You are on page 1of 9

Performance of radial piston type reciprocating

expander for CO
2
refrigeration cycle
Mitsuhiro Fukuta
a,
*, Fumiya Anzai
b
, Masaaki Motozawa
a
,
Hiroyuki Terawaki
b
, Tadashi Yanagisawa
a
a
Department of Mechanical Engineering, Shizuoka University, 3-5-1 Johoku Naka-ku, Hamamatsu 432-8561, Japan
b
Graduate School of Engineering, Shizuoka University, 3-5-1 Johoku Naka-ku, Hamamatsu 432-8561, Japan
a r t i c l e i n f o
Article history:
Received 26 December 2013
Received in revised form
11 February 2014
Accepted 14 February 2014
Available online 24 February 2014
Keywords:
Expander
CO
2
refrigeration cycle
Power recovery
Throttling loss
Reciprocating expander
a b s t r a c t
Various types of expanders have been investigated to recover a throttling loss and to
improve the performance of CO
2
refrigeration cycle. However, the capacity of the expander
studied so far is so large that it cannot be used for a small cycle such as a vending machine,
since the performance of small expander tends to get worse due to inuence of leakage. In
this paper, a novel reciprocating expander which can be applied to the small cycle is
designed and its performance is examined. The expander has four cylinders and pistons
arranged radially, and controls supply and discharge of refrigerant by a reciprocating
motion of an adjacent piston. It is found that the developed expander can be operated with
a small ow rate and the total efciency attains to 0.4 over the wide range of rotational
speed.
2014 Elsevier Ltd and IIR. All rights reserved.
Performance du de tendeur a` piston radial pour un cycle
frigorique au CO
2
Mots cles : De tendeur ; Cycle frigorique au CO
2
; Re cupe ration de nergie ; Perte due au laminage ; De tendeur a` piston
1. Introduction
Carbon dioxide (CO
2
) is natural refrigerant and an alternative
candidate to hydro uorocarbon (HFC) refrigerants in refrig-
eration or heat pump cycles, because it has no ammability
and no toxicity. In recent years, the CO
2
cycle is used for water
heaters and vending machines. However, the inherent
performance of air cooled CO
2
cycle is lower than that of HFCs
due to large throttling loss occurred in the isenthalpic
expansion process (Lorentzen, 1995; Robinson and Groll, 1998)
and it is important to recover the loss in order to improve the
performance of the CO
2
cycle. Using an expander as an
expansion device is one way to recover the throttling loss.
Various types of expander have been developed and
investigated for the CO
2
cycle. A scroll mechanism is suited to
* Corresponding author. Tel./fax: 81 53 478 1054.
E-mail address: tmmfuku@ipc.shizuoka.ac.jp (M. Fukuta).
www. i i i r . or g
Available online at www.sciencedirect.com
ScienceDirect
j ournal homepage: www. el sevi er. com/ l ocat e/ i j ref ri g
i nt e r na t i o na l j o ur na l of r e f r i ge r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6
http://dx.doi.org/10.1016/j.ijrefrig.2014.02.005
0140-7007/ 2014 Elsevier Ltd and IIR. All rights reserved.
the expander and many researchers studied about the scroll
expander for the CO
2
cycle (Huff et al., 2003; Fukuta et al., 2006;
Kohsokabe et al., 2008a; Hiwata et al., 2009). Vane expanders
have the almost the same structure with vane compressors,
and therefore have simple structure. The performance of the
vane expander is examined by Fukuta et al. (2009), Yang et al.
(2009) and Jia et al. (2009). Two-cylinder rotary expander is
proposed in which one of the cylinders is used to control a
supply process of the expander and design parameters are
optimized (Matsui et al., 2009). A revolving vane expander is
developed by Subiantoro and Ooi (2009, 2012a,b) and its per-
formance is investigated both experimentally and theoreti-
cally. Although the control of the supply and discharge
process is necessary in a reciprocating mechanism in case of
the expander, reciprocating machines are considered as the
expander (Kruse et al., 2006; Baek et al., 2002, 2005a,b). Free
piston expander is a kind of the reciprocating expander and
developed by Heyl and Quack (1999), Nickl et al. (2002, 2005)
and Zhang et al. (2007). Driving an auxiliary compressor
directly by the recovered power of the expander or using the
recovered power as a part of driving power of the compressor
is discussed by combining the expander with the compressor
(Fukuta et al., 2001; Okamoto et al., 2005; Kim et al., 2006;
Kohsokabe et al., 2006, 2008b; Kakuda et al., 2009).
The expanders studied so far has relatively big capacity
and cannot be applied to a refrigeration cycle having small
