Development and test of a Stirling engine driven by waste gases [622725]

Development and test of a Stirling engine driven by waste gases
for the micro-CHP system
Tie Lia,*, DaWei Tanga, Zhigang Lia, Jinglong Dua, Tian Zhoub, Yu Jiab
aInstitute of Engineering Thermophysics, Chinese Academy of Sciences, No. 11, BeiSiHuanXi Road, Beijing 100190, China
bBeijing ShiYanTianQiang Technology company, Beijing 100076, China
article info
Article history:
Received 20 July 2011Accepted 13 September 2011
Available online 21 September 2011
Keywords:
Micro-CHP
Stirling engine
Waste gasesTestOutput powerPressureabstract
In recent years, micro-CHP systems are attracting world attention. As one kind of external heating
engines, Stirling engines could be applied to the micro-CHP systems driven by solar, biogas, mid-high
temperature waste gases and many other heat sources. The development of a Stirling engine drivenby mid-high temperature waste gases is presented first. The thermodynamic design method, the key
parameters of the designed Stirling engine and its combustion chamber adapted for waste gases are
described in detail. Then the performance test of the Stirling engine is carried out. During the test, the
temperature of the heater head is monitored by thermocouples, and the pressure of the working fluid
helium in the Stirling engine is monitored by pressure sensors. The relationships among the output shaft
power, torque and speed are studied, and the pressure losses of the working fluid in the heat exchanger
system are also analyzed. The test results demonstrate that the output shaft power could reach 3476 W at
1248 RPM, which is in good agreement with the predicted value of 3901 W at 1500 RPM. The test results
confirm the fact that Stirling engines driven by mid-high temperature waste gases are able to achieve
a valuable output power for engineering application.
Crown Copyright /C2112011 Published by Elsevier Ltd. All rights reserved.
1. Introduction
Sharp fluctuations in oil prices and energy supply security risks
exist all over the world today. Environmental damage and the
effects of atmospheric warming are increasingly serious. In order
for sustainable development of economic, resources, environment,
the way of producing and using powers should be transformed
urgently. In this background, the CHP(Combined Heat and Power)
system that can be driven by clean energy such as solar, biogas,
exhaust heat and so on has been attracting more and more atten-
tion of many countries governments, especially in the European
Union, the United States, Canada, Japan and other developed
countries.
The engines for CHP systems are the key equipments, such as
gas turbines, Diesel engines, gasoline engines, Stirling engines and
so on. Different engines should be used for different CHP systems
depending on the magnitudes of the required output power.
According to different characters of the engines mentioned above
and the application area, the heat source could be various, forexample, solar, biogas, waste gases with mid-high temperature. It
could be called micro-CHP system if the power supply of the system
is less than 5 kW. The Stirling engine, due to its characters of
multiple heat source applicability, high thermal effi ciency and low
noise, is ideal for micro-CHP system, which could be driven by cleanenergy [1,2] . It is already reported that the international famous
Stirling engine that named SOLO/Cleanergy V-161 has been studiedfor the use into micro-CHP systems which were driven by natural
gases [3e5]. Many study work have also been done on the
b-type
Stirilng engines, especially the analytical, simulated, and testingwork [6e15].According to these work, the
b-type is proved to be
a good structure of the Stirling engine because it could avoid theside force from the piston acting to the cylinder ’s inner wall by
using the rhombic drive mechanism. But so far no experimentalstudy on Stirling engines driven by mid-high temperature waste
gases is found from the public literatures.
Exhaust from many industrial equipments are kinds of high-
grade heat, such as the waste gases from forging and billet heat-
ing furnaces at the temperature between 900 and 1200
/C14C, and
from dry-process cement kilns at the temperature between 600and 800
/C14C, and from the glass furnaces at the temperature
between 650 and 900/C14C[16]. These waste gases can be wisely used
as the heat sources to drive CHP systems.*Corresponding author. Tel.: ț86 10 82543092; fax: ț86 10 82543022.
E-mail address: litie@iet.cn (T. Li).
