You are on page 1of 9

Fuel 213 (2018) 106–114

Contents lists available at ScienceDirect

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

Full Length Article

Influence of injection timing and split injection strategies on performance, MARK


emissions, and combustion characteristics of diesel engine fueled with
biodiesel blended fuels
⁎ ⁎
H.G. Howa,b, , H.H. Masjukia, , M.A. Kalama, Y.H. Teohc
a
Centre for Energy Sciences, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
b
Department of Engineering, School of Engineering, Computing and Built Environment, KDU Penang University College, 32, Jalan Anson, 10400 Georgetown, Penang,
Malaysia
c
School of Mechanical Engineering, Universiti Sains Malaysia, Engineering Campus, 14300 Nibong Tebal, Penang, Malaysia

A R T I C L E I N F O A B S T R A C T

Keywords: Biodiesel can be used as alternative to replace fossil diesel. However, usage of biodiesel in an unmodified diesel
Diesel engine engine can cause higher in nitrogen oxides (NOx) emission. In order to reduce the harmful emission, certain
NOx injection strategies can be carried out. In this paper, the effects of biodiesel blends, fuel injection timing and split
Injection timing injection schemes on the engine performance, emissions and combustion characteristics of a medium-duty diesel
Split injection
engine are investigated. Parametric studies relating with start of injection timing variation and multiple injection
Biodiesel
schemes using B20 and B50 biodiesel blends were performed and benchmarked with petroleum diesel fuel as
baseline. A remarkably lower NOx level below 100 ppm can be obtained by retard start of injection (SOI) timing
for both of the B20 and B50 fuel operations and with triple injection scheme. It was found that with the use of
B50, simultaneous NOx and smoke suppression from the levels of petroleum diesel fuel is attainable in parallel
with the implementation of late SOI timing and triple injection scheme in a diesel engine. In conclusion, multiple
split injections is a practical strategies to simultaneously decrease NOx and smoke emissions when the SOI
timing is fine-tuned and is an ideal alternative to operate with biodiesel fuel.

1. Introduction out later to lessen the air temperature even though at the expense of
increasing the amount of CO, HC and smoke emitted due to incomplete
The widespread use of diesel engine has caused air pollution pro- combustion.
blems. This is due to their higher exhaust discharges of nitrogen oxides Another strategy can be implemented to diesel engines to attain
(NOx), particulate matter (PM) and smoke in comparison with that of a lower emission limit is by split injection. It can be carried out to reduce
gasoline engine [1]. The air pollutants jeopardize human health in engine noise and amount of NOx emitted. Furthermore, accurately
different ways, necessitating the needs to curb this problem [2–4]. To performed split injection schemes can be favorable in reducing com-
minimize this impact, research effort are being focused on injection bustion noise, waste emissions and diesel consumption and therefore,
strategies such as variable injection timing, split injection, variable they are effective tools [7]. Besides, particulate emissions can be re-
injection pressure, variable nozzle configuration, and others [5,6]. In- duced substantially without a great increase in NOx emissions [8,9].
jection timing optimization can be performed in order to produce a This is because high heat release rate (HRR) can be prevented at the
suitable ignition delay as well as to reduce the amount of exhaust beginning of combustion, hence decrease the flame temperature and
emission in diesel engine. For instance, advancing the injection timing permit better fuel and air mixing to enhance in-cylinder charge
reduces the amount of carbon monoxide (CO), hydrocarbon (HC) and homogeneity. Besides, energy demands of the world are increasing
smoke while increases the amount of NOx emitted [5]. More time is nowadays. Depletion of fossil diesel fuel can be slowed by adopting
available for oxidation when injection is advanced, thus reducing the renewable source of energy such as biodiesel. Biodiesel can be made
amount of CO, HC and smoke, but with higher amount of NOx emitted. from vegetable oil, animal fat or waste materials such as spent coffee
In order to reduce amount of NOx emitted, the injection can be carried grounds [10]. Also, it is nontoxic, renewable and biodegradable


Corresponding authors at: Centre for Energy Sciences, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia.
E-mail addresses: heoygeok.how@kdupg.edu.my, heoygeok@gmail.com (H.G. How), masjuki@um.edu.my (H.H. Masjuki), kalam@um.edu.my (M.A. Kalam),
yewhengteoh@usm.my (Y.H. Teoh).

http://dx.doi.org/10.1016/j.fuel.2017.10.102
Received 8 August 2017; Received in revised form 15 October 2017; Accepted 24 October 2017
Available online 02 November 2017
0016-2361/ © 2017 Elsevier Ltd. All rights reserved.
H.G. How et al. Fuel 213 (2018) 106–114

