Numerical Investigation on Effect of

Spark Plug Configuration on

Performance in an Engine Cylinder
A numerical investigation of combustion inside single and twin-spark
engines was performed to study the effect of a spark plug, positions and
Bakhshi Mehul
spark timings on engine performance. Improvement in engine performance is
Student
Department of Mechanical Engineering one of the automotive industry's primary research areas. Consequently, the
K.K. Birla Goa Campus, BITS Pilani study's results can be utilised to optimise engine configurations to achieve
India maximum performance. The investigation was conducted using a finite
volume-based open-source software, OpenFOAM, for computational
Pritanshu Ranjan
simulations. Simulations were conducted using the XiEngineFOAM solver
Assistant Professor
Department of Mechanical Engineering with a transport equation for modelling flame fronts. The Standard k-ε
K.K. Birla Goa Campus, BITS Pilani turbulence model was used to predict turbulence parameters. The simulation
India was conducted for compression and power stroke (crank angle between -
180° and 180°), assuming an even distribution of the air-fuel mixture within
Anuj Kumar Shukla
the pentroof 4-valve engine cylinder. Simulations were conducted for four
Assistant Professor
Department of Mechanical Engineering cases, including variations in the position and timing of spark plugs in
NIT Raipur single-spark and twin-spark engines. According to the results of the
India simulations, the single-spark engine provides the best performance when the
spark plug is ignited early and positioned at the cylinder's centre. When
placed at an optimal position determined by flame travel and collision, the
twin-spark engine gives the best performance at the highest difference
between the spark timings of the two spark plugs.

Keywords: OpenFOAM, premixed turbulent combustion, twin-spark, Kiva

test, flame propagation, indicated power, PIMPLE.

1. INTRODUCTION reaction progression were observed to be higher.

Research on the twin-spark engine has revealed that the
The spark plug plays an important role in the ignition of position of the spark plug has little effect on the
the air-fuel mixture in an internal combustion engine. pressure regime [3]. However, for a direct injection
They convert the high-voltage electricity into an methanol engine with a medium compression ratio, the
artificial bolt of lightning known as a spark, which greatest results were obtained with a 0.65 ratio between
paves the way to the initial ignition of the highly the distance of the twin spark plugs and the cylinder
flammable air-fuel mixture. Since the invention of spark diameter [4]. An experimental investigation of a twin
plugs, it has been widely used in engines and has also spark engine with three various ignition timings and
found some applications in industry to ignite fuel in load conditions determined that the same spark timings
furnaces, boilers and combustion chambers. The of the spark plugs resulted in the highest brake thermal
configurations of spark plugs, such as position and efficiency, volumetric efficiency, and lowest CO and
timing, can significantly impact engine performance. UBHC emissions [5]. There have been some other past
The timing of the spark plug can have an impact on the studies as well, discussing the variation of spark plug
knock tendency and combustion duration, while the position [6, 7], ignition timing [8, 9, 10] and flame front
position of the spark plug can have an effect on the collision [11, 12] in twin spark engines within the
homogeneity and turbulence of the air-fuel mixture engine cylinder. All of these past studies provides us
inside the engine cylinder. with an understanding of how spark plug parameters
The effect of spark plug parameters on engine can affect engine performance. However, we found a
performance has been investigated in a considerable lack of literature that explicitly discusses together the
number of past studies. Past research has demonstrated variation of ignition position and timing through flame
that a larger engine has a greater impact on spark plug front propagation and interaction for single and twin
placement than a smaller engine. When a spark plug spark plug engines.
was positioned in the centre of the engine cylinder [1, 2] The present investigation seeks to fill this gap in the
or had a greater spark advance [3], peak pressure and literature by conducting numerical investigations on the
effect of spark plug configurations on engine cylinder
Received: June 2023, Accepted: September 2023 performance and a thorough comparison of the
Correspondence to: Bakhshi Mehul, Department of variations. Additionally, the research seeks to determine
Mechanical Engineering, K.K. Birla Goa Campus, BITS the optimal spark position and timing for both single
Pilani, India, E-mail: [email protected] and twin-spark engines and discuss the effect of flame
doi: 10.5937/fme2304585M front propagation and collision inside the cylinder. The
© Faculty of Mechanical Engineering, Belgrade. All rights reserved FME Transactions (2023) 51, 585-594 585
investigation was conducted using the simulation
software OpenFOAM. Simulations were conducted to
examine the impact of spark plug configurations on a
variety of engine parameters, including in-cylinder
pressure, in-cylinder temperature, indicated power, and
flame propagation. As the automotive industry strives to
satisfy increasingly stringent environmental regulations
while also enhancing the performance and efficiency of
internal combustion engines, this research is particularly
pertinent today. By obtaining a deeper understanding of
the effect of spark plug configurations on engine per–
formance, it will be possible to identify spark plug con–
figurations that concurrently improve engine perfor– Figure 2. Computational domain for simulation of engine
mance and reduce emissions. combustion.
This paper will commence by describing the study's
2.2 Meshing Details
methodology, including simulation setup and validation.
The paper will conclude with an analysis of the nume–
The computational domain was dynamically meshed
rical simulation results, giving the optimal arrangement
using an automatic mesh motion solver with layered
of spark plug positions and spark timings in an ideally
meshing. The mesh used in this study was identical to
homogenous charge for best results in both single and
that of the KIVA-3V 4 valve model [14] used in
twin spark engines.
OpenFOAM with the kivaToFoam grid conversion
2. COMPUTATIONAL METHODOLOGY utility. Table 2 and Figure 3 provides information and a
visual representation of the mesh at the beginning of the
2.1 Computational Domain and Engine Details simulation, i.e. bottom dead centre (BDC).
Table 2. Mesh details at the start of the simulation
The pentroof-4-valve engine was used in the present
study. This cylinder design has proved to be effective Mesh Statistics Value/Detail
for fast burn time and producing high horsepower. The Cell type Hexahedral
geometry is kept the same as that of the already present Number of cells 27,544
tutorial of XiEngineFOAM (Kiva Test) in OpenFOAM. Number of points 30,742
Faces on piston 1,326
Faces on liner 2,710
Faces on the cylinder head 2,184
Internal faces 79,522
Maximum aspect ratio 34.032
Maximum skewness 3.739

