Revue des Energies Renouvelables CICME08 Sousse (2008) 1 - 10

1
Thermal performances of Stirling engine solar driven
M. Abbas
1*
, N. Sad
2
and B. Boumeddane
3

1
Unit de Dveloppement des Equipements Solaires, UDES
Route Nationale N11, B.P. 365, Bou Ismail, 42415, Tipaza, Algrie
2
Centre de Dveloppement des Energies Renouvelables, CDER
B.P. 62, Route de lObservatoire, Bouzarah, Algrie
3
Dpartement de Mcanique, Universit Sad Dahleb,
Route de Souma, Blida, Algrie

Abstract - Solar energy is one of the more attractive renewable energy sources. The
conversion of the latter per thermal way into electricity is a major energy stake. The
current systems are primarily based on technology known as 'solar dish/Stirling', which
uses Stirling engines placed at the focal plan of a parabolic concentrator. The Stirling
engine presents an excellent theoretical output equivalent to the output of Carnot one. It
is with external combustion, less pollutant, silencer and request little maintenance.
Thanks to these advantages which the Stirling engine is very interesting to study. The dish
Stirling system studies consist on three parts; the thermal modelling of Stirling engine,
optical study of parabolic concentrator and finally the thermal study of the receiver. The
present study is dedicated only to a thermal modeling of the Stirling engine based on the
decoupled method. We evaluate, starting from an ideal adiabatic analysis, the thermal
and mechanical powers exchanged, that we correct then by calculating the various losses
within the machine. This model led to the writing of important set of equations algebro-
differentials. The calculation programme worked out under FORTRAN to solve this
system, makes allow calculate the performances of any types of the Stirling engines,
according to the kinematics used, the types of regenerators, the exchangers, as well as the
various working liquids used.
Keywords: Stirling engine - Solar energy - Regenerator - Parabolic concentrator -
conversion.

1. INTRODUCTION
The rapid depletion of our natural resources has focused attention on new energy
sources and on effective means of energy conversion. One of the means of effective
energy conservation is the Stirling engine [1].
A Stirling cycle machine operates on a closed regenerative thermodynamic cycle
using a working gas, and subjects the gas to expansion and compression processes at
different temperatures. Since Stirling engines are externally heated, environmentally
very clean engine having high theoretical cycle efficiency, they can be powered using a
wide variety of fuels and heat sources such as, combustible materials, solar radiation,
geothermal hot water, radioisotope energy [2].
The Stirling heat engine was first patented in 1816 by Robert Stirling. Since then,
several Stirling engines based on his invention have been built in many forms and sizes
[3].
In solar modules, Stirling-Dish, the solar radiation is converted to electricity in three
stages. In the first stage, radiation is converted to heat by focusing the solar radiation

*
abbasdreams2003@yahoo.ca
M. Abbas et al.

2
onto a light absorbing heat pipe by means of a parabolic reflector. In the second stage,
the heat is converted to mechanical power by a Stirling engine. In the final stage, the
mechanical power is converted to electricity by an alternator.
The techniques of analysis for Stirling engines can be categorized with Martini's [4]
nomenclature as follows:
Zero Order Analysis: as Beale formula or West formula.
First Order Analysis: (Schmidt analysis) was done in 1871 by Gustav Schmidt in
which he obtained closed-form solutions for the special case of sinusoidal volume
variations and isothermal hot and cold spaces.
Second Order Analysis (uncoupled methods): This level of analysis may be based
on an adiabatic analysis that subtracts losses caused by heat transfer and flow power
losses. It relies on a modified Schmidt analysis and requires nonlinear time integration
of the model equations. It assumes adiabatic expansion and compression regions.
Third Order Analysis (coupled methods): uses control volumes or nodes to directly
solve one-dimensional (2D and 3D) governing equations.
At this level of modeling, the use of the computer codes is very necessary such as
GLIMPS, HFAST, CAST, FLUENT, STAR, CFX.
In the present study the thermodynamic modeling of a Gamma type Stirling-cycle
engine based on the decoupled method is performed. We evaluate, starting from an ideal
adiabatic analysis, the thermal and mechanical powers exchanged, we correct them by
calculating the various losses within the machine [5].
The engine is divided into 5 control volumes; in addition the regenerator is divided
into three subdivisions, as shown in Figure 1 [6]. The engine consists respectively of an
expansion space, E; heater H, regenerator, R; cooler, K; and a compression space, C.
Heat is transferred from the external heat source to the working fluid in the heater
section, cyclically stored and recovered in the regenerator, and rejected by the working
fluid to the external heat sink in the cooler section.