capacity such as vending machines. The same structures of
the expander described above are hardly adaptable to the
small size expander, since the inuence of leakage becomes
severe in the small expander. Among the expander types, the
reciprocating expander seems to be feasible to have a good
performance in the small capacity refrigeration cycle, since
some kinds of seal device such as a piston ring can be appli-
cable to reduce the leakage. On the other hand, the controls of
supply and discharge process are needed in the reciprocating
expander.
In this study, a novel reciprocating expander which can be
operated with small ow rate is developed. It has four pistons
and cylinders arranged radially. The controls of supply and
discharge of refrigerant are done by reciprocating motion of
an adjacent piston as a spool valve. The performance of the
expander is measured and examined experimentally.
2. Experiment
In this study, the reciprocating expander is developed and the
performance of expander is measured with a test CO
2
refrig-
eration cycle.
2.1. Reciprocating expander
Fig. 1 shows appearance of the radial piston type expander
and a piston with a connecting plate. Fig. 2 shows a schematic
of inner structure of the expander and working principle. The
four cylinders labeled C1eC4 are arranged radially. The pis-
tons are connected to a one crank arm via the connecting
plate. Each cylinder has a ow channel connected to a side
wall of an adjacent cylinder. The piston has a groove sealed by
O-rings at both ends of the groove. The reciprocating motion
of the adjacent piston controls supply and discharge of
refrigerant as a spool valve. An inlet port is connected to four
Nomenclature
Latin letters
d diameter [m]
G mass ow rate [kg s
1
]
l length of connecting rod [m]
n rotational speed [s
1
]
P pressure [Pa]
r crank radius [m]
T torque [Nm]
V volume [m
3
]
W PeV work [J]
Greeks
h efciency
q rotational angle [rad]
r density [kg m
3
]
u angular velocity [rad s
1
]
Subscripts
cyl cylinder
exp expander, experiment
i ith cylinder
in inlet
ind indicated
is isentropic
m mechanical
out outlet
st stroke
t total
th theoretical
top top clearance
v volumetric
Fig. 1 e Appearance of radial piston type expander and
piston.
i nt e r na t i ona l j o ur na l o f r e f r i g e r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 49
supply ports labeled P1eP4. Acrank chamber is connected to a
discharge port, and pressure inside the crank chamber is low
pressure. A crank shaft comes out to the atmosphere through
a side plate. An O-ring is used as a shaft seal. The crank shaft
is supported by ball bearings at both side plates. A thrust force
acting on the shaft is supported by a thrust bearing.
Fig. 2(a)e(d) show the supply and discharge process of the
refrigerant with focusing on the cylinder C1 and the piston 1.
(a) High pressure refrigerant is supplied to the cylinder C1
fromthe supply port P1 through the groove of the piston
2 and the owchannel. The pressure inside the cylinder
C1 becomes high and pushes down the piston 1.
(b) When the piston 1 reaches the bottom dead center
(BDC), the piston 2 closes the supply port P1 and the
supply of refrigerant is stopped.
(c) During the process when the piston 1 moves from BDC
to the top dead center (TDC), the cylinder 1 is connected
the crank chamber through the ow channel. The
refrigerant in the cylinder 1 is discharged to the
discharge port through the crank chamber.
(d) When the piston 1 reaches TDC, the ow channel from
the cylinder 1 is disconnected from the crank chamber
by the piston 2, and the discharge process terminates.
This process (a)e(d) is repeated. All pistons are connected
to the crank arm and the reciprocating motion of the piston is
converted into a rotational motionof the crank shaft. It should
be noted that since the supply process is continued until BDC,
the expander does not have an expansion process and a built-
in expansion volume ratio of the expander is 1.0. Specica-
tions of the expander are shown in Table 1. Stroke volume of
one cylinder is 96.2 mm
3
.
2.2. Performance measurement
The radial piston expander is connected to a CO
2
refrigeration
cycle as shown in Fig. 3. A CO
2
scroll compressor for a domestic
water heater is used. Agas cooler andanevaporator are double-
tube type. Pressure and temperature at several points in the
cycle are measured by Bourdon tube pressure gauges and T
type thermocouples respectively. The accuracy of the pressure
gauge is 0.5% of full scale, and that of the thermocouple is
0.5

