Anda di halaman 1dari 6

98

Performance Characterization of Sunpower Free-Piston


Stirling Engines

Seon-Young Kim,* James Huth,.†, and James G. Wood‡


Sunpower, Inc., Athens, Ohio, 45701

Over the past thirty years, Sunpower, Inc., the originator of the linear free-piston Stirling
engine (FPSE), has developed core Stirling engine modules for a variety of applications.
These range from domestic to military, terrestrial to space, and stationary to portable
applications, with outputs from tens of Watts to 7.5 kW. This paper will present the recent
results of a thermodynamic performance characterization study for three models of current
Sunpower free-piston Stirling engines: 35 We, 80 We and 1.1 kWe. The data was measured
over a wide range of acceptor and rejector temperatures, and power output levels.

I. Introduction

S UNPOWER, Inc. is the world leading developer of free-piston machines including Stirling engine, Stirling
coolers and cryocoolers, and linear compressors. Sunpower machines share many common features that enable
them to achieve high efficiency, reliable operation, and long life from a compact device using relatively simple
components. Sunpower has a successful history of commercialization of these machines.

Key technical features common to all Sunpower’s engines include:


• Non-contact gas bearings
• High-efficiency low-mass permanent magnet linear alternator
• Compliant displacer springing arrangement
• Planar spring for resonating the displacer
• Constant bore cylinder design
• Workspace diameter independent of displacer spring diameter

The main components of the Sunpower free-piston Stirling engine are shown in Fig. 1. The addition of heat to
the working fluid causes an increase in the average pressure of the gas in the engine. Conversely, the rejection of
heat causes a corresponding decrease in gas pressure. The displacer is attached to a planar spring forming a simple
resonant spring-mass system. The oscillating motion of the displacer shuttles gas back and forth from between
regions where heat is accepted or rejected by the working fluid. The resulting pressure wave acts upon the face of
the piston, causing the piston and attached ring of magnets to swing through the coil of wire within the alternator,
producing an AC current output from the engine.

*
Engineer

Project Engineer, AIAA Member Professional Grade

Senior Analyst, AIAA Member Professional Grade

Copyright © 2005 by Sunpower, Inc. 3rd International Energy Conversion Engineering Conference, 15-18 August 2005, San Francisco, California
Heat Heat
In Out Piston

Th Tc

Regenerator Planar Spring


Displacer Alternator
Figure 1. FPSE Components.

While Sunpower has designed and built many different models and sizes of free-piston Stirling engines1 2 3, this
paper will focus on three current models which have been developed for current programs (See Fig. 2): the EE-35,
EE-80, and EG-1000 with nominal outputs of 35 We, 80 We, and 1.1 kWe, respectively.

Figure 2. EG-1000, EE-80, and EE-35 Units.

EG-1000 FPSE This engine module was developed for a European domestic cogeneration appliance which
produces 1 kW of electric power in a home with a fuel-to-electric conversion efficiency over 20%. The heat
rejected from the engine is recovered to heat water for space heating, resulting in an overall fuel-to-combined heat
and electric power conversion efficiency of 90%. Summary performance data is listed in Table 1.

EE-35 FPSE The 35We unit was developed by Sunpower with the support of the NASA Glenn Research Center
and NASA SBIR funding. These FPSE units could eventually be coupled to radioisotope heat sources to provide
power for deep space missions. EE-35 evaluation units were delivered to NASA Glenn in 2004. This hardware has
passed NASA launch vibration qualification tests. Sunpower has explored other potential terrestrial applications for
the EE-35 engine modules. In July 2004, Sunpower began work on a project sponsored by DARPA Palm Power to
integrate the EE-35 into a soldier-borne power source which would operate using JP-8 fuel. Six months into that
program the team, which includes Sunpower, Yale University, and Precision Combustion, Inc., successfully

2
demonstrated the breadboard system using JP-8. In early 2006, the team will demonstrate the system in an
integrated, compact, standalone form for DARPA. The system has already demonstrated 21% JP-8 to electric
conversion efficiency (gross, without parasitics). The final system is projected to achieve 21% net fuel to electric
conversion efficiency. 4

EE-80 FPSE The 80We unit was developed by Sunpower and a subcontractor, Boeing/ Rocketdyne, under a
NASA NRA program. The EE-80 is also intended for use in space power applications using radioisotope heat
sources. An early EE-80 engine demonstrated 36% thermal-to-electric conversion efficiency (AC electrical out/ heat
into the heater head) at a temperature ratio of 3.0. The Phase II effort under this program will develop an Advanced
Stirling Convertor (ASC) version of the EE-80 design. The ASC unit is projected to approach 40% thermal-to-
electric conversion efficiency. 5