Fig. 1: Stirling engine configuration
CICME2008: Thermal performances of Stirling engine solar driven

3
The control volumes of compression and expansion spaces are variable volumes,
and the volumes of H, R and K have fixed volumes [7]. The conservation equation of
mass and energy are written for each control volume of the Engine [6]. A computer
program is written in FORTRAN, and the equations are solved iteratively.
2. THEORY
2.1 Ideal adiabatic model
The adiabatic model relies on the following usual main assumptions [6]:
- The processes are considered to be reversible.
- The ideal gas law can be applied to the working fluid.
- The instantaneous pressure is uniform in the engine.
- The compression and expansion space are adiabatic.
The general approach for deriving the equation set is to apply the equations of
energy and state to each of the control volume. The resulting equations are linked by
applying the continuity equation across the entire system [5].
In this ideal model, the fluid temperature in the heat exchangers is taken as uniform
and constant, in the regenerator a constant linear fluid temperature profile is assumed
between
H
T and
K
T .
The mean effective fluid temperature in the three parts of regenerator is thus
obtained as the logarithmic mean temperature between
H
T ,
1
T ,
2
T and
K
T such as:

=
K
1
K 1
1 R
T
T
ln
T T
T (1)
The variable volume expansion and compression spaces C and E depend on the
kinematics used, in the case of Swash plate, its expressed as:
( ) ( ) + + + = cos 1
2
V
V V
SWE
DEAE E
(2)
( ) ( ) + + = cos 1
2
V
V V
SWC
DEAC C
(3)
The equation of state is given: T R m V P = (4)
The total mass of gas is constant and equal to:
E H 3 R 2 R 1 R K C
m m m m m m m M + + + + + + = (5)
The instantaneous pressure of the engine is:
1
E
E
H
H
R
R
K
K
C
C
T
V
T
V
T
V
T
V
T
V
. R . M P

+ + + + = (6)
The equations of the mass flow rate of the interfaces are obtained by the method
suggested by W R. Martini [6]. They are given for the
th
i control volume by:
( )
t d
m d
m m
1 i , i
i 1 i
+
+
= & & (7)
For example: ( )
C CK
m d m = & (8)
The energy equation applied to a generalised control volume is written as:
( ) ( ) T . C . m d W d T . m . C T . m . C Q d
v out out p in in p
+ = + (9)
This equation is written for each control volume to calculate the amount of heat
transferred in each element of Stirling engine:
M. Abbas et al.

4
( )
KR K CK CK p v K K
m . T m . T . C
R
p d
. C . V Q d & & = (10)
The work done in the compression and expansion cells is:
( )

+ = sin .
2
V
P W
SWE
E
(11)
( )

= sin .
2
V
P W
SWC
C
(12)
We define the indicated efficiency by:
H
Q
W
= (13)
2.2 Decoupled analysis
In this section we adopt a "Quasi-Steady Flow" approach, in that we assume that at
each instant of the cycle the fluid behaves as though it is in steady flow.
2.2.1 Non - ideal heater and cooler
We consider that the fluid temperature is different from the wall temperature in both
exchangers. We evaluate the new values of gas temperatures
K
T and
H
T by:
wgk K
K
WK K
A . h
Q
T T = (14)
wgh H
H
WH H
A . h
Q
T T = (15)
2.2.2 Non - ideal regenerator
The relationship of regenerator effectiveness and thermal efficiency is given
by:
( )

=
1 .
Q
Q
1
Hi
Ri
i
(16)
We can assimilate the regenerator as a counter-flow exchanger, its effectiveness is:
NUT 1
NUT
+
= (17)
With
2
A
A
. N
NUT
R
WR
ST

= (18)
2.2.3 Actual thermal power
The actual thermal power is the sum of basic thermal power and different thermal
losses:

+ + + + =
SHUT PUMP COND R i Act
Q Q Q Q Q Q
& & & & & &
(19)
2.2.3.1 Shuttle heat transfer loss
Given by:
CICME2008: Thermal performances of Stirling engine solar driven

5
( )
p g
C E g
2
L
SHUT
L . x
T T . p d . k . S . 4 . 0
Q

=
&
(20)
2.2.3.2 Regenerator net enthalpy loss
( ) ( ) = 1 . T T . C . m Q
WK WH p R R
&
&
(21)
2.2.3.3 Regenerator wall heat leakage
Can be modelled by:
( )
j
froid chaud
j j COND
L
T T
. A . k Q

