Fuel 207 (2017) 719–728

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Engine performance, exhaust emission and combustion analysis

of a 4-stroke spark ignited engine using dual fuel injection
D. Ramasamy a,⇑, C.Y. Goh b,c, K. Kadirgama a, F. Benedict a, M.M. Noor a, G. Najafi d, A.P. Carlucci e
a
Faculty of Mechanical Engineering, University Malaysia Pahang, Pekan, Pahang 26600
b
USM School of Mechanical Engineering, Engineering Campus, Malaysia
c
Focus Applied Technologies, 6305 Highland PI, Black Forest, CO 80908, USA
d
Tarbiat Modares University, Tehran, Iran
e
University of Salento, Lecce, Italy

h i g h l i g h t s

Comparing gasoline-CNG mixture to CNG, G35 shows best performance with acceptable emission.

G35 (35% gasoline and 65% CNG) has provided higher combustion pressure.

Engine torque and power increased by 8.6%, the BSEC and engine efficiency did not change when compared to CNG.

The HC, CO emissions increased and NOx decreased with using dual fuel.

Overall burning angle has been decreased when compared to CNG.

a r t i c l e i n f o a b s t r a c t

Article history: The world natural gas reserve is plentiful. Instead of using gasoline to power ground vehicles, usage of
Received 11 January 2017 Compressed Natural Gas (CNG) can improve the environment and reduce energy cost. However, most
Received in revised form 10 June 2017 engines that runs on CNG are converted from gasoline based engine. Hence, the engines are not optimized
Accepted 12 June 2017
for CNG. Other problems of converted engines are loss of power due to slower burn rate of CNG and gas
Available online 23 June 2017
displacement effect of CNG, hotter exhaust gas which degrade engine’s reliability, low mileage per tank
and higher NOx. Since, converted engines are gasoline base engine, a dual fuel injection engine system can
Keywords:
be developed to reduce the stated problems. Are activity controlled approach was incorporated where
Compressed natural gas
Spark ignition engine
both gasoline and CNG are mixed before going into the combustion chamber. With this technique, when
Emissions HC, CO, NOx a blend of high reactive 35% gasoline and 65% CNG was used, the engine had its engine performance such
torque, power and efficiency increasing by 10%. Also, the engine emissions such as hydrocarbon was
reduced by 50% and carbon monoxide emission was reduced by 75% and NOx emission was reduced
by 50% when compared with CNG baseline. Combustion of spark ignition engines converted to bi-fuel
CNG is unstable and proper air and fuel mixing strategy is a concern here.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction
The energy needs are higher and higher every year due to the
demand from industrial and transportation sector especially from
some high population part of the world [1,2]. Most transportation
fuels are hydrocarbon base. These fuels are derived from petroleum
which is a non-renewable resource. This implies that someday pet-
roleum having high potential will be depleted [3,4] and this will
⇑ Corresponding author.
E-mail addresses: deva@ump.edu.my (D. Ramasamy), paolo.carlucci@unisalento.
it (A.P. Carlucci).
cause energy crisis to the world [5]. Currently, there is a growing
development trend for electric vehicles. However, with today’s bat-
tery technology, the mileage of such vehicle is poor when compare
to internal combustion engine. The most advanced battery has the
energy density of 180Wh/kg. This is miniscule when comparing it
to gasoline which could provide energy density of 1800Wh/kg after
taking into account the efficiency and losses of the system [6].
Given time and money for research and development, battery
technology will be advanced enough to compete with internal
combustion engine in term of energy density. However, at that
given time there will be a transition period when gasoline will
be too expensive to be used as fuel and electric car with acceptable