Table 1 Performance of Sunpower FPSE Units at Design Point


Temp Ratio Power Output % Carnot*1 Specific Power *2
EG-1000 2.7 1000We 58 % 45 We/kg
EE-35 2.6 43We 57% 90 We/kg
EE-80 3.0 86We 60 % 100 We/kg
*1 mechanical power to the alternator/ heat input to the head
*2 predicted with hermetically sealed housing

II. Performance Results for the EG-1000


In the following tests the nominal engine charge pressure was fixed. The performance of the FPSE subject to
combinations of operational parameters including acceptor/rejector temperature and load (piston amplitude) were
observed.

The current version of the EG-1000 was designed for use with a propane or natural gas burner, although the
acceptor heat exchanger could be optimized for other sources of heat. For these tests, a simple non-recuperated
propane ring burner is used to provide heat to the engine. A continuous-flow closed loop system provides a
specified constant water temperature at the inlet of the engine-mounted water jacket cooler.

The engine alternator leads were connected to an external “grid” which locks the frequency of the piston motion.
The piston amplitude can be modulated via an external variac (variable-turns voltage transformer). The amplitude
of the piston is proportional to the RMS voltage at the alternator terminals. The engine can be operated anywhere
between 130 and 260 volts while connected to a 50 Hz electrical grid. The engine is designed to produce power of 1
kWe at the design acceptor/rejector temperatures of 600°C/50°C, and has a design piston amplitude of 9 mm with a
mean charge pressure of 3.0 MPa. Performance test results at different operating conditions are presented in Fig. 3.

1600 50 1600 50
Power
Efficiency
1300 40 1300 40
E ffic ie n c y (% )

E ffic ie n c y (% )
P o w er (W e )

P o w e r (W e )

1000 30 1000 30

700 20 700 20
Thot=600℃
Power
Trej=50℃ Thot=600℃
Efficiency
400 10 400 Trej=50℃ 10

100 0 100 0
5 6 7 8 9 10 11 12 40 50 60 70 80
Piston Amplitude (mm) Reject Temperature (℃ )

Figure 3. Performance of EG-1000

3
It is very difficult to accurately measure the heat input to the acceptor because the EG-1000 engine has a non-
insulated burner-to-engine interface. Furthermore, the heat loss through the entire engine pressure vessel wall must
be taken into account for the calculation of engine efficiency. The electrical power output and the heat reject to
water are very easily measured. The heat loss through the pressure vessel wall can be measured by a special test in
which power is applied to the alternator to move the piston and therefore establish the typical gas circulation and
convective heat transfer conditions within the back end of the engine. An electric heater is placed within the engine
pressure vessel to simulate the heat provided from the working space of the engine. The power applied to the
electric heater is adjusted until the temperature at the pressure vessel wall is achieved which is equivalent to that
observed during normal running tests of the engine. The power dissipated by the alternator is added to the power
supplied to the electric heater to obtain the steady-state heat loss from the pressure vessel, Qloss. The heat rejected to
the cooling water is represented by Qreject. The electrical power output from the engine is Pout. Therefore, the total
heat input, Qin, to the thermodynamic cycle of the FPSE is calculated by Equation (1). The engine efficiency is
obtained by Equation (2).

Qin = Poutput + Qreject + Qloss


(1)
Poutput
η engine =
Qin (2)

The engine thermal-mechanical efficiency (mechanical power to the alternator/ heat input to the head) is
equivalent to 58% of Carnot efficiency at the design acceptor/rejector temperature of 600/ 50°C and the design
piston amplitude of 9mm. The overall engine efficiency is 33% which includes the efficiency of the alternator, 90%.

III. Performance Results of EE-35 and EE-80


Unlike the EG-1000, the EE-35 and EE-80 are tested using electric cartridge heaters to provide heat to the
acceptor. The acceptor is surrounded by insulation to reduce heat loss to the surrounding air. It is possible to
estimate the insulation heat loss by first measuring the electrical heater power input, Pheater, required to maintain a
constant operation head temperature under static (non-running) conditions. The conduction loss is subtracted from
the input power to the heater to obtain the insulation loss, Qinsulation. The total heat input to the acceptor is then
calculated as shown in Equation (3). The engine efficiency is then obtained by Equation (4)

Qin = Pheater − Qinsulation (3)


Poutput
η engine =
Qin (4)

The EE-35 operates at 105 Hz and produces 43 We at the design acceptor/ rejector temperature of 650/ 80°C and
the design piston amplitude of 4mm with a mean charge pressure of 2.6 MPa. Typical measured performance at
different operating conditions is presented in Fig. 4. At the design point, the EE-35 achieves 57% of Carnot for the
engine thermal-mechanical efficiency.

