Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol II

IMECS 2009, March 18 - 20, 2009, Hong Kong

Rapid Thermodynamic Simulation Model of an

Internal Combustion Engine on Alternate Fuels
Sundeep Ramachandran
Abstract—A thermodynamic model for the simulation of a
spark ignition engine running on alternate hydrocarbon fuel is
presented. This paper aims to develop a simple, fast and
accurate engine simulation model without the need for a great
deal of computational power or knowledge of precise engine
geometrical data. The model is based on the classical two-zone
approach, wherein parameters like heat transfer from the
cylinder, blowby energy loss and heat release rate are also
considered. The general fuel is specified by way of its C-H-O-N
values. Curve-fit coefficients are then employed to simulate air
and fuel data along with frozen composition and practical
chemical equilibrium routines. The calculated data is then used
to plot the various thermodynamic parameters with respect to
crank angle.
Index Terms—Internal Combustion Engines, SI Engine
Simulation, Thermodynamic Modeling of Engines.
the cylinder volume is divided into burned and unburned
zones by an infinitesimally thin flame-front with a spherical
shape. The burned gases are assumed to be in chemical
equilibrium during combustion and for the main expansion
stroke, while near the end of expansion stroke the mixture is
assumed frozen [7], [10], [18]. A Wiebe function specifies
the fuel burn rate and controls the rate at which mixtures
from the unburned zone is converted to the burned zone
[10]. Mass and energy conservation relations and equations
of state form the principle governing equations. Also
considering crank angle as the independent variable, we thus
form the base of our thermodynamic model.
A. Mass and Energy Balance
The equation of state for an ideal gas is
PV = mRT (1)

The rate of change of mass within any open system is the

I. INTRODUCTION net flux of mass across the system boundaries. Hence for a
€
The present trend is towards the development of control volume enclosing the air-fuel mixture, we have
comprehensive 3-D models, which describes the functioning m˙ = ∑m ˙k (2)
of engines at a very high level of detail and accuracy; k

however, these require substantial computational power.

Also, the need for precise experimental input makes the The first law of thermodynamics to an open system yields
process significantly complicated and time consuming. the energy equation as
€
There are several instances where theoretical methods, E˙ = Q
˙ − W˙ + ∑ m˙ k hk (3)
which are based on a limited set of experimental data, are k

preferred. From these considerations, the need for a simple, Equations (2) and (3) can be written as
fast and accurate engine simulation model is quite evident. dm dm k (4)
=∑
Hence, this paper aims to combine the advantages of various
€ dθ k dθ
known models to achieve this goal.
A two-zone, Zero-dimensional model was used to d( mu) dQ dV dm˙k (5)
= −p + ∑ hk
simulate the engine operations. The most important € dθ dθ dθ k
dθ
assumptions were that, a) The working medium was
considered, in general, to be a mixture of 14 species (O2, N2, Equation (5) neglects changes of kinetic and potential
CO2, H2O, H2, OH, NO, CO, O, H, N, Ar, NO2, HO2) and energy in the control volume.
fuel vapor. b) All 14 species were considered as ideal€gases.
and c) The alternate fuels are limited to C-H-O-N species. B. Air and Combustion Products Data
Gordon and McBride [8], proposed the following
II. THERMODYNAMIC MODEL expressions that were curve-fitted to the tabulated JANAF
For the present study, a Zero-dimensional combustion Thermchemical tables [11].
cp
model is employed. The combustion chamber is divided into = a1 + a2T + a3T 2 + a4 T 3 + a5T 4 (6)
two zones consisting of unburned gas (mixture of fuel, air R
and residuals) and burned gas (mixture of 14 product
species), each under uniform composition. This model h a a a a a (7)
= a1 + 2 T + 3 T 2 + 4 T 3 + 5 T 4 + 6
assumes that at any instant of time during the combustion, RT 2 3 4 5 T
€
Manuscript received December 6, 2008. s a a a (8)
= a1 lnT + a2T + 3 T 2 + 4 T 3 + 5 T 4 + a7
Sundeep Ramachandran is a dual-degree student with the Department of R 2 3 4
Mechanical Engineering, Birla Institute of Technology & Science,€Pilani -
Dubai, International Academic City, PB 17965, Dubai, UAE (phone:
00971-50-2278780; e-mail: rsundeep@gmail.com). where cp is the specific heat at constant pressure, h is the
specific enthalpy and s is the specific entropy.
€

