Anda di halaman 1dari 11

Applied Thermal Engineering 142 (2018) 68–78

Contents lists available at ScienceDirect

Applied Thermal Engineering


journal homepage: www.elsevier.com/locate/apthermeng

Research Paper

Effects of the swirl ratio and injector hole number on the combustion and T
emission characteristics of a light duty diesel engine

Sungjun Yoona, Seungpil Leea, Hyuckmo Kwonb, Joonkyu Leeb, Sungwook Parkc,
a
Graduate School of Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea
b
Powertrain Thermo-Fluid CAE Team, Hyundai Motor Company, 150 Hyundai yeonguso-ro, Hwaseong-si, Gyeonggi 44-706, Republic of Korea
c
School of Mechanical Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea

H I GH L IG H T S

• Effects of swirl ratio and injector hole number on combustion and emissions were studied.
• ACorrelations
single cylinder diesel engine and KIVA code were used.
• PM and CO emissions
between swirl control valve and swirl ratio were calculated.
• were reduced with increases in swirl ratio and hole number.

A R T I C LE I N FO A B S T R A C T

Keywords: In this study, the effects of swirl and the injector hole number were investigated to enhance the mixing of fuel
Swirl ratio and air. Both experimental and analytical methods were employed. The KIVA code coupled with the CHEMKIN
Injector hole number chemistry solver was used to analyze the swirl intensity at various angles of the swirl control valve and ex-
Single cylinder diesel engine perimental results. For the experimental investigation, a 498.75 cc single cylinder engine and a 55 kW AC dy-
Light-duty diesel engine
namometer were employed. The experiments were conducted at IMEP values of 6 and 10 bar with 7, 8, 9, and 10
hole injectors. The swirl intensity was varied by opening of the swirl control valve from 0° to 90°.
The experimental results demonstrated that the swirl intensity has an effect on the combustion and emission
characteristics. An enhanced swirl intensity reduced the PM and CO emissions but increased the NOx emissions
and ISFC. An increase of the injector hole number resulted in decreases of the PM and CO emissions along with a
an increase of NOx emissions and no significant difference of output. However, the exhaust characteristics with
the 10 hole injector deteriorated as the interference increased.

1. Introduction In this study, the effects of swirl and the injector hole number were
investigated to improve the internal flow and the atomization char-
Diesel was once regarded as an environmentally friendly auto- acteristics of the fuel for better mixing [10,11]. Swirl, which is the
mobile fuel because of its low CO2 emissions, which is known as a rotary motion of the air about the axis of the cylinder, can be reinforced
greenhouse gas (GHG) [1]. However, due to high PM and NOx emis- by increasing the flow rate of a pipe by attaching a valve to one of the
sions, many issues have been raised [2]. As a result, regulations re- two intake pipes. Previous studies have shown that increasing the swirl
garding exhaust gas have been continuously strengthened and in recent reduces exhaust emissions due to increased in-cylinder mixing [12–18],
years, exhaust gas regulations for roads have been added. Accordingly, but decreases the power due to increased pumping losses and increased
various studies have been conducted to reduce exhaust emissions to wall heat loss [19–21]. If the injector hole number is increased, the
satisfy these regulations [3,4]. nozzle hole size decreases. As a result, the fuel injected per hole de-
In the direct injection (DI) diesel engine, because fuel is directly creases and atomization characteristics are improved [22–25]. There-
injected into the cylinder at the end of the compression stroke, the time fore, the PM, THC, and CO emissions would be reduced with a higher
for mixing the fuel and air is very short. Thus, increasing the mixture of injector hole number which has smaller hole sizes [14,25–27]. On the
fuel and air improves the combustion and exhaust characteristics [5–9]. other hand, it is known that increasing the hole number causes worse


Corresponding author.
E-mail address: parks@hanyang.ac.kr (S. Park).

https://doi.org/10.1016/j.applthermaleng.2018.06.076
Received 9 March 2018; Received in revised form 21 May 2018; Accepted 25 June 2018
Available online 25 June 2018
1359-4311/ © 2018 Elsevier Ltd. All rights reserved.
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

