Cryogenics 95 (2018) 69–75

Contents lists available at ScienceDirect

Cryogenics
journal homepage: www.elsevier.com/locate/cryogenics

Research paper

Experimental study of a cascade pulse tube cryocooler with a displacer T

a,b a,⁎ a,b a a,⁎ a
Jingyuan Xu , Jianying Hu , Jiangfeng Hu , Limin Zhang , Ercang Luo , Bo Gao
a
CAS Key Laboratory of Cryogenics, Technical Institute of Physics and Chemistry, Beijing 100190, China
b
University of Chinese Academy of Sciences, Beijing 100049, China

A R T I C LE I N FO A B S T R A C T

Keywords: Recovering the expansion power in pulse tube cryocooler is of great utility in improving cooling efficiency. Using
Pulse tube cryocooler a second-stage cooler after a primary cooler to produce extra cooling power is an effective way especially when
Displacer the cooling temperature is not very low. In the configuration, the two coolers are connected by a displacer which
Power recovery is used as a phase shifter. In this paper, experimental investigations were conducted to study this system. Firstly,
Cascade cryocooler
the performance of the overall system and separated cooler was respectively presented. To better understand the
displacer, phase relation, mechanical resistance and displacement were then clarified. In addition, the power
consumption distribution of the cascade cryocooler was discussed. Finally, both numerical and experimental
comparisons were made on the displacer-type and tube-type cryocooler. The experimental results show that the
displacer-type cryocooler has superior performance due to the better phase-modulation capability and less
power loss. With the input electric power of 1.9 kW and cooling temperature of 130 K, the overall system
achieved a cooling power of 371 W and a relative Carnot efficiency of 24.5%.

1. Introduction implemented by Gan et al. when they realized it was hard to apply the
Swift’s configuration in practice. When compared with a single-stage
No moving components at the cold head endows Stirling-type pulse SPTC, 33% increase in cooling efficiency was achieved at 233 K [9,10].
tube cryocooler (SPTC) with remarkable advantages of high reliability A cascade SPTC using a displacer as a phase shifter was proposed
and low vibration, which shows great prospects in applications such as and designed by our group [11]. The previous work mainly focused on
cooling infrared sensor, high temperature superconductivity, and small- cryocooler design and theoretical analysis, so only a few preliminary
scale gas liquefier. However, when compared with Stirling cryocooler experiment was conducted at a quite limited working condition which
which has a theoretical efficiency of Carnot efficiency, the intrinsic is aimed at verifying the feasibility of configuration; Furthermore, the
lower efficiency is an obstacle for the extensive usage of the SPTC. This compressor design has not been considered in the previous work, as it
is because an inertance tube is always employed in the SPTC at the only concerned on performance of the cryocooler; Finally, there is a
expense of transferring all the expansion work into a waste heat. For lack of theoretical and experimental comparisons with tube-type cas-
this reason, recovering the expansion power is of great help to improve cade cryocooler. For these reasons, this paper presents the detailed
the cooling efficiency. experimental results of the overall system at different working condi-
There are two feasible methods to recovering the expansion power. tions and compares results with tube-type cryocooler. Firstly, the
One approach, which has been widely studied, is to recover the power compressor was carefully designed and the system performance is
from a feedback loop between the cryocooler and the compressor [1–7]. presented. The phase relationship, mechanical resistance and dis-
This configuration is endowed with a theoretical efficiency of Carnot placement of the displacer are then investigated. Furthermore, power
efficiency. Another approach, which lacks sufficient study, is to use a consumption distribution of the cryocooler is presented. Finally, com-
cascade SPTC. In this configuration, the expansion work is recovered parison is made between the cascade cryocoolers using two types of
into the cascading cooler to produce extra cooling power. The Carnot phase shifters.
efficiency is theoretically obtained when cascading infinite stages. In
2011, a two-stage cascade SPTC, with a secondary cooler directly
connected after a quarter-wavelength primary pulse tube, was proposed
and demonstrated by Swift et al. [8]. In 2015, an improved version was