ISBN: 978-988-17012-7-5 IMECS 2009

 Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol II
IMECS 2009, March 18 - 20, 2009, Hong Kong

The coefficients a1 to a7 are calculated over two different Olikara and Borman [14] included two additional mole
temperature ranges: 1) 300 <T< 1000 K; and 2) 1000 <T< fractions in the products: N and Ar. They also made
5000 K and can be sourced from [8]. provision to include user specified quality of air. Depcik [6]
The most complete models are based on the assumption modified the Olikara and Borman [14] model by including
that the unburned mixture is frozen in composition and the the following two reactions
burned mixture is in equilibrium [10]. The following are 1 (16)
NO + O 2 ← → NO 2
species of interest during combustion: CO2, H2O, N2, O2, 2
CO, H2, H, O, OH and NO [7].
1 (17)
C. Fuel Data O2 + H2 ←
→ HO 2
2
Heywood [10] has represented the thermodynamic €
properties of fuels (in vapor phase) using curves that slightly The combustion reaction now changes to
differ from (6), (7), (8).  O2 + Q N 2 N 2 +
cp 1 €   
2 3
= a1 + a2T + a3T + a4 T + a5 2 (9)
R T  n + 0.25m - 0.5l Q Ar Ar + 
x13 C n H m O lN k +
φ  Q CO 2 CO 2 +  (18)
  
h a a a 1 a (10)
= a1 + 2 T + 3 T 2 + 4 T 3 − a5 2 + 6  Q H 2 OH 2O 
RT 2 3 4 T T
€ → x1H + x 2O + x 3N + x 4 H 2 + x 5OH + x 6CO +
s a a a 1 (11) x 7NO + x 8O 2 + x 9H 2O + x10CO 2 + x11N 2 + x12 Ar +
= a1 lnT + a2T + 3 T 2 + 4 T 3 − 5 2 + a7
R 2 3 2 T x14 NO 2 + x15HO 2
€
Data for calculating the coefficients can be obtained from Here x and Q denote mole fraction and quality respectively.
[11], [5] or [8]. Reference [9], has discussed about the  
€
methods for calculating the thermodynamic properties € of For clarity, we define: ψ =  n + 0.25m − 0.5l 
 φ 
user-defined fuels.
The following relation is proposed for deriving the The atom balance for the various elements gives
properties like specific heats and enthalpies for various C : x 6 + x10 = x13 (ψQ CO + n) (19)
2

species, with ξ referring to the property and x is the burnt €

mass fraction. H : x1 + 2x 4 + x 5 + 2x 9 + x15 = x13 (2ψQ H 2 O + m) (20)
N
ξ mixture = ∑ x k ⋅ ξ k € (12)
O : x 2 + x 5 + x 6 + x 7 + 2x 8 + x 9 + 2x10 + 2x14 + 2x15 =
k =1 (21)
€
D. Equivalence Ratio € (
2x13 ψ (1+ Q CO 2 + 0.5Q H 2 O ) + 0.5l )
When modeling with a single fuel, the equivalence ratio is
€ given by [7] N : x 3 + x 7 + 2x11 + x14 = 2x13 (ψQ N 2 + 0.5k ) (22)
 [F ]   [F ]  € (13)
φ =   
 Air  Act.  Air St. Ar : x12 = x13ψQ Ar (23)
where subscript Act. refers to actual and St. refers to €
stoichiometric. The constraint that the mole fraction of all the products adds
For blending of hydrocarbons and/or alcohol€ with up to unity requires that
€ 12
hydrogen fuel, the equivalence ratio changes to [16] (24)
 
