Fuel 159 (2015) 470–475

Contents lists available at ScienceDirect

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

Experimental study on the two stage injection of diesel and gasoline

blends on a common rail injection system
Dong Han ⇑, Yaozong Duan, Chunhai Wang, He Lin ⇑, Zhen Huang ⇑
Key Laboratory of Power Machinery and Engineering, Ministry of Education, Shanghai Jiao Tong University, Shanghai 200240, China

h i g h l i g h t s

 Two stage injection of diesel–gasoline blends on a common rail system is studied.

 Injection dynamics and characteristics are studied with changed parameters.
 Diesel/gasoline blends and diesel have similar cycle injection rate and quantity.
 Extended pilot injection pulse causes slightly reduced main injection mass.

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

Article history: This paper characterizes the two stage injection processes of pure diesel and two diesel–gasoline blends
Received 19 March 2015 using a diesel common rail injection system. The investigated fuel injection characteristics include the
Received in revised form 5 June 2015 cycle injection quantity, injection rate and injector inlet pressure characteristics, with changes in gasoline
Accepted 1 July 2015
proportion and pilot injection energizing pulse. Increased gasoline proportion has negligible influences
Available online 9 July 2015
on the cycle injection rate and mass, as well as the pressure dynamics during the injection process. In
addition, the extended pilot injection energizing pulse leads to increased pressure drop at the injector
Keywords:
inlet during the main injection and causes reduced main injection mass.
Two stage injection
Injection characteristics
Ó 2015 Elsevier Ltd. All rights reserved.
Diesel
Gasoline
Common rail

1. Introduction well-maintained combustion temperature could avoid the com-

The pursuit of high efficiency and clean internal combustion
engines significantly promoted the development of novel engine
combustion strategies. In the past two decades, novel combustion
modes, e.g. premixed low temperature combustion (LTC) have
been extensively studied, accomplishing simultaneous reduction
in NOx and soot emissions with deteriorated thermal efficiency
[1,2]. Fuel properties are crucial for the combustion process and
emissions formation in the advanced engine combustion strategies
[3,4]. The optimized fuel properties are proved to promote uniform
mixture formation, which could restrict the in-cylinder
soot-favorable region and avoid heavy usage of exhaust gas recir-
culation, as employed in the conventional LTC mode, to reduce
combustion temperature below the soot formation threshold. The
⇑ Corresponding authors at: Institute of Internal Combustion Engine, 800
Dongchuan Road, Shanghai Jiao Tong University, Shanghai 200240, China. Fax:
+86 21 34205553.
E-mail addresses: dong_han@sjtu.edu.cn (D. Han), linhe@sjtu.edu.cn (H. Lin),
z-huang@sjtu.edu.cn (Z. Huang).
bustion deterioration as in the conventional LTC mode. Some
researchers considered using blends of diesel and gasoline instead
of diesel in advanced compression ignition (CI) engine combustion
modes, and remarkable advantages in emissions control and fuel
efficiency improvements were observed due to the extended igni-
tion delay and improved in-cylinder fuel/air distribution [5–10].
Meanwhile, compared to gasoline LTC, using blends of diesel and
gasoline does not cause combustion instability problems at low
load and low speed conditions, as reported by Weall and Collings
[11,12].
However, as blends of diesel and gasoline are proposed to the
novel engine combustion strategies, except for their extended igni-
tion delay, their changed physical properties may potentially affect
the injection and spray characteristics, which are equally impor-
tant for the mixture formation and engine combustion. Therefore,
it behooves researchers to elucidate the injection and spray charac-
teristics of diesel and gasoline blends. Some consensuses include
that at non-reactive conditions, gasoline or diesel and gasoline
blends produce similar spray tip penetration distances as diesel

http://dx.doi.org/10.1016/j.fuel.2015.07.005
0016-2361/Ó 2015 Elsevier Ltd. All rights reserved.
 D. Han et al. / Fuel 159 (2015) 470–475 471

[13–15], and the spray droplet size decreases with increased gaso- Table 1
line proportion in diesel and gasoline blends [15,16]. Further, Payri Experimental apparatus.

