International Communications in Heat and Mass Transfer 33 (2006) 1264 – 1272

[Link]/locate/ichmt

Thermodynamic modeling of spark-ignition engine: Effect of

temperature dependent specific heats ☆
E. Abu-Nada ⁎, I. Al-Hinti, A. Al-Sarkhi, B. Akash
Department of Mechanical Engineering, Hashemite University, Zarqa, 13115, Jordan

Available online 18 July 2006

Abstract

This paper presents thermodynamics analysis of spark-ignition (SI) engine. A theoretical model of air-standard Otto cycle
having temperature dependent specific heats has been implemented. It was compared to that which uses constant temperature
specific heats. Wide range of engine parameters was studied. In most cases there were significant variations between the results
obtained by using temperature dependent specific heats with those obtained at constant specific heats especially at high engine
speeds. Therefore, it is more realistic to use temperature dependent specific heat. This should be considered in cycle analysis;
especially that temperature variation in the actual cycles is quite large.
© 2006 Elsevier Ltd. All rights reserved.

Keywords: Temperature dependent specific heats; SI engine; Otto cycle

1. Introduction

In thermal design of the internal combustion engines most researchers use air-standard power cycle models to
perform their thermodynamic analyses. Such models are used for comparison reasons in order to show the effect of
varying engine parameters, conditions, fluid properties, etc. In most previous studies on air-standard power cycles, air
was assumed as the working fluid as an ideal gas with constant specific heats without taking into consideration
temperature dependence of the specific heats of the working fluid [1–10]. However, due to the high rise in combustion
temperature this assumption becomes less realistic. Although air-standard power cycle analysis gives only
approximation to the actual conditions and outputs [11], it would be very useful to compare the performance of air-
standard power cycles using constant- and variable-specific heats assumptions. In a recent study the effect of various
internal combustion engines parameters in the SI engine were studied [12]. Some studies presented the effect of having
temperature dependent specific heats on various air-standard cycles such as Otto, Diesel, and Miller [13–15]. However,
the model used for temperature dependent specific heats was a linear model. Constant specific heat models may be used
for very small temperature variations. Also, linear models can be applied with moderate temperature changes.
However, for large changes in temperature, more accurate models are needed. In this study a more realistic approach on
the behavior of variable specific heats will be implemented on the performance evaluation of the SI engine.

☆
Communicated by W. J. Minkowycz.
⁎ Corresponding author. Fax: +962 5 382 6613.
E-mail address: eiyad@[Link] (E. Abu-Nada).

0735-1933/$ - see front matter © 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/[Link].2006.06.014
 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272 1265

Nomenclature

A Heat transfer area, m2

AF Air–fuel ratio, dimensionless
Cp Constant pressure specific heat, kJ/kgK
Cv Constant volume specific heat, kJ/kgK
D Cylinder diameter, m
hcg Heat transfer coefficient for gases in the cylinder, W/m2K
k Specific heat ratio, dimensionless
LHV Lower heating value, kJ/kg
l Connecting rod length, m
m Mass of cylinder contents, kg
mf Mass of burned fuel, kg
P Pressure inside cylinder, Pa
Pi Inlet pressure, Pa
Q Heat transfer, kJ
Qin Heat added from burning fuel, kJ
R Crank radius, m
Rg Gas constant, kJ/kgK
T Gas temperature in the cylinder, K
Ti Inlet temperature, K
Tw Cylinder temperature, K
U Internal energy, kJ
Up Piston speed, m/s
V Cylinder volume, m3
Vc Clearance volume, m3
Vd Displacement volume, m3
x Distance from top dead center, m
xb Burning rate of the fuel, dimensionless
w Average cylinder gas velocity, m/s
θ Angle, degree
θs Start of combustion or heat addition, degree
Δθ Duration of combustion, degree

2. Thermodynamics properties of air

In most air-standard power cycle models air is assumed to behave as an ideal gas with constant specific heats. The
values of specific heats are usually used as cold properties. This assumption can be valid only for small temperature
differences. However, the assumption would produce greater error in all air-standard power cycles. In order to account
for the large temperature difference encountered in air-standard power cycles, constant average values of specific heats
and specific heat ratios are sometimes used. These average values are evaluated using the extreme temperatures of the
cycle, and are believed to yield better results. Obviously, this remains a rough simplification and can result in
significant deviations from reality. Thus, the incorporation of variable specific heats in air-standard power cycle models
can improve their predictions and bring them closer to reality.
One model was introduced by a team of researchers called the NASA polynomial equation [16]. It has the following
formula for the temperature range of 200–1000 K:

Cp
¼ 3:56839−6:788729  10−4 T þ 1:5537  10−6 T 2 −3:29937  10−12 T 3 −466:395  10−15 T 4 ð1aÞ
Rg
1266 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272

Fig. 1. Cp and k versus temperature for air (dotted lines representing Eqs. (1a) and (1b); solid lines representing Eq. (2)).

However, for the temperature range of 1000–6000 K, the above equation is written as:
Cp
¼ 3:08793 þ 12:4597  10−4 T −0:42372  10−6 T 2 þ 67:4775  10−12 T 3 −3:97077  10−15 T 4 ð1bÞ
Rg

Moreover, there is an alternative equation that can be used, except for lower temperature range. It is obtained from
Sonntag et al. [17]. It is based on the assumption that air is an ideal gas mixture containing 78.1% nitrogen, 20.95%
oxygen, 0.92% argon, and 0.03% carbon dioxide (on mole basis). It is presented in the following equation:

Cp ¼ 2:506  10−11 T 2 þ 1:454  10−7 T 1:5 −4:246  10−7 T þ 3:162  10−5 T 0:5 þ 1:3303−1:512
 104 T −1:5 þ 3:063  105 T −2 −2:212  107 T −3 ð2Þ

Fig. 1 presents comparisons between plots of Eqs. (1a), (1b) and (2). Eqs. (1a) and (1b) can be applied to a wide
temperature range of 200–6000 K. However, Eq. (2) covers lower temperature range of 300–3500 K. It is found from
the figure that specific heat at constant pressure increases with temperature from about 1.0 kJ/kg-K at 300 K to about
1.3 kJ/kg-K at about 3000 K. Surely, such difference should be taken into consideration. Similarly, the specific heat
ratio, k, decreases from 1.40 to about 1.28 within the same temperature range.

3. Thermodynamic analysis of air-standard Otto cycle

For a closed system, a small change of the process, the first law of thermodynamics is simply written as:
yQ−yW ¼ dU ð3Þ

Therefore, by using the definition of work, the first law can be expressed as:

yQ−PdV ¼ dU ð4Þ

For an ideal gas the equation of state is expressed as:

PV ¼ mRg T ð5Þ

By differentiating Eq. (5), we can get:

PdV þ V dP ¼ mRg dT ð6Þ

Also, for an ideal gas with constant specific heats the change in internal energy is expressed as:

dU ¼ mCv dT ð7Þ
 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272 1267

By substituting Eq. (7) into Eq. (6) then:

Cv
dU ¼ ðPdV þ V d PÞ ð8Þ
Rg

By substituting Eq. (8) into Eq. (4), dU can be eliminated and thus the following equation is obtained:

Cv
yQ−PdV ¼ ðPdV þ V d PÞ ð9Þ
Rg

Eq. (9) can be rewritten in terms of heat addition and heat loss as:
Cv
yQin −yQloss −PdV ¼ ðPdV þ V d PÞ ð10Þ
Rg

The total amount of heat input to the cylinder by combustion of fuel in one cycle is:
Qin ¼ mf LHV ð11Þ

The total heat added from the fuel to the system until the crank position reaches angle θ is given as:

QðhÞ ¼ Qin xb ð12Þ

Where xb is the Weibe function that is used to determine the combustion rate of the fuel and is expressed as:
  
h−hs
xb ¼ 1−exp −a ð13Þ
Dh

Also, the total amount of heat loss from the system when the crank moves an increment of δθ:

hcg A
Qloss ¼ ðT −Tw Þyh ð14Þ
x

By substituting Eqs. (11), (12), (13), and (14) into Eq. (10) followed by differentiation with respect to crank angle
(θ), the following equation is obtained:
 
dp k−1 dxb hcg A k p dV
¼ Qin − ðT −Tw Þ −k ð15Þ
dh V dh x 180 V dh

Where, k = Cp / Cv and Rg = Cp − Cv.