Figure 1. Isometric-view (left) and top-view (right) of exha–

ust and inlet valves of pent-roof 4-valve engine.

The geometry has three boundary conditions, i.e.

cylinder head, liner and piston [13], where the piston is
moving with every time step, as depicted in Figure 2.
Other details of the engine are mentioned in Table 1.
Table 1. Engine details
Figure 3. (a) Meshing of the engine cylinder during the start
S. No. Engine Specifications Value/Detail of the simulation. (b) Cross-sectional view of the meshing
1 Bore Diameter (D), mm 92 domain.
2 Connecting Rod Length, mm 147
3 Stroke Length (L), mm 84.23 2.3 Governing Equations
4 Clearance, mm 1.15
5 Compression Ratio 8.82 EngineFoam solver [15] was used to simulate
6 Engine Speed (N), rpm 1500 combustion in our study. The propagation of the flame
7 Fuel IsoOctane front is represented by a progress variable. The progress
8 Stoichiometric Air Fuel Mass 15.0336 variable can be denoted with the help of any quantity,
Ratio like temperature or react mass fraction. In our case, we
9 The diameter of the central 2 used temperature to calculate the progress variable.
electrode of the spark plug, mm
10 Spark Plug Duration, CA 20 T  Tu
c (1)
11 Strength of Spark Plug, mJ 4.5 Tb  Tu

586▪ VOL. 51, No 4, 2023 FME Transactions

 The value of c lies between 0 and 1. Where 0 repre– Table 4. Discretisation schemes used in the present
simulation
sents a completely fresh air-fuel mixture, and 1 represents
a completely burnt air-fuel mixture. The value of c Mathematical term Discretisation scheme
between 0 and 1 represents an ongoing combustion Time derivative (∂⁄∂t, ∂2⁄∂2t) Euler
reaction. The flame front is modelled using the transport Gradient (  ) Gauss linear
equation, where a regress variable is defined as having Divergence (  ) Gauss upwind, gauss
exactly the opposite physical meaning of the progress linear, gauss linear
variable. limited
Laplacian (  2) Gauss linear limited
b  1c (2) corrected
  b  t  Point-to-point interpolation of Linear
      u  b           Sc (3)
t  Sct  b  values
Component of gradient normal Limited corrected
The reaction source term can be represented as to a cell face