C. The expander is connected to an oil pump as a load to
control the rotational speed. A torque meter is inserted be-
tween the expander and the oil pump to measure the rotational
torque of the expander. The error of the torque meter is within
0.2% of full scale. Pressure inside the cylinder is measured by a
piezo-electric pressure transducer as shown in Fig. 1, whose
accuracy is 3% of full scale. The lubrication oil is fed to the
Fig. 2 e Schematic of inner structure of expander and working principle.
Table 1 e Specications.
Cylinder inner diameter 5.0 mm
Stroke 4.9 mm
Shaft diameter 6.0 mm
Connecting plate length 11 mm
Piston seal O-ring
i nt e r na t i o na l j o ur na l of r e f r i ge r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 50
expander fromthe compressor oil sump by pressure difference
between the compressor outlet and the expander inlet. The
mass owrate of the refrigerant is measured by a Coriolis mass
ow meter. The ow rate of the oil is measured by a gear type
ow meter. These accuracies are 0.5% of read scale. An oil cir-
culation ratio is set about 1e2 wt.%.
The experimental conditions are shown in Table 2. An
inverter control of the compressor and a bypass valve is used
to control the operating condition of the cycle. The mass ow
rates of refrigerant and oil, the output torque, the inlet tem-
perature, and the pressure change in the cylinder are
measured after the cycle reaches steady state.
3. Theoretical efciency of power recovery
3.1. Theoretical indicated efciency
A dot line in Fig. 4 shows a PeV diagram of an isentropic
expansion process under the condition that the inlet pressure
is 8.4 MPa, the outlet pressure is 3.2 MPa and the inlet tem-
perature is 10

C, while a broken line shows a theoretical PeV
diagramof the expander having the expansionvolume ratio of
1.0 under the same condition. The PeV diagrams show that
the expander cannot recover a part of the isentropic expan-
sion work in a two phase region. Ratio of the work theoreti-
cally recovered by the expander to the isentropic work is
dened as a theoretical indicated efciency h
ind,th
.
h
ind;th

W
th
W
is
(1)
It is found that from Fig. 4 that the portion which is not
recovered by the expander is relatively small in the CO
2
transcritical refrigeration cycle even if the expander does not
have the expansion process. In case of Fig. 4, the theoretical
indicated efciency attains to 94%.
3.2. Theoretical volumetric efciency
The high pressure refrigerant is supplied to the cylinder
through the ow channel which connects the cylinder with
the supply port. Volume of a top clearance and the ow
channel is dened as V
top
. When the supply process starts at
TDC, the high pressure refrigerant is supplied to V
top
and the
pressure in V
top
increases from the outlet pressure to the inlet
pressure as shown in Fig. 5. The mass of refrigerant supplied
at this moment does not give any effective work but increases
useless owrate through the expander. Therefore, it results in
decreases of theoretical power by unit mass ow rate and a
volumetric efciency of the expander. The theoretical volu-
metric efciency h
v,th
is dened as follows.
h
v;th

V
st
r
in
V
st
r
in
V
top
r
in
r
out

(2)
where V
st
is a stroke volume, r
in
is inlet density and r
out
is
outlet density. The top clearance of the expander is 12 mm
3
.
3.3. Theoretical efciency
FromEqs. (1) and (2), a theoretical efciency of power recovery
is dened as h
th
.
h
th
h
ind;th
h
v;th
(3)
Fig. 6 shows the theoretical efciency of power recovery
with a solid line against the inlet temperature. The theoretical
indicated efciency and the theoretical volumetric efciency
are also plotted in Fig. 6 with a broken line and a dotted line,
respectively. The inuence of the top clearance on the theo-
retical volumetric efciency is relatively small, while the
theoretical indicated efciency and the theoretical efciency
of power recovery decrease with the inlet temperature.
Fig. 3 e Experimental cycle.
Table 2 e Experimental conditions.
Inlet pressure Pin 8.5 0.1 MPa
Discharge pressure Pout 3.2 0.1 MPa
Inlet temperature Tin 26 1.0