Eq. (15) can be solved by using explicit finite difference technique with second order accurate differentiation. The
result is given as:

4 3 k−1 2 k−1 1
pðhÞ ¼ pðh−1Þ− pðh−2Þ þ Qin ð3xb ðhÞ−4xb ðh−1Þ þ xb ðh−2ÞÞ− ðhcg ðhÞAðhÞðT −Tw ÞÞ
3 1 3V  3 3V x
ð16Þ
2kpðh−1Þ V ðh þ 1Þ−V ðh−1Þ
−
3V ðhÞ 2Dh
Where, dP / dθ was expressed as:

dp 3PðhÞ−4Pðh−1Þ þ Pðh−2Þ
¼ ð17Þ
dh 2Dh
1268 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272

The instantaneous cylinder volume, area, and displacement are given by the slider crank model as [18]:

kD2
V ðhÞ ¼ Vc þ xðhÞ ð18Þ
4

kD2 kDS
AðhÞ ¼ þ ½R þ 1−cosðhÞ þ ðR2 −sin2 ðhÞÞ1=2  ð19Þ
4 2

1=2
xðhÞ ¼ ðS þ RÞ−½RcosðhÞ þ fS 2 −R2 sin2 ðhÞg  ð20Þ

Once the pressure is calculated, the temperature of the gases in the cylinder can be calculated using the equation of
state as:
PðhÞV ðhÞ
T¼ ð21Þ
mRg

The convective heat transfer coefficient in Eq. (15) hcg is given by the Woschni model as [19,20]:

hcg ¼ 3:26D−0:2 P 0:8 T −0:55 w0:8 ð22Þ

Where, w is the velocity of the burned gas and is given as:

 
Vd Tr
wðhÞ ¼ C1 Up þ C2 ½PðhÞ−Pm  ð23Þ
pr V r

In the above equation, the displacement volume is Vd, and Vr, Tr, and Pr are reference state properties at closing of
inlet valve and Pm is the pressure at same position to get P without combustion (pressure values in cranking). The
values of C1 and C2 are given as:
For compression: C1 = 2.28, C2 = 0 and for combustion and expansion: C1 = 2.28, C2 = 0.00324.
Eq. (16) is solved for each crank angle for −180 ≤θ ≤ 180 using a step size δθ = 1°. The values of θ =± 180 correspond to
BDC whereas the value of θ = 0 corresponds to TDC. The heat addition in Eq. (16) is only valid for θs b θ b (θs +Δθ), i.e.,
during the period of combustion. In solving Eq. (16), notice that k, P, T, and hcg are coupled, i.e., solution of one of these
variables depends on the solution of others. The solution procedure is as follows: The temperature of the gases is first
calculated using Eq. (21). Then, the value of the pressure is determined using Eq. (16). Once the value of the pressure is
obtained, the temperature dependent properties Cp(T) is calculated using Eq. (2). The value of Cv(T ) then is determined
from the relation [Cv(T)=Cp(T)−R]. Thereafter, the value of k is calculated as k(T )=Cp(T ) /Cv(T ). Finally, the heat transfer
coefficient is calculated using the Woschni model given by Eq. (22). The mentioned procedure is repeated for each value of
θ many times until the change between two successive iterations for all variables (T, P, k, and hcg) is less than 10− 4.

Table 1
Engine and operational specifications used in simulation
Fuel C8H18
Compression ratio 8.3
Cylinder bore (m) 0.0864
Stroke (m) 0.0674
Connecting rod length (m) 0.13
Crank radius (m) 0.0337
Clearance volume (m3) 5.41 × 10− 5
Swept volume (m3) 3.95 × 10− 4
Engine speed (rpm) 2000–5000
Inlet pressure (bar) 1
Equivalence ratio 1
Ignition timing − 25° BTDC
Duration of combustion 70°
Wall temperature (K) 400
 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272 1269

Fig. 2. Variation of cylinder pressure versus volume for SI engine using constant-specific heats running at 2000 rpm and 5000 rpm at air–fuel ratio
of 15.

Parametric studies have been performed based on the numerical solution of Eq. (16). The study covers wide range of
dependent variables such as engine speed, air–fuel ratio and others, taking into consideration the variation of the
specific heat with temperature. Engine specifications, dimensions, and other constants used in the parametric study are
listed in Table 1.