  Sc   u  Su   b The PIMPLE (Pressure Implicit with Splitting of

  b  t  Operators Multi-Pressure-Linked Equations) algorithm
      u  b       (4) was used to solve the discretised equations. Combining
t  Sct  b 
the SIMPLE and PISO algorithms, this algorithm is
   u  Su   b ideally adapted for solving turbulent phenomena. In the
For simulating the flow field, the RANS equation current simulation, the value of correctors for the
with the standard k-ε turbulence model was used. In the PIMPLE algorithm is set to 2, corresponding to the
present study, our major focus was on flame pro– number of times the algorithm solves the pressure and
pagation and its effects on engine performance. Further momentum in each time step. The value of non-
details of continuity, momentum, turbulent kinetic orthogonal correctors is set to 1. In the current
energy and dissipation, internal energy, energy of state simulation, the moment predictor was utilised. The
and chemical reaction can be found in [16, 17] as well momentum predictor is a first approximation of the
as in the appendix section. velocity field obtained from the solution of a
momentum equation assembled with the previous time-
2.4 Initial and Boundary Conditions step pressure field.
The absolute convergence criterion (tolerance) for
During the beginning of the numerical simulation, the various parameters is listed for the current simulations
values of various parameters were specified, as shown in Table 5.
in Table 3. The initial velocity is regarded to vary Table 5. Converging criteria specified in the present
linearly, from zero at the cylinder head to the piston simulation
speed at the piston head. The liner and cylinder head Parameter Tolerance
were assumed to have no-slip conditions. The boundary Pressure (p) 10-6
layers were presumed to be turbulent and followed the Density (ρ) 10-5
law-of-the-wall velocity profile to aid in calculating Velocity (U) 10-5
heat and momentum losses. At the onset of the Flame wrinkling factor (ξ) 10-5
simulation, the air-fuel mixture was assumed to be Regress variable (b) 10-5
evenly distributed throughout the cylinder and at rest. Turbulent kinetic energy dissipation rate (ε) 10-5
Table 3. Values of different parameters during the start of
the simulation The progression of the simulations was measured in
terms of crank angle, with the bottom dead centre
Parameter Initial Value
(BDC) at -180° and the top dead centre (TDC) at 0°. To
Turbulent thermal diffusivity (αt) 0 m2/s
precisely capture the flame propagation, the time step
Regress variable (b) 1
was maintained at 0.25° and decreased to 0.025° at the
Turbulent kinetic energy dissipation 450 m2/s2
rate (ε) ignition's onset. The time steps met the CFL criteria
Turbulent kinetic energy (k) 4 m2/s2 effectively.
Turbulent viscosity (μt) 0 m2/s
2.6 Validation
Laminar flame speed (Su) 0.4 m/s
In-cylinder pressure (p) 1.9 bar
Internal temperature of the cylinder (T) 600 K
Figure 4 shows the comparison of the simulation with
Temperature at the walls (T) 450 K the findings of Anetor et al. [16]. Initial pressure and
Flame wrinkling factor (ξ) 1 temperatures were maintained at the same levels as
specified above. The spark was given 15° before the top
dead centre. Thus, we received pressure readings similar
2.5 Numerical Schemes and Solver
to the reference until the spark plug ignited. Following
the ignition, there seemed to be pressure value
The governing equations were discretised using the
fluctuations with a maximum error of 11.69%, which
schemes mentioned in Table 4. More details of the
was still within the permitted error margins. Besides
schemes can be found in [18].