compared to conventional diesel [11]. Generally, CO, smoke and par- adopted in commercial vehicles. Thus a research gap remains in these
ticulate matter concentration emitted when biodiesel is used are lower fields which are addressed in this research study. The outcome due to
compared to conventional diesel [12,13]. However, NOx emission of changes in biodiesel blends, injection timing and split (with up to three
biodiesel can be higher or lower for different types of biodiesel and injections of equal fuel quantity) injection schemes on a four-cylinder
operating condition [12,14]. Biodiesel has cetane number which is medium-duty diesel engine will be investigated in the present study.
higher than ordinary petroleum diesel, which implies that it has a better From the results, the effect of each of the injection strategy will be
ignition properties and higher combustion efficiency [12,15,16]. Bio- determined to enhance the research in developing biodiesel to over-
diesel evaporates, atomizes and breaks up slower because it has a come the air pollution and fossil fuels depletion issues.
higher kinematic viscosity and surface tension. Hence, it is important to
apply a suitable injection strategies to overcome this problem.
The increasing of world population has rapidly increasing the en- 2. Experimental apparatus and procedure
ergy demand from non-renewable resources [17]. Biodiesel fuels have
gained increasing attention worldwide as blending components or di- 2.1. Apparatus setup
rect substitution for diesel engines [18]. However, the utilization of
biodiesel in diesel engines has usually produced higher NOx and brake The experimental study was implemented using three types of fuel
specific fuel consumption [19–21]. Using alternative fuels and optimi- sample. The samples consist of diesel, B20 (in a volume fraction of 20%
zation of fuel injection parameters in diesel engine can be reliable biodiesel blends and 80% diesel) and B50 (in a volume fraction of 50%
methods to solving this problem. According to Shivakumar et al. [22] in biodiesel blends and 50% diesel) of coconut oil biodiesel (COB) blends.
the research of effects of biodiesel and injection timing on single-cy- A four-cylinder diesel engine with Delphi common-rail fuel injection
linder diesel engine performance, NOx emission of waste cooking oil system and turbocharger set up was utilized in the research. Variation
blended fuel is relatively higher than that of baseline diesel. It is found of speeds and loads was controlled using eddy current engine dynam-
that the NOx emission will be lower at retarded injection timing for ometer with rating of 150 kW. Measurement of engine fuel consump-
baseline diesel and biodiesel. The smoke emission of biodiesel is less tion was done by employing Kobold fuel flow meter. Temperatures of
than baseline diesel and the advanced injection timing will reduce the surrounding air, engine lubricant oil, exhaust gas emitted and engine
smoke reduction of both kinds of diesel. Qi et al. [23] showed that NOx coolant were obtained by using K-type thermocouples. Table 1 shows
emission always decreases with the retarding injection timing when the specifications and information of test engine.
using biodiesel produced from soybean as energy source in six-cylinder Commercially available microcontroller was used as the engine
diesel engine. Sayin et al. [24] and Ganapathy et al. [25] also obtained control module (ECM). The microcontroller employed three interrupt
the same results as Shivakumar et al. [22] when Jatropha biodiesel are service routines in order to receive the signals from incremental en-
used correspondingly in their researches by using single-cylinder diesel coder and engine camshaft. Besides, a C programming language was
engine. Recently, modern engines are trending toward multiple fuel used in programming coding. The codes were uploaded to micro-
injection events as a means to decrease emissions and improve engine controller through serial communication with personal computer. A
performance [26,27]. In the investigation of effects of split injection on graphic user interface (GUI) program was created using LabVIEW to
the emissions of biodiesel, Fang et al. [28] found that NOx emission of manipulate and examine the engine parameters including start of in-
biodiesel can be 34% lower than baseline diesel under certain injection jection (SOI) timing, engine speed, opening pulse-width (PW) and
scheme at specific condition in single-cylinder diesel engine. The in- number of injections (single, double and triple injection) and closed-
jection strategies used in their study included double injection of a loop engine speed control mode selection. Quantity of diesel injected
small first injection with an early pre-TDC (top dead center) timing and was regulated by a dedicated engine speed controller to maintain en-
followed by a main injection at or after TDC. Joonho Jeon et al. [29] gine rpm to within ± 10 rpm (revolution per minute) from the set
discovered that retardation of pilot injection timing will cause in- point. This engine speed controller comprised a fine-tuned propor-
creasing amount of NOx emission when soybean biodiesel is used in tional-integral (PI) control loop. By deploying this approach, the speed
single-cylinder diesel engine. Kuen Yehliu et al. [30] showed that under controller could reject a huge amount of disturbance and minor steady-
single injection scheme in four-cylinder diesel engine, the amount of state error spanning the whole engine operating range. In addition,
emission of NOx is higher for biodiesel compared to baseline diesel. programmable peak and hold pulse-width-modulation (PWM) was in-
However, when split (pilot and main with non-equal fuel quantity) corporated in engine controller to control the solenoid injectors for
injection scheme is carried out, the NOx emission is the lowest for common-rail direct injection efficiently. Engine parameters could be
biodiesel. The drawback is that there will be an increase in particulate fully controlled by this specifically designed control unit.
matter emitted when biodiesel and baseline diesel are used when split To execute the combustion process analysis, the gas pressure in
injection scheme is carried out. Su Han Park et al. [31] studied about cylinder was obtained with a Kistler 6058A piezoelectric sensor and
the relationship of injection timing and split injection (with up to two high speed data acquisition system was used to record its signal. Glow
injections per combustion cycle) with the performance of biodiesel by plug adapter was utilized so that pressure sensor could be fixed in the
using a single-cylinder diesel engine. They found that NOx emission first cylinder’s head. DAQ-Charge-B charge amplifier was employed to
increases with advanced injection timing. Multiple injection scheme enlarge the signal from pressure sensor. Incremental encoder with the
cause less soot to be emitted compared to single injection scheme ex-
Table 1
cept at highly advanced injection timing. The retardation of pilot in-
Specifications and information of the test engine.
jection timing reduces the amount of soot emitted. Besides, multiple
injection strategy has been proposed as an effective way to reduce Type of Engine Diesel, 4-stroke, turbocharged direct injection
unburned emissions and noise [32,33]. engine
From the previous works, it is evident that most of the studies have Fuel injection supply system Diesel common-rail with rail pressure 140 MPa
max.
been focused on experimental research of biodiesels in which few ex- Cylinder 4
periments were conducted with injection timing variation and split Valve per each cylinder 2
injection, specifically dealing with two or more fuel injections of equal Bore x stroke 76.0 mm × 80.5 mm
quantities per combustion cycle. Besides, although the above men- Connecting rod length 135 mm
Displacement 1461 cm3
tioned literature reported the studies of split injection engine combus-
Compression ratio 18.25–1
tion, but most of them have been performed on single-cylinder research Maximum power & torque 48 kW at 4000 rpm & 160 Nm at 2000 rpm
engine, which is not practical representative of the production engine

107
H.G. How et al. Fuel 213 (2018) 106–114

Fig. 1. The arrangement of experimental setup.