fuel efficiency [28,29]. However, the distance between the plumes of Table 1
injected fuel decreases and plume interference occurs as the swirl in- Specifications of the test engine and injector.
tensity increases. If plume interference occurs, a locally rich region is Description Specification
generated which adversely affects the exhaust characteristics [30].
Therefore, the purpose of this study was to investigate the influence of Engine Type Single cylinder DI engine
Bore × Stroke (mm) 84 × 90
the swirl and injector hole number on a light duty diesel engine ac-
Displacement volume (cc) 498.75
cording to the load conditions. Compression ratio 16
In this study, experimental and analytical methods were applied Valve type DOHC 4
simultaneously to investigate the influence of swirl and the injector Injector Number of injector holes 8
hole number. The KIVA code was used to analyze the swirl intensity Direct injection (DI) system Common-rail
according to the angle of the swirl control valve with the full 360° mesh Nozzle hole diameter (mm) 0.117
including the intake port and swirl control valve using ICEM-CFD which Hydraulic flow rate (cc at 100 bar, 30 s) 350
is computational grid generating tool and to analyze the experimental
results. Experiments were conducted using a 498.75 cc single cylinder
controlled using NI Compact Rio. The combustion characteristics were
engine. In addition to the conventional 8 hole injector, the effects of the
measured using a Kistler pressure sensor installed in the glow plug
injector hole number were analyzed using 7, 9, and 10 hole injectors.
position and the exhaust was measured using a AVL Smoke Meter and
Horriba Emission Bench. All data was saved using the NI DAQ board
2. Experimental setup and test conditions
(see Fig. 1).
In order to investigate the effect of swirl and the injector hole
In this study, a manual swirl control valve (SCV) was used to control
number on the light-duty diesel engine, experiments were carried out
the swirl intensity. The valve was installed in one of the two pipes of the
under the conditions of IMEP 6 bar, which is the low load, and IMEP
intake port to form a swirl flow and the opening degree was confirmed
10 bar, which is the high load, as shown in Table 2. Conventional
by a protractor installed on the right side. In this study, the intake port
conditions were used for each experiment where the conditions for each
flow was modeled using ICEM CFD which is computational grid gen-
case are shown in Table 2. In order to investigate the effect of the swirl
erating tool and KIVA code to predict the swirl ratio.
intensity, a swirl control valve (SCV) was used and the SCV swing ex-
Simulated EGR was used to simulate the intake flow in a real engine
periment was performed at 10° intervals from 0° to 90°. In addition to
and flow meters were installed for each of the compressed air, N2, and
the conventional injector, 8, 7, 9 and 10 hole injectors were used to
CO2 gases. An air dryer was installed to remove the moisture present in
study the effects of the injector hole number.
the compressed air. The intake composition was calculated assuming an
infinite loop using the fresh air flow rate and fuel injection amount. The
calculation formulas are shown below. 3. Numerical models
O2 Residual
In this study, the numerical analysis was carried out using modified
Z = A∗y + B∗k (1)
KIVA-3V release 2 code coupled with CHEMKIN chemistry [31]. The
x KIVA-3V code calculates the in-cylinder flow, spray behavior, turbulent
Z = A∗y + ∗A∗k
1−x (2) flow, pressure, and temperature of each computational cell. The
CHEMKIN chemistry solver solves the chemical reactions in the com-
x Z1−3.474F ⎞ putational cells and calculates the change of the species’ mole fractions
Z = A∗y + ∗A∗ ⎛
⎜ ⎟

1−x ⎝ C ⎠ (3) and internal energies. To predict the ignition and combustion phe-
nomenon, the CHEMKIN chemistry solver was used and the reduced
Z = A∗y + x ∗ (Z −3.474F ) (4)
normal heptane mechanism was applied for computational cost effi-
x 1−x ciency. Patel et al. [32] developed a reduced chemical reaction me-
Z = A∗y + ∗A∗ ∗ (Z −3.474F ) chanism of normal heptane from an existing n-heptane mechanism to
1−x A (5)
simulate diesel fuel chemistry. It consisted of 29 species and 52 reac-
Z = (A∗y−3.474∗F ∗x )/(1−x ) (6) tions and agreed well with the experiment results. In this paper, to
CO2 and H2O Residuals predict the swirl ratio according to the swirl valve position, a compu-
tational 360° full mesh was generated, including the intake, exhaust
Z = A∗y + B∗k (7) valve and port, and swirl control valve from 0° (opening) to 90°
(closing) in 10° intervals, as shown in Fig. 2. Calculating the swirl ratio
Z = A∗y + x ∗ (Z + 3.111F ) (8)
using a full mesh proceeded from the intake valve opening (IVO) to the
Z = (A∗y + 3.111∗F ∗x )/(1−x ) (9) exhaust valve opening (EVO) timing. When calculating the ignition and
combustion process, for cost and time efficiency, a sector computational
Z = A∗y + x ∗ (Z1 + 3.111F ) (10) mesh was used instead of a full mesh, as shown in Fig. 3, and the si-
mulation proceeded from the intake valve closing (IVC) timing to EVO
A: Fresh air flow (g/stroke) timing.
B: EGR flow (g/stroke) In order to predict the high-pressure direct diesel injection spray
C: A + B break-up, the Kelvin Helmholtz (KH)–Rayleigh Taylor (RT) hybrid
F: Fuel quantity (g) break-up model was used [33]. The KH break-up model predicts the
x: EGR rate (%) break-up of larger blobs into smaller child droplets, which is caused by
y: Mass fractions of O2, H2O, or CO2 in the fresh air (%) the instability created by the sheer force between ambient air and the
k: Mass fractions of O2, H2O, or CO2 in the EGR flow (%) spray referred to as primary breakup. Moreover, the KH break-up model
predicts the normal velocity of the child droplets. Secondary breakup,
A 498.75 cc single cylinder DI engine was used and the engine was called the RT break-up model, is induced by instability at the interface
driven by an AC dynamometer. The detailed specifications of the engine between two different fluids which have a normal acceleration direc-
are summarized in Table 1. A separate pump cart was made to exclude tion.
the effects of lost power for fuel pressurization and the engine was The amount of hydrocarbon (HC) was calculated as the sum of the

69
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

Fig. 1. The single cylinder engine experimental apparatus.

Table 2
Engine test conditions.
Description Case 1 Case 2

Engine speed (rpm) 1500 1750


IMEP (bar) 6 10
Injection pressure (bar) 725 1000
EGR rate (%) 33.8 34.2
Air flow (kg/h) 52.2 95.9
SOE (bTDC deg) 22.2/13.2/2 (2 15.0/2.9/−8.7 (1 Pilot & 1
Pilot) Post)
Injection Quantity (mg/stk) 1.2/1.4/10.9 1.5/20.5/2.4
Injector Hole Number 7, 8, 9, 10
Swirl Valve Angle (deg) 0, 10, 20, 30, 40, 50, 60, 70, 80, 90

total quantities of species that contain both hydrogen and carbon.