∑x k + x14 + x15 = 1
k=1
 
 [ F]   [F ]  (14)
φ =    To solve for the unknowns, we need equations that are
 [ Air ] − [H]   Air St. provided by the criteria of equilibrium among the products,
€

 (
[H]
)
Air st. 
which are expressed by the following hypothetical relations
1 x ⋅ p 0.5
H2 ←→ H K1 = 1 0.5 (25a)
E. Practical Chemical Equilibrium 2 x4
Under the assumption of atmospheric air composition
€ (21%v Oxygen and 79%v Nitrogen), and provided < 3, the 1 x 2 ⋅ p 0.5
O2 ←
→ O K2 = (25b)
0.5
only species that are important because of dissociation
€ are 2 x8
O, H, OH, and NO [7]. The combustion reaction thus
becomes 1 x 3 ⋅ p 0.5 (25c)
N2 ←
→ N K3 =
εφCα H β O γ Nδ + 0.21O 2 + 0.79N 2 → x1CO 2 + 2 x11
0.5
€
x H O + x N + x O + x CO + x H + (15)
2 2 3 2 4 2 5 6 2

x 7H + x 8O + x 9OH + x10NO 1 1 x5 (25d)

H2 + O2 ←
→ OH K5 = 0.5 0.5
€ 2 2 x4 ⋅ x8
where x1 to x10 represents mole fractions of the products.
€
€
ISBN: 978-988-17012-7-5 IMECS 2009
 Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol II
IMECS 2009, March 18 - 20, 2009, Hong Kong

1 1 x7 (25e)
O2 + N2 ←
→ NO K7 =
2 2 x8
0.5
⋅ x11
0.5
O : φεγ + 2W + 2Z = 2x1 + x 2 (34)

1 x9 φεδ + 2X (35)
H2 + O2 ←
→ H 2O K9 = (25f) N : φεδ + 2X = 2x 3 ⇒ x3 =
0.5
2 x4 ⋅ x8 0.5
⋅p € 2
€

1 x10 Ar : Y = x 7 ⇒ x7 = Y (36)
CO + O2 ←
→ CO 2 K10 = (25g)
0.5 0.5
€ 2 x6 ⋅ x8 ⋅p €
W (37)
ε=
β γ
1 x14 € (25h) α+ −
NO + O2 ←
→ NO 2 K14 = 0.5 0.5 4 2
€ 2 x7 ⋅ x8 ⋅p
Hence the stoichiometric fuel-air ratio is
1 x15 (25i)  [ F ] 
O2 + H2 ←
→ HO 2 K15 = ε(12.011α + 1.008β + 16.0γ + 14.01δ ) (38)
€ 2 x 8 ⋅ x 4 0.5 ⋅ p 0.5€   =
 Air  st. 31.998W + 28.012X + 38.948Y + 44.009Z
The equilibrium constant K, for the above are determined
Equation (38) is used in (13) or (14) to calculate the
from the Gibbs free energy as
€ equivalence ratio. Similar formulation can be used for lean
ΔG  €
(26)
− = lnK p and rich mixture cases.
RT

where ΔG is the standard-state Gibbs free energy. G. Thermal Properties
The values of the equilibrium constants can be obtained We consider the unburned and burnt mixture zones as
€ from [11]. Equations (25a) to (25i) reduces to the following separate open systems. Therefore, the specific internal
which is then solved using Newton-Raphson iteration energy, u and specific volume, v is expressed as
€ scheme. U (39)
u= = xub + (1− x ) uu
x1 + 2x 4 + x 5 + 2x 9 + x15 −
(2ψQ H2O + m) ( x + x ) = 0 (27) m

(ψQ CO2 + n) 6 10 V (40)