⁎
Corresponding authors.
E-mail addresses: jyhu@mail.ipc.ac.cn (J. Hu), ecluo@mail.ipc.ac.cn (E. Luo).

https://doi.org/10.1016/j.cryogenics.2018.09.006
Received 18 April 2018; Received in revised form 24 July 2018; Accepted 12 September 2018
Available online 14 September 2018
0011-2275/ © 2018 Published by Elsevier Ltd.
J. Xu et al. Cryogenics 95 (2018) 69–75

Fig. 1. A two-stage cascade pulse tube cryo-

cooler: (a) schematic drawing (b) experimental
setup. where inertance tube and gas reservoir are
absent in this figure. (b). We is electrical input
power to the compressor Wa is acoustic power
out from the compressor, Wb is acoustic power
out from the 1st cooler, Wc is acoustic power out
from the displacer, Tc1 and Tc2 are cooling tem-
peratures of 1st cooler and 2nd cooler, P1 to P5
are pressure waves at the back space of com-
pressor, the inlet of the 1st cooler, the expansion
space of displacer, the back space of displacer
and the compression space of displacer.

Table 1 Table 2
Detailed geometries of the coolers. Main parameters of the displacer.
Cooler Component Diameter Length Other dimension Piston Diameter Expansion piston 70 mm
Compression piston 43 mm
First-stage MAHX 75 64 26% in porosity, 0.4 in channel
width Spring Constant Mechanical Spring 27 kN/m (Axial)
REG 70 300-mesh stainless-steel screen 4014 kN/m (Radial)
CHX 30 18% in porosity, 0.25 in Gas Spring 33 kN/m
channel width
Space Volume Expansion space 46 cm3
PT 37 156 1.2 in wall thickness
Back space 890 cm3
FS 3 40-mech copper screen
Compression space 43 cm3
Second- MAHX 40 40 20% in porosity, 0.4 in channel
Moving Mass 2.3 kg
stage width
REG 60 300-mesh stainless-steel screen
CHX 25 10% in porosity, 0.25 in
channel width cascade SPTC. The dimensions of the two coolers are listed in Table 1.
PT 22 80 0.8 in wall thickness In the linear compressor, the working gas oscillates due to the re-
FS 3 40-mech copper screen ciprocating motion of the pistons, thus electric power is converted into
acoustic power; The input acoustic power is firstly consumed by the
Listed parameters are in millimeters.
first-stage cooler to obtain primary cooling power; The expansion
power from the first-stage cooler is then transferred to the second-stage
cooler to produce the secondary cooling power; The expansion power
from the second-stage cooler is not recovered. The displacer and in-
ertance tube are used to modulate the phase relation in the cryocooler.
Fig. 2 shows the profile of the displacer. Six springs are fixed by
screw bolts. To ensure sufficient space for deformation of the spring,
there is a flat gasket between two springs. The connecting rod is con-
nected to the springs as well as two pistons, thus undergoing same
oscillation. The main parameters of the displacer are listed in Table 2.
Clearance seal between the piston and the cylinder wall is 0.018 mm.
The natural frequency of the displacer is 15.6 Hz, which deviates far
from the working frequency of 55 Hz. The piston displacement is lim-
ited within 6 mm according to the stress analysis of the springs. The
total axial spring constant is simultaneously produced by mechanical
spring and gas spring from the back space. The gas spring constant is
2
deduced by γ PmA (γ is the specific heat ratio, Pm is the mean pressure, A
VDB
Fig. 2. Profile of the displacer. is the area difference on two piston sides and VDB is the back space
volume of the displacer).
2. Experimental system
2.2. Experimental measurements
2.1. Experimental setup
The measuring points of oscillating pressure and temperature are
Fig. 1 shows a schematic drawing and experimental setup of a shown in Fig. 1. Five high-precision dynamic pressure sensors produced