http://dx.doi.org/10.1016/j.fuel.2017.06.065
0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.
720
mileage will be too expensive to purchase. This is where com-
pressed natural gas (CNG) aids the transition period that is from
hydrocarbon fuel to electric.
The petroleum production in has peaked in the year 2015 [7], as
shown in Fig. 1, according to the prediction curve in a case study,
Malaysia a net oil exporter will become a petroleum importer in
the year 2015 [8]. The petroleum production is subjected to change
when new oil fields are found. However, the consumption of petro-
leum will always increase due to several reasons. The first reason is
population growth which will increase the domestic consumption
in the country. Second reason is economic growth of the country
which increases transportation activity. This activity requires
ample amount of fuel. With such factors considered, the increasing
consumption of petroleum will counter any new oil field found.
In Fig. 2, we can see that the use of natural gas has not peaked as
compared to petroleum. Hence with all this encouraging factors, it
makes natural gas a good alternative transportation fuel for auto-
motive in countries such as Malaysia [9–11]. The rise in usage of
natural gas is also an indication of the shift in highlight towards
better energy sustainability by not relying too much on petroleum
production [12,13]. With a large natural gas reserves and the low
running cost of CNG much CNG retrofit kits are being developed.
However, these retrofit kits are bi-fuel system that is the engine
can run either CNG or gasoline without much modification. The
baseline engine was optimized for gasoline. Therefore, when it
runs on CNG, the engine will be less efficient when compared to
gasoline. Also, the power of the engine significantly dropped after
switching to CNG [10]. Since gasoline injectors are already inside
the engine, a dual fuel injection system can be implemented to
improve the efficiency of the engine. Dual fuel injection means that
the engine will be running on both gasoline and CNG together at a
certain ratio [14]. Hence, performance analysis is crucial for this
dual fuel injection system.
Another contributing factor that encourages this improvement
study is that CNG vehicles require a bigger fuel tank to have the
equivalent mileage of a gasoline vehicle. For example, if the vehicle
is using a fuel mixture of 25% gasoline and 75% CNG, the vehicle
will have 33% more mileage per tank for CNG and 400% more mile-
age per tank for gasoline. The parameters that will be studied are
fuel composition and air fuel ratio. These parameters will be stud-
ied to provide maximum performance.
Engine that runs on CNG are notorious for NOx emission [15].
This is due to their slow burn rate of the fuel [16]. When the engine
1000
900
Petroleum (Thousand barrels/day)
800
700
600
500
400
300
200
100
Fig. 1. Crude oil production, consumption and exports/imports in Malaysia [8].
D. Ramasamy et al. / Fuel 207 (2017) 719–728
is running on gasoline, its NOx level is lower. This is since, gasoline
burns faster than CNG. To illustrate this, all fuel is burnt after the
expansion stroke when using gasoline. However, for CNG the fuel
is still burning during exhaust stroke. Therefore, there is not
enough time for the exhaust gases to cool down and reduce the
NOx level, hence the creation of thermal NOx. Since the engine
was designed for gasoline, there will be reliability issue when the
engine is running on CNG. When gasoline is used, it burns quickly
in the combustion chamber before opening the exhaust valves.
However this changes when CNG is used [17]. Natural gas burns
slowly in the combustion chamber and still may burn even after
the exhaust valves are open. This will create a reliability problem
for the exhaust valves and higher operating temperature. Also,
gasoline is a liquid fuel hence it will provide a cooling effect to
the engine’s combustion chamber making it cooler to operate.
When the engine uses CNG alone, it does not have such cooling
capabilities of the gasoline [18]. Therefore, the engine might run
with higher operating temperature which may cause reliability
issue.
Dual fuel injection is injecting 2 types of fuel into the combus-
tion chamber together. This technique is similar with injecting
hydrogen into the mixture however instead of injecting hydrogen,
it uses any other fuel [19]. This technique offers many technical
advantages and economy advantages for vehicle equipped with
CNG injection system. Firstly, there is no major engine hardware
modification to implement this technique. This technique requires
a CNG retrofit kit which includes a gas tank, pressure regulator,
refuelling valve, gas injectors and the electronic system to control
the injection duration. Most taxi in are powered by CNG due to
high operation costs [20]. By implementing this technique, the taxi
drivers do not require major technical modification to get the vehi-
cle running on CNG-gasoline. Using the original fuel injection sys-
tem, the only changes required is on the electronic system which is
plug and play.
Secondly, this technique increases the engine performance and
the combustion performance. Studies has shown that dual fuel
injection technique that involves hydrogen-CNG mixture on spark
ignited engine decrease the ignition delay, burn duration and
increase engine performance [21]. Research on different type of
fuel mixture like ethanol-gasoline mixture, which was done at
South America where alcohol powered vehicle are common, shows
similar results that is performance improvement over gasoline
[22]. Other fuel mixture such as gasoline-CNG was done and there
Production Consumption
Exports/Imports (-) Production Prediction
Consumption Prediction Exports/Imports (-) Prediction
 D. Ramasamy et al. / Fuel 207 (2017) 719–728 721

6,000
Petroleum use

5,000 Natural Gas use

Coal use

Energy Use, Mtoe/yr

4,000 Nuclear use

3,000

2,000

1,000

0
1960 1970 1980 1990 2000 2010 2020 2030 2040 2050
Year

Fig. 2. World energy used [12].

was improvement in power and efficiency. However the study was Table 1
done on 8 valve, 4 cylinder engine with 1.3 litre capacity from FIAT Engine specification.