4. Results and discussion

In order to examine the validity and sensitivity of the presented model, cylinder pressure is presented in Fig. 2. It shows the
variation of cylinder pressure versus volume for SI engine using constant-average specific heats running at piston speeds of 2000 and
5000 rpm at a given air–fuel ratio of 15. It is obvious that cylinder pressure is higher at higher engine speeds. For example, a
maximum pressure of 48 and 53 bars is reported at engine speeds of 2000 and 5000 rpm, respectively.
In order to study the effect of temperature dependent specific heats, Fig. 3 is presented. It shows variation of pressure and gas
temperature versus crank angle using variable and constant-average specific heats running at engine speed of 5000 rpm and in-
cylinder air–fuel ratio of 15. It is obvious that there is some difference when temperature dependent specific heat is used instead of
constant specific heat. Although they have similar trends, the maximum temperature and pressure with constant specific heats are
significantly over-estimated in comparison with results obtained with variable specific heats.
The effects of engine speed and air–fuel ratios on the gas outlet temperature, using constant and variable specific heat models, are
presented in Fig. 4. Fig. 4(a) presents outlet gas temperature versus engine speed at air–fuel ratios of 14 (rich mixture) and 16 (lean
mixture). However, Fig. 4(b) presents outlet gas temperature versus air–fuel ratio at engine speeds of 2000 rpm (low speed) and
5000 rpm (high speed). Higher outlet temperatures are obtained at higher engine speeds and lower air–fuel ratios. It is interesting to
note that the effect of temperature dependent specific heat is very significant on the gas outlet temperature. Similarly, the effects of
engine speed and air–fuel ratios on the maximum gas temperature, using constant and variable specific heat models, are presented in

Fig. 3. Variation of gas temperature and cylinder pressure versus crank for SI engine using variable and constant-specific heats running at 5000 rpm
and air–fuel ratio of 15.
1270 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272

Fig. 4. (a) Outlet gas temperature versus engine speed at different air–fuel ratios using constant and variable specific heat models. (b) Outlet gas
temperature versus air–fuel ratio at different engine speeds using constant and variable specific heat models.

Fig. 5. Fig. 5(a) presents the maximum gas temperature versus engine speed at air–fuel ratios of 14 (rich mixture) and 16 (lean
mixture). On the other hand, Fig. 5(b) presents maximum gas temperature versus air–fuel ratio at various engine speeds of 2000 rpm
(low speed) and 5000 rpm (high speed). Higher values of maximum temperatures are obtained at higher engine speeds and lower
air–fuel ratios. Again, as noted previously, the effect of temperature dependent specific heat is very significant on the reported
maximum gas temperatures. The effects of engine speed and air–fuel ratios on brake mean effective pressure (BMEP), using
constant and variable specific heat models, are presented in Fig. 6(a), which presents BMEP versus air–fuel ratio at engine speeds of
2000 rpm (low speed) and 5000 rpm (high speed). Higher values of BMEP are obtained at higher engine speeds and lower air–fuel
ratios. Unlike temperatures, BMEP was not affected significantly with temperature dependent specific heat.
Finally, effect of temperature dependence was investigated on cycle efficiency. The results are presented in Fig. 6(b). It is found
that higher thermal efficiencies were produced at higher engine speeds and higher air–fuel ratios. Also, it was found that at low
engine speeds using constant or variable specific heat models makes no significant difference on efficiency. However, when the
engine runs on high speed it makes a difference whether constant specific heat or variable specific heat models were used; i.e., at
higher engine speeds variable specific heat model becomes more considerable.

5. Conclusion

It is concluded that engine working parameters are affected by variable specific heats, significantly. The results show
that there is a great effect of the temperature dependent specific heat of the working fluid on the performance of the air-

Fig. 5. (a) Maximum gas temperature versus engine speed at different air–fuel ratios using constant and variable specific heat models. (b) Maximum
gas temperature versus air–fuel ratio at different engine speeds using constant and variable specific heat models.
 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272 1271

Fig. 6. (a) Brake mean effective pressure versus air–fuel ratio at different engine speeds using constant and variable specific heat models. (b)
Efficiency versus air–fuel ratio at different engine speeds using constant and variable specific heat models.

standard Otto cycle. Therefore, it is more realistic to use temperature dependent specific heat during the investigation of
air-standard power cycles. This should be considered in the practical cycle analysis, especially, in the actual cycles the
temperature variations are quite large. The results are expected to provide significant guidance for the performance
evaluation and improvement of real SI engines.