resolution of 0.125°CA (crank angle) was used to measure the crank- injection. Firstly, single injection was carried out in the fuel injection
shaft rotation angle. Cylinder pressure data for 100 consecutive engine event. During second approach, the fuel injection process was per-
revolutions were collected and averaged for each test. Fig. 1 shows the formed by two similar injection pulses, while in the third fuel injection
schematic diagram of the experiment setup. Regarding pollutant event three equal amount injection pulses were introduced. It has to be
emission evaluation, concentrations of NOx was measured by using AVL emphasized that the strategy of splitting the single injection pulse into
DICOM 4000 gas analyser while AVL DiSmoke 4000 was set up to ob- two and three identical pulses was that the sum of entire engine power
tain smoke opacity. was fixed for every profile and SOI timing. In each test scheme, a
commercial diesel fuel was used as baseline fuel for comparison pur-
poses. When biodiesel blends were used to operate the engine at room
2.2. Experimental procedure temperature, it functioned adequately over the entire test and there was
no starting difficulty. The tests were conducted when the steady-state
Usually, engines used for medium-duty diesel powered urban ve- conditions were achieved. Exhaust gas was warmed sufficiently and
hicles are operated under partial load condition. For this reason, part water coolant was maintained at suitable temperature. The important
load operation is considered in choosing the experimental cases. In the properties of the petroleum diesel, B20 and B50 are listed in Table 2.
present engine testing, the engine speed is held constant at 2000 rpm Every test case was repeated for two times to obtain average value in
and with 60 Nm of torque. The influences of biodiesel blended fuels on order to improve the accuracy in the study. Repeatability was as high as
engine out-responses, such as combustion, tail-pipe emissions and 95% for every case tested.
performance characteristics under different split injection approach
(single, double and triple) and injection timing (−12°ATDC to 2°ATDC)
conditions were studied. Split injection approach can be implemented 2.3. Biodiesel property test
by dividing the single injection event into two or three consecutive
equal injection events for each engine combustion cycle with certain After transesterification process was completed, the methyl ester
dwell timing between consecutive injections. The advantages of this produced was thoroughly inspected its fuel properties and comparison
injection strategy is that it reduces combustion flame temperature and with biodiesel standards was made. Table 2 shows the detail of the
enables ample fuel and air mixing so that charge homogeneity can be crucial physicochemical properties possessed by the converted neat
improved. Fig. 2 delineates the timing schemes for single, double and COB compared with ASTM (American Society for Testing and Mate-
triple injection strategies studied in the present work. Obviously, the rials) standard. The essential properties of fossil diesel fuel are also
investigated approaches vary in the splitting schematization of main shown in that table. ASTM D6751 has been used as biodiesel standard
where measured physiochemical properties of biodiesel produced were
Dwell Dwell benchmarked against it. It shows that the converted COB has all the
Triple 1st 2nd 3rd physicochemical properties complied with the ASTM and biodiesel
Injeciton 33.3% 33.3% 33.3%
standard. Notably, the kinematic viscosity of the transesterified coconut
Dwell oil was improved considerably. However, it was marginally greater
Double
1st 2nd Dwell= 1.3ms than that of conventional diesel. Besides, COB had a higher flash point
Injeciton
50% 50% than conventional diesel and therefore suitable to be utilized as trans-
portation fuel. The drawback is that COB possess lower calorific value
Single
Injeciton 100% compared to petroleum diesel fuel. Another factor which influences
engine combustion characteristics and performance is the distillation
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 temperature of fuel. Commonly, distillation temperature can act as a
Time ASOI (ms) parameter for checking fuel quality while information about volatility,
flash point and fatty acid composition can be obtained by examining the
Note: ASOI: After Start of Injection
distribution range. In this research, the entire ranges of distillation
Fig. 2. Timing chart for various split injection strategies. Note: ASOI: After Start of temperatures of the fuel sample Tx, in which “x” means distillation
Injection.
temperatures corresponding to x vol% of the distilled and condensed

108
H.G. How et al. Fuel 213 (2018) 106–114

Table 2
Physicochemical properties of baseline fossil diesel and neat COB.

Parameters Units Standards COB Biodiesel Limit (ASTM D6751) Diesel

Kinematic viscosity @ 40 °C mm2s−1 ASTM D445 4.02 1.9–6.0 2.99


Density @ 40 °C kg/m3 ASTM D1298 856.0 – 825.6
Flash point °C ASTM D93 145.5 130 min 71.5
Cloud point °C ASTM D2500 4 Not specified 3
Pour point °C ASTM D97 3 Not specified 0
CFPP °C ASTM D6371 7 Not specified 5
Calorific value MJ/kg ASTM D240 39.92 – 45.21
Acid value mg KOH/g ASTM D664 0.05 0.5 max –
Cetane number – ASTM D6890 56.7 47 min 52.4
Oxidation stability h EN ISO 14112 15.8 3 min > 100.0
Carbon %wt ASTM D 5291 73.2 – 86.1
Hydrogen %wt 12.5 – 13.8
Nitrogen %wt < 0.1 – < 0.1
Oxygen %wt Calculation 14.3 – 0.1
Water content %wt EN ISO 12937 0.021 < 500 ppm 0.012
Copper Strip Corrosion (3 h @ 50 °C) – ASTM D 130 1a 1 1a
Distillation: °C D86 Distillation temperature, 90% recovered (T90) = 360 °C max 165.5
Initial boiling point 92
Recovery of 5% 240 220
Recovery of 10% 249 240
Recovery of 20% 260 262.5
Recovery of 30% recovery 269 276.5
Recovery of 40% 276 288.5
Recovery of 50% 284 298.5
Recovery of 60% 294 309
Recovery of 70% 312 320
Recovery of 80% 321 333
Recovery of 90% 324 351
Final boiling point 324 374