= xv b + (1− x )v u
v=
€ m
x 2 + x 5 + x 6 + x 7 + 2x 8 + x 9 + 2x10 + 2x14 + x15 − subscripts b and u refers to burnt gas and unburned gas
(28) respectively.
€ (
2 ψ (1+ Q CO 2 + 0.5Q H 2 O ) + 0.5l )
( x 6 + x10 ) €
=0 Going by our assumption that the pressures of burnt and
(ψQ CO2 + n) unburned gases are equal, vb and vu are functions of Tb, Tu
and p. Hence
2(ψQ N 2 + 0.5k ) dv b ∂v b dTb ∂v b dp (41)
x 3 + x 7 + 2x11 + x14 − ( x 6 + x10 ) = 0 (29) dθ = ∂Tb dθ + ∂p dθ
€ ( ψQ CO 2 +n )
dv u ∂v u dTu ∂v u dp (42)
x1 + x 2 + x 3 + x 4 + x 5 + x 6 + x 7 + x 8 + x 9 + x10 + = +
€ (30) dθ ∂Tu dθ ∂p dθ
€ ψQ Ar
x11 + x14 + x15 + ( x 6 + x10 ) −1 = 0
(ψQ CO 2 + n) Back substituting the logarithmic derivatives from Depcik
model [6] on (41) and (42), we have
F. Frozen Composition €
dv b v b ∂ lnv b dTb v b ∂ lnv b dp
= + (43)
Depcik model [6] is applicable only over the range of dθ Tb ∂ lnTb dθ p ∂ ln p dθ
€ 600-4000 K. We use the Ferguson model [7] that calculates
the species of combustion in the low ranges of temperature dv u v u ∂ lnv u dTu v u ∂ lnv u dp
from 300-600 K. This model assumes that the only species = + (44)
d θ T ∂ lnT d θ p ∂ ln p d θ
prevalent in the exhaust are: CO2, H2O, N2, Ar, CO, € O2 and u u

H2. The general reaction is given by

Similarly, the internal energies of both the burnt and
εφCα H β O γ Nδ + (WO 2 + XN 2 + YAr + ZCO 2 ) → (31) unburned gases, under the same pressure condition and
x1CO 2 + x 2H 2O + x 3N 2 + x 4 O 2 + x 5CO + x 6H 2 + x€7 Ar including the logarithmic derivatives can be written as
dub  pv b ∂ lnv b  dTb  ∂ lnv b ∂ lnv b  dp
(45)
Equation (31) is formulated for stoichiometric case, i.e. = c p b −  − v b + 
dθ  Tb ∂ lnTb  dθ  ∂ lnTb ∂ ln p  dθ
( ) wherein x4=x5=x6=0. For lean, i.e. ( ), x5=x6=0
€
and for rich, i.e. ( ), x4=0. Performing atom balance for
duu  pv u ∂ lnv u  dTu  ∂ lnv u ∂ lnv u  dp
(46)
stoichiometric case and solving for the coefficients give = c p u −  − vu  + 
€ dθ  Tu ∂ lnTu  dθ  ∂ lnTu ∂ ln p  dθ
C : φεα + Z = x1 ⇒ x1 = φεα + Z (32)
H. Trapped Mass in Control Volume
φεβ (33) The trapped mass in control volume is defined at various
H : φεβ = 2x 2 ⇒ x2 =
2 €
€

ISBN: 978-988-17012-7-5 IMECS 2009

€
 Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol II
IMECS 2009, March 18 - 20, 2009, Hong Kong

periods by [4] Abi = Ai x 0.5 (55)