et al. [13] revealed that gasoline produced lower mass flow rate Apparatus Model Specification
than diesel with fully open needle valve, while Han et al. [15] Common rail Rail pressure up to 140 MPa
found that diesel and gasoline blends with up to 40% gasoline vol- injection system Rail volume: 30 cm3
umetric proportion did not show significant differences in mass Injector Multiple-hole Hole number  diameter:
injection rate compared to pure diesel. injector 7  0.157 mm
Mono injection EFS8246 Measurement range:
Some researchers also investigated effects of two-stage qualifier 0–600 mm3
split-injection strategies on LTC engines fuelled with diesel and Relative accuracy: 0.1%
gasoline blends or gasoline-like fuels [17,18], because of the Current sensor Tektronix A622 Max input current: 100 A
advantages of split-injection strategies in the combustion phase Output voltage: 10 mV A1,
100 mV A1
control and combustion noise reduction [19]. In the two-stage
Fuel pressure Kistler Measurement range:
split-injection strategies, the pilot-main injection quantity ratio is transducer/amplifier 4067A3000/4068 0–300 MPa
a key factor influencing the mixture formation and heat release Data acquisition Yokogawa DL750 Sample frequency: 50,000 Hz
process. However, the main injection quantity is very sensitive to
the instantaneous fuel pressure magnitude [20,21]. As diesel and
gasoline blends are fuelled, the changed fuel physical properties
may influence the fuel pressure propagation characteristics in the
fuel pipe and common rail, thus impacting the main injection test apparatus are shown in Table 1. More information about the
quantity. Therefore, this study aims to compare the two-stage experimental apparatus could be referred to Ref. [22].
injection processes of pure diesel and two diesel–gasoline blends
on a high pressure common rail injection system. With changed 2.2. Test fuels and conditions
pilot injection energizing times, the effects of gasoline proportion
in fuel blends on the injection dynamics and the main injection The test fuels in this study are pure diesel and gasoline–diesel
quantities in the two-stage injection processes are illustrated. blends named G0, G20 and G40, which represent that the gasoline
volumetric percentages are 0%, 20% and 40%, respectively. The
physical properties of test fuels are shown in Table 2. Gasoline
2. Test method and test fuels has lower density, viscosity and surface tension than diesel, so
with increased gasoline percentage, the fuel blends’ density, vis-
2.1. Experimental apparatus cosity and surface tension are reduced.
The experiment was carried out at the room temperature 298 K
The schematic of the fuel injection test bench is shown in Fig. 1, and atmospheric pressure 1 atm. The speed of fuel pump test
which includes a high-pressure common rail injection system, an bench was fixed at 600 rpm. Injection pressures were held at
electronic control unit (ECU) and a data acquisition system. The 40 MPa and 100 MPa, respectively. The pilot injection pulse widths
fuel supply, the rail pressure and the injection pulse profile were were 0.3 ms, 0.5 ms and 0.8 ms and the injection intervals were 20
controlled by the ECU. Cycle fuel supply and instantaneous injec- and 30CAD (Crank Angle Degree). During the experiment, the main
tion rate were measured by the mono injection qualifier. A current injection timing was fixed on 3CAD BTDC (Before Top Dead
sensor was used to capture the injector energizing current, and a Center). The injection interval was adjusted via changing the start
fuel pressure transducer was mounted at the injector inlet to mon- of pilot injection. The data of 60 continuous injection cycles were
itor the fuel pressure trace. The data acquisition device was used to collected in each operation condition. The cyclic coefficient of vari-
store the fuel injection rate, the energizing current and the inlet ation of fuel injection quantity was calculated by the equation as
pressure of the injector. The model and specification of the main following:

Fig. 1. Schematic of the fuel injection test system.

472 D. Han et al. / Fuel 159 (2015) 470–475

Table 2
Physical properties of test fuels.

Fuel G0 G20 G40

1
Dencity@313 K/g mL 0.8045 0.7895 0.7765
Kinematic viscosity@313 K/mm2 s1 2.3878 1.6788 1.1275
Surface tension@298 K/N m1 0.0271 0.0264 0.0253

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 XN
1
r¼ ðx  lÞ2
l N i¼1 i
Herein, N is the number of collected cycles; xi is the fuel injection
quantity of every single cycle; l is the mean cyclic fuel injection
quantity.