and such engine does not represent today’s high performance auto- Engine specification
motive engine [23]. Also, the study lack combustion analysis and Engine Name K3-VE
emission study. Valve Mechanism DOHC with DVVT
Thirdly it improves emission. Studies on hydrogen-CNG mix- Valves 16 Valves
ture had shown that the hydrocarbon, carbon monoxide and NOx Bore 72.0 mm
Stroke 79.7 mm
emission were reduce when ignition timing was advanced [24].
Displacement 1298 CC
Another study on this technique which involves CNG and reformer No. of Cylinder 4
gas-gasoline mixture concluded that as the concentration of refor- Compression Ratio 10:1
mer gas increase in the mixture, there was a strong reduction in Maximum Power 67 kW
Maximum Torque 117 Nm
hydrocarbon and NOx emission [21]. Studies on ethanol-gasoline
Ignition Timing 20 Deg BTDC@4000 rpm
mixture also show that the hydrocarbon and carbon monoxide Fuel System Electronic Fuel Injection
emission were decreased [22]. Dual fuel injection technique is
superior for vehicle which is retrofitted to run on CNG. This tech-
nique is technically feasible, economical, improve performance
valve, CNG control unit, high pressure pipe, low pressure pipe,
and improve emission. Hence, research needs to be done on the
CNG injector, 2 temperature sensors, air pressure sensor and CNG
effects of gasoline addition into CNG on the engine performance,
tank pressure indicator. Fig. 3 shows the mechanical setup of the
combustion performance and emission on a spark ignited automo-
system.
tive engine.
When CNG mode is switched on, the control unit will switch on
This paper will concentrate on developing an injection system
the solenoid valve on the CNG tank and cut the signal going into
and study the effects of the various parameters on the engine per-
formance. The primary objectives of this study are to develop a
dual fuel injection system for gasoline & CNG and study the effects
of dual fuel injection system on the engine performance.

2. Experimental setup

The engine used in this experiment was the K3-VE engine. The
engine is a 16-valve double overhead cam engine (DOHC) with
dynamic variable valve timing (DVVT). This engine was chosen
for this experiment because high number of production on many
vehicle types such as Toyota Passo, Daihatsu Sirion and Perodua
Myvi. The specification of the engine is shown in Table 1.
There are 2 types of CNG conversion kit in the market. One of it
is the carburettor type and the sequential type. The carburettor
type CNG conversion uses a gas carburettor to deliver the fuel into
the engine. Although it is cheap, it does not provide full control of
the fuel injected into the engine. The sequential type uses gas
injector control by a CNG control unit. This setup is costlier but it
provides full control of fuel injected into the engine. Also it is cap-
able of doing fine tuning to the fuel injection when there is a differ-
ence of manifold intake and CNG temperature [25]. The CNG kit
has a pressure regulator, refuelling valve, control valve, solenoid Fig. 3. Mechanical setup of the CNG kit.
722 D. Ramasamy et al. / Fuel 207 (2017) 719–728
the gasoline injector. The pressurize fuelfrom the CNG tank will
flow into a pressure regulator which regulated the pressure to
170 kPa. Hence there will be a constant supply of 170 kPa methane
to the CNG injectors. The CNG injector is controlled by the CNG
control unit which is supplied in this kit. The CNG control unit
operates by taking the signal from the engine gasoline fuel injec-
tors, CNG pressure and temperature in the injector, CNG pressure
regulator temperature and manifold air pressure to calculate the
required injector opening time for CNG injector. The CNG is
injected into the cylinder the moment the CNG control unit detects
signal from the gasoline injector. Fig. 4 shows the electrical setup
of the CNG kit. Fig. 5 shows the operation flow of the CNG kit [26].
The experimental setup is shown in Fig. 6, it starts with the Kis-
tler 6052C sensor, it is used to measure cylinder pressure. This sen-
sor can operate at high engine combustion temperatures and
couples with the spark plug. A wide band oxygen sensor was used
in this experiment to provide accurate reading for air fuel ratio. The
wide band oxygen sensor used in the experiment was a Motorsport
3737 Bosch Wideband O2 sensor. The gaseous fuel CNG weighing
machine used in this experiment was from NAPCO. The capacity
of the scale is 50 kg with a resolution of 0.001 kg. The crank enco-
der used in this experiment was a Kubler incremental encoder, the
crank encoder used in this experiment will provide 1 degree reso-
Fig. 4. Electrical setup of the CNG kit.
Fig. 5. Operation flow of the CNG kit.
lution required for combustion analysis. An Agilent U2542A data
acquisition unit (DAQ) was used to log the pressure and encoder
signals. It features 4 simultaneous analog input channels with sam-
pling rate up to 500,000 samples per second (Sa/s). This DAQ could
communicate with Agilent Measurement Manager (AMM) soft-
ware and can save the data in comma-separated values (CSV) file
for offline analysis. A minimum of 100 cycles was taken for each
test for a good combustion analysis [27]. The results for engine
emissions were recorded via an emission analyzer Kane AUTO 5–3.
3. Dual fuel injection system
In the market, there is no dual fuel injection system available.
Hence, a dual fuel injection system was designed for this experi-
ment. When designing this system, many considerations must be
considered. The injectors used by K3-VE were high impedance type
hence the injector drivers must be designed accordingly. Also, the
signal going into the fuel injector is 12 V and has high current flow-
ing through it. Therefore, the signal detection circuit must be iso-
lated to prevent damage to the microcontroller. Furthermore, the
signal detection of the system cannot disturb the original signal.
If the system disturbs the original signal, it will cause the engine
to run erratically. In case of microcontroller failure or any elec-
tronic part failure, the system has to stop inject fuel into the engine
to prevent hydraulic locking in the engine [28]. Fig. 7 shows the
consideration taken in designing the dual fuel injection system.
With the base consideration, the circuit is designed and a fail proof
mechanism is targeted. The system complements the existing ECU
to have better control of the injectors of CNG and gasoline.
Fig. 8 shows the physical appearance of the system. The system
has a microcontroller to control its fuel injection timing and injec-
tor opening time. It has an isolated signal detection system and iso-
lated injector driver to protect the microcontroller, expensive
electronic parts in the system and engine electrical component in
case of overvoltage, over current or voltage spike. Fig. 9 shows
the schematics of the dual fuel injection system algorithm prior
to the circuit design.
This system uses the signal from the gasoline injectors from the
engine as fuel injection timing for the dual fuel injection system.
The injector opening time was controlled by the user with a com-
puter. With the gasoline injector curve, precise control of the fuel
was done. The microcontroller programming and the computer
interface to the hardware was programmed [29]. The flow of the
program is shown in Fig. 10.
The engine testing measures parameters such as torque, engine
speed, engine power, fuel consumption, brake specific fuel con-
sumption (BSFC), engine efficiency and emissions. The engine
 D. Ramasamy et al. / Fuel 207 (2017) 719–728 723