Nitrogen oxide (NOx) formation was calculated by the reduced Gas
Research Institute (GRI) NO mechanism, which is composed of 12
chemical reactions and 12 species. A modified 2-step phenomenological Fig. 3. Sector computational mesh employed for validation.
soot formation and oxidation model was used to predict the soot

Fig. 2. Computational 360° full mesh including the intake, exhaust port, and swirl control valve (right top: 0°, bottom: 90°).

70
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

4.0 2.8
0º (Fully-opened)
3.5 20º 40º 60º
2.6 Case 1
90º (Closed)
3.0
Case 2
2.4
2.5
Swirl ratio (-)

2.2

Swirl ratio (-)


2.0
2.0
1.5
1.8
1.0
1.6
0.5
1.4
0.0
0 10
10 20 30 40 50 60
60 70
70 80 90
400 500 600 700 800 1.2
Swirl Valve Angle (deg)
IVO Crank angle (aTDC deg.) EVO
1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58
Fig. 4. Swirl ratio according to the crank angle for various swirl valve positions Swirl Ratio
for case 1.
Fig. 5. Swirl ratio as a function of the swirl control valve position for case 1 and
case 2.
emissions. The soot formation model was suggested by Hiroyasu et al.
[34] and the oxidation model was suggested by Nagle and Strickland-

Incylinder Pressure (MPa)


Constable (NSC) [35]. Han and Reitz [36] suggested a temperature wall 8 Case1, SCV Angle Solid line : Exp. 140
function model to predict the wall heat transfer by convection for 20 deg. Dashed line : Sim.
40 deg. 120
variable density turbulent flow.
6 40 deg.

ROHR (J/deg)
100
70 deg.
4. Results and discussion
90 deg. 80
4.1. Model validation 4
60

4.1.1. Effects of SCV (Swirl control Valve) position on swirl ratio 40


In order to use the swirl ratio as the initial computation input value 2
20
in the sector mesh calculation, combustion, and emission simulation,
the swirl ratio prediction according to the swirl valve position was 0
previously conducted. To predict the swirl ratio according to the swirl 0
valve position, numerical analysis was conducted using KIVA-3V code -50 -40 -30 -20 -10 0 10 20 30 40 50
applying the full mesh, as shown in Fig. 2. Using the operating and Crank angle degree (deg aTDC)
boundary conditions of case 1 and case 2, prediction of the swirl ratio
(a) Case 1
was conducted and Fig. 4 shows the results of the swirl ratio for case 1.
As the swirl valve is closed, the swirl ratio increases because of the
increasing flow velocity and helical port.
Incylinder Pressure (MPa)

8 Solid line : Exp. 140


Using these results, the swirl ratios at IVC timing according to the Dashed line : Sim.
120
swirl control valve position were calculated, as shown in Fig. 5. Al- Case2, SCV Angle
though the operating conditions of case 1 and case 2 are different such 6 20 deg.

ROHR (J/deg)
100
as the engine speed, intake port pressure, and fraction of air, the swirl 40 deg.
ratios are very similar. It can be conjectured based on these results that 40 deg. 80
the swirl ratio is independent of the operating conditions. 70 deg.
4
90 deg. 60

4.1.2. Combustion model validation 40


Fig. 6 shows a comparison of the in-cylinder pressure and heat re- 2
lease rate between the experimental results and simulations according 20
to the swirl control valve position for case 1 and case 2. The solid lines
0
indicate the experiment results and the dashed lines correspond to the 0
simulation results. In both cases, the pressure obtained from the si- -50 -40 -30 -20 -10 0 10 20 30 40 50
mulation is slightly higher. However, the initiation timing is slightly Crank angle degree (deg aTDC)
different for each case. In case 1, the pilot ignition timing is slightly
retarded and the main ignition timing is advanced whereas in case 2, (b) Case 2
the ignition timing and combustion phasing agree well with the ex-
Fig. 6. Model validation for the in-cylinder pressure and heat release rate for
periment results. However, these slight discrepancies are reasonable to
with the 8 hole injector for case 1 and 2.
predict the combustion and emissions characteristics.
In Fig. 7, the predicted emissions (e.g., soot, NOx, CO, and THC) are
compared to the experimental results according to the swirl valve po- NOx prediction is also agree well with the experiment results. Although,
sition for case 1 and case 2. As shown in Fig. 7, in spite of the com- it seems that there is a discrepancy in quantitative prediction, it is
plexity of the soot formation and oxidation process, the soot quantity because of the scaling of the graph. It shows more clearly the change of
and trend are very similar to the experimental results. In addition, the the emissions’ values for various swirl control valve angle. And the

71
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

0 10 20 30 40 50 60 70 80 90 30
1.0 MFB 50 - 90
Case 1 Symbol : Exp. 29
0.8 MFB 10 - 50
Case 2 MFB 90

Soot (g/kW-hr)
Symbol + line : Sim.

MFB 10, 50, and 90 (deg)


28
0.6
27
0.4
0.2 26

1.2 0.0 25

24 MFB 50
NOx (g/kW-hr)

1.0
10
0.8 5
MFB 10
0.6 0
0 10 20 30 40 50 60 70 80 90
-5
0.4 Swirl Valve Angle (deg)
2.0 1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58
Swirl Ratio
1.5

THC (g/kW-hr)
Fig. 8. MFB 10, 50, and 90 and the combustion duration.
1.0
700
0.5

Accumulated Heat Release Rate (J)