for θ IVC ≥ θ ≥ −360  CA (intake)
(47) A = A 1− x 0.5 (56)
V (θ ) ui i( )
m=
vu €
πb 2 (Hemispherical cylinder head) (57)
Ah =
for θ EVO ≥ θ ≥ θ IVC (valve - closed) 2
€ (48)
€ m = m IVC e -C b (θ −θ IVC ) ω
πb 2 (Flat piston crown) (58)
Ap =
for 360  CA ≥ θ ≥ θ EVO (blowdown and exhaust) € 4
(49)
€ V (θ )
m= 4V (θ ) (Linear surface area exposed to gases) (59)
vb Al =
€ b
IVC refers to intake valve open and EVO refers to exhaust subscript i refers to intake.
valve open.
The volume of the cylinder at any crank angle instant is K. Heat Transfer Correlation
€ €
given by [7] In (53) and (54), h is the instantaneous heat transfer
 r -1  1  coefficient. Woschni proposed a correlation of the form [18]
V (θ ) = Vc 1+
 2  ε [ 0.5
]
1− cosθ + 1− (1− ε 2 sin 2 θ )  (50)
 Nu = 0.035Re 0.8 (60)
Vc is clearance volume, r is the compression ratio and
stroke where Nu is Nusselt number and Re is Reynolds number.
ε= Now, assuming a local average gas velocity in the cylinder,
2 x length of connecting rod €
€ becomes
0.8
h = 0.82b−0.2 ( p ⋅10−3 ⋅ c ) T −0.53 (61)
I. Fuel Burning Rate Model
€ Many experiments show that the burning rate depends where h is the heat transfer coefficient, p is pressure.
mostly on the combustion chamber shape and the position of c = 6.18 cm (For gas exchange process) (62)
the spark plug. The Wiebe function represents the€ mass
fraction burned, xb versus crank angle and defined as [10] T = xTb + (1− x )Tu (63)
m +1
{
x b (θ ) = 1− exp −a[(θ − θ 0 ) Δθ ] } € (51)
L. Blowby Energy Loss
Enthalpy loss due to blowby is expressed as [7]
The present simulation uses a=5 and m=2 [10]. € This (64)
hl = (1− x 2 ) hu + x 2 hb
model is capable of representing many combustion chamber
€
shapes with different positions of spark plugs by adjusting a
and m. The constant a is responsible for maximum value of which indicates that more leaking is due to the unburned gas
xb which is unity [12]. Hence a may represent the positions compared with the burnt gas in the early stage of
€
and/or number of spark plugs. By varying m, he observed combustion.
the starting point and end point of xb are independent of m. M. Principle Governing Equations
Hence even though, for a particular a, the amount of burned Differentiating (40) with respect to crank angle and also
mass fraction xb is independent of m, its rate closely depends
considering (43) and (44), we have [7]
on m. Hence as m increases, xb is found to decrease and 1 dV VCb v ∂ lnv b dTb v ∂ lnv u dTu
hence m may represent combustion chamber shape-factor. + =x b + (1− x ) u +
m dθ m ω Tb ∂ lnTb dθ Tu ∂ lnTu dθ (65)
J. Heat Transfer Model  v b ∂ lnv b v ∂ lnv u  dp dx
Heat transfer into a thermodynamic system is expressed x + (1− x ) u  + (v b − v u )
 p ∂ ln p p ∂ ln p  dθ dθ
in terms of heat loss
dQ -Ql -Q ˙ -Q˙
= = b u (52) here Cb is the blowby coefficient and given by
dθ ω ω Cb = m˙ l m and m˙ l is the leakage due to blowby.
€
where
(53) Expressing the heat loss of burnt and unburned gases as a
Q˙ = h ∑ A (T − T )
€ b bi b wi function of the rate of change of specific entropy gives
i= h,p,l
€ −Q ˙ b =€mωxTb dsb (66)
dθ
Q˙ = h ∑ A (T − T ) (54)
u ui u wi
i= h,p,l
€
−Q ˙ u = mω (1− x )Tu dsu (67)
€gases d θ
Here Abi and Aui are the areas of burnt and unburned
in contact with the combustion chamber component at where
€
temperature Twi, and subscripts h, p, l refers to cylinder head,  c p  dT v ∂ lnv dp
€ ds (68)
piston crown and liner, respectively. We have the following b
= b b − b b

relations [7] dθ  Tb  dθ Tb ∂ lnTb dθ

ISBN: 978-988-17012-7-5 € IMECS 2009

 Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol II
IMECS 2009, March 18 - 20, 2009, Hong Kong