=
&
(22)
2.2.3.4 Pumping losses
Can be expressed by Leo formula [5]:
( ) ( )
( )
6 . 2
g
6 . 1
C E
p min max C E p
6 . 0
g
p
PUMP
x .
T T . R
C . freq . P P 2
.
5 . 1
T T . L . 2
.
k
d .
Q

=
&
(23)
2.2.4 Actual mechanical power
The actual mechanical power is equal to:

=
fric W i Act
W W W W
& & & &
(24)
Pressure drop loss
We can determine the new value of work done by integrating over the complete
cycle, and isolate the pumping loss term as follows as:
( ) ( ) ( ) ( )


= + = W W V d . P V d V d P W
i E E c
(25)
where:
( )

=
2
0
5
0 i
E
i
d .
d
V d
. P W (26)
The pressure drop in the 3 heat exchangers can be expressed as:
A . d
V . u . . C . 2
P
2
h
ref

= (27)
ref
C can be according to the following equation:
f ref
C . Re C = (28)
3. NUMERICAL METHOD
We must solve a system of 30 equations non linear among which 10 differential
equations that should be integrated numerically for specific configurations and
operating conditions.
Because of its cyclic nature, the system can be formed as an initial value problem by
assigning arbitrary initial conditions, and integrating the equations through several
complete cycles until a cyclic steady state has been attained. According to Ureili and
Berchowitz [5], the most sensitive measure of convergence to cyclic steady state is the
residual regenerator heat
R
Q at the end of the cycle, which should be zero.
M. Abbas et al.

6
4. RESULTS AND DISCUSSIONS
The computer program prepared in Fortran is applied to a Stirling engine known as
Organ machine. Air is used as the working gas, the compression and expansion space
initial gas temperatures are taken as 875 and 285 K, respectively. The results are
obtained using the parameters given in Table 1.
Table 1: Data used to analysis the Stirling engine [6, 8]
Type engine Gamma
Work fluid Hydrogen
Engine speed 2000 rpm
Average pressure 15 Mpa
Phase angle 90 degree
Swept volume per cylinder 134 cc
Compression dead volume 28 cc
Expansion dead volume 33 cc
The simulation of a kinematics Stirling engine led to the PV diagram given in
figure 2.
We can notice however that the diagram obtained differs appreciably from that of
the theoretical cycle of Stirling, although it does not correspond yet to the real diagram.

Fig. 2: P-V diagram
The variation of temperatures of the control volumes with crank angle are shown in
figure 3. We can remark that:
- The expansion control volume temperature is on the average lower than that of the
wall of the heater but there is however a part of the cycle where it is higher.
- The cyclic variation of the gas temperature in the expansion space is larger that
space compression one.
Figure 4 show the mass flow rate variation in different spaces as a function of the
crank angle, the figure show that the variation amplitude of flow rates is higher in the
cold space.
The positive flow rate period is more important that the negative flow rate one and
there are values of the phase for which all the flow rates are equal. This phenomenon
occurs at the time where pressure is at extreme.
CICME2008: Thermal performances of Stirling engine solar driven

7

Fig. 3: Temperatures variation in Stirling engine

Fig. 4: Flow rates in controls volume of engine

Fig. 5: Heat accumulated in engine
Figure 5 shows the accumulated heats transferred over the cycle. The most
significant aspect of this diagram are:
M. Abbas et al.

8
- Considerable amount of heat transferred in the regenerator over the cycle. This
tends to indicate that the engine performance depends critically on the regenerator
effectiveness.
- The energy rejected by the gas to the regenerator matrix in the first half of the
cycle is equal to the energy absorbed by the gas from the matrix in the second half of
the cycle, thus the net heat transfer to the regenerator over a cycle is zero.
4.1 Influence of losses on the Stirling engine performances
Figure 6 illustrates the influence of the imperfection of cooler and heater on the
Stirling engine performances.