600 SCV Angle
15 0.0
20 deg
500
10 50 deg
CO (g/kW-hr)

5 400 70 deg Increasing SCV Valve


90 deg
0 300

200
0 10 20 30 40 50 60 70 80 90
Swirl Valve Angle (deg) 100

Fig. 7. Model validation for emissions (soot, NOx, THC, and CO) for case 1 and 0
case 2 with the 8 hole injector.
-100
-20 -10 0 10 20 30 40 50 60
discrepancy of the emissions can be seen as the reasonable error. Fig. 7
Crank Angle Degree (aTDC deg)
also shows comparisons of the THC and CO emissions between the
experiments and simulation results. The simulation results of THC and Fig. 9. Accumulated heat release rate for case 1 with the 9 hole injector.
CO of case 1 were slightly different in terms of quantity. However,
because the measured values were very low and the range of the figure also shows the same result. The accumulated heat release rate (ACHR)
is scaled and a lot of assumption to simplify the calculation, it is in- is almost the same at 0–10°, which corresponds to premixed combustion
evitable that the simulation results could not match perfectly the ex- but thereafter, the slope becomes steeper with increasing swirl in-
periment results. Thus it is demonstrated that the emission model is in tensity. Since diffusion combustion is controlled by air-fuel mixing,
good agreement with the experiments. combustion is promoted by the increasing swirl and the maximum
value of ACHR is also increased.
4.2. Effects of swirl ratio and injector hole number on combustion From the above results, it can be expected that the output will also
characteristics and fuel consumption decrease with increasing swirl, which can be observed in the indicated
specific fuel consumption (ISFC) values shown in Fig. 10. Fig. 10 shows
As shown in Fig. 6, the swirl intensity does not significantly affect the indicated specific fuel consumption (ISFC) results for case 1 for all
the in-cylinder pressure. While the combustion phase is almost the injectors. The black line represents the average value of the four in-
same, even if the swirl is different, the maximum pressure of the jectors. According to the results of Fig. 10, the ISFC values do not differ
combustion of the main injection slightly increases with increasing greatly depending on the hole number, but they increase with in-
swirl intensity. creasing swirl. It is expected that this result stems from an increase of
In Fig. 8, the top of the bar graph shows the Mass Fraction Burned wall heat loss and an increase of pumping loss due to the increase of
(MFB) 90 angle, the middle is the MFB 50 angle, and the bottom is the internal flow.
MFB 10 angle. MFB 10, 50, and 90 represent the crank angles corre- Fig. 11 shows the accumulated heat release rate (AHRR) and wall
sponding to 10%, 50%, and 90% of cumulative heat generation, re- heat transfer loss for case 1 and case 2 using the KIVA-3V code ac-
spectively. In this study, MFB10 is defined as the combustion start point cording to the swirl valve angle. The black bar corresponds to the AHRR
and MFB90 is defined as the combustion end point. Therefore, the and red bar is the wall heat transfer loss. The values on the bar refer to
difference between MFB 10 and MFB 90 represents the combustion the percentage of the wall heat transfer loss in the AHRR. As the swirl
duration. As shown in Fig. 8, MFB 10 and 50 are similar despite swirl valve closes, the swirl ratio increases and both the total heat release and
changes, but the combustion ending point, MFB 90, advances by ap- the percentage of wall heat transfer loss increase slightly. A high in-
proximately 4° and thus, the combustion duration decreases. This in- tensity of the swirl ratio improves the air-fuel mixing, which makes the
dicates that the swirl strength does not significantly affect premixed combustion more active. On the other hand, active combustion causes
combustion but instead has an effect on the diffusion combustion. Fig. 9 an increase of the in-cylinder temperature and thus, a slight increase of

72
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

178 3.3
7 Holes 9 Holes Mean Value 9 Hole Inj Case 1
176 3.2
8 Holes 10 Holes Case 2
174 3.1

Heat Loss (kW)


172 3.0
ISFC (g/kW-hr)

170

168
ISFCnet=5.82
1.6

166
1.4
164
0 10 20 30 40 50 60 70 80 90
162 1.2
Swirl Valve Angle (deg) 0 10 20 30 40 50 60 70 80 90
1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58 Swirl Valve Angle (deg)
Swirl Ratio Fig. 12. Heat Loss using the 9 hole injector for case 1 and case 2.
Fig. 10. Indicated specific fuel consumption with 7, 8, 9, and 10 hole injectors
for case 1. the wall heat transfer loss.
Since the heat loss cannot be directly obtained through the experi-
660 ment, the heat loss was obtained by using the energy balance in this
Total heat release Case 1 study. The energy of the injected fuel is used as indicated power, ex-
640 Wall heat loss
haust gas enthalpy flux due to incomplete combustion, exhaust gas
620 enthalpy flux, heat transfer rate to wall. In order to consider the in-
Accumulated heat (J)

600 complete combustion, the energy of the calculated fuel was multiplied
150 by the combustion efficiency. Indicated power is easily calculated and
the exhaust gas enthalpy flux is obtained by referring to the enthalpy
140 table of the diesel engine exhaust gas temperature as shown in Perry's
Chemical Engineers' Handbook by Perry, R.H. The heat loss value thus
130 obtained is shown in Fig. 12. As a result of numerical analysis, it was
20.57 20.58 20.57
20.39 20.49 found that heat loss increases with increasing swirl intensity [37,38].
120 20.02 19.99 20.09 20.24 Fig. 13 shows the pumping mean effective pressure and fuel con-
19.9
version efficiency to analyze the pumping loss and output. Each solid
110
0 10 20 30 40 50 60 70 80 90 line represents a trend line of the pumping mean effective pressure
100 (PMEP) and fuel conversion efficiency. PMEP is the difference between
Swirl Valve Angle (deg) the gross IMEP and net IMEP, which increases by about 0.015 bar as the
1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58 swirl valve opening increases. The fuel conversion efficiency is a mea-
Swirl ratio sure of the efficiency of the injected fuel in terms of the change of heat.
It decreases by 0.8% as the swirl strength increases. In conclusion, the
(a) Case 1 output decreases slightly due to pumping losses and the heat loss in-
creases with increasing swirl strength. Changing the number of holes
1200
Total heat release does not have a significant effect on the output.
Case 2
Wall heat loss
1100
Accumulated heat (J)