C
Rotational speed 180e400 rpm
Fig. 4 e Theoretical PeV diagram of radial piston expander
and isentropic expansion PeV diagram.
i nt e r na t i ona l j o ur na l o f r e f r i g e r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 51
However, since the inlet temperature of the expander, i.e.
outlet temperature of a gas cooler, in case of a CO
2
cycle for
water heater is not so high, the theoretical efciency of power
recovery by the expander is acceptable even if its build-in
volume ratio is 1.0.
4. Expander performance
4.1. Denition of efciencies
The performance of the radial piston expander is examined
based on the mass ow rate through the expander, the PeV
diagram and the output power. The performance is expressed
by efciencies dened as follows.
The theoretical mass ow rate G
th
is given by Eq. (4).
G
th
4n
__
V
st
V
top
_
r
in
V
top
r
out

(4)
where n is the rotational speed. The ratio of the theoretical
ow rate to an experimental ow rate, G
exp
, is dened as
volumetric efciency, h
v
.
h
v

G
th
G
exp
(5)
PeV expansion work is obtained by integrating the area
enveloped by the PeV diagram. From the ratio of the PeV
works obtained theoretically (W
th
) and experimentally (W
exp
),
an indicated efciency, h
ind
, is dened as follows.
h
ind

W
exp
W
th
(6)
Atheoretical shaft torque is calculated geometrically based
on the pressure change inside the each cylinder. An instan-
taneous torque T
i
(q) generated by one piston with a rotational
angle, q, is expressed as follows.
T
i
q r
_
_
P
cyl
P
out
_
$
_
pd
2
4
__
$
_
_
_sin q
r
l
sin2q
2

1
_
r
l
sin q
_
2
_
_
_
_ (7)
where q is the rotational angle of the crank arm from TDC,
d is a piston diameter, r is a length of crank arm and l is a
length of connecting plate. The shaft torque is obtained by
adding the torques for all cylinders and averaging it in one
rotation. The ratio of an experimental shaft torque, T
exp
,
measured by the torque meter to the theoretical torque, T
th
,
obtained from the pressure change in the cylinder is a me-
chanical efciency, h
m
.
h
m

T
exp
T
th
(8)
Consequently, a total efciency, h
t
, is dened as the ratio of
the measured power to power that is theoretically obtainable
with the theoretical ow rate by this expander.
h
t

T
exp
u
nW
exp
Gexp
G
th

G
th
G
exp
nW
th
nW
exp
T
exp
u
T
th
u
h
v
h
ind
h
m
(9)
Fig. 5 e Inuence of top clearance on volumetric efciency.
Fig. 6 e Theoretical efciency of power recovery,
theoretical indicated efciency and theoretical volumetric
efciency.
Fig. 7 e Experimental PeV diagram (Rotational speed is
270 rpm.).
i nt e r na t i o na l j o ur na l of r e f r i ge r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 52
It should be noted that the total efciency expresses the
performance of this expander and does not include the
theoretical efciency of power recovery dened by Eq. (3).
4.2. Experimental results
Fig. 7 shows the PeV diagram under the condition that the
inlet pressure is 8.4 MPa(abs), the outlet pressure is
3.2 MPa(abs), the inlet temperature is 26.3