Contents lists available at SciVerse ScienceDirect
Applied Thermal Engineering
journal homepage: www.elsevier.com/locate/apthermeng
1359-4311/$ esee front matter Crown Copyright /C2112011 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2011.09.020Applied Thermal Engineering 33-34 (2012) 119 e123

We have developed a single-cylinder, b-type Stirilng engine
prototype with output shaft power of several kilowatts, which
could be driven by mid-high temperature waste gases. And the
results of our test work con firm the fact. We have also studied the
pressure variation of the working fluid in the heat exchanger
systems of the Stirling engine during the test process.2. Design of the b-type stirling engine
The single-cylinder, b-type Stirling engine developed by us is
designed with the rhombic drive mechanism, which was invented
by Philips Company in Netherlands [17,18] . Its main structure and
key parts are shown in Fig. 1 . The heater was made by 46 U-shape
stainless steel tubes, in order to use the heat carried by waste gases
wisely.
The thermodynamic design is based on the classic Schmidt
analysis method. In order to simplify the design and analysis work,
the Schmidt analysis method firstly assumes that the heat source
and the heat sink are both isothermal, the working fluid obeys the
perfect gas equation, no working fluid leaks, and so on. Then to get
the pressure changes of the working fluid in a cycle according to the
given key parameters and designed working state, and finally to
compute the work done in a cycle by using the first law of ther-
modynamics. The indicated output power could also be calculated
at the designed speed. It is found that the experimental output
power are usually no more than 55% of the results calculated by the
above Schmidt analysis method [19e21], so we modify the classic
Schmidt method by considering the experimental data of the GPU-3 Stirling engine developed by the General Motor Company, which
is also a
b-type Stirling engine with the rhombic drive mechanism,
in order to make the designed/predicted output power of our
Stirling engine more reliable than that computed by the classic
Fig. 1. The structure of the designed Stirling engine driven by waste gases.
Table 1
Some key design parameters of the Stirling engine development by the authors.
Parameter Value
Mechanical drive system Rhombic drive length 72 mm
Crank radius 26 mm
Eccentricity 26 mm
Displacer/pistonswept area81.71 cm
2
Displacer stroke 5.65 cm
Piston stroke 4.44 cm
Expansion cavityswept volume462.01 mL
Compression cavityswept volume362.81 mL
Phase angle advance 115
/C14
Heat exchanger
systemheater Number of heater tubes 46
Heat transfer length
of single tube313 mm(outside)
305 mm(inside)
Equivalent diameter 27.13 mmDesigned walltemperature550
/C14C
Total volume 178.62 mL
regenerator Structure Layers of
porous metal
Axial length 40 mm
Number of wires per inch 200 /C2200
Filler factor 27.04%
Volume for working fluid 217.64 mL
Equivalent workingtemperature204
/C14C
cooler Number of cooler tubes 190
Heat transfer lengthof single tube98 mm
Equivalent diameter 27.56 mmDesigned walltemperature15
/C14C
Total volume 58.50 mL
Working state Working fluid Helium
Designed Speed 1500 RPM
Mean pressure 2 MPa
Required heat 15,003 W
Indicated power output 3901 WThermal ef ficiency
(predicted)0.26
Fig. 2. The complete Stirling engine with its combustion chamber for waste gases and
the radiator.
Fig. 3. The structure chart of the designed combustion chamber for waste gases.T. Li et al. / Applied Thermal Engineering 33-34 (2012) 119 e123 120

Schmidt analysis method. The modi fication method is to get the
ratio of the experimental values to the computed values by the
Schmidt analysis method under some working states (the ratio
should be a range due to different working states of the GPU-3
Stirling engine under experimental testing), and then to use it as
a multiplying factor to the output power computed by the Schmidt
method, so the final designed value would agree well with the real
output power obtained from the experimental test. Some keydesign parameters of the Stirling engine are shown in Table 1 .The
designed/predicted output shaft power is 3901 W, while therequired input heat is 15,003 W at the designed working state.