70
J. Xu et al. Cryogenics 95 (2018) 69–75

by PCB Piezotronics were used to measure the oscillating pressures in T0

ηcas _cooler = (Qc1 + Qc 2)( −1)/ Wa
the back space of compressor (P1), the inlet of the first-stage cooler (P2), Tc (11)
the inlet of the displacer (P3), the back space of the displacer (P4), the
T0
outlet of the displacer (P5). Two calibrated platinum resistance ther- η1st _cooler = Qc1 ( −1)/(Wa−Wb)
Tc (12)
mometers (PT100) with an accuracy of ± 0.1 K were installed at cold
heads to measure the cooling temperatures of the first-stage cooler (Tc1) T0
η2nd _cooler = Qc 2 ( −1)/ Wc
and the second-stage cooler (Tc2). The mean pressure was measured by Tc (13)
a pressure transmitter with an accuracy of 0.5%. High-precision voltage
and current sensors supplied by Tektronix were used to measure the where Qc1, Qc2, T0 and Tc are the cooling capacities of the first-stage and
voltage and current of the compressor. Based on the thermal balance, second-stage coolers, ambient temperature and cooling temperature,
cooling power was simulated by the heating power on constantan wires respectively.
mounted on the cold heads. In the experiment, the ambient heat ex- The overall relative Carnot efficiency ηoverall calculated with input
changer and inertance tube were cooled by cooling water. The re- electric power as the reference is defined as:
generator, cold head, and pulse tube were enclosed in a vacuum T0
ηoverall = ηcom ·ηcas _cooler = (Qc1 + Qc 2)( −1)/ We
chamber to decrease the ambient losses. Tc (14)
Some crucial parameters need to be deducted from the measured
parameters. The volume flow rate at the compressor’s piston Ucom and
3. Experimental results of the cryocooler
the difference of volume flow rate between two displacer’s piston sides
ΔUdis are expressed as:
In the experiment, the mean pressure was 3 MPa, system frequency
iωVcom was 55 Hz, temperature of cooling water was 293 K for two ambient
Ucom = p
γPm 1 (1) heat exchangers, and the cooling temperatures of two cold heads were
set to be the same.
iωVdis
ΔUdis=− p
γPm 4 (2) 3.1. System performance
where ω is the angular frequency, Vcom is the volume of the back space
of the compressor, Vdis is the volume of the back space of the displacer, Fig. 3 shows the overall system performance at natural gas lique-
Pm is the mean pressure, P1 and P4 are the pressure waves shown in faction range. The overall system has a higher efficiency and larger
Fig. 1, and γ is the specific heat ratio. cooling capacity with the increased input electric power. When in-
For the displacer, ΔUdis can be also expressed as: creasing the electric power to 1.9 kW, an overall efficiency of 24.5%
were achieved at cooling temperature of 130 K. Fig. 4 shows theoretical
ΔUdis = ω (AC −AE ) Xdis i (3) and experimental results of the cascade cryocooler performance. The
where AE and AC represent the expression piston area and compression highest cryocooler efficiency of 37.3% and a cooling power of 330 W
piston area, Xdis represents the displacement of the displacer. are achieved at 130 K. The experimental results are qualitatively in
Combining Eqs. (2) and (3), Xdis can be obtained as: agreement with the calculated results although they are a little quan-
titatively smaller than them. Taking a deeper look at the performance of
P4 Vdis the compressor in Fig. 5, the compressor efficiency ranges from 60% to
Xdis =
(AE −AC ) γPm (4) 70%. Fig. 6 shows the efficiency of the 1st and 2nd cooler at 130 K. The
1st and 2nd coolers showed the similar trend with the increased input
The volume flow rate at the inlet of the displacer and at the outlet of
electric power. When compared with the 2nd cooler, the 1st cooler had
the displacer can be written as:
a much higher relative Carnot efficiency as a consequence of power
ωAE P4 Vdis recovery. Taking an input electric power of 1.7 kW as an example, the
Udis _in = ωAE Xdis i = i
(AE −AC ) γPm (5) relative Carnot efficiency for the 1st cooler was 46.1% while that for the
2nd cooler was 23.7%.
ωAC P4 Vdis
Udis _out = ωAC Xdis i = i Fig. 7 shows the cooling-power recovery ratio under different
(AE −AC ) γPm (6) cooling temperatures and input electric power. Here, the cooling-power
As shown in Fig. 1, Wa, Wb and Wc are calculated as: recovery ratio is defined as the ratio of the cooling power of the 2nd
cooler to that of the 1st cooler. Higher cooling temperature was helpful
1 to obtain higher cooling-power recovery ratio: for the cooling
Wa = |p ||Ucom | cosθpU
2 2 (7)