C and the rota-
tional speed is 270 rpm. Solid line shows the experimental
PeV diagramand broken line shows the theoretical one. It can
be seen that there is the pressure loss at supply process during
the piston moves from TDC to BDC and the ow resistance
becomes larger when the refrigerant inside the cylinder be-
comes two-phase condition. Although the experimental PeV
expansion work is smaller than the theoretical one, the dif-
ference between them is relatively small and the indicated
efciency is 0.87 under this condition.
The pressure change inside the cylinder is shown against
the rotational angle with broken line in Fig. 8, while the
theoretical torque calculated from the pressure change and
the experimental torque are plotted with dotted line and solid
line respectively. The operating condition is the same as that
of Fig. 7. The difference between the theoretical torque and
the experimental one is caused by the mechanical loss mainly
due to the O-ring at the piston and the shaft seals. The me-
chanical efciency in this case is 0.55.
Fig. 9 shows the experimental mass ow rate through the
expander (,) and the theoretical mass ow rate (A) against
the rotational speed. The volumetric efciency (B) calculated
from both ow rate is plotted as well. It is found that the
expander developed in this study can be operated with small
mass ow rate, and the minimum ow rate in this study is
about 76 g min
1
(4.6 kg h
1
) at 180 rpm. Although a clearance
between the piston and the cylinder is sealed by O-ring, a
certain leakage exists at the piston O-ring, since squeeze of
the piston O-ring is set to be smaller than general to reduce
Fig. 8 e Pressure and torque change against rotational
angle (rotational speed is 270 rpm.).
Fig. 9 e Mass ow rates and volumetric efciency.
Fig. 10 e PeV works and indicated efciency.
Fig. 11 e Output powers and mechanical efciency.
i nt e r na t i ona l j o ur na l o f r e f r i g e r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 53
the mechanical loss. The difference betweenthe experimental
ow rate and the theoretical one is, therefore, caused by the
leakage. The amount of leakage is almost constant regardless
of the rotational speed, which results in an increase of the
volumetric efciency with increasing the rotational speed
since the inuence of the leakage on the mass ow rate be-
comes small at higher rotational speed. The volumetric ef-
ciency increases from 0.7 to 0.9 with the increase of rotational
speed.
Fig. 10 shows the expansion work obtained from the
experimental PeV diagram (,), the theoretical work (A) and
the indicated efciency (B) calculated by Eq. (6). Both the
experimental and the theoretical work are almost the same at
each rotational speed. The indicated efciency is approxi-
mately 0.9 over the range tested in this study except for an
extremely low rotational speed of 180 rpm. When the rota-
tional speed is less than 200 rpm, the expander cannot be
operated stationary.
The experimental shaft power (,) and the theoretical one
(A) derived as the product of the torque and the angular ve-
locity are shown in Fig. 11 as well as the mechanical efciency
(B). When the rotational speed increases, friction losses at the
shaft seal and the piston seal become large. Difference be-
tween the theoretical and the experimental power increases
and, therefore, the mechanical efciency decreases with the
rotational speed. To improve the mechanical efciency, the
mechanical loss is needed to reduce.
Fig. 12 shows the total efciency against the rotational
speed. The total efciency is a product of three efciencies as
expressed by Eq. (9) and is about 0.4 over the rotational speed
tested in this study. This value is relatively high for such small
expander.
4.3. Mechanical loss
From the performance discussed above, it is found that the
mechanical loss is a dominant factor which reduces the per-
formance. Therefore, the frictional losses at the piston seal
and the shaft seal are measured individually.
An experimental setup is shown in Fig. 13. The expander is
connected to the torque meter and the rotational torque by
the frictional loss is measured with or without the O-ring at
the piston and the crank shaft.
Fig. 14 shows the rotational torque caused by the frictional
loss by the O-ring in the cases without the shaft seal and the
piston seal, with the O-ring only at the piston and with the O-
ring only at the shaft seal. The torque is measured under
Fig. 12 e Total efciency.
Fig. 13 e Experimental set up for measurement of
mechanical loss.
Fig. 14 e Mechanical losses at each portion.
Fig. 15 e Frictional torque against inside pressure.
i nt e r na t i o na l j o ur na l of r e f r i ge r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 54
atmospheric pressure. The torque without the O-ring is
caused by the friction at the bearing and a viscous force of oil,
and is negligible compared with other losses. As compared the
piston seal with the shaft seal, the frictional loss at the shaft
seal is greater than that at piston seal. This is because the
squeeze of the piston O-ring is set to be small to reduce the
mechanical loss as mentioned previously.
The friction by the O-ring may change according to pres-