1000
0.36
PMEP 51.2
0.35 Fuel Converion
Fuel Conversion Efficiency (%)

210
Efficiency 51.0

0.34
18.08 18.14
PMEP (bar)

200 18.06 18 50.8


17.98 17.97
17.86 17.83 17.83 17.91
0.33
50.6
190
0.32
0 10 20 30 40 50 60 70 80 90 50.4
180
Swirl Valve Angle (deg) 0.31 50.2
1.42 1.47 1.50 1.57 1.77 2.08 2.32 2.53 2.57 2.60 0 10 20 30 40 50 60 70 80 90
Swirl ratio 0.30 50.0
Swirl Valve Angle (deg)
(b) Case 2 1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58
Swirl Ratio
Fig. 11. Accumulated heat release rate and wall heat loss at different swirl
valve angles for (a) case 1 and (b) case 2. Fig. 13. The pumping mean effective pressure and fuel conversion efficiency
for case 1 with the 9 hole injector at various swirl valve angles.

73
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

1.0 0.8
7 Holes SCV=90º
Particulate Matter (g/kW-hr)

0.8 8 Holes SCV=20º


9 Holes 0.6

0.6
10 Holes

Soot (g/kW-hr)
0.4
0.4

0.2
0.2 SCV = 20 deg. (swirl ratio = 1.48)
SCV = 40 deg. (1.76)
SCV = 60 deg. (2.30)
0.0 0.0 SCV = 70 deg. (2.51)
0 10 20 30 40 50 60 70 80 90 SCV = 90 deg. (2.58)
Swirl Valve Angle (deg) 0 20 40 60 80 100 120
1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58 Crank angle (deg aTDC)
Swirl Ratio
(a) Case 1
(a) Case 1
0.8
1.0 SCV=90º
7 Holes SCV=20º
0.6
0.8 8 Holes
Particulate Matter (g/kW-hr)

9 Holes Soot (g/kW-hr)


0.6 10 Holes 0.4

0.4
0.2
SCV=20 deg. (swirl ratio = 1.50)
0.2 SCV=40 deg. (1.77)
SCV=60 deg. (2.32)
0.0 SCV=70 deg. (2.53)
0.0 SCV=90 deg. (2.60)
0 10 20 30 40 50 60 70 80 90
0 20 40 60 80 100 120
Swirl Valve Angle (deg)
Crank angle (deg aTDC)
1.42 1.47 1.50 1.57 1.77 2.08 2.32 2.53 2.57 2.60
Swirl Ratio (b) Case 2
(b) Case 2 Fig. 15. Soot formation and oxidation results for (a) cases 1 and (b) case 2 with
the 8 hole injector.
Fig. 14. Particulate matter emissions for (a) case 1 and (b) case 2 with 7, 8, 9,
and 10 hole injectors.
result, there are no big differences between the 9 hole injector and 10
hole injector. In case 2, as the injection amount increases, the effect of
4.3. Effects of swirl ratio and injector hole number on exhaust emissions
interplume interference increases and the differences of PM emissions
between the 9 hole and 10 hole injectors increase.
In Fig. 14, the influences of the swirl intensity and injector hole
The relationship between the swirl intensity and soot formation and
number were investigated. Case 1 is the low load condition with IMEP
oxidation is shown in Fig. 15. Fig. 15 shows the soot histories of cases 1
6 bar at 1500 rpm and case 2 is the high load case with IMEP 10 bar at
and 2 calculated by KIVA-3V code according to the crank angle. As
1750 rpm. Comparing case 1 and case 2, the PM emissions decreased as
shown in Fig. 15, the soot is reduced when the swirl intensity is in-
the load increased. This is probably due to post injection in case 2
creased. The soot formation rate and quantities are nearly the same at
unlike case 1, in which only the pilot injects twice without post injec-
various swirl intensities. However, the soot oxidation rate and quan-
tion. Moreover, the high EGR rate of case 1 is also the reason for high
tities are different depending on the swirl intensity. When the swirl
PM emissions. In both cases, as the swirl strength increases, the PM
intensity is increased, the in-cylinder flow is strengthened and it im-
emissions decrease because the improved air-fuel mixing causes an
proves the air-fuel mixing rate to reduce the local equivalence ratio.
increase of the oxygen utilization rate and thus, the PM production
Thus, the soot oxidation rate is increased by improved internal flow.
decreases and the oxidation amount increases. The PM emissions de-
Based on these results, we can conclude that the swirl intensity affects
creased with increasing number of holes up to 9 holes and the PM
the soot oxidation rate more than the formation rate.
emissions of the 10 hole injector were slightly higher for case 1 com-
Fig. 16 shows the NOx emissions for case 1 and case 2. The NOx
pared to 9 holes in addition to being higher at all SCV angles for case 2.
emissions of case 1 were lower than those of case 2 because of the high
This is because plume-to-plume interference occurs where the inter-
EGR rate. The NOx emissions increased with increasing swirl intensity
hole distance decreases with increasing injector hole and the injected
in both cases. In particular, case 2, which is a high load condition, was
fuel is refracted as the swirl increases, creating a locally rich region in
more influenced by swirl than case 1. Unlike the PM emissions, the NOx
contact with the lateral plume. In case 1, the injection amount is low
emissions were not influenced by the injector hole number and there
and thus, the effect of interplume interference is relatively low. As a
were no significant differences.