25

21 24
23
26 18
17
22 20 3
19
12
13
31 14
29
2
27 1
5
4
15
28 16
11

30

6 9

7 8

10

1. Test Engine 11. Crank Speed pickup 21. CNG 200bar 25L Tank
2. Dynamometer 12. Crank 36 tooth gear 22. Fuel Scale
3. Dynamometer Controller 13. Crank Encoder 23. Check Valve
4. Pressure Transducer 14. Crank Angle Sensor 24. Three Stage Pressure Regulator
5. Spark Signal 15. TDC Sensor 25. Gas Filter
6. Oscilloscope 16. Signal Conditioner 26. Common Rail Gas Injector
7. Charge Amplifier 17. Digital Voltmeter 27. Cooling Tower
8. Data Acquisition System 18. TPS Sensor 28. Digital Thermocouple Reader
9. Computer 19. Gasoline Fuel Tank 29. RS232 Connector
10. Printer 20. Gasoline Filter 30. Emissions Analyzer
31. Gasoline Injector

Fig. 6. Test Cell Layout.

torque was measured from a strain gauge mounted on the the period between each pulse and the number of the gear teeth
dynamometer [30]. The strain gauge provides a voltage signal in Eq. (2).
which is the converted into torque through a formula as in Eq. (1).
60
N¼ ð2Þ
T ¼kV ð1Þ tn
where, N is the engine speed (RPM), t is the period of the pulse (s)
where T is the torque (Nm), k is the constant (Nm
V
) and V is the volt- and n is the number of gear teeth. The measured torque was then
age output (V). Engine speed was measured from the speed sensor converted to power which is the product of the engine torque and
mounted on the dynamometer. It can be calculated by measuring angular speed as in Eq. (3).

Fig. 7. Consideration taken in designing the dual fuel injection system.

