
its advantage compared with the compression cycle from the
overall energy saving point of view. The proper shift temper-
ature for the novel system is about
5  C while considering the
cooling capacity, overall EER and annual overall energy saving
simultaneously. When the shift temperature is
5  C, the
annual energy saving can reach to 1500 kW h. It provides an
alternative approach to modifying the traditional air-cooled
compressor systems by combining with a liquid pump cycle.
The novel hybrid system can be shifted to the compression
cycle mode when the ambient temperature is higher than

5  C, and turn to the pump cycle mode when the ambient
temperature is below
5  C.
4. Conclusion
A novel energy saving cooling system driven by a liquid
refrigerant pump and vapor compressor with R22 was pro-
posed for the data centers. The system has two operation
modes, i.e. one is the compression cycle mode driven by a
18 i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 4 9 ( 2 0 1 5 ) 1 1 e1 8
compressor when the ambient temperature is high and the
other is the pump cycle mode driven by the liquid refrigerant
pump when the ambient temperature is low. Based on the
multi-purpose test bench, the performance of the novel
hybrid cooling system was experimentally evaluated under
various operation conditions and compared with the tradi-
tional compression cycle systems. The effects of the cycle
working condition and shift temperature on the overall cycle
performance were investigated. The annual overall energy
saving was analyzed using Bin hours method based on the
experimental test data. The results showed that the novel
system has a higher energy efficiency ratio as compared to the
traditional compressor system when the ambient tempera-
ture is lower due to the lower input power. The input power
the novel system can be decreased by 50% when compared to
that of the compressor system. The experimental results also
indicated that the proposed novel system can obtain nearly
the same cooling capacity as that of the traditional
compressor system when the ambient air temperature is

5  C. The pressure drop of the connected pipe between the
evaporator and condenser has a significant impact on to the
total system pressure drop due to the two-phase region under
the working conditions. The novel system EER is higher than
that of the compressor cycle when the ambient temperature is
below 7  C. The comparative annual energy saving analysis
showed that the lower annual bin temperature is, more en-
ergy will be saved. The proper shift temperature is about
5  C
from the system EER, cooling capacity and annual overall
energy saving point of view. Overall, this study shows the
proposed novel cycle could provide a way to modify the
traditional air-cooled compressor system and show strong
energy saving potential for its application in green data
centers.
references
Brown, R., 2008. Report to Congress on Server and Data Center
Energy Efficiency Public Law 109-431. U.S. Lawrence Berkeley
National Laboratory.
Crepinsek, Michael, Park, Chanwoo, 2012. Experimental analysis
of pump-assisted and capillary-driven dual-evaporators two-
phase cooling loop. Appl. Therm. Eng. 38, 133e142.
Glassman, B., Kuravi, S., Du, J., et al., 2004. A fluid management
system for a multiple nozzle array spray cooler. In: 37th AIAA
Thermo-physics Conference, Portland, Oregon, 2004-2574.
Hannemann, R., Marsala, J., Pitasi, M., 2004. Pumped liquid
multiphase cooling. In: IMECEe Int. Mechanical
EngineeringCongress and Exposition, Anaheim, CA, USA.
Paper 60669.
Kline, S.J., McClintock, F.A., 1953. Describing uncertainties in
single-sample experiments. Mech. Eng. 75, 3e8.
Lemmon, E.W., Huber, M.L., McLinden, M.O., 2007. NIST Standard
Reference Database 23: Reference Fluid Thermaldynamic and
Transport Properties-REFPROP, Version 8.0. National Institute
of Standard and Technology. Standard Reference Data
Program.
Lin, L.C., Ponnappan, R., Yerkes, K., 2006. Actively pumped two-
phase loop for spray cooling. J. Thermophy. heat Transf. 20,
107e110.
Liu, Y., 1993. Measuring Retrofit Savings in Commercial Buildings
with Pre-retrofit Utility Billing Data and Post-retrofit Sub-
metered Data. M.S. Thesis, Energy Systems Laboratory Report,
no. ESL-TH-93/12-04. Texas A&M University.
Marcinichen, J.B., Thome, J.R., Michel, B., 2010. Cooling of
microprocessors with micro-evaporation: a novel two-phase
cooling cycle. Int. J. Refrigeration 33, 1264e1276.
Marcinichen, J.B., Olivier, J.A., Thom, J.R., 2012. On-chip two-
phase cooling of datacenters: cooling system and energy
recovery evaluation. Appl. Therm. Eng. 41, 36e51.
Mongia, R., Masahiro, K., DiStefano, E., et al., 2006. Small Scale
Refrigeration System for Electronics Cooling within a
Notebook Computer. ITHERM, San Diego, CA.
Mudawar, I., Bharathan, D., Kelly, K., et al., 2009. Two-phase spray
cooling of hybrid vehicle electronics. IEEE Trans. Components
Packag. Technol. 32, 501e512.
Ohadi, M.M., Dessiatoun, S.V., Choo, K., et al., 2012. A comparison
analysis of air, liquid, and two phases cooling of data centers.
In: 28th IEEE SEMI-THERM Symposium. IEEE, pp. 58e63.
Park, Chanwoo, Vallury, Aparna, Zuo, John, et al., 2007.
Electronics thermal management using advanced hibrid two-
phase loop technology. In: ASME-JSME Thermal Engineering
Summer Heat Transfer Conference, July 8e12, Vancouver,
British Columbia, Canada. HT2007-321962.
Silk, E.A., Golliher, E.L., Selvam, R.P., 2008. Spray cooling heat
transfer: technology overview and assessment of future
challenges for micro-gravity application. Energy Cover.
Manag. 49, 453e468.
Trutassanawin, S., Groll, E.A., Garimella, S.V., et al., 2006.
Experimental investigation of a miniature-scale refrigeration
system for electronics cooling. IEEE Transact. Components
Packag. Technol. 29 (3), 267e687.
Wu, D., Marcinichen, J.B., Thome, J.R., 2013. Experimental
evaluation of a controlled hybrid two-phase multi-
microchannel cooling and heat recovery system driven by
liquid pump and vapor compressor. Int. J. Refrigeration 36,
375e385.
Zhou, R., Zhang, T., Catano, J., Tong, T., Chang, J., Prasher, R.,
Jensen, M.K., 2010. The steady-state modeling and
optimization of a refrigeration system for high heat flux
removal. Appl. Therm. Eng. 30 (16), 2347e2356.