III. SIMULATION RESULTS

dsu  c p u  dTu v u ∂ lnv u dp (69) Having formulated the mathematical framework, we
=  −
dθ  Tu  dθ Tu ∂ lnTu dθ simulate Pressure, Temperature, Work and Heat Transfer for
an single cylinder, four stroke SI Engine running on Ethanol
Expressing the heat loss of burnt and unburned gases as a (CH3CH2OH). We have assumed a bore of 0.1m, stroke of
function of the rate of change of specific entropy by 0.08m and compression ratio of 10. The engine is assumed
€ to operate at 2000 rpm, with an equivalence ratio of 0.8.
combining (53)-(56) and (66)-(69)
−h ∑ Abi (Tb − Twi )
dTb ∂ lnv b dp i= h,p,l (70)
cpb − vb =
dθ ∂ lnTb dθ mωx

−h ∑ A (T ui u − Twi )
dTu ∂ lnv u dp i= h,p,l (71)
€ cpu − vu =
dθ ∂ lnTu dθ mω (1− x )

Differentiating Equations (50) and (51) and incorporating

with Equations (4), (39), (40), (43)-(46) and (52)-(64) into
€
Equation (5), we have the following relations [7]
dp f + f2 + f3 (72)
= 1
dθ f4 + f5

−h ∑ A (T bi b − Twi )
Figure 1: Pressure v/s Crank Angle
dTb i= h,p,l v b ∂ lnv b dp
€ = + + (73)
dθ m ωc p b x c p b ∂ lnTb dθ
hu − hb  dx C 
− (x − x 2) b 
xc p b dθ ω

−h ∑ A (T ui u − Twi )
dTu i= h,p,l v u ∂ lnv u dp (74)
€ = +
dθ mωc p u (1− x ) c p u ∂ lnTu dθ

where
1  dV VCb  (75)
€ f1 =  + 
m  dθ ω 

 ∑ Abi (Tb − Twi ) 

 vb ∂ lnv b i= h,p,l
+ Figure 2: Temperature v/s Crank Angle
€
h c p b ∂ lnTb Tb  (76)
f2 =  
mω 
v ∂ lnv u ∑i= h,p,l Aui (Tu − Twi ) 
 u 
c p u ∂ lnTu Tu 

dx ∂ lnv b hu − hb  dx C 
f 3 = −(v b − v u ) − vb − ( x − x 2 ) b (77)
€ dθ ∂ lnTb c p bTb dθ ω

 2  2 
v b ∂ lnv b  v b ∂ lnv b (78)
f4 = x    + 
€ c p bTb  ∂ lnTb  p ∂ ln p 

 2  2 
v ∂ lnv u  v u ∂ lnv u (79)
f 5 = (1− x ) u   + 
€ c p u Tu  ∂ lnTu  p ∂ ln p  Figure 3: Work v/s Crank Angle

Equations (75)-(79) are functions of , p, Tb and Tu and are

€ solved using the Runge-Kutta method.

ISBN: 978-988-17012-7-5 IMECS 2009

Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol II
IMECS 2009, March 18 - 20, 2009, Hong Kong