Fig. 6: Influence of the imperfection of heater and cooler

Fig. 7: Pressure drop in heat exchangers
Notice that the mean temperature of the gas in the heater space is 53.9 degrees
below that of the heater wall, and similarly the mean temperature of the gas in the
cooler space is 11 degrees above that of the cooler wall. This lower temperature range
of operation reduced the output power, and the thermal efficiency.
Figure 7 illustrates the pressure drop in the 3 heat exchangers in Stirling engine.
Note the relative magnitude of the regenerator pressure drop with respect to those of the
heater and cooler.
CICME2008: Thermal performances of Stirling engine solar driven

9

Fig. 8: Regenerator effectiveness influence
Figure 8 shows the effect of regenerator effectiveness on thermal efficiency .
Notice that the thermal efficiency drops from more than 60 % to less than 10 %.
Thus we see that for highly effective regenerators a 1 % reduction in regenerator
effectiveness results in a more than 5 % reduction in thermal efficiency .
Furthermore we see that for a regenerator with an effectiveness of 0.8, the thermal
efficiency drops by around 30 %. This means a significantly less efficient machine.
Obviously we need to have a means of determining the actual regenerator effectiveness
in any specific machine [7].
5. CONCLUSION
The following conclusions are derived from the analysis:
The analysis provides the necessary data for the comparison of several aspects of
the Stirling-cycle engine.
The Stirling engine can be designed using the results of this analysis.
The regenerator effectiveness must be more than 80 %.
The progression of the gas work pressure must be ensured by a good sealing.
The phase angle must be 90 degrees.
A sensitivity study of different parameters is necessary to the study of the Stirling
engine performances.
NOMENCLATURE
A Area (section), m
2
NUT Number unit transfer
f
C Friction coefficient
st
N Stanton number
ref
C Reynolds friction coefficient R Ideal gas constant, J .kg
-1
.K
-1

p
C Pressure constant heat capacity,
J .kg
-1
.K
-1

Re Reynolds number
v
C Volume constant heat capacity,
J .kg
-1
.K
-1

l
S Piston course, m
m Masse, kg Q Heat, J
m& Flow rate, kg/s W Work, J
M. Abbas et al.

10
P Pressure, Pa
g
x Displacers-cylinder clearance,m
Subscripts and superscripts
Act Actual Dea Dead volume
C Compression space E Expansion space
CK Interface between compression
space and cooler
HE Interface between heater and
expansion space
g
Gas H Heater
Fric Friction i Ideal
K Cooler Pump Pump age
R Regenerator
1
R Part 1 of regenerator
2
R Part 2 of regenerator
3
R Part 3 of regenerator
2 1
R R Interface between part 1 and
part 2 of regenerator
3 2
R R Interface between part 2 and
part 3 of regenerator
KR Interface between part 1 of
regenerator and cooler
H R
3
Interface between part 3 of
regenerator and heater
sw Swept volume Shut Shuttle losses
wg
Wetted area Wk Cooler wall
Wh Heater wall
Greek symbols
Phase angle piston-displacer Regenerator effectiveness

Engine thermal efficiency Rotation angle, rad
REFERENCES
[1] C. inar and H. Karabulut, Manufacturing and Testing of a Gamma Type Stirling Engine,
Renewable Energy, Vol. 30, N1, pp. 57 66, 2005.
[2] . Ercan Ataer and H. Karabulut, Thermodynamic Analysis of the V-Type Stirling-Cycle
Refrigerator, International J ournal of Refrigeration, Vol. 28, N2, pp. 183 189, 2005.
[3] B. Kongtragool and S. Wongwises, A review of solar-powered Stirling engines and low
temperature differential Stirling engines, Renewable and Sustainable Energy Reviews, Vol.
7, N2, pp. 131 154, 2003.
[4] R.W. Dyson, S.D. Wilson and R.C. Tew, Review of Computational Stirling Analysis
Methods, NASA/TM2004-213300, AIAA20045582.
[5] I. Ureili and D.M. Berchowitz, Stirling Cycle Analysis, Published by Adam Hilger Ltd.
Bristol. Version actualise de ce livre disponible sur le site personnel du professeur Isral
Ureili: www.ent.ohiou.edu
[6] M. Abbas, Simulation Numrique dun cycle de moteur Stirling solaire de type gamma par la
mthode dcouple, Mmoire de Magister, Universit de Blida, 2006.
[7] S. Bonnet, Moteurs thermiques apport de chaleur externe: tude dun moteur Stirling et
dun moteur Ericsson, Thse de Doctorat, Universit de Pau et des Pays de lAdour, 2005.
[8] P. Stouffs, Conception et ralisation dun moteur Stirling de 1kW lectrique destin la
production combine dlectricit et deau chaude partir dnergie solaire, J ourne
dtudes SFT, France, 14 Octobre 1999.