FME Transactions VOL. 51, No 4, 2023▪587

this, the numerical schemes used have been previously Indicated Power (IP) is another parameter that we
validated in [19, 20, 21]. used to analyse engine performance. As IP is directly
related to Indicated Efficiency (IE) and inversely pro–
3. RESULTS portional to Indicated Specific Fuel Consumption
(ISFC), the greatest IP corresponds to the highest IE and
Simulations were conducted for the engine's compre– lowest ISFC. IP was derived from indicated mean
ssion and power strokes (-180° to 180°) to understand effective pressure (IMEP) using the formula:
the effect of spark plug position and spark timing.
IMEP  bar  * L *  * D2 * N *100
3.1 Effect of spark plug position on single spark IP  kW   (5)
240* n
engine
Here, the value of n is 2 for a four-stroke engine.
The variation in the position of the spark plug was IMEP signifies average pressure over an engine cycle.
analysed by placing the spark plug at four different Thus, it was calculated using the p-v diagram of the
positions, i.e. 0.03 m, 0.02 m, 0.01 m and 0 m from the compression and power stroke. A p-v diagram for the
centre of the engine cylinder as depicted in Figure 5, spark plug position 0.03 m from the centre is shown in
adapted from [22]. Figure 7. The diagram illustrates the pressure and disp–
lacement volume for compression and power stroke,
while the intake and exhaust strokes were not taken into
consideration. The area enclosed by the curve is equi–
valent to the product of volume and IMEP. Table 6
displays the I.P. variance for different spark plug
locations. It can be inferred that the spark plug posi–
tioned in the middle of the engine cylinder achieves the
maximum I.P.

Figure 4. Comparison of in-cylinder pressure of present

simulations with the results of Anetor et al. [9]

The ignition timing for all the cases was kept

constant at 25° before TDC. Figure 6 depicts the in-
cylinder pressure vs. crank angle for the four different
spark plug locations. When the spark plug is positioned
in the centre of the engine cylinder, a peak pressure of
slightly more than 40 bar is recorded. The peak pressure
drops when the spark plug goes farther from the centre. Figure 6. Pressure (Pa) vs. crank angle (degree) for diffe–
rent positions of the spark plug in the single-spark engine.
The reason behind this drop in pressure is associated
with flame propagation, which is explained below.

Figure 7. Pressure vs. displacement volume for compres-

sion and power stroke in the single spark engine cylinder.
Figure 5. The spark plug placed between the exhaust and Spark plug position at 0.03 m from the centre.
inlet valves.

588▪ VOL. 51, No 4, 2023 FME Transactions

Table 6. Indicated Power for different positions of spark case. This results in increased pressure and temperature,
plug
which is supported by the piston that continuously com–
Distance from the centre (m) Indicated Power (kW) presses the air-fuel mixture. However, it is necessary to
0.03 6.85 determine the optimal spark timing, as igniting
0.02 7.90 significantly earlier can lead to knocking due to a rapid
0.01 8.31 increase in pressure and temperature.
0 8.61

The results demonstrate that the optimum engine

performance is achieved when the spark plug is
positioned in the cylinder's centre. This can be
explained by the even flame propagation and
combustion that occurs when the spark is positioned in
the geometric centre of the combustion chamber. Any
position other than the centre of the engine cylinder
produces an uneven flame, hence decreasing engine
performance. Figures 8(a) and 8(b) depict flame
propagation for spark plugs located 0.03m from the
centre and at the centre, respectively. When the ignition
is initiated at -25° for both cases, the flame covers the
entire engine cylinder at 50° for the case where the
spark plug is positioned in the centre, unlike the
position away from the centre. This demonstrates that Figure 9. Pressure (Pa) vs crank angle (degree) for different
flame propagation triggered by a spark plug at the spark timings in a single spark engine
engine's centre results in more rapid combustion than a Table 7. Indicated Power for different spark timings of
spark initiated away from the engine's centre. spark plug

Crank Angle Indicated Power (kW)