724 D. Ramasamy et al. / Fuel 207 (2017) 719–728

transfer to wall (J). The mass fraction burned curve can be converted
from gross heat release, by dividing the mass of fuel within the
combustion chamber times its lower heating value in Eq. (8). From
here rate of heat release (ROHR) was estimated adding the gross
heat release rate [33].
Q ch Q
xb ¼ ¼ ch ð8Þ
mf  Q lhv Q in
where, xb is the mass fraction burn, Q ch is the gross heat release (J),
mf is the fuel mass (g), Q lhv is the fuel lower heating value (J/kg) and
Q in is the heat input from fuel mass (J). The process is assumed adi-
abatic for reducing the heat to wall condition. This can be rewritten
in Eq. (9) with respect to crank angle [34],
 m
dQ ch dxb Q h  hs
¼ Q in ¼ ma in ð1  xb Þ ð9Þ
dh dh dh hd
Fig. 8. Dual fuel injection system.
and finally the Weibe cumulative heat release function as in Eq.
(10) [27]:
2pNT   m 
P¼ ð3Þ
60 h  ho
xb ðhÞ ¼ 1  exp a ð10Þ
hd
where, P is the engine power (W) and N is the engine speed (RPM).
The analysis also measured the fuel mass by using a weighing scale. where, h is the crank angle, ho is the start of combustion, hd is the
Fuel consumption (FC) of the engine was measured by the fuel combustion duration, a is a efficiency parameter and m is a form
weight used during a test over the time required for the engine to factor.
consume that amount of fuel as in Eq. (4). From the Weibe Eq. (10), the flame-development angle is the
m interval between start of ignition to a time when 10% of fuel in
_f ¼
FC ¼ m ð4Þ the cylinder is burnt. The flame-development is given by Eq. (11).
t
where, m _ f is the fuel consumption (gs), m is the fuel mass used (g) hd ¼ h10  h0 ð11Þ
and t is the time required to use m amount of fuel mass (s). Based where, hd is the flame-development angle, h10 is the crank angle
g
on the power and FC, brake specific fuel consumption (BSFC) (kW:hr ) when mass burn fraction reaches 10% burn and h0 is the start of
was also calculated to have better comparison between engine ignition
speeds in Eq. (5). The rapid-burn angle is the duration taken to burn the whole
_ f  60  60
m bulk of charge. It is given by Eq. (12).
BSFC ¼ ð5Þ
P  1000 Dhb ¼ h90  h10 ð12Þ
Brake specific energy consumption is energy flow rate per unit where Dhb is the rapid-burn angle and h90 is the crank angle when
power. It measures how efficient is the engine is using the energy mass burn fraction reaches 90% burn.
supplied to produce work. The formula for brake specific energy is The duration for the burning process is called overall burn angle
shown in Eq. (6) [31]. and is given by Eq. (13).
ðFC 1  Q lhv 1 þ FC 2  Q lhv 2 Þ  60  60 Dho ¼ hd þ hb ð13Þ
BSEC ¼ ð6Þ
P  1000
where Dho is the overall burn angle.
where The experiment was divided into two categories which ran on
J
BSEC – Brake specific energy consumption (kW:hr ) SAE standards [35]. The categories which were divided to baseline
FC 1 – Fuel consumption for energytype 1 (gs) engine tests and Gasoline-CNG mixture reaction. At baseline
Q lhv 1 – Lower heating value of energytype 1 (kgJ ) stage, experiment was done on the engine using either gasoline
FC 2 – Fuel consumption for energytype 2 (gs) or CNG. These experiments were repeated 3 times to provide
average values. These baseline data were used later to compare
Q lhv 2 – Lower heating value of energytype 2 (kgJ ))
with the modified engine data. Also with the baseline data, tar-
P – Engine power (kW) gets can be set to choose the fuel mixture with the best engine
performance with acceptable engine emissions and engine tem-
where, g is the engine efficiency and Q lhv is the lower heating perature. At gasoline-CNG mixture stage, the experiment was
value of fuel (kgJ ). Combustion pressure data from the experiment done on the engine using gasoline-CNG mixture. The mixture
was acquired by reading the pressure sensor. The pressure sensor started from 15% gasoline and 85% CNG and increased by 10%
provides a voltage which was recorded by a data acquisition sys- at each succession that was 25%, 35% and 45%. The mixture vari-
tem [32]. The pressure trace was used to investigate the many ation stops at 45%.
aspects of the engine such as gross heat release, mass fraction burn,
flame development angle, rapid burn angle and the overall burn 4. Results and discussions
angle. Netheat release was given by Eq. (7),
dQ ch c dV 1 dP dQ ht Engine torque when running on gasoline-CNG mixture with on
¼ P þ V þ ð7Þ CNG is tested. Fig. 11 shows the torque data of the engine running
dh c  1 dh c  1 dh dh
on CNG and gasoline-CNG mixture. The G15 mixture had poor tor-
where, dQch is the net heat release (J), c is the gas constant, P is the que when compared to CNG. The power increase by 2.1% as shown
pressure (Pa), V is the cylinder volume (m3 ) and dQ ht is the heat in Fig. 12, however the efficiency decreases by 5% as in Fig. 13. This
 D. Ramasamy et al. / Fuel 207 (2017) 719–728 725

Fig. 9. Dual fuel injection system algorithm.

was due to the poor mixing of gasoline and air in the mixture and
the concentration of gasoline was not enough to influence the
flame speed of the combustion due to large volumetric displace-
ments [36]. The engine BSEC was obtained in Fig. 14. However,
when the engine runs on 25% gasoline, the torque increase by
3.8% on average and BSEC and efficiency deteriorate by 1.3%.
When the engine runs on 35% gasoline, the torque increase by
8.6% on average and the BSEC and efficiency improved by 1.5%.
The torque of the engine increase by 10.5% on average and the
BSEC and efficiency increase by 3.8% on average when the runs
on 45% gasoline and 55% CNG mixture. At higher gasoline mixture,
the gasoline in the mixture could influence the combustion speed
hence improving the torque, power, BSEC and efficiency.
4.1. Effects of Varying fuel mixture onEngine emission
Figs. 15–17 shows the emission data of the gasoline-CNG mix-
ture at different engine speed. From Figs. 15 and 16, the hydrocar-
bon emission and carbon monoxide emissions of the combustion
mixture fuel have higher CO values as compared to the baseline
CNG alone but lower than the gasoline at G0 [37]. This is due to
addition of complex hydrocarbon chain fuel into the simple CNG
fuel which consists mainly of methane. This complex hydrocarbon
chain has high carbon content in its molecule making it easier to
form carbon monoxide. Liquid fuel, even after atomization, sticks
to the cylinder wall and cannot be burnt because of wall quench-
ing. This makes the combustion flame front burns relatively cooler
726 D. Ramasamy et al. / Fuel 207 (2017) 719–728

18000 G0
17500 G15
17000 G25

BSEC (kJ/kWhr)
16500 G35
16000
G45
15500
15000
14500
14000
13500
13000
0 1000 2000 3000 4000 5000 6000
Engine Speed (RPM)

Fig. 14. Engine Power Vs Engine Speed.