Table 3 number and the biodiesel blending ratio was obtained by Acharya et al.
The key fuel physicochemical properties of petroleum diesel, neat COB, B20 and B50. [34], and has been considered in this study. The cetane number (Y) of
the biodiesel blends was calculated from the% volume fraction of bio-
Properties Units Diesel COB B20 B50 Test method
diesel blend (X) using the following formula:
Calorific Value MJ/kg 45.21 39.92 44.60 42.79 ASTM D240
Cetane number – 52.4 56.7 53.3a 54.6a ASTM D6890 Y = 0.043X + 52.4 (1)
Density @ 40 °C kg/m3 825.6 856.0 831.3 840.1 ASTM D1298
Flash Point °C 71.5 145.5 74 80.5 ASTM D93
Kinematic viscosity mm2s−1 2.985 4.02 3.204 3.491 ASTM D445
at 40 °C
2.4. Statistical and equipment uncertainty analysis
Oxidation Stability h > 100.0 15.8 89.48 51.44 EN ISO14112

a
: calculated by assuming a linear increase in cetane number with increase in biodiesel Generally, all experimental measurement are subject to some un-
blend ratio. certainties or errors. The uncertainty in an experimental result can arise
from sensor selection, condition, calibration, observation and test pro-
liquid fuel, were obtained by using distillation temperature analyzer. It cedure. The summary of the equipment used in this study including the
is observed that the distillation temperatures of T50 of COB and diesel measurement range and accuracy of the instruments is given in Table 4.
fuel are 284 °C and 298.5 °C, respectively. In this study, two type of fuel Uncertainty analysis is essential to verify the accuracy of the experi-
blends with different percentage of methyl ester blends, which were ments. Hence, the percentage uncertainties of various parameters such
B20 and B50 were formulated and tested. The important physico- as brake thermal efficiency (BTE) and brake specific fuel consumption
chemical properties of the conventional diesel and biodiesel blends are (BSFC) were determined based on the percentage uncertainties of the
tabulated in Table 3. It can be clearly seen that mixing of biodiesel with instruments used in the experiments. The uncertainty analysis was
conventional diesel could greatly improve the resultant biodiesel blend performed using the method described by [35]. The overall experi-
properties. Particularly, the kinematic viscosity was reduced with the mental uncertainty was determined using the following equation:
increase of the ratio of conventional diesel in the blends. In additional,
the flash points of both of the B20 and B50 biodiesel blends were Overall experimental uncertainty
comparably greater than that of pure diesel. This makes them suitable = Square root of [(uncertainty of Fuel Flow Rate) 2
candidates to be used as transportation fuel. However, biodiesel blends
have a lower calorific value compared to conventional diesel. Another + (uncertainty of BSFC)2 + (uncertainty of BTE)2
key fuel property that substantially affects engine operating char- + (uncertainty of NOx )2 + (uncertainty of Smoke)2
acteristics is the cetane number of fuel. The result of cetane number was + (uncertainty of Pressure sensor)2
tested according to ASTM D6890 method and it can be seen that the
neat COB has a greater cetane number than petroleum diesel fuels. + (uncertainty of Crank angle encoder)2]
Those results are shown in Table 2. Besides, the variation of cetane
number with biodiesel blend ratio is illustrated in Table 3. It is observed = Square root of [(0.5)2 + (1.5)2 + (1.7)2 + (1.3)2 + (1)2 + (0.5)2
that the cetane number increases linearly with an increase in the bio-
diesel blending ratio. The linear relationship between the cetane + (0.03)2] = ± 2.9%

109
H.G. How et al. Fuel 213 (2018) 106–114

Table 4
List of measurement range, accuracy and percentage uncertainties.

Measurement Measurement range Accuracy Measurement techniques % Uncertainty

Load ± 600 Nm ± 0.1 Nm Strain gauge type load cell ± 0.25


Speed 0–10,000 rpm ± 1 rpm Magnetic pick up type ± 0.1
Time – ± 0.1 s – ± 0.2
Fuel flow measurement 0.5–36 L/hr ± 0.04 L/hr Positive displacement gear wheel flow meter ± 0.5
NOx 0–5000 ppm ± 1 ppm Electrochemical ± 1.3
Smoke 0–100% ± 0.1% Photodiode detector ±1
Pressure sensor 0–25,000 kPa ± 10 kPa Piezoelectric crystal type ± 0.5
Crank angle encoder 0–12,000 rpm ± 0.125° Incremental optical encoder ± 0.03

Computed
BSFC – ± 5 g/kWhr – ± 1.5
BTE – ± 0.5% – ± 1.7

Fig. 3. BTE and BSFC for various fuels, start of


Single Injection Scheme Double Injection Scheme Triple Injection Scheme injection timings and split injection schemes.
36.0 450
34.0 430
32.0 410
30.0 390
28.0 D (BTE) 370
26.0 B20 (BTE) 350
BTE (%)

B50 (BTE)
24.0 D (BSFC) 330

BSFC (g/kWhr)
22.0 B20 (BSFC) 310
B50 (BSFC)
20.0 290
18.0 270
16.0 250
14.0 230
-12 -10 -8 -6 -4 -2 0 2 -12 -10 -8 -6 -4 -2 0 2 -12 -10 -8 -6 -4 -2 0 2
Start of Injection (SOI) timing for 1st pulse, °ATDC