-5° 5.94
-10° 6.61
-15° 7.30
-20° 7.98
-25° 8.61

3.3 Effect of spark plug position on twin-spark

engine

We attempted to determine the impact of spark plug

configurations on a twin-spark engine. Even flame pro–
pagation was ensured by positioning the spark plugs in
symmetrically opposite positions between the inlet and
exhaust valves, as shown in Figure 11. Considering the
results of a single spark engine, the spark plugs were
Figure 8. Flame propagation for different spark plug posi– ignited simultaneously 25° before TDC. Three hypot–
tions. (a) Spark plug at 0.03 m from the centre. (b) Spark
plug at the centre of the engine cylinder. hetical spark plug positions for computational simu–
lations were analysed: 0.02 m, 0.01 m, and 0.005 m
3.2 Effect of spark time on the single-spark engine from the engine's centre.
Figure 12 and Table 8 demonstrate little difference
For the single-spark engine, the spark timing was in performance between the three positions. However,
changed between 5° and 25° before TDC while main– the intermediate position (0.01 m) produced slightly
taining the spark plug in the middle of the engine higher in-cylinder peak pressure and indicated power
cylinder. It was noted that the pressure within the cy– than the other two configurations.
linder increased as one moved away from the TDC. The This behaviour is explicable by the flame propa–
spark plug ignited 25° before TDC recorded the highest gation and flame front collision phenomena. When spark
peak pressure of over 40 bar, as depicted in Figure 9. plugs are positioned 0.02 m from the centre, as depicted
Similarly, Table 7 shows a rise in I.P. as we spark away in Figure 13(c), the flame fronts have to travel a greater
from the TDC. distance to reach the centre and merge, as shown in
The results of this case study demonstrate that Figure 13(c) at 10°. Due to the increasing volume of the
optimal engine performance is attained when the spark engine cylinder after 0°, the flame gets weaker as it
plug ignites significantly sooner in the compression moves towards the centre. This decreases cylinder
stroke. This provides ample time for the flame front to pressure and power compared to the 0.01 m spark plug
propagate, i.e., the combustion process. Figure 10 position, as shown in Figure 12 and Table 8. However, a
illustrates an approximately 25° earlier completion of decrease in pressure and temperature can also be
the combustion process for a -25° case than for a -5°

FME Transactions VOL. 51, No 4, 2023▪589

observed in Figure 12 and Table 8 when spark plugs are Table 8. Indicated power (kW) for varying positions of
spark plugs in twin spark engine
positioned too close at a distance of 0.005 m from the
engine's centre. Even though the flame front has to travel Distance from the centre (m) Indicated Power (kW)
a very short distance to reach the centre of the engine 0.02-0.02 8.90
cylinder, the flame fronts collide at a very early stage, as 0.01-0.01 9.0
depicted in Figure 13(a) at -10°, thus resulting in lower 0.005-0.005 8.90
in-cylinder peak pressure and power. This causes
irregular flame propagation and decreased combustion
efficiency. Therefore, the optimal place–ment of twin
spark plugs in the engine requires a ba–lance between
flame front propagation and collision, such as the case of
a spark plug positioned 0.01m from the centre when
flame collides at 0°, as illustrated in Figure 13(b).

Figure 13. Collision of flame fronts at different time steps

for different positions of twin spark plugs. The three
different positions of spark plugs from the centre are (a)
0.005 m, (b) 0.01 m, (c) 0.02 m.

3.4 Effect of spark time in twin spark engine

Figure 10. Reaction regress variable (b) vs crank angle The variation in spark timings was investigated by ma–
(degree) for spark ignition angle of -25° and -5°.
intaining the spark timing of one spark plug at -25°
before TDC and varying the spark timing of the second
spark plug between -5° and -25°, with a 5° interval. The
spark plugs were positioned at 0.01 m for optimal
placement results.
After simulating the combustion within the cylinder
at various spark timings, it was determined that the
spark configuration of -5° and -25° produced the
maximum peak pressure, as shown in Figure 14.
Figure 15 and Table 9 depict that the in-cylinder
temperature and indicated power are also highest for the
-25°: -5° configuration.
We obtain the best performance when the difference
between the spark timing of two spark plugs is the
greatest because the combustion and compression pro–
Figure 11. The spark plugs are placed at symmetrically cesses are applied for a prolonged period. When the first
opposite positions between the exhaust and inlet valves.
spark plug is ignited, the combustion process has
sufficient time to spread, resulting in increased pressure
and temperature, which are further increased by the
piston, which consistently compresses the air-fuel mix–
ture as it rises. Consequently, igniting the second spark
plug adds more energy to the already heated and
compressed air-fuel mixture, increasing cylinder pres–
sure, temperature, and power.
Table 9. Indicated Power for different spark timings of
spark plugs

Crank Angle Indicated Power (kW)

-25°: -5° 10.01
-25°: -10° 9.37
-25°: -15° 8.79
-25°: -20° 8.70
Figure 12. Pressure (Pa) vs. crank angle (degree) for -25°: -25° 8.99
different positions of the spark plug in a twin-spark engine.