3. Results and discussion

3.1. Definition of the injection process
The definition of the two stage injection process is shown in
Fig. 2, which are named pilot injection and main injection, respec-
tively. ‘‘0 ms’’ on the time axis is defined as the starting point of the
pilot injection energizing. On the fuel injection rate trace, the injec-
tion start point was defined as the timing when the instantaneous
fuel injection rate reaches 1 mg ms1, and the injection ending
point is defined as the timing when the instantaneous fuel injec-
tion rate decreases to 0 mg ms1. The time interval between the
energizing current starting time of the pilot injection and main
injection is defined as the injection interval. A pressure transducer
was mounted at the inlet of the injector to monitor the pressure
characteristics during the injection process. In this study, the pres-
sure between 5 ms before and 30 ms after the pilot injection was
monitored.
3.2. Influences of physical properties on injection rate and dynamics
Fig. 3 describes the instantaneous mass fuel injection rates ver-
sus time for G0, G20 and G40 at 40 MPa and 100 MPa injection
pressures. When the needle valve is fully open, the differences in
injection rates of all the test fuels are quite negligible. The
steady-state fuel injection rate is primarily influenced by discharge
coefficient, fuel density, flow area and pressure drop, as shown in
Eq. (1).
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
_ f ¼ Cd  A 
m 2  qf  DP ð1Þ
Fig. 2. Definition of the two stage injection process (G0, injection pressure:
100 MPa, pilot injection energizing pulse: 0.3 ms, injection dwell time: 20CAD,
main injection energizing pulse: 1.0 ms).
Fig. 3. Comparison of mass injection rates for three test fuels (pilot injection
energizing pulse: 0.5 ms, injection dwell time: 20CAD, main injection energizing
pulse: 2.0 ms).
where m _ f is the fuel mass flow rate, Cd is the discharge coefficient, A
is the orifice flow area, qf is the fuel density and DP is the pressure
drop. In the study by Payri et al. [13], gasoline and diesel were found
to produce quite similar discharge coefficients, so given the injector
geometry and pressure drop, the higher density of diesel fuel led to
higher fuel flow rate when the needle valve fully opened. Similarly,
in this study, the discharge coefficients for different test fuels are
also found to trivially vary within a small range of 0.78–0.81, but
in contrast to the relatively noticeable difference in the densities
of diesel and gasoline [13], the relative difference in fuel density
in this study is below 3.5%, so the differences in injection rates of
different test fuels are quite negligible.
3.3. Pressure fluctuation at injector inlet
Fig. 4 describes the pressure characteristics at the injector inlet
for G0, G20 and G40 during and after the two stage injection pro-
cess, with the injection pressure, pilot/main injection energizing
pulses and injection dwell time held at 100 MPa, 0.5 ms, 1.0 ms
and 20CAD, respectively. When the pilot injection starts, the nee-
dle valve begins to steadily rise, so the pressure at the injector inlet
declines accordingly. Afterward, as the fuel pressure reaches a local
minimum value, the pressure rises again due to the throttle effect
of the nozzle. Meanwhile, this compression wave produced propa-
gates to the common rail. When the pilot injection ends, the needle
valve falls back. Consequently, a compression wave due to the
water hammer effect oscillates and damps in the fuel pipe and
common rail. Similar pressure waves also appear during the main
injection process, and the pressure wave caused by the pilot injec-
tion will influence the pressure fluctuation produced by the main
injection. Fig. 4a shows the pressure fluctuation at the injector inlet
during the injection process. During the injection period, three test
fuels show tiny differences in the pressure fluctuation. Therefore, it
could be considered that as different test fuels are used, the injec-
tor inlet fuel pressure, as long as the test condition is given, poses
little influences on the injection rate. Fig. 4b illustrates the pres-
sure wave propagation and damping process at the injector inlet
after the injection. G0 produces the lowest oscillation magnitude
than the other two test fuels, indicating that the pressure wave
of G0 decays faster.
Fig. 5 compares the pressure traces at the injector inlet for dif-
ferent pilot injection energizing pulses, while the main energizing
time is held at 1.0 ms and the injection dwell time is held at
30CAD. Herein the single injection is considered a two stage
 D. Han et al. / Fuel 159 (2015) 470–475 473

Fig. 4. Effects of fuel type on the injector inlet pressure fluctuation during the two stage injection process (injection pressure: 100 MPa, pilot injection energizing pulse:
0.5 ms, injection dwell time: 20CAD, main injection energizing pulse: 1.0 ms) (a) pressure fluctuation during injection and (b) the pressure decay after injection.