3. Results and discussion smaller effective pressure will be produced. Another observation is that
when single injection scheme is implemented at SOI of −12°ATDC, it is
3.1. Performance characteristics found that BTE decreases with increasing of biodiesel blend. This trend
is discovered for other SOI and split injection scheme, which is in ac-
Effects of biodiesel blends, injection timing variation and split in- cordance with other researches [30,37]. The lower BTE can be due to
jection scheme on BTE and BSFC are discussed in the following section. the fact that diesel engine used in the present study is not purposely
Fig. 3 delineates the changes in BTE with various SOI timings and split designed for the use of biodiesel. Besides, with the addition of biodiesel
injection schemes of the engine fuelled with B20, B50 and baseline fuel in the blend, the oxygen content is richer than that of conventional
diesel. It was found that the BTE for baseline diesel is constantly higher diesel. This causes biodiesel blended fuels to have a lower calorific
than that of B20 and B50 at every SOI timing. Literally, peak BTE of value in comparison to baseline diesel properties as shown in Table 2.
baseline diesel, B20 and B50 are observed to be 33.5%, 32.6% and Another important engine performance characterisitic is the BSFC.
32.1% respectively at SOI timing equals −12°ATDC and with single According to the results, BSFC values obtained when B20 and B50
injection scheme. Besides, the outcomes also show that difference in biodiesel blended fuels are used are consistently larger than that of
SOI timing profoundly affects BTE. Advanced SOI timings causes in- baseline diesel at all SOI timing and injection schemes. The greater
crement in BTE for all fuel type tested. The improvement is attributable BSFC value of both B20 and B50 indicates that more fuel was needed to
to the longer ignition delay (physical delay) which in turn brings about generate the same power amount. This expectation is made due to the
a better mixing. This results in more efficient combustion and a greater low heating value of B20 and B50 than that of petroleum diesel, which
BTE. Another explanation can be made where at advanced injection was about 1.4% and 5.4%, respectively less than heating value of pet-
timing, the engine attains peak combustion pressure near to TDC. roleum diesel fuel. Moreover, it is discovered that the changes in in-
Hence, greater effective pressure can be produced to generate useful jection timing has caused impact on the magnitude of the BSFC. Re-
work [36]. Nonetheless, for the corresponding SOI timing, there is a duction in BSFC can be achieved by advancing the SOI timing from
perpetual decline of BTE by an average of 5.2% and 13.1% for the case 2°ATDC when different types of diesel are used. When SOI was im-
of double and triple injection, respectively, in comparison to the BTE of plemented earlier, there was continuous improvement in combustion
single injection operation using baseline diesel. It may be due to the effectiveness and quality. This explains the decrement in BSFC. As BSFC
combustion process occurs over a longer period for double and triple decreases under fixed value of brake power output, less amount of fuel
split injection. Thus, the heat losses through the cylinder wall is in- is being injected to enable a more efficient combustion process to occur.
creases due to longer combustion duration, and consequently less useful This occurs particularly when earlier SOI timing is set. In addition, it
mechanical work is generated. Another explanation is that the power can be noticed that for the respective SOI timing, the BSFC is increased
output is reduced because a greater quantity of fuel is combusted when with the increasing in number of split injections for all fuels. These
the cylinder expansion occurs. Subsequently, the cylinder pressure in- results are consistent with other researchers [38,39]. By diving the
creases only when volume of cylinder is expanding quickly. Therefore, a main fuel injection into two and three parts reduced the peak

110
H.G. How et al. Fuel 213 (2018) 106–114

Fig. 4. NOx and smoke emission for various fuels, SOI


Single Injection Scheme Double Injection Scheme Triple Injection Scheme timings and injection strategies.
500 20

Smoke Emission (%)


16
400 12
8
NOx (ppm)

300 4
0
200 -4
D (NOx)
B20 (NOx) -8
B50 (NOx)
100 D (Smoke)
-12
B20 (Smoke) -16
B50 (Smoke)
0 -20
-12 -10 -8 -6 -4 -2 0 2 -12 -10 -8 -6 -4 -2 0 2 -12 -10 -8 -6 -4 -2 0 2
Start of Injection (SOI) timing for 1st Pulse, °ATDC

combustion pressure, thus lower the amount of fuel injected when TDC and the absence or lesser quantity of aromatics hydrocarbon in bio-
is reached. This causes reduction in work done. Another possible ex- diesel blends. The results also indicate that with advanced SOI timings,
planation is that it may be due to the longer combustion duration, thus the smoke emissions were reduced when double injection strategy is
increased the heat loss when higher number of split injections is being carried out, and somewhat remained unchanged during operation of
used. As a result, BSFC has to be increased to supply more energy. single injection strategy. This is because when SOI timing is advanced,
gas temperature developed will be higher, in which fuel and oxygen
3.2. Emissions characteristics reaction will be improved. As a result, both injection schemes will yield
lower emission level of smoke. The second possible reason is the pre-
Effects of biodiesel blends, injection timing and split injection sence of enough time for the fuel to vaporize and form mixture with air,
scheme on NOx and smoke emissions amount are discussed in this thus allowing effective mixing and complete combustion. However, the
section. The trend of NOx emissions when various test fuels are used at triple injection scheme operation shows slightly higher smoke emission
different SOI timings as well as split injection schemes is displayed in with earlier SOI timings. It may be due to the decrease in the in-cylinder
Fig. 4. It is found that advanced SOI timing causes an increase in NOx combustion temperature (as shown in Fig. 7) that prevent high burn
emission amount for all fuel types and multiple injection scheme. In- rates, which leads to incomplete combustion with increased smoke
creasing trend is observed in NOx emissions when SOI timing is ad- emissions. This implies that multiple injections are able to jeopardize
vanced. This condition suggests that the ignition and combustion of smoke emissions level when the optimization of injection timing is not
mixture occur earlier and result in early formation of peak pressure implemented well. By operating the engine with B50 fuel, smoke
around TDC. This causes the combustion to happen at a higher tem- emission can be diminished without affecting the reduction effect in
perature and promotes the thermal or Zeldovich NOx formation me- NOx emissions amount. The results show that a similar smoke emission
chanism. The results also imply that both B20 and B50 biodiesel blends level (i.e. below 6% for diesel operation with single injection scheme)
tend to diminish NOx emissions over all SOI timings and split injection can be obtained when the B50 is used along with the strategies of re-
strategies. This is due to the rather greater cetane number and smaller tarded SOI timing and triple injection scheme. Therefore, simultaneous
calorific value of the B20 and B50 in comparison with baseline diesel as decrease in NOx and smoke emission level from the baseline levels of
depicted in Table 3. The factors subsequently reduce the heat release pure diesel is achievable with the usage of B50 biodiesel blended fuel in
rate during the premix combustion phase and lower the peak combus- conjunction with the application of retarded SOI timing and triple in-
tion temperature. The discovery can be explained by comparing the in- jection scheme in a diesel engine.
cylinder mean gas temperature results depicted in Fig. 7. Besides, a
considerably lower NOx emission level under 100 ppm can be attained 3.3. Combustion characteristics
by applying late SOI timing for B20 and B50 fuel operations and with
triple injection scheme. With injection performed lately, the retardation To investigate the influence of injection scheme and types of fuel on
of combustion phasing will happen progressively in the expansion combustion characteristics, 100 consecutive cycles of cylinder com-
stroke and moves away from TDC. This effect causes a decrease in bustion pressure were logged and processed. The average values were
combustion gas temperature and reducing NOx formation. Meanwhile, calculated and compared. Fig. 5 depicts the graphs of injector current
dividing the main injection into two and three parts lowered the NOx profile, combustion pressure and HRR of the engine, utilizing baseline
emissions. This can be attributed to the improvements made in the diesel at SOI of −6°ATDC (nominal timing) and with various injection
injection process of modulating the injection rate for multiple injection strategies. From the injector current profile, it can be seen that the fuel
strategy, thus splitting the heat release process into separate portions injection event was formed by two and three identical injection pulses
and reduced in peak rate of heat release. This shows that split injection for double and triple injection schemes, respectively. Besides, the en-
is effective in reducing NOx emissions. The smoke formed due to the gine operated with multiple injection schemes had significant impact
unfinished combustion of the hydrocarbon in the fuel and partial oxi- on the combustion process. The crank angle position of peak pressure
dation of the carbon content in the fuel. The amount of smoke emissions was advanced towards the compression stroke with greater number of
of each test fuel at different SOI timings and injection schemes is de- injection schemes. Also, the location of start of combustion (SOC)
lineated in Fig. 4. Overall, it is discovered that drop in level of smoke timing for double and triple injection scheme occurred 0.5°CA and 1°CA
emission occurs when B20 and B50 are employed at every SOI timing. respectively, more advanced than that of single injection scheme. Ad-
Less smoke was emitted than when diesel fuel is utilized at each SOI ditionally, in double injection test case, peak pressure undergoes a
timing, mainly due to richer fuel-borne oxygen, less carbon content, minor decline in the range of 1.7 bar. However, triple injection strategy

