Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
An Open Cycle Simulation of DI Diesel Engine Flow
Field Effect on Spray Processes
2012-01-0696
Published
04/16/2012
Kohei Fukuda, Abbas Ghasemi, Ronald Barron and Ram Balachandar
University of Windsor
Copyright © 2012 SAE International
doi:10.4271/2012-01-0696
ABSTRACT
Clean diesel engines are one of the fuel efficient and low
emission engines of interest in the automotive industry. The
combustion chamber flow field and its effect on fuel spray
characteristics plays an important role in improving the
efficiency and reducing the pollutant emission in a direct
injection diesel engine, in terms of influencing processes of
breakup, evaporation mixture formation, ignition, combustion
and pollutant formation. Ultra-high injection pressure fuel
sprays have benefits in jet atomization, penetration and air
entrainment, which promote better fuel-air mixture and
combustion. CFD modeling is a valuable tool to acquire
detailed information about these important processes. In this
research, the characteristics of ultra-high injection pressure
diesel fuel sprays are simulated and validated in a quiescent
constant volume chamber. A profile function is utilized in
order to apply variable velocity and mass flow rate at the
nozzle exit. The CFD model is also applied to an open cycle
engine model to study the effects of engine flow field features
such as swirl and tumble motions on the spray behavior. In
particular, the effect of the above mentioned parameters on
spray penetration Sauter mean diameter (SMD) and fuel
distribution in the chamber are extensively discussed.
INTRODUCTION
Demands on clean diesel engines are continuously growing in
the automotive sector. One of the solutions advanced in diesel
engine applications is to improve fuel droplet atomization and
air-fuel mixing. The development of high pressure injectors
will result in finer atomization and provide for better air-fuel
mixture [1]. Recently, the injection pressure of the injection
systems has reached to 300 MPa or above in these
applications and a growing number of researchers have
shown interest in the performance of the ultra-high injectors.
Kato et al. [2] and Yokota et al. [3] have reported on
experiments in the ultra-high injection system at the early
stage of the research. They examined the effects of the
injection pressure, ranging from 55 to 250 MPa, and also the
variations of nozzle orifice and injection duration. From their
studies they concluded that the Sauter mean diameter is
correlated with the average injection pressure and the
transition of the injection pressure by time. Moreover, a
shorter combustion process and reduced soot formation are
realized by utilizing ultra-high injection pressure and smaller
orifice diameter.
Nishida's research group of University of Hiroshima has
conducted numerous experiments using various ultra-high
injection pressures, micro-hole nozzles, spray wallimpingement, and diesel and alternative diesel fuels
[4,5,6,7,8,9,10]. The combination of 300 MPa injection
pressure and 0.08 mm nozzle-hole diameter reportedly gave
the best performance in terms of turbulent mixture rate and
droplet size reduction to decrease the mixture process and
lean mixture formation.
Lee et al. [11] experimentally and numerically investigated
free sprays at ultra-high injection pressure in the range of 150
to 355 MPa. The limitation of Sauter mean diameter after
reaching an injection pressure of 300 MPa and the reduced
growth rate of the penetration length are reported.
Tao and Bergstrand [12] studied the effect of ultra-high
injection pressures on engine ignition and combustion using
three-dimensional numerical simulations. The advantage of
high pressure injection in producing reduced ignition delay,
short combustion phase, and fast flame propagation is
reported. Additionally, three different rates of injection
profiles were examined. Rate falling injection results were
found to shorten at the early stage of combustion and expand
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
at the late stage, and rate rising injection performs inversely.
On the other hand, rate rising injection estimates a wider
flame area at high temperature and reduces NO formation due
to faster cooling after combustion. Flame lift-off lengths were
observed to be constant at different injection pressures, in
contrast to the case of injection in a constant volume
chamber.
To study the characteristics of ultra-high injection pressure
sprays numerically, it is important to understand the effect of
spray modeling. Comprehensive reviews of droplet
phenomena have been presented by Lin and Reitz [13] and
Jiang et al. [14]. The difference between the popular breakup
models has been discussed by Djavareshkian and Ghasemi
[15] and Hossainpour and Binesh [16]. They reported on
implementation of WAVE and KH-RT models, and found
better agreement with experimental data using KH-RT. The
interaction of the mesh, turbulence model and spray has been
studied by Karrholm and Nordin [17] in a constant volume
chamber.
The effect of spray-in-cylinder-flow interaction is realized in
the combustion process [18] and studied by a number of
researchers. Choi et al. [19] found that the flow pattern
around the jet is similar at different injection pressures, but
strong flow recirculation is observed at higher injection
pressure. Spray characteristics in crossflow was studied by
Desantes et al. [20], McCracken and Abraham [21], and Park
et al. [22] to observe the effect on particle size and mixing
process. Correlation of penetration and dispersion of gas jet
and sprays were examined by Iyer and Abraham [23]. The
effects of gas density and vaporization on penetration,
injection condition and dispersion of spray are discussed by
Naber and Siebers [24], Kennaird et al. [25], and Post et al.
[26]. Jagus et al. [27] assessed injection and mixing using
LES turbulence modeling.
Not many numerical studies have been conducted with ultrahigh injection pressures. The objective of this research is to
take advantage of numerical simulation to investigate the
effect of high pressure injection on atomization and fuel
mixing with in-cylinder flow. As a first step, the numerical
setup is optimized and validation of the spray model is
achieved in the constant volume vessel with different grid
sizes. The experimental setup and data for the validation are
acquired from Wang et al [7]. In the next stage, an engine
model with vertical ports is meshed and the flow structure is
verified. Finally, the spray models are introduced into the
engine simulation with three different injection pressures and
two inlet pressure cases. The spray characteristics are
investigated and the interaction of the in-cylinder flow with
the sprays is studied.