References

[1] B. Akash, Effect of heat transfer on the performance of an air-standard diesel cycle, International Communications in Heat and Mass Transfer 28
(1) (2001) 87–95.
[2] A. Al-Sarkhi, B. Akash, J. Jaber, M. Mohsen, E. Abu-Nada, Efficiency of Miller engine at maximum power density, International
Communications in Heat and Mass Transfer 29 (8) (2002) 1159–1167.
[3] Y. Ge, L. Chen, F. Sun, C. Wu, Effect of heat transfer and friction on the performance of an irreversible air-standard miller cycle, International
Communications in Heat and Mass Transfer 32 (2005) 1045–1056.
[4] L. Chen, C. Wu, F. Sun, S. Cao, Heat transfer effects on the net work output and efficiency characteristics for an air-standard Otto cycle, Energy
Conversion and Management 39 (1998) 643–648.
[5] A. Parlak, Comparative performance analysis of irreversible dual and diesel cycles under maximum power conditions, Energy Conversion and
Management 46 (3) (2005) 351–359.
[6] A. Parlak, The effect of heat transfer on performance of the diesel cycle and exergy of the exhaust gas stream in a LHR Diesel engine at the
optimum injection timing, Energy Conversion and Management 46 (2) (2005) 167–179.
[7] S. Hou, Heat transfer effects on the performance of an air standard dual cycle, Energy Conversion and Management 45 (2004) 3003–3015.
[8] A. Jafari, S. Hannani, Effect of fuel and engine operational characteristics on the heat loss from combustion chamber surfaces of SI engines,
International Communications in Heat and Mass Transfer 33 (2006) 122–124.
[9] O. Ozsoysal, Heat loss as a percentage of fuel's energy in air standard Otto and Diesel cycles, Energy Conversion and Management 47 (2006)
1051–1062.
[10] P. Wang, S. Hou, Performance analysis and comparison of an Atkinson cycle coupled to variable temperature heat reservoirs under maximum
power and maximum power density conditions, Energy Conversion and Management 46 (2005) 2637–2655.
[11] W. Pulkrabek, Engineering Fundamentals of the Internal Combustion Engine, second ed., Pearson Prentice-Hall, Upper Saddle River, New Jersey, 2004.
[12] M. Karamangil, O. Kaynakli, A. Surmen, Parametric investigation of cylinder and jacket side convective heat transfer coefficients of gasoline
engines, Energy Conversion and Management 47 (2006) 800–816.
[13] A. Al-Sarkhi, J. Jaber, M. Abu-Qudais, S. Probert, Effects of friction and temperature-dependent specific-heat of the working fluid on the
performance of a diesel-engine, Applied Energy 83 (2) (2006) 153–165.
[14] A. Al-Sarkhi, J. Jaber, S. Probert, Efficiency of a Miller engine, Applied Energy 83 (4) (2006) 343–351.
[15] Y. Ge, L. Chen, F. Sun, C. Wu, Thermodynamic simulation of performance of Otto cycle with heat transfer and variable specific heats of
working fluid, International Journal of Thermal Sciences 44 (5) (2005) 506–511.
[16] A. Burcat, B. Ruscic, Third Millennium Ideal Gas and Condensed Phase Thermochemical Database for Combustion with Updates from Active
Thermochemical Tables, Argonne National Laboratory, (2005), Report number ANL-05/20, [Link]
[Link].
1272 E. Abu-Nada et al. / International Communications in Heat and Mass Transfer 33 (2006) 1264–1272

[17] R. Sonntag, C. Borgnakke, G. Van Wylen, Fundamentals of Thermodynamics, fifth [Link], New York, 1998.
[18] C. Ferguson, A. Kirkpatrick, Internal Combustion Engines: Applied Thermosciences, Wiley, New York, 2001.
[19] G. Woschni, Universally Applicable Equation for the Instantaneous Heat Transfer Coefficient in Internal Combustion Engine, SAE Paper
670931 (1967).
[20] R. Stone, Introduction to Internal Combustion Engines, third ed., Antony Row, UK, 1999.