angular velocity ( )
equivalance ratio

REFERENCES
[1] G. H. Abd Alla, “Computer simulation of a four stroke spark ignition
engine” En. Convers. Mng., vol. 43, 2002, pp. 1043-1061.
[2] I. Arsie, C. Pianese and G. Rizzo, “Models for the Prediction of
Performance and Emissions in a Spark Ignition Engine – A
Sequentially Structured Approach”, SAE Paper 980779, 1998.
[3] D. R. Buttsworth, “Spark Ignition Internal Combustion Engine
Modelling using Matlab”, Faculty of Engineering & Surveying
Technical Reports, University of Southern Queensland, 2002.
[4] S. H. Chan and J. Zhu, “Modelling of engine in-cylinder
thermodynamics under high values of ignition retard”, Int. J. Therm.
Sci., vol. 40, 1999, pp. 94-103.
[5] “CHEMKIN Thermodynamic Database”, Sandia Report
Figure 4: Heat Transfer v/s Crank Angle SAND87-8215 B, 1991.
[6] C. Depcik, “Open-Ended Thermodynamic Cycle Simulation”, M.S.
Thesis, University of Michigan, Ann Arbor, 2000.
IV. CONCLUSION [7] C. R. Ferguson, Internal Combustion Engines, Applied
The present work achieves its goal by being a simple, fast Thermosciences. New York: John Wiley and Sons, 1986.
[8] S. Gordon and B. J. McBride, “Computer Program for Calculation of
and accurate engine simulation model. The results obtained Complex Chemical Equilibrium Composition, Rocket Performance,
can be used as a first-degree approximation and is useful in Incident and Reflected Shocks, and Chapman Jouguet Detonations”,
numerous engineering applications including general design NASA publication SP-273, 1971.
[9] M. Grill, M. Chiodi. H. Berner and M. Bargende, “Calculating the
predictions. The model can predict an array of Thermodynamic Properties of Burnt Gas and Vapor Fuel for
thermodynamic and combustion parameters and easily adapt User-Defined Fuels”, MTZ 05|2007, 2007, pp. 398.
to any combustion chamber shape. Due to its simplicity and [10] J. B. Heywood, Internal Combustion Engine Fundamentals. New
computational efficiency, the model can also be used as a York: McGraw-Hill, 1988.
[11] “JANAF Thermochemical Tables”, United States National Bureau of
preliminary test on a wide range of alternate hydrocarbon Standards Publications NSRDS-NBS 37, 1971.
fuels. [12] Z. H. Kodah, H. S. Soliman, M. Abu Qudais and Z. A. Jahmany,
“Combustion in a spark-ignition engine”, Appl. Energy, vol. 66, 2000,
pp. 237-250.
NOMENCLATURE [13] B. J. McBride, S. Gordon and M.A. Reno, “Coefficients for
Calculating Thermodynamic and Transport Properties of Individual
Species”, NASA Technical Memorandum 4513, 1993.
area exposed to heat transfer (m2) [14] C. Olikara and G. L. Borman, “Calculating Properties of Equilibrium
bore of cylinder (m) Combustion Products with Some Application to I.C. Engines”, SAE
Paper 750468, 1975.
specific heat at constant pressure ( ) [15] R. R. Raine, C. R. Stone and J. Gould, “Modelling of Nitric Oxide
Formation in Spark Ignition Engines with a Multi-zone Burned Gas”,
blowby coefficient ( ) Combustion and Flame, vol. 102, 1995, pp. 241-255.
E total energy (kJ) [16] E. Sher and Y. Hacohen, “On the Modeling of a SI 4-Stroke Cycle
Engine Fueled with Hydrogen-Enriched Gasoline”, Int. J. Hydrogen
standard-state Gibbs free energy Energy, vol. 12, 1987, pp. 773-781.
h specific enthalpy ( ) or [17] R. R. Stone, Introduction to Internal Combustion Engines. Macmillan
Press, 1999.
heat transfer coefficient ( ) [18] G. Woschni, “A Universally Applicable Equation for the
Instantaneous Heat Transfer Coefficient in the Internal Combustion
K equilibrium const. Engine”, SAE Paper 670931, 1967.
m mass (kg)
Nu Nusselt number
p pressure (Pa)
Q heat transfer (kJ)
r compression ratio
R Gas constant
Re Reynolds number
s specific entropy ( )
T temperature (K)
u specific internal energy ( )
U internal energy (kJ)
v specific volume ( )
V volume ( )
W work done (kJ)
x burnt mass fraction
crank angle ( )
start of combustion ( )
total combustion duration ( )

ISBN: 978-988-17012-7-5 IMECS 2009