Finally, the Stirling engine prototype has been fabricated as shown
inFig. 2 .
3. Design of the combustion chamber for waste gases
In the traditional design, the waste gases flow straight-forwardly
across the heater of a Stirling engine. The waste gases in the
combustion chamber are of low turbulence, and the convection
boundary layer is thick, which is disadvantageous for the convectiveheat transfer. Neither the thermal effi ciency nor the speci fic output
power of the Stirling engine is high.
In view of these disadvantages, we designed a combustion
chamber that allows the gases swirling in it, in order to enhance theheat transfer between the waste gases and the heater tubes of the
Stirling engine. The structure of this combustion chamber is shown in
Fig. 3 . The combustion chamber mainly includes a cylindrical shell,
a thermal insulating interlayer, an inlet with guide vanes, and a tan-gential outlet. The mid-high temperature waste gases could obtain
a tangential velocity component after flowing through the inlet guide
vanes, and then swirling flow is generated in the chamber, thus to
intensify the flow turbulence. After releasing heat to the heater tubes,
the waste gases flow from the bottom gaps of the interlayer into the
outer shell, and then exit the chamber from the tangential outlet. This
design could signi ficantly enhance the convective heat transfer
between the waste gases and the heater, so as to increase both thespecific output power and the engine ef ficiency due to the high ef fi-
ciency of retrieving the heat carried by waste gases.
4. Performance test process of the stirling engine
The layout of the test system is shown in Fig. 4 . The mid-high
temperature exhaust gases from a gasoline engine are used to
Fig. 4. The layout of the test system for the Stirling engine driven by waste gases. 1-
Holder of the gasoline engine; 2 – Radiator for the gasoline engine; 3-Gasoline
engine; 4-Bridge for waste gases; 5-Connecting conduit; 6-Test data recorder; 7-Channels of circulating cooling waters; 8-Radiator of cooling waters; 9-Pump ofcooling waters; 10-electric eddy current dynamometer; 11-Flywheel of the Stirlingengine; 12-Body of the Stirling engine.
Fig. 5. The test data of the temperature of a heater tube.
Fig. 6. The test data of the pressure in the heat exchanger system of the Stirling engine.T. Li et al. / Applied Thermal Engineering 33-34 (2012) 119 e123 121

drive the Stirling engine in the test. When the gasoline engine is
running, the waste gases enters into the combustion chamber
through a connecting conduit to release heat to the heater tubes of
the Stirling engine, and then flow out from the tangential outlet
mentioned above. The output shaft across the center of theflywheel is coupled to an eddy current dynamometer, by which the
output power, torque and speed were measured simultaneouslyduring the whole test time. The circulating cooling waters are
driven by a pump, in order to cool both the cooler of the Stirling
engine and the eddy current dynamometer. A radiator dissipates
the heat from the cooling water to the ambient air. In the test, three
pressure sensors are used to monitor the pressure of the working
fluid within the heat exchanger system of the Stirling engine at
three different points. Pressure sensor 1 is located at the joint gapbetween the cooler and the inlet of the compression chamber.Pressure sensor 2 is located at the joint gap between the cooler and
the regenerator. Pressure sensor 3 is located at the joint gap
between the regenerator and the heater. Only one heater tube is
under temperature monitoring by 4 K-type thermocouples in the
test, because all the 46 tubes are arranged axisymmetrically. The
required data in the test including temperature, pressure, power,
speed, and torque are collected by separate data collection instru-
ments through each corresponding lines. All the data were trans-
ferred to the same one computer for displaying and post-
processing.
Before operating the test, the indicators of all the collection
instruments and the tightness of the Stirling engine are checked out
to ensure they were OK. When all is ready, the Stirling engine is
made nearly vacuous, and then helium is filled until the pressure
sensor averagely reads 2 MPa. Then the valves are closed and thefilling process is finished.The gasoline engine is started at no-load state and the readings
of all the 4 thermocouples are paid special attention to. When the
average temperature reaches 350
/C14C, the cooling water pump is
opened, and then the starting motor is intermittently operateduntil the Stirling engine begin to run itself.