590▪ VOL. 51, No 4, 2023 FME Transactions

 4. A higher difference in spark timings between the two
sparks leads to higher in-cylinder pressure, tem–
perature and power, which again needs to be kept in
check with the knocking tendency of the engine.
5. A single-spark engine's highest pressure, tempe–
rature, and power are obtained when the spark plug
is positioned in the centre and ignited as early as
possible while avoiding knocking.
6. The best results are obtained in a twin-spark engine
when the spark plugs are optimally positioned and
have the greatest spark timing difference.

REFERENCES
[1] Joshi, A., Borse, S.L., Alblawi, A. and Alshqirate,
A.A.: Effect of spark plug position on performance
of engine using openFoam. International Journal of
Figure 14. Pressure (Pa) vs. crank angle (degree) for Mechanical Engineering and Technology (IJMET)
different spark timings in a twin-spark engine Vol. 9 Issue 11 (2018), pp. 114-120.
[2] Altin, I. and Bilgin, A.: A parametric study on the
performance parameters of a twin-spark S.I. engine.
Energy Conversion and Management 50 (2009), pp.
1902-1907.
[3] Joshi, A.R. and Borse, S.L.: Study of The Effect of
Spark Advance, Engine Speed Variation and
Number of Spark Plugs on Engine Performance
Using CFD Software. Journal of Ocean,
Mechanical and Aerospace Science and
Engineering Vol 48 (2017), pp. 1-9.
[4] Gong, C., Sun, J., and Liu, F.: Numerical study of
twin-spark plug arrangement effects on flame,
combustion and emissions of a medium
compression ratio direct-injection methanol engine.
Fuel 279 (2020) 118427.
Figure 15. Temperature (K) vs. crank angle (degree) for [5] Bailkeri, N., Prasad, K.S., and Rao, S.B.R.:
different spark timings in a twin spark engine (Same
legends as of Figure 14). Performance Study on Twin Plug Spark Ignition
Engine at Different Ignition Timings. International
4. CONCLUSION Journal of Science and Research (IJSR) Vol. 2
Issue 8 (2013), pp. 231-236.
Computational simulations were performed to examine [6] Altin, I., Bilgin, A., Sezer, I.: Theoretical investiga-
the effect of spark plug configuration variation (spark tion on combustion characteristics of ethanol-fueled
timing and position) in this study. The simulations were dual-plug S.I. engine. Fuel 257 (2019), 116068.
performed for both single and twin-spark engines. The [7] Yan, X., Feng, H., Zuo, Z., Jia, B., Zhang, Z. and
following conclusion can be derived from the obtained Wu, L.: Research on the influence of dual spark
simulation results. ignition strategy at combustion process for dual
1. As a result of the rapid combustion of the air-fuel cylinder free piston generator under direct
mixture within the engine cylinder, optimal engine injection. Fuel 299 (2021), 120911.
performance is achieved by a flame that spreads [8] Gong, C., Sun, J., Chen, Y. and Liu, F.: Numerical
evenly. study of cold-start performances of a medium
2. Due to the extra time given for flame propagation, compression ratio direct-injection twin-spark plug
i.e. combustion, prior to the power stroke, an early synchronous ignition engine fueled with methanol.
ignition can increase cylinder pressure, tempe– Fuel 285 (2021), 119235.
rature, and power. However, the required pressure [9] Altin, I., Bilgin, A and Ceper, B.A.: Parametric
within the cylinder must be verified by the study on some combustion characteristics in a
knocking tendency of the engine. natural gas fueled dual plug S.I. engine. Energy 139
3. The timing of flame front collision in twin spark (2017), pp. 1237-1242..
engines must be optimal. Neither too early, which [10] Gong, C., Sun, J., Chen, Y. and Liu, F.: Numerical
would result in irregular flame propagation, nor too research on combustion and emissions behaviors of
late, which would result in the collision of wea– a medium compression ratio direct-injection twin-
kened flame fronts. spark plug synchronous ignition methanol engine