1 26%
Wb = |p ||Udis _in | cosθpU 110 K
2 3 (8) 25% 130 K
150 K
1 24%
Wc = |p ||Udis _out | cosθpU
2 5 (9)
23%
overall

where θpU is the phase difference between the pressure wave and the
volume flow rate. 22%
We define several efficiency to present the performance of different 21%
assemblies. The compressor electric-to-acoustic efficiency ηcom is de-
fined as: 20%

ηcom = Wa/ We (10) 19%

800 1200 1600 2000 2400
where We is the input electric power to the compressor. We(Watts)
The cascade cryocooler relative Carnot efficiency ηcas_cooler, the 1st
cooler relative Carnot efficiency η1st_cooler and the 2nd cooler relative Fig. 3. Overall system efficiency under different cooling temperature and input
Carnot efficiency η2nd_cooler are defined as: electric power (Experiment).

71
J. Xu et al. Cryogenics 95 (2018) 69–75

42%
400
40%
350
38%

(Qc1+Qc2) /W
300

cas_cooler
36%
250
110 K-Experiment 110 K-Experiment
34% 110 K-Simulation 200 110 K-Simulation
130 K-Experiment 130 K-Experiment
32% 130 K-Simulation
150
130 K-Simulation
150 K-Experiment
150 K-Experiment
150 K-Simulation 150 K-Simulation
30% 100
400 600 800 1000 1200 1400 1600 400 600 800 1000 1200 1400 1600
Wa(Watts) Wa (Watts)
(a) Relative Carnot efficiency (b) Overall cooling power
Fig. 4. Performance of the cascade cryocooler under different cooling temperature and input acoustic power (Experiment & Simulation).

70% 25%
110K
68% 130K
150K
66% 20%

Qc2/Qc1
com

64%

62% 15%
110 K
60% 130 K
150 K

58% 10%
800 1200 1600 2000 2400 800 1200 1600 2000
We(Watts) We(Watts)
Fig. 5. Efficiency of the compressor (Experiment). Fig. 7. Cooling-power recovery ratio (Experiment).

50% 1.24
Tc=130 K st
1 cooler
nd
2 cooler
40%
Pressure ratio

1.20
1 /2 _cooler

st
1 cooler
nd
30% 2 cooler
nd
st

1.16

20% Tc=130 K

1.12
800 1000 1200 1400 1600 1800 2000 800 1000 1200 1400 1600 1800 2000
We(Watts) We(Watts)
Fig. 6. Efficiency of the 1st cooler and 2nd cooler (Experiment). Fig. 8. Pressure ratio at the inlet of the two coolers (Experiment).

temperatures of 150 K, 130 K and 110 K, the highest cooling-power oscillation; while adding with the expansion- and compression-spaces,
recovery ratio were 21.9%, 18.2%, 14.7% respectively. Fig. 8 shows the phase of the volume flow rate also changes. Fig. 9 shows the
pressure ratio at the inlets of the 1st cooler and 2nd cooler. The pressure pressure waves at two sides of the displacer. The pressure wave phase
ratio increased with the increased input electric power. When the input changes 106.8 degree under the working condition. Fig. 10 shows the
electric power was 1.9 kW, the pressure ratios for two coolers exceeded phase difference between the pressure wave and volume flow rate at
1.225. two sides of the displacer. The reversed phase relation verifies that the
displacer modulates the phase relation, which makes it possible for the
3.2. Phase relation at the displacer efficient power recovery.