111
H.G. How et al. Fuel 213 (2018) 106–114

Fig. 5. In-cylinder combustion pressure, HRR and injector current patterns for baseline
diesel with various injection strategy at SOI of −6°ATDC.

has resulted in a slight increases in the range of 1 bar was discovered for
peak pressure. Two and three notable HRR peaks can be noticed for
double and triple injection strategies, respectively. Besides, it can be
seen that the occurring position for the first peak of HRR curve, when
double and triple injection strategies are conducted, is shifted earlier by
0.5°CA and 1°CA respectively, in comparison with single injection
strategy. The main cause for the first HRR peak timing to occur early is
the advanced occurrence in SOC timing and subsequently leads to the
earlier HRR rise. Besides, the two diffusion burns resulting from the
subsequent fuel pulses occur at about the same time after start of in-
jection for both timings (ignition delay_2 is approximately same with
ignition delay_3) but lower intensities are evidently seen. The second
and third fuel pulse ignite almost immediately soon after the SOI_2 and
SOI_3 since they are directly injected into the main combustion zones.
Fig. 6 indicates the plot of combustion pressure and HRR variations
as a function of crank angle at various SOI timings, single injection
strategy and with all types of fuel tested. Overall, the pressure peak of
combustion constantly rises and advances towards the location of TDC
with earlier SOI timing for all fuels. Greater effective work can be
generated via the larger pressure and this helps in enhancing BTE and
reducing BSFC. Besides, the HRR plots also exhibit a resembling pat-
terns as that of combustion pressure in which first peak of HRR asso-
ciated with the premixed combustion phase was advanced towards Fig. 6. In-cylinder combustion pressure and HRR plots for (a) baseline diesel, (b) B20, (c)
compression stroke with earlier SOI timing when different fuels are B50 at different SOI timings and single injection strategy.

utilized. With SOI timing advancement toward the TDC point in cy-
linder expansion stroke, the peak HRR correspond to the premixed
combustion phase became initially reduce and kept unchanged. With
the increasing of biodiesel in the fuel blend, the HRR pattern was al-
most the same as that of conventional diesel, however it can be noticed
that larger fraction of fuel was consumed during the mixing controlled
combustion phase. This can be shown by the wider plateau after the
first peak of HRR. In fact, this phenomenon can be observed clearly in
the late SOI condition of 2°ATDC, in comparison with earlier SOI
timing. The reason is that biodiesel has a greater cetane number than
that of baseline diesel as depicted in Table 3. This causes ignition delay
to be shortened, thereby producing a lower peak of HRR. In general, it
is discovered that B20 and B50 combustion release 1.4 J/°CA and 4.3 J/
°CA, respectively lower in HRR peak compared to baseline diesel over
the entire range of SOI timings.
Effects of biodiesel blends, injection timing and split injection
scheme on peak mean gas temperature (PMGT) and peak heat release
rate (PHRR) are discussed in the following section. Fig. 7 shows the
changes in PMGT with various SOI timings and split injection strategies
Fig. 7. In-cylinder combustion pressure and HRR profiles for (a) baseline diesel, (b) B20,
of the engine operated with B20, B50 and baseline fuel. The PMGT of
(c) B50 at different SOI timings and single injection strategy.
B20 and B50 are always found lower than that of baseline diesel over
the entire range of SOI timings. Indeed, the results show that the
highest PMGT of baseline diesel, B20 and B50 are 1964 K (at
SOI = −12°ATDC), 1922 K (at SOI = 2°ATDC) and 1955 K (at