FME Transactions VOL. 51, No 4, 2023▪591

 under steady-state lean-burn conditions. Energy Tb Burned gas temperature (K)
215 (2021), 119193. b Regress variable
[11] Altin, I., Bilgin, A.: Quasi-dimensional mode–ling ũ Filtered velocity of flame (m/s)
of a fast-burn combustion dual-plug spark-ignition Reaction source term (net rate of production or
engine with complex combustion chamber geomet– Sc consumption of species due to chemical
ries. Applied Thermal Engineering 87 (2015), pp. reactions)
678-687. Turbulent Schmidt number (ratio of
Sct momentum diffusivity to scalar diffusivity in a
[12] Zhang, S. et al.: Numerical investigation the effects
turbulent flow)
of the twin-spark plugs coupled with EGR on the
Su Laminar flame speed (m/s)
combustion process and emissions characteristics in
a lean burn natural gas S.I. engine. Energy 206
Greek symbols
(2020), 118181.
[13] Kolambe, K.S. and Borse, S.L, Engine combustion ρ Mean gas density (kg/m3)
simulation using openFoam.: International Engi– Wrinkling factor (ratio of turbulent flame
ξ
neering Research Journal (IERJ) Special Issue velocity to laminar flame velocity)
(2016), pp. 885-891 ρu Density of unburned mixture (kg/m3)
[14] Anthony A. A., KIVA-3V: A Block-Structured
APPENDIX
KIVA Program for Engines with Vertical or Canted
Valves, Los Alamos National Laboratory, Los Continuity Equation:
Alamos, New Mexico 87545 LA-13313-MS, UC-
1412 (July 1997).   m     m 
     mu       D      cm
[15] Wadekar, S., engineFoam tutorial with different t    
flame wrinkling (Xi) model, in: Proceedings of
CFD with OpenSource Software, 2018, Edited by Here ρm is the density of species m
Nilsson. H., http://dx.doi.org/10.17196/OS_CFD# ρ is the total mass density

YEAR_2018. u is fluid velocity
D is the coefficient of diffusion
[16] Anetor, L., Edward, E.O., Harris, K.: Simulation
Studies of Combustion in Spark Ignition Engine ρcm is the source term due to chemistry
using Openfoam. FME Transactions (2020) 48, pp.
Summing all the species results in
787-799.
[17] Anetor, L., Edward, E.O., Odetunde, C.: Compa-  
    u   0
rative analysis of combustion dynamics using three t
reaction source models. Australian Journal of Mec–
hanical Engineering (2021). Momentum Equation:
[18] Christopher J. Greenshields, OpenFOAM.: The   u   2  
Open Source CFD Toolbox Programmer's Guide,      uu   p            g
t 3 
Version 3.0.1, OpenFOAM Foundation Ltd,
December 2015. Here, p is fluid pressure.
[19] Jasak, H. and Gosman, A.D.: Automatic resolution σ stress tensor is Newtonian in form,
control for the Finite Volume Method. Part 1: A-   
posteriori error estimates, Numerical Heat Transfer,    eff u   u      uI
 
Part B, 38(3) (September 2000), pp. 237–256.
[20] Jasak, H., Gosman, A.D.: Residual error estimate
for the Finite Volume Method, Int. Journal of Turbulent Kinetic Energy and Dissipation Models:
Numerical Methods and Fluids 39 (2001), pp. 1-9.
  k  2   
[21] Weller, H.G., Tabor, G., Jasak, H.: and Fureby, C.:      uk      k    u   : u 
t 3 
A tensorial approach to computational continuum
mechanics using object orientated techniques,   eff  
Computers in Physics 12(6) (1998), pp. 620-631.    k   
 P rk  
[22] F., Claudio, Bianchi G.M., Corti, E. and Fantoni S.:
Evaluation of the effects of a Twin Spark Ignition Here, k is turbulent kinetic energy
System on combustion stability of a high ε is the dissipation rate of k
performance PFI engine, in: Processings of 69th
Conference of the Italian Thermal Engineering
Association, 10-13.09.2014, Milan, pp. 897-906.      2  
     u     c 1  c 3    u 
t 3 
NOMENCLATURE   eff    
c Progress variable
      c 1 : u  c 2  
 P r   k
Tu Unburned gas temperature (K)