When the Stirling engine is running normally, the pressure of
working fluid is controlled by the valves, through which helium
could flow into or out of the engine. The wall temperature of the
heater tubes is controlled by the throttle of the gasoline engine,
through which the flow rate and the temperature of the waste
gases could be adjusted. These operations are continued until theStirling engine maintains a relatively stable working state. Data are
recorded in all the test time.
When all the test work is finished, the gasoline engine is
stopped firstly. After keeping the Stirling engine running for next
2e3 min, the bleed channels are opened until the pressure of the
helium inside the Stirling engine dropped to about 1 MPa. Then the
deflation is stopped. Finally the circulating cooling waters system is
closed after next 20 min, and the test is ended.
5. Test results and discussion
Fig. 5 shows the test data of the wall temperature of the heater
tube under monitoring in a period of about 50 s during testing. The
temperature captured by No. 4 thermocouple is the highest, which is
located near the central axis of the 46 U-shape tubes, while that of
No. 1 thermocouple is the lowest, which is located close to the
bottom of the heater tube. These results show that the temperature
distribution agrees with our expectation in that the tubes at inner
side were heated by waste gases firstly while those at the outer were
heated subsequently. The combustion chamber functions well. The
overall temperature difference is less than 200 ﾟC during the heating
processes. In the process of reducing heating, the temperature isalmost stable, and the maximum temperature difference is around
100 C (between No. 1 and No. 4 thermocouple).
Fig. 6 shows the pressure change of the helium in the heat
exchanger system when the Stirling engine is working at the state
that the mean pressure of it is about 2 MPa in a period of 21 s with
a recording time interval of 0.5 s. Each pressure sensor ’s location isTable 2
The average pressure.
Pressure point No. Average value
1 1.945 MPa
2 1.964 MPa3 1.919 MPa
Fig. 7. The test data of the power, torque and speed of the Stirling engine.T. Li et al. / Applied Thermal Engineering 33-34 (2012) 119 e123 122

mentioned in the section of performance test process of the Stirling
engine. As shown in Table 2 , the pressure loss of the helium is
0.02 MPa when it flows through the cooler, as observed from the
data recorded by the sensor 1 and sensor 2, while the pressure loss
is 0.045 MPa when it flows through the regenerator, as observed
from the data recorded by the sensor 2 and sensor 3. It also suggeststhat the pressure difference ratio is 2.3% between the inlet and
outlet of the regenerator, which is more than twice of that between
the inlet and outlet of the cooler, and the latter is 1%. The results
indicate that the porous metal regenerator mainly causes the
pressure loss of the working fluid in the heat exchanger system of
the Stirling engine. Meanwhile, the shape of the curves shown inFig. 6 told that the pressure variation processes of the helium are
approximately sinusoidal periodic, and the phase difference amongthe three pressure test points is almost zero.
Fig. 7 shows the relationships among the torque, power and
speed of the Stirling engine when it is working. During the test, the
maximum torque has reached 26.6 Nm, while the maximum output
shaft power has reached 3476 W, and the corresponding speed is
1248 RPM. The results satisfy the corresponding designed/pre-
dicted values shown in Table 1 , although the speed does not reach
the designed value of 1500 RPM. The test data also show that
a certain amount of mid-high temperature waste gases could
indeed drive some Stirling engines to generate considerable output
shaft power to further drive generators or other mechanical
machines like pumps. The output power is valuable from a practical
point of view. From the upward trend at the end of the curve shown
inFig. 7 , the output power and torque have not reached their
maximum yet. When the heat carried by waste gases increases, it isexpected to output more power and torque.
6. Conclusions
The development and test work of a Stirling engine driven by
waste gases for the micro-CHP system is presented in this paper.
The maximum output power has reached 3476 W in the test
process. And the test work also con firmed the fact that Stirling
engines could be indeed driven by mid-high temperature wastegases, and the output power is valuable from a practical point
of view.