74
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

3.5
0.55 7 Holes
7 Holes
8 Holes 3.0 8 Holes
0.50 9 Holes

Carbon Monoxide (g/kW-hr)


9 Holes
10 Holes 2.5
10 Holes
NOx (g/kW-hr)

0.45

2.0
0.40

1.5
0.35

1.0
0.30

0.25 0.5
0 10 20 30 40 50 60 70 80 90
0 10 20 30 40 50 60 70 80 90
Swirl Valve Angle (deg) 0.0
Swirl Valve Angle (deg)
1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58
1.39 1.44 1.48 1.56 1.76 2.06 2.30 2.51 2.55 2.58
Swirl Ratio Swirl Ratio
(a) Case 1
(a) Case 1
1.15

1.10
7 Holes 3.5
8 Holes
1.05 3.0
9 Holes

Carbon Monoxide (g/kW-hr)


10 Holes
NOx (g/kW-hr)

1.00 2.5

0.95 7 Holes
2.0
8 Holes
0.90
1.5 9 Holes
0.85 10 Holes
0.80 1.0

0.75 0 10 20 30 40 50 60 70 80 90 0.5
Swirl Valve Angle (deg) 0 10 20 30 40 50 60 70 80 90
0.0
1.42 1.47 1.50 1.57 1.77 2.08 2.32 2.53 2.57 2.60 Swirl Valve Angle (deg)
Swirl Ratio 1.42 1.47 1.50 1.57 1.77 2.08 2.32 2.53 2.57 2.60
(b) Case 2 Swirl Ratio

Fig. 16. NOx emissions for (a) case 1 and (b) case 2 with various injector hole (b) Case 2
numbers.
Fig. 17. Carbon monoxide emissions for (a) case 1 and (b) case 2 with various
injector hole numbers.
Fig. 17 shows the carbon monoxide emissions for case 1 and case 2.
The CO emissions were strongly influenced by the swirl intensity and
injector hole number. The CO emissions decreased significantly as the of the swirl flow because the high swirl flow affects the axial direction
oxygen utilization rate increased with increasing swirl intensity and momentum of the spray. The high swirl flow disturbs the spray of the
decreased with increasing number of injector holes. As evidenced by axial direction and the radial direction momentum is increased. In
the PM emissions, spray interference in case 2 was more severe than in particular, the z-plane distributions at the C.A 10 ATDC degree show
case 1. Thus, the CO emissions with 10 holes were higher than the CO obvious results in which the high swirl flow enhanced the radial mo-
emissions obtained with 9 holes. Unlike case 1, in which the CO mentum of the spray plume, shortened the spray penetration, and led to
emissions gradually decreased with increasing injector holes, there good atomization of the fuel in addition to fuel-air mixing. As a result,
were no significant differences in the results of 8, 9, and 10 hole in- the fuel concentration in the bowl is reduced and the locally fuel rich
jectors for case 2, whereas the 7 hole injector shows very large emis- region is also reduced. Also, the reduced soot as SCV closes can be
sions. THC emissions were expected to be similar to the CO emissions explained by the above reasons.
but the emissions were so small that only very small amounts were As shown in Fig. 18, the equivalence ratio shows the spray plum
emitted over the whole range without being affected by swirl and the behavior. Thus it can be observed the plum interference possibility.
number of injector holes. When the SCV angle is increased the spray plum is bended more, for the
Fig. 18 shows the equivalence ratio distribution of the in-cylinder reasons as mentioned above, and possibility of interference is slightly
according to the swirl valve position for case 2 with 8 hole injector increased. However the direct fuel interference is not observed even
based on the numerical results. As shown in Fig. 18, when the swirl though the SCV angle is increased in 9 hole injector. The possibility of
valve is closed, the fuel spray penetration is shortened and the dis- interference is increased when injector hole is increased to 10. Because
tribution of the fuel is widened at the same crank angle. At the C.A 10 the injector hole is increased, the space of each spray plum is shorten.
ATDC degree, the spray plume is divided into two regions, which are And the combustion and emission results are deteriorated by plum in-
the bowl and squish at a SCV of 20°. When the SCV is closed, the spray teraction as shown in Figs. 14 and 17.
plume is shortened and it reached the bowl tip due to the high intensity Fig. 19 shows the comparison of the CO distribution according to

75
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

SCV = 20° 40° 60° 90°

Crank angle =
6deg., ATDC

Crank angle =
10deg., ATDC

Crank angle =
35deg., ATDC

Fig. 18. Comparison of the equivalence ratio according to the swirl valve position for case 2 with 8 hole injector.

the swirl valve position for case 2 with 8 hole injector at EVO. As shown high in-cylinder temperature distribution. As the SCV is closed, the
in Fig. 19, the CO distribution is reduced when the SCV closes. At an mean temperature increased and combustion is more active by high
SCV of 20°, high CO emissions are distributed in various regions such as intensity of swirl flow. The high intensity of swirl flow induced the
in the bowl, center of the cylinder, and near the piston head. However, homogeneous mixing of air-fuel mixture and rapid dispersion of spray
as the swirl ratio is increased, the CO concentration decreases. In par- plum in in-cylinder. And it leads to decrease combustion duration,
ticular, the concentrations near the center of the cylinder and near the which increases in-cylinder temperature. Thus when SCV is increased,
bowl were dramatically reduced by the high swirl flow. As mentioned the NOx emission concentration is increased. At the C.A 35 ATDC de-
above, the high swirl flow enhances the mixing of fuel and air and the gree, the regions that experience high temperatures increase when the
combustion phase, reduces the locally rich regions, and leads to a re- SCV is closed and the distribution of the high temperature in the central
duction of the possibility of incomplete combustion, which is the cause region of the cylinder also increased. As a results, the high intensity of
of the formation of the CO. In spite of this advantage, the high swirl swirl ratio increase the combustion speed and NOx formation.
flow can cause interference of the combustion flame. When SCV was
increased above 60°, the soot and CO emissions slightly increased due 5. Conclusions
to the combustion flame interference, as shown in Fig. 18. Fig. 20 shows
a comparison of the in-cylinder temperature distribution for case 2 with In this study, numerical and the experimental methods were used to
8 hole injector. As shown in Fig. 20, the temperature distribution show study the effects of swirl and the number of injector holes in a light-duty
the evidence of NOx generation that the increasing SCV angle leads to diesel engine. 7, 8, 9, and 10 hole injectors were compared and the SCV