sure difference across the O-ring. The O-ring is equipped only
at the shaft seal and the rotational torque by the shaft seal is
measured with pressurizing inside the expander. The fric-
tional torque is plotted against the inside pressure in Fig. 15.
The frictional torque obtained when the trust force acting on
the shaft is supported by the radial bearing without the thrust
bearing is also plotted in Fig. 15. The frictional torque in-
creases with the inside pressure and the thrust bearing is
effective to reduce the frictional loss. Although the frictional
torque at the shaft seal is large, this loss can be eliminated if
the expander and a generator are combined in a shell.
5. Conclusions
In this study, a novel reciprocating expander which has four
cylinder arranged radially was developed and the perfor-
mance of the expander is investigated. Conclusions obtained
in this study are summarized as follows.
(1) The radial piston type expander developed here can be
operated adequately with small mass ow rate. It is,
therefore, available in a small capacity refrigeration
cycle.
(2) Since built-in volume ratio of the expander is 1.0, it does
not have an expansion process. However, an expansion
work in two-phase region which is not recovered is
small in a CO
2
refrigeration cycle. Especially when inlet
temperature, i.e. temperature at a gas cooler exit, is low
in case of a water heater, a theoretical efciency of
power recovery of the expander is acceptable.
(3) In an operating range tested in this study, the indicated
efciency is 0.8e0.9, the volumetric efciency increases
from 0.7 to 0.9 as a rotational speed increases and the
mechanical efciency decreases from 0.6 to 0.5 with an
increase of the rotational speed. Consequently, the total
efciency of the expander is about 0.4 over the range of
rotational speed. This value is relatively good as such
small expander.
(4) It is found that mechanical loss caused by O-rings is a
dominant factor which reduces the performance of the
expander. In order to improve the performance, there-
fore, it is necessary to reduce the frictional loss at the
seal devices.
r e f e r e n c e s
Baek, J.S., Groll, E.A., Lawless, P.B., 2002. Development of a Piston-
Cylinder Expansion Device for the Transcritical Carbon
Dioxide Cycle, pp. 457e466. Proceedings of the 9th
International Refrigeration and Conditioning Conference at
Purdue.
Baek, J.S., Groll, E.A., Lawless, P.B., 2005a. Piston-cylinder work
producing expansion device in a transcritical carbon dioxide
cycle. Part I: experimental investigation. Int. J. Refrigeration
28, 141e151.
Baek, J.S., Groll, E.A., Lawless, P.B., 2005b. Piston-cylinder work
producing expansion device in a transcritical carbon dioxide
cycle. Part II: theoretical model. Int. J. Refrigeration 28,
152e164.
Fukuta, M., Yanagisawa, T., Higashiyama, M., Ogi, Y., 2009.
Performance of vane-type CO
2
expander and characteristics of
transcritical expansion process. HVAC&R Res. 15, 711e727.
Fukuta, M., Yanagisawa, T., Kosuda, O., Ogi, Y., 2006. Performance
of scroll Expander for CO
2
Refrigeration Cycle, p. C109.
Proceedings of the 18th International Compressor Engineering
Conference at Purdue.
Fukuta, M., Yanagisawa, T., Ogi, Y., Radermacher, R., 2001. Cycle
Performance of CO
2
Cycle With Vane Compressor-Expander
Combination, pp. 315e324. Proceedings of the International
Conference on Compressors and Their Systems.
Heyl, P., Quack, H., 1999. Free Piston ExpandereCompressor for
CO
2
e Design, Applications and Results, pp. 2612e2619.
Proceedings of the 20th International Congress of
Refrigeration.
Hiwata, A., Ikeda, A., Morimoto, T., Kosuda, O., Matsui, M., 2009.
Axial and radial force for a CO
2
scroll expander. HVAC&R Res.
15, 759e770.
Huff, H., Radermacher, R., Preissner, M., 2003. Experimental
Investigation of a Scroll Expander in a Carbon Dioxide air-
Conditioning System. Proceedings of the 21st International
Congress of Refrigeration.
Jia, X., Zhang, B., Yang, B., Peng, X., 2009. Study of a rotary vane
expander for the transcritical CO
2
cycle e part II: theoretical
modeling. HVAC&R Res. 15, 689e709.
Kakuda, M., Nagata, H., Ishizono, F., 2009. Development of a scroll
expander for the CO
2
refrigeration cycle. HVAC&R Res. 15,
771e783.
Kim, H.J., Ahn, J.M., Cho, S.O., Cho, K.R., 2006. Numerical
Simulation on Scroll Expander-Compressor Unit for CO
2
Trans-critical Cycles, p. C104. Proceedings of the 18th
International Compressor Engineering Conference at Purdue.
Kohsokabe, H., Koyama, M., Tojo, K., Matsunaga, M.,
Nakayama, S., 2008a. Performance Characteristics of Scroll
Expander for CO
2
Refrigeration Cycles, p. 1239. Proceedings of
the 19th International Compressor Engineering Conference at
Purdue.
Kohsokabe, H., Funakoshi, S., Tojo, K., Nakayama, S., Kohno, K.,
Kurashige, K., 2008b. Operating Characteristics of CO
2
Chiller
Cycles with Expander-compressor Unit. Proceeding of the 8th
IIR Gustav Lorentzen Conference on Natural Working Fluids.
Kohsokabe, H., Funakoshi, S., Tojo, K., Nakayama, S., Kohno, K.,
Kurashige, K., 2006. Basic Operating Characteristics of CO
2
Refrigeration Cycles With Expander-Compressor Unit, p. R159.
Proceedings of the 11th International Refrigeration and Air-
conditioning Conference at Purdue.
Kruse, H., RU