In an efficient pulse tube cryocooler, the pressure wave always lags 3.3. Mechanical resistance and displacement of the displacer
the volume flow rate at the inlet of the cooler, contrary to the phase
relationship at the outlet of the pulse tube. So, it is necessary to employ The displacer is a crucial component not only for phase shifting but
a phase shifter while coupling two coolers. As for the two-sides at also for power transmission. To have less power loss, the mechanical
displacer piston, only the phase of pressure wave changes while the resistance of the displacer should be low. Actually, the mechanical re-
phase of volume flow rate remains the same due to the same piston sistance is determined by the available technique and processing level,

72
J. Xu et al. Cryogenics 95 (2018) 69–75

4 6.0
Inlet of displacer 110 K
Outlet of displacer 130 K
5.5 150 K
Pressure wave/ bar

Xdis / mm
5.0
0
4.5

-2
4.0
106.8° Wa=1110 W
Tc=130 K
-4
0.012 0.016 0.020 0.024 0.028 600 900 1200 1500 1800 2100
Time (Seconds) We (Watts)

Fig. 9. Pressure waves at two ends of the displacer (Experiment). Fig. 12. Displacement of the displacer (Experiment).

90 Compressor
Tc=130 K
f =55 Hz st
1 Cooler
60 We=1.7 kW Displacer
o
Phase difference/

nd
Tc=130 K 2 Cooler
30 43.22%
T0=293 K 3.4%
0
Inlet of displacer 18.66%
Outlet of displacer
-30

-60
800 1000 1200 1400 1600 1800 2000 34.72%
We (Watts)
Fig. 13. Experimental results of power consumption distributions of the cryo-
cooler (Experiment).
Fig. 10. Phase difference at two ends of the displacer (Experiment).

3.4. Power consumption distribution

40
110 K_Mechanical Resistance 99%
130 K_Mechanical Resistance
35 150 K_Mechanical Resistance
96%
Fig. 13 shows the power consumption ratios of the compressor, 1st
cooler, the displacer and the 2nd cooler at the input electric power of
30 93% 1.7 kW, frequency of 55 Hz and cooling temperature of 130 K. Herein,
Rmech / kg/s

the power consumption ratio is defined as a ratio of the consumed

25 90%
dis

power to the input electric power. The compressor and the 1st cooler
20 87% which produced the primary cooling power was the most significant
contributors to the power consumption. The displacer took a small
15 110 K_Displacer Efficiency 84%
130 K_Displacer Efficiency portion of power consumption, so it was efficient in power transmis-
150 K_Displacer Efficiency 81% sion.
10
600 900 1200 1500 1800 2100
We (Watts)
4. Comparison with the tube-type cascade cryocooler
Fig. 11. Mechanical resistance and power loss ratio of the displacer
(Experiment). In a cascade cryocooler, the phase shifter between two adjacent
coolers plays important roles in both phase shifting and power trans-
mission. To obtain high efficiency, it is necessary to have good phase-
which is hard to estimate in advance. Fig. 11 shows the displacer effi-
modulation ability as well as low power loss. In the previous studies
ciency ηdis (ratio of Wc to Wb) and mechanical resistance Rmech (cal-
2(W − W ) [8–10], as shown in Fig. 14, long tube was usually employed between
culated by Rmech = (ωXb )2c ) of the displacer. As the input power in- two coolers as a phase shifter. This configuration maintains the ad-
dis
creasing, mechanical resistance increased within a small range from vantage of no moving parts and high reliability. In this section, both
23 kg/s to 35 kg/s. The displacer efficiency is within the ranges of numerical and experimental investigations are conducted to compare
83–88%, which means it is efficient to transmit acoustic power. Fig. 12 the displacer-type cryocooler with such tube-type cryocooler.
shows the displacer displacement under different cooling temperatures The comparison is based on the same cooler dimensions and same
and input acoustic power. Increasing trend of the displacement was working conditions. That is to say, the mean pressure was 3 MPa,
observed at the increased input acoustic power and cooling tempera- system frequency was 55 Hz, temperature of cooling water was 293 K
tures due to the larger transferred power. To limit the piston dis- for two ambient heat exchangers, and the cooling temperatures of two
placement, the input acoustic power should be controlled at higher cold heads were 130 K. The input acoustic power to the cryocooler is
cooling temperature. If the displacement is beyond allowable value, the 2 kW. Since the tube-type cryocooler has different coupling character-
stress will be too large to break the spring. In the experiment, the dis- istic with the compressor because of its different acoustic impedance,
placer displacement was limited below 6 mm. our major comparison target was the performance of the cascade

73
J. Xu et al. Cryogenics 95 (2018) 69–75

Fig. 14. Schematic of two-stage tube-type cascade cryocooler.