Fig. 10. Dual fuel injection system program flow.

2000
1800 G0

105 1600 G15

HC (ppm Vol)
1400 G25
100 G0 1200
G35
95 G15 1000
800 G45
Torque (Nm)

90 G25
600
85 G35
400
80 G45 200
75 0
0 1000 2000 3000 4000 5000 6000
70 Engine Speed (RPM)
65
Fig. 15. Hydrocarbon emission data of the gasoline-CNG mixture.
60
0 1000 2000 3000 4000 5000 6000
Engine Speed (RPM)
2.5
Fig. 11. Engine Torque Vs Engine Speed. G0
2
G15
CO (% Vol)

G25
60 1.5
G0 G35
50 G15 1 G45
G25
Power (kW)

40
G35 0.5
30 G45
0
20 0 1000 2000 3000 4000 5000 6000 7000
Engine Speed (RPM)
10
Fig. 16. Carbon Monoxide emission gasoline-CNG mixture.
0
0 1000 2000 3000 4000 5000 6000
Engine Speed (RPM) 2500
G0
Fig. 12. Engine Power Vs Engine Speed.
2000 G15
NOx (ppm Vol)

G25
1500
G35
G0
27 1000 G45
G15
26
G25 500
25 G35
Efficiency (%)

24 G45 0
0 1000 2000 3000 4000 5000 6000 7000
23
Engine Speed (RPM)
22
21 Fig. 17. NOx emission of the gasoline-CNG mixture.

20
19 at the walls of the combustion chamber which is caused by heat
0 1000 2000 3000 4000 5000 6000
Engine Speed (RPM) transfer of the combustion heat to the combustion chamber wall.
This explains the increase of hydrocarbon emission when the fuel
Fig. 13. Engine Power Vs Engine Speed. mixture had higher concentration of gasoline in it. The NOx values
 D. Ramasamy et al. / Fuel 207 (2017) 719–728 727

60 G0 (a)
G15 Combuson advancing Flame Development Angle
G25 25
50 24
G35
Pressure (Bar)

G45 23
40 Max. P. G0 22 G15

Angle (°)
Max. P. G15 21
30 Max. P. G25 20 G25
Max. P. G35 19
Max. P. G45 G35
20 18
TDC 17 G45
10 16
15
0 1000 2000 3000 4000 5000 6000 7000
0
300 320 340 360 380 400 420 440 460 Engine Speed (RPM)
Crank Angle (°)
(b) Rapid-Burn Angle
Fig. 18. Combustion pressure of gasoline-CNG mixtures and CNG. 32
30
in Fig. 17 reduces as the gasoline addition is done from G15. This
explains that the cooling properties of the gasoline reducing the 28

Angle (°)
G15
thermal NOx. As CNG burns dry the gasoline addition propagates 26 G25
with more HC and complete combustion of Carbon to induce more G35
24
water in the by-products. This water makes cooler peak tempera- G45
tures and as such reducing the thermal NOx. However, more addi- 22
CNG
tion of gasoline to 25% till 45% only increases the NOx again. This 20
may occur as more air is displaced for fuels in the system and 0 1000 2000 3000 4000 5000 6000 7000
hence the prompt NOx from more chemical reaction is increased. Engine Speed (RPM)
(c) Overall Burn Angle
4.2. Effects of varying fuel mixture on engine combustion 60

Fig. 18 shows the combustion pressure of gasoline-CNG mix- 55

Angle (°)