CO SCV = 20° 40° 60° 90°

Crank angle
= 126°,
ATDC
(EVO)

Fig. 19. Comparison of the CO emissions according to the swirl valve position for case 2 with 8 hole injector at EVO.

76
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

SCV = 20° 40° 60° 90°

Crank angle =
6deg., ATDC

Crank angle =
10deg., ATDC

Crank angle =
15deg., ATDC

Crank angle =
20deg., ATDC

Crank angle =
35deg., ATDC

Fig. 20. Comparison of the temperature distribution according to the swirl valve position for case 2 with 8 hole injector.

angle was swung from 0 to 90°. The results can be summarized as fol- resource from the Ministry of Trade, Industry and Energy, South Korea
lows. (No. 20164010200860).

– The increase of the swirl intensity caused a decrease of the com- References
bustion duration. Premixed combustion remains constant, but the
increase in mixing due to swirl accelerates the diffusion combustion, [1] W. Knecht, Diesel engine development in view of reduced emission standards,
leading to advanced combustion end timing. Energy 33 (2008) 264–271.
[2] S. Lee, S. Park, Optimization of the piston bowl geometry and the operating con-
– As the swirl intensity was increased, the ISFC value increased by an ditions of a gasoline-diesel dual-fuel engine based on a compression ignition engine,
average of 5.82 because the heat loss to the cylinder wall and the Energy 121 (2017) 433–448.
pumping loss increased. On the other hand, the number of injector [3] J. Cha, S. Yoon, S. Lee, S. Park, Effects of intake oxygen mole fraction on the near-
stoichiometric combustion and emission characteristics of a CI (compression igni-
holes did not significantly affect the fuel efficiency. tion) engine, Energy 80 (2015) 677–686.
– The increase of swirl and number of injector holes caused a decrease [4] S. Yoon, H. Kim, D. Kim, S. Park, Effect of the fuel injection strategy on diesel
of PM and CO emissions and an increase of NOx emissions. The THC particulate filter regeneration in a single-cylinder diesel engine, J. Eng. Gas
Turbines Power 138 (2016) 102810–102810.
emissions were too small to be affected. However, in the 10-hole [5] X.-R. Li, W. Yang, L.-W. Su, F.-S. Liu, Mixing and combustion mechanisms within
injector, inter-hole interference occurred and thus the opposite lateral swirl combustion system (LSCS) in a DI diesel engine, Appl. Therm. Eng. 123
tendency was shown. On the other hand, no inter-hall interference (2017) 7–18.
[6] X. Li, Z. Qiao, L. Su, X. Li, F. Liu, The combustion and emission characteristics of a
due to increased swirl intensity was found. This is because the boar
multi-swirl combustion system in a DI diesel engine, Appl. Therm. Eng. 115 (2017)
size is very small compared to the penetration in the light-duty 1203–1212.
engine. [7] T. Kim, S. Park, Effects of spray patterns on the mixture formation process in multi-
– According to the numerical analysis results, the PM production hole-type direct injection spark ignition (DISI) gasoline engines, Atomiz. Sprays 26
(2016) 1151–1175.
amount does not affect the increase of the swirl intensity, but the PM [8] T. Kim, J. Song, S. Park, Effects of turbulence enhancement on combustion process
oxidation amount increases relatively with the intensification of the using a double injection strategy in direct-injection spark-ignition (DISI) gasoline
flow inside the cylinder. engines, Int. J. Heat Fluid Flow 56 (2015) 124–136.
[9] J. Song, T. Kim, J. Jang, S. Park, Effects of the injection strategy on the mixture
formation and combustion characteristics in a DISI (direct injection spark ignition)
Acknowledgments optical engine, Energy 93 (2015) 1758–1768.
[10] C. Espey, J.A. Pinson, T.A. Litzinger, Swirl Effects on Mixing and Flame Evolution in
a Research DI Diesel Engine, SAE Paper 902076, 1990.
This work was supported by Hyundai Motor Company and “Human [11] H. Hiroyasu, K. Nishida, M. Arai, S. Yoshikawa, A Visual Study of D.I. Diesel
Resources Program in Energy Technology” of the Korea Institute of Combustion from the Under and Lateral Sides of an Engine, SAE Paper 861182,
Energy Technology Evaluation and Planning (KETEP), granted financial 1986.