Comparing with the test results, the classic Schmidt method
modi fied by some experimental data could gain a design value of
practical signi ficance, which is close to the reality.
The pressure change of the working fluid in the heat exchanger
system of the designed Stirling engine is also studied in theexperimental way. The loss between the inlet and outlet of the
regenerator is more than twice of that between the inlet and outletof the cooler. The porous metal regenerator mainly causes thepressure loss of the working fluid. The pressure variation is
approximately sinusoidal periodic, and the phase difference amongthe three different locations is almost zero.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (Grant No. 51161140332).
References
[1] G. Walker, Stirling Engines. Oxford University Press, Britain, 1980.
[2] D.H. Jin, The Technology of Stirling Engines. Harbin Engineering University
Press, China, 2009.
[3] B. Thomas, Benchmark testing of micro-CHP units, Appl. Therm. Eng. 28
(2008) 2049 e2054.
[4] V. Kuhn, J. Klemes, I. Bulatov, Micro CHP: overview of selected technologies,
products and field test results, Appl. Therm. Eng. 28 (2008) 2039 e2048.
[5] MD. Paepe, P. D ’Herdt, D. Mertens, Micro-CHP systems for residential appli-
cations, Energy Convers. Manage 47 (2006) 3435 e3446.
[6] JI. Prieto, A.B. Stefanovskiy, Dimensional analysis of leakage and mechanical
power losses of kinematic Stirling engines, J. Mech. Eng. Sci. 217 (2003)
917e934.
[7] JI. Prieto, M.A. González, C. González, J. Fano, A new equation representing the
performance of kinematic Stirling engines, J. Mech. Eng. Sci. 214 (2000)449e464.
[8] JI. Prieto, J. Fano, C. González, M.A. González, R. Diaz, Preliminary design of the
kinematic Stirling engine using dynamic similarity and quasi-static simula-tion, J. Mech. Eng. Sci. 211 (1997) 229 e238.
[9] T. Finkelstein, Generalized thermodynamic analysis of Stirling engine, SAE
Paper 118 B (1960).
[10] G. Walker, M.I. Kahn, Theoretical performance of Stirling cycle engines, SAE
Paper 949 A (1965).
[11] A. Shock, Nodal analysis of Stirling cycle Devices, Proceeding of the 13th IECEC
Paper, No.789191, 1978, 1771 e1779.
[12] I. Urieli, A computer simulation of Stirling cycle machines, PhD thesis,
University of the Witwatersrand, South Africa, 1977.
[13] F.Y. Lee, Computer Simulation of Stirling Engines, MSc thesis University of
Calgary, Canada, 1976.
[14] T.J. Sullivan, NASA Lewis Stirling Engine Computer Code Evaluation, DOE/
NASA-4105 e4, 1989.
[15] J.C. Kim, an analytical and experimental study of heat transfer and flow-
Friction Characteristics for Periodically Reversing flow through the porous
Matrix of thermal Regenarators PhD thesis Purdue University USA, 1970.
[16] D.F. Che, Y.H. Liu, Cascade, Utilization of the Heat from Smoke Gases. Chemical
Industry Press, China, 2006.
[17] CM. Hargreaves, The Philips Stirling Engine. Elsevier science publishers,
Netherlands, 1991.
[18] RJ. Meijer, The Philips hot-gas engine with rhombic drive mechanism, Philips
Tech. Rev. 20 (1959) 245 e276.
[19] L.G. Thieme, Low-power Baseline Test Results for the GPU 3 Stirling Engine,
DOE/NASA/1040 e79/6, 1979.
[20] L.G. Thieme, High-power Baseline and Motoring Test Results for the GPU-3
Stirling Engine, DOE/NASA/51040 e31, 1981.
[21] I. Urieli, D.M. Berchowitz, Stirling Cycle Engine Analysis. Adam Hilger Ltd
Bristol, Britain, 1984.T. Li et al. / Applied Thermal Engineering 33-34 (2012) 119 e123 123