tures and CNG for engine speed of 4000 rpm. From the figure, as 50
G15
the concentration of gasoline increased, the maximum pressure G25
increased and the peak pressure shifted towards TDC. This is some 45 G35
sort of advancing of the burn from the fuel mixture. The advancing G45
40
works to enable cooler burning in the engine. This can also be seen CNG
in Fig. 19 of the mass burn curve. The gradient increase is a sign the 35
engine was responding to the fuel additive. Fig. 20 compiles the 0 1000 2000 3000 4000 5000 6000 7000
flame-development angle, rapid-burn angle and overall burn angle Engine Speed (RPM)
of gasoline-CNG mixture and CNG from the burning curve. CNG
Fig. 20. Flame development angle (a), rapid-burn angle (b) and overall burn angle
alone burns slower than gasoline however when CNG is mixed
(c) of CNG and gasoline-CNG mixture at different engine speeds.
with gasoline the fuel mixture burns faster than CNG. The stoi-
chiometry AFR of CNG and gasoline is respectively 17.2 and 14.7.
and G25 and better emission when compared to G45. G45 had car-
As the gasoline concentration increases the AFR decreases. For
bon monoxide emission more than 2% therefore it cannot meet the
G15, the AFR is 16.9. This AFR value shows that there is more air
emission target that had been set.
in the combustion chamber for the fuel to combust. Since more fuel
can be inserted into combustion chamber and combust at stoi-
chiometry, more energy will be released. Also, when there is less 5. Conclusion
CNG fuel in the mixture, there will be less air displacement effect
of the CNG fuel. As the air displacement effect of CNG fuel is The study was carried out for the engine fuelled with dual fuel
reduced, more air can be used for stoichiometry combustion hence of gasoline and CNG. When comparing gasoline-CNG mixture to
increasing the torque of the engine. CNG, G35 shows best performance with acceptable emission. The
G35 was chosen to proceed as the best gasoline addition engine torque and power has increased by 8.6% on average and
because of the increased performance when compared with G15 the BSEC and engine efficiency did not change when compared to
CNG. The hydrocarbon increased however did not exceed
1.0 2000 ppm Vol. The carbon monoxide emission increased but did
0.9 G0 Burn Curve not exceed 2% Vol. The nitrogen oxides emission decreases and
Mass Burn Fracon