77
S. Yoon et al. Applied Thermal Engineering 142 (2018) 68–78

[12] H. Dembinski, H.-E. Angstrom, Swirl and Injection Pressure Impact on After- pressure and micro-hole nozzle on flame structure and soot formation of impinging
Oxidation in Diesel Combustion, Examined with Simultaneous Combustion Image diesel spray, Appl. Energy 88 (2011) 1620–1628.
Velocimetry and Two Colour Optical Method, SAE Paper 2013-01-0913, 2013. [26] C. Sayin, M. Gumus, M. Canakci, Influence of injector hole number on the perfor-
[13] J.J. Lopez, J. Martin, A. Garcia, D. Villalta, A. Warey, V. Domenech, mance and emissions of a DI diesel engine fueled with biodiesel–diesel fuel blends,
Characterization of In-Cylinder Soot Oxidation Using Two-Color Pyrometry in a Appl. Therm. Eng. 61 (2013) 121–128.
Production Light-Duty Diesel Engine, SAE Paper 2016-01-0735, 2016. [27] T.R. Fuchs, C.J. Rutland, Intake Flow Effects on Combustion and Emissions in a
[14] H.-W. Ge, Y. Shi, R.D. Reitz, D.D. Wickman, G. Zhu, H. Zhang, Y. Kalish, Heavy- Diesel Engine, SAE Paper 980508, 1998.
Duty Diesel Combustion Optimization Using Multi-Objective Genetic Algorithm and [28] B.-S. Kim, W.H. Yoon, S.H. Ryu, J.S. Ha, Effect of the Injector Nozzle Hole Diameter
Multi-Dimensional Modeling, SAE Paper 2009-01-0716, 2009. and Number on the Spray Characteristics and the Combustion Performance in
[15] Y. Sun, K. Sun, Z. Lu, T. Wang, M. Jia, Selection of Swirl Ratio in Diesel Engines Medium-Speed Diesel Marine Engines, SAE Paper 2005-01-3853, 2005.
Based on Droplet Trajectory Analysis, SAE Paper 2017-01-0813, 2017. [29] F. Millo, E. Pautasso, D. Delneri, M. Troberg, A DoE Analysis on the Effects of
[16] M.E. McCracken, J. Abraham, Swirl-Spray Interactions in a Diesel Engine, SAE Compression Ratio, Injection Timing, Injector Nozzle Hole Size and Number on
Paper 2001-01-0996, 2001. Performance and Emissions in a Diesel Marine Engine, SAE Paper 2007-01-0670,
[17] S. Shundoh, T. Kakegawa, K. Tsujimura, S. Kobayashi, The Effect of Injection 2007.
Parameters and Swirl on Diesel Combustion with High Pressure Fuel Injection, SAE [30] J. Benajes, S. Molina, K.D. Rudder, D. Maroteaux, H.B.H. Hamouda, The use of
Paper 970627, 1991. micro-orifice nozzles and swirl in a small HSDI engine operating at a late split-
[18] K.Y. Kang, R.D. Reitz, The effect of intake valve alignment on swirl generation in a injection LTC regime, Proc. Instit. Mech. Eng., Part D: J. Automobile Eng. 220
DI diesel engine, Exp. Therm Fluid Sci. 20 (1999) 94–103. (2006) 1807–1817.
[19] J. Benajes, J. Martín, A. García, D. Villalta, A. Warey, Swirl ratio and post injection [31] A.A. Amsden, KIVA-3V, Release 2: Improvements to KIVA-3V, in Other Information:
strategies to improve late cycle diffusion combustion in a light-duty diesel engine, PBD: 1 May 1999, 1999, pp. Medium: P; Size: vp.
Appl. Therm. Eng. 123 (2017) 365–376. [32] A. Patel, S.-C. Kong, R.D. Reitz, Development and Validation of a Reduced Reaction
[20] P. Olmeda, J. Martin, A. Garcia, D. Villalta, A. Warey, V. Domenech, A Combination Mechanism for HCCI Engine Simulations, SAE Paper 2004-01-0558, 2004.
of Swirl Ratio and Injection Strategy to Increase Engine Efficiency, SAE Paper 2017- [33] M.A. Patterson, R.D. Reitz, Modeling the Effects of Fuel Spray Characteristics on
01-0722, 2017. Diesel Engine Combustion and Emission, SAE Paper 980131, 1998.
[21] P.C. Miles, The Influence of Swirl on HSDI Diesel Combustion at Moderate Speed [34] H. Hiroyasu, T. Kadota, M. Arai, Development and use of a spray combustion
and Load, SAE Paper 2000-01-1829, 2000. modeling to predict diesel engine efficiency and pollutant emissions: part 2 com-
[22] S. Lahane, K.A. Subramanian, Impact of nozzle holes configuration on fuel spray, putational procedure and parametric study, Bull. JSME 26 (1983) 576–583.
wall impingement and NOx emission of a diesel engine for biodiesel–diesel blend [35] J.R. Walls, R.F. Strickland-Constable, Oxidation of carbon between 1000–2400°C,
(B20), Appl. Therm. Eng. 64 (2014) 307–314. Carbon 1 (1964) 333–338.
[23] S. Kang, W. Cho, C. Bae, Y. Kim, Influence of the Injector Geometry at 250 MPa [36] Z. Han, R.D. Reitz, A temperature wall function formulation for variable-density
Injection in a Light-Duty Diesel Engine, SAE Paper 2017-01-0693, 2017. turbulent flows with application to engine convective heat transfer modeling, Int. J.
[24] N. Ladommatos, R.A. Balian, R. Stone, Analysis of Swirl in Unsteady Flow and its Heat Mass Transf. 40 (1997) 613–625.
Effect on Diesel Combustion, SAE Paper 921643, 1992. [37] J.B. Heywood, Internal Combustion Engine Fundamentals, McGRAW-HILL, 1988.
[25] X. Wang, Z. Huang, W. Zhang, O.A. Kuti, K. Nishida, Effects of ultra-high injection [38] R.H. Perry, Perry's Chemical Engineers' Handbook, McGRAW-HILL, 1984.

78

Anda mungkin juga menyukai