650 628 W
42%
Dtube=14 mm Dtube=14 mm
Displacer-type 39.4 %
600 Dtube=16 mm
39% Dtube=16 mm Displacer-type
Dtube=18 mm Dtube=18 mm
(Qc1+Qc2) /W

Dtube=20 mm Tube-type
550 36% Dtube=20 mm Tube-type

cas-cooler
33%
500
30%
450
27%
400
24%
5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5
Ltube (Meters) Ltube(Meters)

(a) overall cooling power (b) Efficiency of cascade cryocooler

Fig. 15. Tube-type cascade cryocooler performance under different tube dimensions (Simulation). Input acoustic power is 2 kW, frequency is 55 Hz, cooling tem-
peratures for both coolers are 130 K. Asterisk is for the optimal performance of displacer-type cryocooler).

Table 3
Performance of two types of cryocoolers (Simulation). Input acoustic power is
2 kW, frequency is 55 Hz, cooling temperatures for both coolers are 130 K.
Displacer-type Tube-type

Qc1 505 W 484 W

Qc2 123 W 66 W
Qc1 + Qc2 628 W 550 W
ηcas _cooler 39.4% 34.5%
η1st _cooler 46.3% 43.5%
η2nd_cooler 32.6% 32.6%
ΔWdis 170 W 346 W

cryocooler—not the coupling characteristics with the compressor. We

therefore optimized the dimensions of the long tube and inertance tube
with an aim of achieving high cascade-cryocooler efficiency under the
fixed input acoustic power. The optimization results are shown in
Fig. 15. For each tube diameter, there is a different tube length in terms
of achieving the same optimal efficiency. In our study, the tube di-
mension was chosen to be 18 mm in diameter and 7350 mm in length.
Taking the respective optimal dimensions of two phase shifters, the
comparison results are presented in Table 3. The displacer-type cryo-
cooler obviously has superior cooling performance, where the cooling
power and cryocooler relative Carnot efficiency were improved by
approximately 14%. From Table 3, the superior performance of the
displacer-type cryocooler is mainly attributed to two reasons: one
reason is that a displacer has better phase-modulation capability so that Fig. 16. Photo of the tube-type cascade cryocooler.
the 1st cooler has higher efficiency; a second reason is that the displacer
suffers less power loss. power of the 1st cooler than that of the tube-type. This verified that the
Experimental investigations were conducted to verify the simulation displacer-type cryocooler has better phase modulation capability.
results. Fig. 16 shows the experimental setup of the tube-type cascade Fig. 19 shows the power consumption ratio of two cryocooler types
cryocooler. The tube dimension was mentioned above, and the other under the typical conditions. As for the tube-type cryocooler, the
components were the same to the previous experiments. Fig. 17 shows acoustic power at the inlet and the outlet of the long tube is hard to
the experimental and theoretical performance of the tube-type cryo- directly obtain in the experiment, they were therefore calculated ac-
cooler. At the cooling temperature of 130 K, the highest relative Carnot cording to the cooler’s relative Carnot efficiency and the cooling power:
efficiency of the tube-type cryocooler was only 31.3%, which is much the power at inlet of the tube is deduced from the 1st cooler’s efficiency,
lower than 37.1% of the displacer-type. Taking a closer look at cooling the 1st cooling power and the input power to the 1st cooler; the power
power in Fig. 18, the displacer-type cryocooler achieved higher cooling at outlet of the tube is deduced from the 2nd cooler’s efficiency and the

74
J. Xu et al. Cryogenics 95 (2018) 69–75

38% 500

36%
400

(Qc1+Qc2) /W
cas_cooler
34% 110K-Experiment
110K-Simulation
130K-Experiment 300
32% 130K-Simulation
110K-Experiment
150K-Simulation
150K-Simulation 110K-Simulation
200 130K-Experiment
30% 130K-Simulation
150K-Simulation
150K-Simulation
28% 100
400 800 1200 1600 2000 2400 400 800 1200 1600 2000 2400
Wa (Watts) Wa (Watts)

(a) Relative Carnot efficiency (b) Qverall cooling power

Fig. 17. Overall cooling performance of the tube-type cascade cryocooler (Experiment & Simulation).