gradient
0.8 did not exceed more than 1000 ppm Vol. Engine exhaust and cool-
G15
0.7 ant temperature has remained at gasoline levels. As for maximum
0.6 G25 combustion pressure, G35 has provided higher combustion pres-
0.5 sure and its peak pressure shifted towards TDC. Overall burning
G35
0.4 angle is decreased when compared to CNG. This is the main reason
0.3 G45
for the increased efficiency by using gasoline-CNG addition.
0.2
0.1
Acknowledgement
0.0
320 340 360 380 400 420
Crank Angle (°) The financial support to the authors by Ministry of Education
Malaysia under FRGS grant RDU140125, RDU160152, UMP Internal
Fig. 19. Combustion mass burn of gasoline-CNG mixtures and CNG. Grant RDU1403124, RDU160319, University Malaysia Pahang,
728 D. Ramasamy et al. / Fuel 207 (2017) 719–728
Automotive Excellence Center and Faculty of Mechanical Engineer-
ing is gratefully acknowledged.
References
[1] Damirchi H, Najafi G, Alizadehnia S, Mamat R, Nor Azwadi CS, Azmi WH, et al.
Micro combined heat and power to provide heat and electrical power using
biomass and gamma-type stirling engine. Appl Therm Eng 2016;103:1460–9.
[2] Awad OI, Ali OM, Mamat R, Abdullah AA, Najafi G, Kamarulzaman MK, et al.
Using fusel oil as a blend in gasoline to improve SI engine efficiencies: a
comprehensive review. Renew Sustain Energy Rev 2017;69:1232–42.
[3] Noor MM, Wandel AP, Yusaf T. Mild combustion: the future for lean and clean
combustion. Int Rev Mech Eng 2013;8:251–7.
[4] Shafiee S, Topal E. When will fossil fuel reserves be diminished? Energy Policy
2009;37:181–9.
[5] Noor MM. Experimental and numerical study of MILD combustion in an open-
end furnace with exhaust gas recirculation using methane and biogas [Ph.D.
thesis]. Australia: University of Southern Queensland; 2015.
[6] Hirai H, Matsuo Y, Uno H, Nagatomi Y. Current status and future prospects of
electric power as automotive fuel. energy working papers; 2009.
[7] Abas N, Kalair A, Khan N. Review of fossil fuels and future energy technologies.
Futures 2015;69:31–49.
[8] Miller RG. Future oil supply: the changing stance of the International Energy
Agency. Energy Policy 2011;39:1569–74.
[9] Seng LG, Itam OJA, Bangi KI. An analysis of NGV market in Asia Pacific. In: The
Joint International Conference on China Natural Gas Vehicle Technology
Innovation and Industry Development Forum13–14th May 2009, Chongqing,
China; 2009.
[10] Ramasamy D, Zainal ZA, Kadirgama K, Walker-Gitano Briggs H. Effect of
dissimilar valve lift on a bi-fuel CNG engine operation. Energy
2016;112:509–19.
[11] Malaysian Gas A. Malaysia: natural gas industry annual review 2012; 2013.
p. 25.
[12] Randers J. 2052: A global forecast fort he next forty years. Chelsea Green
Publishing; 2012.
[13] Najafi G, Ghobadian B, Yusaf T, Safieddin Ardebili SM, Mamat R. Optimization
of performance and exhaust emission parameters of a SI (spark ignition)
engine with gasoline–ethanol blended fuels using response surface
methodology. Energy 2015;90:1815–29.
[14] Pipitone E, Genchi G. NO X reduction and efficiency improvements by means
of the Double Fuel HCCI combustion of natural gas–gasoline mixtures. Appl
Therm Eng 2016;102:1001–10.
[15] Li W, Liu Z, Wang Z, Xu Y, Wang J. Experimental and theoretical analysis of
effects of N 2, O 2 and Ar in excess air on combustion and NOx emissions of a
turbocharged NG engine. Energy Convers Manage 2015;97:253–64.
[16] Ramasamy D, Aik Soon K, Walker-Gitano Briggs H, Zainal ZA. Variation of
airflow pattern through dissimilar valve lift in a spark ignition engine. J Chin
Inst Eng Trans Chin Inst Eng Series A/Chung-kuo Kung Ch’eng Hsuch K’an
2013;36:1083–96.
[17] Tavakoli S, Ganji DD, Gorji M, Rasekh A, Naeejee S. Different camshaft profile
analyses for natural gas engine performance and emission. J Braz Soc Mech Sci
Eng 2016;38:355–64.
[18] Liu K, Yang J, Jiang W, Li Y, Wang Y, Feng R, et al. Effect of asynchronous valve
timing on combustion characteristic and performance of a high speed SI
marine engine with five valves. Energy Convers Manage 2016;123:185–99.
[19] Aksu C, Kawahara N, Tsuboi K, Kondo M, Tomita E. Extension of PREMIER
combustion operation range using split micro pilot fuel injection in a dual fuel
natural gas compression ignition engine: a performance-based and visual
investigation. Fuel 2016;185:243–53.
[20] Nijboer M. The contribution of natural gas vehicles to sustainable transport.
International Energy Agency. France: OECD/IEA 2010; 2010.
[21] Czerwinski J, Comte P. Addition of CNG and reformer gas to the gasoline
fuelled SI-engine. SAE Technical Paper 2004–01-0973; 2004.
[22] Topgül T, Yücesu HS, Cinar C, Koca A. The effects of ethanol–unleaded gasoline
blends and ignition timing on engine performance and exhaust emissions.
Renewable Energy 2006;31:2534–42.
[23] Pipitone E, Beccari S. Performances improvement of a SI CNG bi-fuel engine by
means of double-fuel injection. SAE Technical Paper 2009–24-0058; 2009.
[24] Ma L, Geng J, Li W, Liu P, Li Z. The development of natural gas as an automotive
fuel in China. Energy Policy 2013;62:531–9.
[25] Ismail MM, Zulkifli FH, Fawzi M, Osman SA. Conversion method of a diesel
engine to a Cng-diesel dual fuel engine and its financial savings; 2006.
[26] Aslam M, Masjuki H, Kalam M, Abdesselam H, Mahlia T, Amalina M. An
experimental investigation of CNG as an alternative fuel for a retrofitted
gasoline vehicle. Fuel 2006;85:717–24.
[27] Ferguson CR, Kirkpatrick AT. Internal combustion engines: applied
thermosciences. John Wiley & Sons; 2015.
[28] Bauer H. Gasoline-engine management. John Wiley & Sons; 2004.
[29] Timmis H. Practical arduino engineering. 1st ed. New York, USA: Apress; 2012.
[30] Ceviz M. Intake plenum volume and its influence on the engine performance,
cyclic variability and emissions. Energy Convers Manage 2007;48:961–6.
[31] Shashikantha S, Parikh P. Spark ignition producer gas engine and dedicated
compressed natural gas engine-technology development and experimental
performance optimisation. SAE Technical Paper 1999–01-3515; 1999.
[32] Tunestal PA. The use of cylinder pressure for estimation of the in-cylinder
air/fuel ratio of an internal combustion engine, University of California,
Berkeley; 2000.
[33] Carlucci AP, de Risi A, Laforgia D, Naccarato F. Experimental investigation and
combustion analysis of a direct injection dual-fuel diesel–natural gas engine.
Energy 2008;33:256–63.
[34] Kornbluth K, McCaffrey Z. Erickson Pa. Incorporating in-cylinder pressure data
to predict NOx emissions from spark-ignition engines fueled with landfill gas/
hydrogen mixtures. Int J Hydrogen Energy 2009;34:9248–57.
[35] Grigoratos T, Fontaras G, Martini G, Pelleto C. Experimental assessment of a
heavy-duty natural gas engine under transient and real life conditions. SAE
Technical Paper; 2015.
[36] Yang S, Kolla H, Swaminathan N. Application of a New Turbulent Flame Speed
Combustion Model on Burn Rate Simulation of Spark Ignition Engines. SAE
Technical Paper 2016–01-0588; 2016.
[37] Verma G, Prasad RK, Agarwal RA, Jain S, Agarwal AK. Experimental
investigations of combustion, performance and emission characteristics of a
hydrogen enriched natural gas fuelled prototype spark ignition engine. Fuel
2016;178:209–17.