112
H.G. How et al. Fuel 213 (2018) 106–114

SOI = −12°ATDC) respectively and with single injection scheme. Be-


sides, it is observed that SOI timing variation profoundly influences the
magnitude of PMGT. For single injection scheme, when SOI timing is set
earlier, the PMGT will increase for all fuel type tested. The incremental
effect is due to the higher peak combustion pressure, which results in a
higher PMGT. However, for the corresponding SOI timing, there is a
continuous reduction of the PMGT by an average of 5.1% and 9.3% for
the case of double and triple injection, respectively, as compared to
those of baseline diesel single injection operation. Another observation
is that when triple injection scheme is implemented at SOI of −12°A-
TDC, it is found that PMGT decreases with biodiesel blended fuels.
Another important combustion characterisitic is the magnitude of
PHRR. According to Fig. 7, it can be seen that B20 and B50 are con-
sistently exhibit lower PHRR compared to baseline diesel for all SOI
timings and injection schemes, except for the single injection operation.
This is because biodiesel has a greater cetane number in comparison
with baseline diesel. Consequently, biodiesel has a shorter ignition
delay and a smaller peak of HRR, as aforementioned. Also, it can be
seen that for a specific SOI timing, the PHRR is decreased with the
increasing in number of split injections for all fuels. By diving the main
fuel injection into two and three parts may reduced the quantity of fuel
combusted in the premixed burn stage, thus lower PHRR.
Multiple injections approach, comprises pilot and post injections are
commonly used to mitigate particulate matter/ smoke, controlling
amount of NOx, managing engine combustion noise and to handling
exhaust pipe after treatment equipment in diesel engine. Combustion
pressure and heat release rate data of (a) single, (b) double, (c) triple
injection at different SOI timings using baseline diesel as fuel can be
obtained from Fig. 8. The beneficial effect of double and triple split
injections will be discussed in this section by directly comparing dif-
ferent injection strategies with the corresponding SOI timing. This al-
lows one to point out the differences in the combustion events that are
responsible for results observed in NOx and smoke emissions. First, the
double injection at later SOI of 2°ATDC is compared to the single in-
jection with similar injection timing in order to further analyze the
benefits of one injection interruption. The heat release data shows that
the two combustion events start deviating once the first fuel pulse of the
double injection has been terminated after a premixed burn of about the
same intensity, as can be clearly seen in Fig. 5. For double injection, as
larger amount of the injected fuel has already been consumed in the Fig. 8. Combustion pressure and heat release rate curves for (a) single, (b) double, (c)
premixed flame. Thus, the following combustion process during the triple injection at various SOI timings and with baseline diesel.
injection dwell (i.e. the duration between the SOI_1 and SOI_2 as de-
picted in Fig. 5) is characterized by a fuel-air mixture that burns at very SOI timing of 2°ATDC. Another interesting observation is that because
high temperatures. Consequently, the second fuel injection in double of the similarity of the second and third fuel injection event, it is no
injection burns as a diffusion flame only. The fuel mixes with the hot surprise that the second and third peak HRR are almost identical for
gases remaining in the combustion chamber and ignites after a very triple injection case (see Fig. 8c).
short ignition delay. Besides, the double injection has a lower peak
mean gas temperature that keeps the NOx production on a lower level, 4. Conclusion
as shown in Fig. 7. The subsequent fuel injection causes another,
slightly lower HRR and flame temperature rise, which lasts for about In the present study, the engine performance, combustion, and ex-
20°CA in the combustion cycle and ensured lower NOx production. haust gas emissions characteristics of baseline diesel, B20 and B50
Further optimization of split injections in terms of NOx and smoke biodiesel blended fuels have been empirically tested in a medium-duty
emissions can be accomplished by using additional injection pulses, common-rail DI diesel engine. By holding the engine speed at 2000 rpm
which is called triple injection. The experimental observation made by and at 60 Nm load as experimental condition, investigation of the effect
Pierpont found that the optimized triple injection, as the most sophis- of SOI timing and multiple injection strategies had been successfully
ticated, but also most complicated injection strategy, can lower emis- carried out. Below are the main conclusions which are inferred from the
sions beyond the capability of double injections [39,40]. The direct present work.
comparison of double and triple injection with the same injection
timing as shown in Fig. 5, reveals that the two combustion events re- (i) The diesel engine emissions, performance and combustion char-
sulting from these strategies differ only in the second half. This com- acteristics are affected by SOI timing and split injection scheme
bustion phase is most responsible for overall smoke emissions. The fuel significantly for all types of fuel tested.
combustion of the third injected fuel during piston expansion cooling (ii) A considerably lower level of NOx, which is below 100 ppm, can be
stroke can be used to ensure a proper oxidation of the fuel-air mixture. attained by retarded SOI timing for B20 and B50 fuel operations
This positive effect of the soot oxidation during the last stage of the and with triple injection scheme.
combustion process can be noticed with the lower smoke emission level (iii) Simultaneous NOx and smoke emission decrement from the
compared to the double injection, as is reflected in Fig. 4 with the late