Fig. 5. Effects of pilot injection energizing pulse on the injector inlet pressure fluctuation during the two stage injection process for different test fuels (injection pressure:
100 MPa, injection dwell time: 30CAD, main injection energizing pulse: 1.0 ms) (a) pilot injection energizing pulse: 0.0 ms, (b) pilot injection energizing pulse: 0.3 ms, (c)
pilot injection energizing pulse: 0.5 ms and (d) pilot injection energizing pulse: 0.8 ms.

injection with 0 ms pilot injection energizing pulse. Firstly, the
pressure trace of 0 ms pilot injection condition (single stage injec-
tion) is shown in Fig. 5a, in which the magnitude of pressure fluc-
tuation caused by the pilot injection is 0 MPa. As the energizing
pulse increases to 0.3 ms, the pilot injection causes a
high-frequency and low-magnitude pressure fluctuation, but this
pressure wave could quickly damp to the target injection pressure
before the main injection, without causing obvious influences on
the pressure fluctuation during the main injection. Therefore, for
an interval of 30CAD between the pilot and main injection events,
the pressure waves during the main injection period are similar for
these two cases. However, as the pilot energizing pulse further
increases, as shown in Fig. 5c and d, the pressure wave caused by
the pilot injection has reduced frequency but increased magnitude,
which cannot fully damp before the main injection and is in a
decreasing trend at the start of the main injection. This pressure
drop overlaps the one by the opening of the needle valve in the
main injection period, thus causing an increased pressure drop
during the start of the main injection period. After the main injec-
tion event, the pressure wave could finally decay to a relatively
474
Fig. 6. Effects of the pilot injection energizing pulse on the main injection mass and coefficient of variance for different test fuels (injection pressure: 100 MPa, injection dwell
time: 30CAD, main injection energizing pulse: 1.0 ms).
steady level. Among the four test cases, this steady pressure level
decreases with prolonged pilot energizing pulse. It is postulated
that this decreased pressure level with prolonged pilot energizing
time might be related to the fuel pressure during the main injec-
tion. As the pilot energizing time increases from 0 ms to 0.8 ms,
the minimum pressure during the main injection decreases from
87 MPa to 80 MPa. As discussed above, the energizing pulse width
of the pilot injection has pronounced effects on the pressure at the
injector inlet during a two stage injection process. The pilot ener-
gizing time is closely related with the pressure frequency and mag-
nitude during the main injection, but a short energizing time could
produce a similar pressure trend to that in the single injection
process.
3.4. Cycle injection quantity
Fig. 6 shows the influences of the pilot injection energizing
pulse on the mass injection quantity and coefficient of variance
(COV) during the two stage injection processes of G0, G20 and
G40. As shown in Fig. 6a, the cycle injection mass by the single
injection is the maximum for a given fuel. For the two stage injec-
tion processes, the cycle injection quantity during the main injec-
tion decreases with extended pilot injection energizing pulse. The
reason is that the extended pilot injection pulse increases the fuel
pressure drop at the injector inlet during the main injection, lead-
ing to decreased instantaneous injection rate during the main
injection, and finally causing the reduced cycle injection quantity.
For any given condition, the test fuels show quite close mass injec-
tion quantities, with the maximum differences below 4%. This
result is consistent with the phenomenon observed in the instanta-
neous injection rate traces for different test fuels, as shown in
Fig. 3. In addition, as shown in Fig. 6b, the COV of main injection
quantities for all test fuels are below 1% with short pilot injection
energizing pulses; however, as the pilot injection energizing pulse
is above 0.5 ms, diesel and gasoline blends have higher COV than
diesel.
4. Conclusions
In this study, the two stage injection process of pure diesel and
two diesel–gasoline blends on a high pressure common rail injec-
tion system was studied. With changed fuel properties and pilot
injection energizing time, the fuel pressure dynamics, cycle injec-
tion rates and quantities of the two stage injection processes are
investigated. Increased gasoline proportion does not significantly
influence the fuel pressure at the injector inlet, the cycle mass
injection rate and injection quantity. However, the pressure wave
of diesel was found to decay faster than diesel and gasoline blends
D. Han et al. / Fuel 159 (2015) 470–475
after injection. For the tested conditions, the extended pilot injec-
tion energizing pulse leads to increased pressure drop at the injec-
tor inlet during the main injection and cause reduced main
injection quantity.
Acknowledgements
The authors would like to thank the support from the National
Natural Science Foundation of China (Grant No. 51306114) and
China Postdoctoral Science Foundation (Grant No. 2013M530195).
References
[1] Lu XC, Han D, Huang Z. Fuel design and management for the control of
advanced compression-ignition combustion modes. Prog Energy Combust
2011;37(6):741–83.
[2] Dec JE. Advanced compression-ignition engines – understanding the in-
cylinder processes. Proc Combust Inst 2009;32:2727–42.
[3] Li T, Moriwaki R, Ogawa H, Kakizaki R, Murase M. Dependence of premixed
low-temperature diesel combustion on fuel ignitability and volatility. Int J
Engine Res 2012;13(1):14–27.
[4] De Ojeda W, Bulicz T, Han X, Zheng M, Cornforth F. Impact of fuel properties on
diesel low temperature combustion. SAE Int J Engines 2011;4(1):188–201.
[5] Han D, Ickes AM, Assanis DN, Huang Z, Bohac SV. Attainment and load
extension of high-efficiency premixed low-temperature combustion with
dieseline in a compression ignition engine. Energy Fuel 2010;24(6):3517–25.
[6] Han D, Ickes AM, Bohac SV, Huang Z, Assanis DN. Premixed low-temperature
combustion of blends of diesel and gasoline in a high speed compression
ignition engine. Proc Combust Inst 2011;33(2):3039–46.
[7] Han D, Ickes AM, Bohac SV, Huang Z, Assanis DN. HC and CO emissions of
premixed low-temperature combustion fueled by blends of diesel and
gasoline. Fuel 2012;99:13–9.
[8] Benajes J, Broatch A, Garcia A, Munoz LM. An experimental investigation of
diesel–gasoline blends effects in a direct-injection compression-ignition
engine operating in PCCI conditions. SAE Technical Paper 2013-01-1676; 2013.
[9] Zhang F, Xu HM, Zhang J, Tian GH, Kalghatgi G. Investigation into light duty
dieseline fuelled partially-premixed compression ignition engine. SAE Int J
Engines 2011;4(1):2124–34.
[10] Valentino G, Corcione FE, Iannuzzi SE. Effects of gasoline–diesel and n-
butanol–diesel blends on performance and emissions of an automotive direct-
injection diesel engine. Int J Engine Res 2012;13(3):199–215.
[11] Weall A, Collings N. Investigation into partially premixed combustion in a
light-duty multi-cylinder diesel engine fuelled with a mixture of gasoline and
diesel. SAE Technical Paper 2007-01-4058; 2007.
[12] Weall A, Collings N. Gasoline fuelled partially premixed compression ignition
in a light duty multi cylinder engine: a study of low load and low speed
operation. SAE Int J Engines 2009;2(1):1574–86.
[13] Payri R, Garcia A, Domenech V, Durrett R, Plazas AH. An experimental study of
gasoline effects on injection rate, momentum flux and spray characteristics
using a common rail diesel injection system. Fuel 2012;97:390–9.
[14] Kim KY, Kim DH, Jung YJ, Bae CS. Spray and combustion characteristics of
gasoline and diesel in a direct injection compression ignition engine. Fuel
2013;109:616–26.
[15] Han D, Wang CH, Duan YZ, Tian ZS, Huang Z. An experimental study of
injection and spray characteristics of diesel and gasoline blends on a common
rail injection system. Energy 2014;75:513–9.
[16] Park SH, Youn IM, Lim Y, Lee CS. Influence of the mixture of gasoline and diesel
fuels on droplet atomization, combustion, and exhaust emission
 D. Han et al. / Fuel 159 (2015) 470–475 475