SSMANN, H., Martin, E., Jakobs, R., 2006. Positive


Displacement Carbon Dioxide Expansion Machines e Chances
and Limitations, p. C138. Proceedings of the 18th International
Compressor Engineering Conference at Purdue.
Lorentzen, G., 1995. The use of natural refrigerants: a complete
solution for the CFC/HCFC predicament. Int. J. Refrigeration
18, 190e197.
Matsui, M., Hasegawa, H., Ogata, T., Wada, M., 2009. Development
of the high-efciency technology of a CO
2
two-stage rotary
expander. HVAC&R Res. 15, 743e758.
Nickl, J., Will, G., Kraus, W.E., Quack, H., 2002. Design
Considerations for a Second Generation CO

2
Expanders,
i nt e r na t i ona l j o ur na l o f r e f r i g e r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 55
pp. 189e196. Proceeding of the 5th IIR Gustav Lorentzen
Conference on Natural Working Fluids.
Nickl, J., Will, G., Kraus, W.E., Quack, H., 2005. Integration of a
three-stage expander into a CO
2
refrigeration system. Int. J.
Refrigeration 28, 1219e1224.
Okamoto, M., Kumakura, E., Okamoto, T., Sakitani, K.,
Moriwaki, M., 2005. Development of Two-Phase Flow
Expander for CO
2
Air-Conditioners. Proceedings of the 2005
JSRAE Annual Conference, C315-1-4 (Japanese).
Robinson, D.M., Groll, E.A., 1998. Efciency of transcritical CO
2
cycle with and without an expansion turbine. Int. J.
Refrigeration 21, 577e589.
Subiantoro, A., Ooi, K.T., 2009. Introduction of the revolving vane
expander. HVAC&R Res. 15, 801e816.
Subiantoro, A., Ooi, K.T., 2012a. Analysis of the revolving vane
(RV-0) expander, part 1: experimental investigations. Int. J.
Refrigeration 35, 1734e1743.
Subiantoro, A., Ooi, K.T., 2012b. Analysis of the revolving vane
(RV-0) expander, part 2: verications of theoretical models.
Int. J. Refrigeration 35, 1744e1756.
Yang, B., Peng, X., Sun, S., Guo, B., Xing, Z., 2009. Study of a rotary
vane expander for the transcritical CO
2
cycle e part I:
experimental investigation. HVAC&R Res. 15, 673e688.
Zhang, B., Peng, X., He, Z., Xing, Z., Shu, P., 2007. Development of
a double acting free piston expander for power recovery in
transcritical CO
2
cycle. Appl. Therm. Eng. 27, 1629e1636.
i nt e r na t i o na l j o ur na l of r e f r i ge r a t i o n 4 2 ( 2 0 1 4 ) 4 8 e5 6 56