400 Displacer-type
recovery is experimentally studied. A displacer sandwiched by two
Qc1 Qc2 coolers acts both roles of phase modulation and power transmission.
Tube-type With the input electric power of 1.9 kW and cooling temperature of
Tc=130 K
300 130 K, it achieved an overall relative Carnot efficiency of 24.5% and
cooling power of 371 W. The compressor electric-to-acoustic efficiency
was 65.8% and cryocooler relative Carnot efficiency was 37.2%. The
Q c/ W

200 cooling-power recovery ratio was higher than 17% at 130 K and ex-
Displacer-type
ceeded 20% as increasing the cooling temperature to 150 K. Reversed
Tube-type
phase relation was observed in the displacer, which is mainly con-
100 tributed by the change of the pressure-wave phase. The power loss of
the displacer occupied only 4–6% of the input acoustic power, which is
efficient in transmitting acoustic power. When compared with the tube-
0 type cryocooler, the displacer-type cryocooler experimentally improved
580 W 1249 W the cooling efficiency by 20%. This is attributed by the better phase-
Wa modulation capability and less power loss of the displacer.

Fig. 18. Cooling power of two cryocoolers (Experiment). Acknowledgements

nd st
2 Cooler Displacer 1 Cooler This work was supported by the National Key Research and
Development Program of China (Contract No. 2017YFB0903603) and
Power comsumption ratio

Displacer-type Tube-type Displacer-type Tube-type

100%
National Natural Science Foundation of China (Grant Nos. 51627809,
51576204 and 51506211).
80%
References
60%
[1] Swift GW, Gardner DL, Backhaus SN. Acoustic recovery of lost power in pulse tube
40% refrigerators. J Acoust Soc Am 1999;105(2):711–24.
[2] Zhu SW. Step piston pulse tube refrigerator. Cryogenics 2014;64:63–9.
[3] Zhu SW. Four operation modes of a pulse tube machine with a step piston com-
20% pressor. Energy 2016;100:190–8.
Tc=130 K
[4] Zhu SW, Nogawa M. Pulse tube stirling machine with warm gas-driven displacer.
0% Cryogenics 2010;50(5):320–30.
580 W 1249 W [5] Wang XT, Zhang YB, Li HB, et al. A high efficiency hybrid stirling-pulse tube
Wa cryocooler. AIP Adv 2015;5:037127.
[6] Hu JY, Luo EC, Zhang LM, et al. A double-acting thermoacoustic cryocooler for high
temperature superconducting electric power grids. Appl Energy 2013;112:1166–70.
Fig. 19. Experimental results of power consumption ratio of two cryocoolers [7] Shi YZ, Zhu SW. Experimental investigation of pulse tube refrigerator with dis-
(Experiment). placer. Int J Refrig 2017;76:1–6.
[8] Swift GW, Gardner DL, Backhaus SN. Quarter-wave pulse tube. Cryogenics
2011;51(10):575–83.
2nd cooling power. This experimental results verified that the displacer [9] Wu M, Wang LY, Song YJ, et al. Study on a cascade pulse tube cooler with work
recovery. Phys Proc 2015;67:524–9.
is more efficient to transmit acoustic power.
[10] Wang LY, Wu M, Sun X, et al. A cascade pulse tube cooler capable of energy re-
covery. Appl Energy 2016;164:572–8.
5. Conclusions [11] Xu JY, Hu JY, Hu JF, et al. Cascade pulse-tube cryocooler using a displacer for
efficient work recovery. Cryogenics 2017;86:112–7.

A two-stage cascade pulse tube cryocooler capable of power

75