characteristics in a compression ignition engine. Fuel Process Technol [20] Catania AE, Ferrari A, Manno M, Spessa E. Experimental investigation of
2013;106:392–401. dynamics effects on multiple-injection common rail system performance. J
[17] Ciatti S, Subramanian SN. An experimental investigation of low-octane Eng Gas Turbines Power 2008;130(3):032806.
gasoline in diesel engines. J Eng Gas Turbines Power 2011;133(9):092802. [21] Catania AE, Ferrari A, Manno M. Development and application of a complete
[18] Rezaei SZ, Zhang F, Xu HM, Ghafourian A, Herreros JM, Shuai SJ. Investigation multijet common-rail injection-system mathematical model for
of two-stage split-injection strategies for a Dieseline fuelled PPCI engine. Fuel hydrodynamic analysis and diagnostics. J Eng Gas Turbines Power
2013;107:299–308. 2008;130:062809.
[19] Mendez S, Thirouard B. Using multiple injection strategies in diesel [22] Han D, Duan YZ, Wang CH, Lin H, Huang Z. Experimental study on injection
combustion: potential to improve emissions, noise and fuel economy trade- characteristics of fatty acid esters on a diesel engine common rail system. Fuel
off in low CR engines. SAE Int J Fuel Lubricant 2009;1(1):662–74. 2014;123:19–25.