592▪ VOL. 51, No 4, 2023 FME Transactions

   k r  k fr   m / Wm 

amr

Here, represents convection of turbulence by the
t m

kbr   m / Wm  mr
resolved velocity field. b

    uk  convection of turbulence by the resolved m
velocity field. 
k fr  AfrT  fr exp E fr / T 
2  
   k    u is a compressibility term kbr  AbrT  br exp  Ebr / T 
3 

 : u represents the production of turbulence by shear Here, Efr and Ebr are activation temperatures.
  eff    eff
Kcr T     m / Wm  mr mr 
  b a
 k  represents self-diffusion with as
P r
 k   
m

diffusivity. where Krc(T) is concentration equilibrium constant.

ρε is the decay of turbulence into kinetic energy. c
m  Wm   amr  bmr r
m

Equations of State: Q   Qr r
c

r
The equations are assumed to be those of an ideal gas
mixture. Here, Qc is chemical heat release

p  RoT   m / Wm 
m
Qr    amr  bmr   h0f
m
 
m

I T     m /  Im T  Here (δhf0)mis the heat of formation of species m at

m absolute temperature zero.
C0   air  c k2 / 
c p T     m /   c pm T    a1m 
m pmR0
  A3 
a2mTm  a3mTm2  a4mTm3  a5mTm4  hm T  
 Im T   R0T / Wm c p
K
Pr
0
sm a a
 a1m lnTm  a2mTm  3m Tm2  4m Tm3  
R0 2 2 D
a5m 4
 Sc
 Tm  a7m
2 Here cμ is the empirical constant with a value of 0.09.
The Sutherland formula used for μair is
Here, R0 is the universal gas constant
Wm is the molecular weight of species m A1T 3/2
Im(T) the specific internal energy of species m air 
cpm is specific heat coefficient at constant T  A2
pressure Here, A1, A2 are constants
s0 is specific entropy A3 is taken as -⅔ for turbulent flow
Pr is Prandalt number
Sc is Schmidt number
Chemical Reaction:

 amr Xm   bmr Xm
m m
НУМЕРИЧКО ИСПИТИВАЊЕ УТИЦАЈА
КОНФИГУРАЦИЈЕ СВЕЋИЦА НА
Here, Xm is one mole of species m ПЕРФОРМАНСЕ У ЦИЛИНДРУ МОТОРА
amr, bmr is integral stoichiometric coefficients for
reaction r. Б. Мехул, П. Ранџан, А.К. Шукла

  amr  bmr Wm  0 Нумеричко испитивање сагоревања у моторима са

m једном и двоструком варницом је извршено да би се
проучавао утицај свећице, положаја и времена вар–
Kinetic reaction r proceeds at a rate ωr which is
нице на перформансе мотора. Побољшање перфор–
манси мотора је једно од примарних области истра–
живања аутомобилске индустрије. Сходно томе,

FME Transactions VOL. 51, No 4, 2023▪593

резултати студије се могу искористити за оптими– равномерне дистрибуције смеше ваздуха и горива
зацију конфигурације мотора како би се постигле унутар цилиндра мотора са 4 вентила. Симулације
максималне перформансе. Истраживање је спро– су спроведене за четири случаја, укључујући вари–
ведено коришћењем софтвера отвореног кода засно– јације у положају и времену свећица у моторима са
ваног на ограниченом волумену, ОпенФОАМ, за једном и двоструком варницом. Према резултатима
рачунарске симулације. Симулације су спроведене симулација, мотор са једном варницом даје најбоље
коришћењем КсиЕнгинеФОАМ решавача са транс– перформансе када се свећица рано упали и постави
портном једначином за моделовање фронтова пла– у центар цилиндра. Када се постави на оптималан
мена. Стандардни к-ε модел турбуленције је положај одређен кретањем пламена и сударом,
коришћен за предвиђање параметара турбуленције. мотор са две свећице даје најбоље перформансе при
Симулација је спроведена за компресију и ход снаге највећој разлици између времена паљења две
(угао курбле између -180° и 180°), уз претпоставку свећице.

594▪ VOL. 51, No 4, 2023 FME Transactions