113
H.G. How et al. Fuel 213 (2018) 106–114

baseline levels of petroleum diesel fuel can be achieved by utilizing emissions of gasoline and LPG in car engines under cold climatic conditions. Energy
Sourc Part A: Recov Utiliz Environ Effects 2010;33(3):244–53.
B50 in conjunction with implementation of retarded SOI timing [18] Demirbas A. Progress and recent trends in biodiesel fuels. Energy Convers Manage
and employment of triple injection scheme in a diesel engine. 2009;50(1):14–34.
(iv) Multiple split injections is a practical strategies to simultaneously [19] Al-Dawody MF, Bhatti SK. Optimization strategies to reduce the biodiesel NOx ef-
fect in diesel engine with experimental verification. Energy Convers Manage
decrease NOx and smoke emissions when the SOI timing is fine- 2013;68:96–104.
tuned and is an ideal alternative to operate with biodiesel fuel. [20] Palash SM, et al. State of the art of NOx mitigation technologies and their effect on
the performance and emission characteristics of biodiesel-fueled Compression
Ignition engines. Energy Convers Manage 2013;76:400–20.
Acknowledgments [21] Rahman SMA, et al. Production of palm and Calophyllum inophyllum based bio-
diesel and investigation of blend performance and exhaust emission in an un-
The authors would like to acknowledge the Ministry of Higher modified diesel engine at high idling conditions. Energy Convers Manage
2013;76:362–7.
Education (MOHE) of Malaysia, Universiti Malaya and Universiti Sains
[22] Shivakumar P, Srinivasa Pai, Shrinivasa Rao BR. Artificial Neural Network based
Malaysia for financial support through short term research grant prediction of performance and emission characteristics of a variable compression
scheme (304/PMEKANIK/60313054). ratio CI engine using WCO as a biodiesel at different injection timings. Appl Energy
2011;88(7):2344–54.
[23] Qi D, et al. Effect of EGR and injection timing on combustion and emission char-
References acteristics of split injection strategy DI-diesel engine fueled with biodiesel. Fuel
2011;90(5):1884–91.
[1] El Diwani G, Attia NK, Hawash SI. Development and evaluation of biodiesel fuel and [24] Sayin C, Gumus M. Impact of compression ratio and injection parameters on the
by-products from jatropha oil. Int J Environ Sci Technol 2009;6(2):219–24. performance and emissions of a DI diesel engine fueled with biodiesel-blended
[2] Attfield MD, et al. The diesel exhaust in miners study: a cohort mortality study with diesel fuel. Appl Therm Eng 2011;31(16):3182–8.
emphasis on lung cancer. JNCI: J Natl Cancer Inst 2012;104(11):869–83. [25] Ganapathy T, Gakkhar RP, Murugesan K. Influence of injection timing on perfor-
[3] Li N, et al. Particulate air pollutants and asthma. A paradigm for the role of oxi- mance, combustion and emission characteristics of Jatropha biodiesel engine. Appl
dative stress in PM-induced adverse health effects. Clin Immunol Energy 2011;88(12):4376–86.
2003;109(3):250–65. [26] Han D, et al. Experimental study of the two-stage injection process of fatty acid
[4] Zhang JJ, et al. Health effects of real-world exposure to diesel exhaust in persons esters on a common rail injection system. Fuel 2016;163(Supplement C):214–22.
with asthma. Res Rep Health Eff Inst 2009;138:5–109. discussion 111-23. [27] Cha J, et al. High pressure and split injection strategies for fuel efficiency and
[5] Mohan B, Yang W, Chou SK. Fuel injection strategies for performance improvement emissions in DI diesel engine. SAE Int 2015.
and emissions reduction in compression ignition engines—A review. Renew Sustain [28] Fang T, Lee C-fF. Bio-diesel effects on combustion processes in an HSDI diesel en-
Energy Rev 2013;28:664–76. gine using advanced injection strategies. Proc Combust Inst 2009;32(2):2785–92.
[6] Park SH, Kim HJ, Lee CS. Effect of multiple injection strategies on combustion and [29] Jeon J, Park S. Effects of pilot injection strategies on the flame temperature and soot
emission characteristics in a diesel engine. Energy Fuels 2016;30(2):810–8. distributions in an optical CI engine fueled with biodiesel and conventional diesel.
[7] Busch S, Zha K, Miles PC. Investigations of closely coupled pilot and main injections Appl Energy 2015;160:581–91.
as a means to reduce combustion noise in a small-bore direct injection Diesel en- [30] Yehliu K, Boehman AL, Armas O. Emissions from different alternative diesel fuels
gine. Int J Engine Res 2015;16(1):13–22. operating with single and split fuel injection. Fuel 2010;89(2):423–37.
[8] Herfatmanesh MR, et al. Experimental investigation into the effects of two-stage [31] Park SH, Yoon SH, Lee CS. Effects of multiple-injection strategies on overall spray
injection on fuel injection quantity, combustion and emissions in a high-speed op- behavior, combustion, and emissions reduction characteristics of biodiesel fuel.
tical common rail diesel engine. Fuel 2013;109:137–47. Appl Energy 2011;88(1):88–98.
[9] Montgomery DT, Reitz RD. Six-mode cycle evaluation of the effect of EGR and [32] Kim D, Bae C. Application of double-injection strategy on gasoline compression
multiple injections on particulate and NOx emissions from a D.I. diesel engine. SAE ignition engine under low load condition. Fuel 2017;203(Supplement C):792–801.
Int 1996. [33] Park W, et al. Use of multiple injection strategies to reduce emission and noise in
[10] Liu Y, et al. Direct transesterification of spent coffee grounds for biodiesel pro- low temperature diesel combustion. SAE Int 2015.
duction. Fuel 2017;199(Supplement C):157–61. [34] Acharya N, et al. Analysis of properties and estimation of optimum blending ratio of
[11] Ma F, Hanna MA. Biodiesel production: a review1Journal Series #12109, blended mahua biodiesel. Eng Sci Technol Int J 2017;20(2):511–7.
Agricultural Research Division, Institute of Agriculture and Natural Resources, [35] How HG, et al. An investigation of the engine performance, emissions and com-
University of Nebraska-Lincoln. 1. Bioresour Technol 1999;70(1):1–15. bustion characteristics of coconut biodiesel in a high-pressure common-rail diesel
[12] Mohamed Shameer P, et al. Effects of fuel injection parameters on emission char- engine. Energy 2014;69(Supplement C):749–59.
acteristics of diesel engines operating on various biodiesel: a review. Renew Sustain [36] Bari S, Yu CW, Lim TH. Effect of fuel injection timing with waste cooking oil as a
Energy Rev 2017;67:1267–81. fuel in a direct injection diesel engine. Proc Inst Mech Eng Part D: J Automob Eng
[13] Liu H, et al. Combustion and emission characteristics of a direct injection diesel 2004;218(1):93–104.
engine fueled with biodiesel and PODE/biodiesel fuel blends. Fuel [37] Gnanasekaran S, Saravanan N, Ilangkumaran M. Influence of injection timing on
2017;209(Supplement C):62–8. performance, emission and combustion characteristics of a DI diesel engine running
[14] Zhang Y, Boehman AL. Impact of biodiesel on NOx emissions in a common rail on fish oil biodiesel. Energy 2016;116:1218–29.
direct injection diesel engine. Energy Fuels 2007;21(4):2003–12. [38] Chen SK. Simultaneous reduction of NOx and particulate emissions by using mul-
[15] Atabani AE, et al. A comparative evaluation of physical and chemical properties of tiple injections in a small diesel engine. SAE Int 2000.
biodiesel synthesized from edible and non-edible oils and study on the effect of [39] Pierpont DA, Montgomery DT, Reitz RD. Reducing particulate and NOx using
biodiesel blending. Energy 2013;58:296–304. multiple injections and EGR in a D.I. Diesel. SAE Int 1995.
[16] Mittelbach M. Diesel fuel derived from vegetable oils, VI: Specifications and quality [40] Bakenhus M, Reitz RD. Two-color combustion visualization of single and split in-
control of biodiesel. Bioresour Technol 1996;56(1):7–11. jections in a single-cylinder heavy-duty D.I. diesel engine using an endoscope-based
[17] Arslan O, Kose R, Ceylan N. Experimental analysis of consumption and exhaust imaging system. SAE Int 1999.

114

You might also like