4
45 0.45 46 0.38
40 Pow er 0.40 44 0.37
35 Eff 0.35
42 0.36

Power (W e)
Power (W e)

Efficiency

Efficiiency
30 0.30
40 Power 0.35
25 0.25
38 Eff 0.34
20 0.20
36 0.33
15 0.15

10 Thot = 650 0C 0.10 34 0.32


Thot = 650 oC
5 Trej = 80 oC 0.05 32 0.31
Stroke = 8 mm
0 0.00 30 0.30
1 2 3 4 5 6 7 8 50 55 60 65 70 75 80
Piston Stroke (mm) Reject Temperature ( C) o

Figure 4. Performance of EE-35.

The EE-80 operates at 105 Hz and produces 86 We at the design acceptor/rejector temperature of 650/ 30°C and
the design piston amplitude of 4.5mm with a mean charge pressure of 3.5 MPa. Fig. 5 presents the measured
performance over a range of conditions. The performance at the design point represents 60% of the Carnot for the
engine thermal-mechanical efficiency.

90 0.45 90 0.45
80 Power 0.40
C o n v e r s io n E f f ic ie n c y

80 0.40
70 Eff. 0.35
Power (W e )

Efficiency
P o w e r (W e)

60 0.30 70 0.35
50 0.25
60 0.30
40 0.20

30 0.15 50 0.25
o Eff
20 0.10 Thot = 650 C
Thot = 650 oC; Trej = 30 oC 40 Power 0.20
10 0.05

0 0.00 30 0.15
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 70 60 50 40 30
o
Piston Amplitude (mm) Trej ( C)

Figure 5. Performance of EE-80.

IV. Conclusions
Sunpower engines achieve overall efficiencies of greater than 55% of Carnot efficiency at their design point as
shown in Fig. 6. The engines require no further development to achieve their performance targets. Efforts are
currently underway to hermetically seal each of these engines. The engines are projected to have specific power
values exceeding 90 W/kg in their final hermetic sealed forms. The engine modules are now being integrated into
power generation systems for both military and commercial systems.

5
0.8
Efficiency (fraction of Carnot)

0.7 Sunpower EG-1000


3

1 (2000)
Curzon-Ahlborn Efficiency
(for reference only)
0.6
Sunpower 80 We
4
5 MTI CPTC (1993) EE-80 (2004)
MTI SPRE (1990)
0.5 Sunpower 35 We
6
EE-35 (2004)
TDC 55 W
(2000)

0.4
7
Sunpower RE-1000
(1979)

0.3

0.2
1.0 1.5 2.0 2.5 3.0 3.5
Temperature Ratio (T heater / Trejector)

Figure 6. Percentage of Carnot Efficiency Performance for Sunpower Engines (without


alternator).5

Acknowledgments
Sunpower wishes to acknowledge Microgen, NASA Glenn Research Center, DARPA Palm Power, and Army
CERDEC/ARO/ARL for their continued support of projects which incorporate these Sunpower engines into
systems. The authors thank Sunpower employees Dave Shade and Faith Knutsen for their efforts with the
performance map testing of the 1 kW engine and editorial support, respectively.

References
1
Lane, N.W. and Beale, W.T., 1996, “A Free-Piston Stirling Engine-Alternator for Solar Electric Power,” 8th International
Symposium on Solar Thermal Concentrating Technologies, Koln, Germany, October 1996.
2
Lane, N.W. and Beale, W.T., 1999, “A Biomass-Fired 1kWe Stirling Engine Generator and Its applications in South
Africa,” 9th International Stirling Engine Conference, South Africa, 1999.
3
Wood, G.W., Lane, N.W. and Beale, W.T., 2001, “Preliminary Design of a 7kWe Free-Piston Stirling Engine with Rotary
Generator Output,” 10th International Stirling Engine Conference, Osnabruck, Germany, 2001.
4
Wood, G.W., and Lane, N.W., 2005, “Development of the Sunpower 35We Free-Piston Stirling Converter,” Space
Technology & Applications International Forum, Albuqurque New Mexico, 2005.
5
Wood, G.W. et. al, “Advanced 80We Stirling Converter Development Progress,” Space Technology & Applications
International Forum, Albuqurque New Mexico, 2004.

Anda mungkin juga menyukai