NUMERICAL METHODOLOGY
MESHING
An engine model to be simulated in ANSYS FLUENT 13.0
[28] is meshed with a hybrid topology (half-model shown in
Fig. 1). The flow domain is divided into four major zones:
chamber, ports, piston layer valve layer. The zones adjacent
to reciprocating boundaries such as the piston and valves are
meshed with quadrilateral cells (structured mesh).
Tetrahedral cells (unstructured mesh) are used in the chamber
zone because the valves move into this zone and its cells
deform and must subsequently be remeshed. Interfaces must
be created between the chamber and valve layer zones to
transfer nodal values from one side to the other.
Figure 1. Engine geometry with vertical ports,
illustrating the mesh at BDC
SPRAY AND BREAKUP MODELS
In CFD, spray mechanisms are represented by mathematical
models. Two approaches, Euler-Lagrange and Euler-Euler,
are used in multiphase flows. In both of these approaches, the
fluid phase is regarded as a continuum and modeled by the
Navier-Stokes equations. For the Euler-Lagrange approach,
the Lagrangian discrete phase model is introduced in general
CFD codes to calculate the disperse phase by tracking
particles, droplets, or parcels [29]. The trajectories of
particles in a turbulent flow field are predicted by the
turbulent dispersion models. To reduce the computational
time of the particle collision calculation, the O'Rourke
algorithm is employed [30]. The outcomes of collisions are
also determined by this algorithm i.e., whether the droplets
coalesce or reflect apart [26].
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
The hydrodynamic model, in which the diffusion of droplet
only controls its vaporization, or the kinetic model, which is
concerned with the molecules' detachment from the surface of
droplet, is utilized in this study [31]. The disintegration of
existing droplets is modeled to numerically simulate different
kinds of breakup modes. WAVE, or Kelvin-Helmholtz (KH)
instability models [32], and Kelvin-Helmholtz-RayleighTaylor (KH-RT) instability model [33], known as a hybrid
model, are favored in high speed high Weber number (We
>100) fuel-injection models. On the other hand, the Taylor
Analogy Breakup (TAB) is more commonly used in low
speed and low Weber number flows. KH-RT incorporates the
effects of aerodynamic breakup and instabilities of droplet
acceleration; thus, it is capable of handling both TAB and
WAVE models. Recently, there have been many hybrid
models developed by combining different breakup models to
estimate the spray characteristics accurately over a variety of
Weber number [22, 34]. In this study, due to the high Weber
number condition, WAVE and KH-RT model are most
suitable. In the next subsections, both of these models are
discussed in details to estimate the coefficients of these
models for good modeling.
WAVE Model
Reitz [32] developed a model called WAVE based on droplet
breakup due to the relative velocity between the gaseous and
liquid phases. The model is formulated from the KelvinHelmholtz instability's wavelength and growth rate to
determine the size of the droplets. The model is limited by
Weber number. We, which must be larger than 100 so that
Kelvin-Helmholtz instability is dominant in droplet breakup.
The maximum wave growth rate (also the most unstable
surface wave), Ω, and corresponding wavelength, Λ, are
defined as:
(1)
(2)
where
,
,
,
,
and
. Z is the Ohnesorge number, T is the Taylor
number and r0 is the radius of the undisturbed jet. We and Re
are the Weber number and the Reynolds number and
subscripts l and g represent liquid and gas phase,
respectively. The radius of a newly formed droplet from a
parent droplet during the breakup process is assumed
proportional to the wavelength ΛKH.
(3)
The constant B0 is set equal to the experimentally determined
value of 0.61. Additionally, the rate of change of the parent
droplet is defined by
(4)
where the breakup time, τKH, is given by
(5)
The breakup time constant B1 is an adjustable variable which
is recommended to be in the range of 1.73 to 60 [32, 35, 36].
Larger value of B1, produces fewer breakups and more
penetration. An estimation of the B1, factor has been
proposed by Liu et al. [37]:
(6)
However, Liu et al. [37] conclude that Eq. (6) does not
determine the valus of this constant qualitatively and
recommend that it should be used only for reference value as
the first guess.
Kelvin-Helmholtz-Rayleigh-Taylor (KH-RT)
Model
The KH-RT instability model is a combination of KelvinHelmholtz (KH) instability and Ravleigh-Tavlor (RT)
instability models. Both Kelvin-Helmholtz and RavleighTaylor models decide droplet breakup by detecting the fastest
growing surface wave on the droplets. The source of the
Kelvin-Helmholtz wave is induced by aerodynamic forces
between gas and liquid phases, whereas the Rayleigh-Taylor
wave is the result of acceleration of shed drops ejected into
free-stream conditions. Hwang et al. [33] showed in their
experiments the sequential breakup process in the
catastrophic breakup regime. A droplet first gets flatten by
the aerodynamic force on it and breaks up due to the
deceleration of the sheet droplet by means of Rayleigh-Taylor
instability model. Further breakups proceeded by the smaller
wavelength of Kelvin-Helmholtz wave found at the edge of
the fragments above. In high Weber number cases with high
droplet acceleration, Rayleigh-Taylor instability grows faster
and dominates the breakup of droplets. For the numerical
model, both instability models are utilized simultaneously
and breakups are determined by the fastest growth rate of
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
waves. In the Rayleigh-Taylor model, the fastest growing
wave's growth rate and its corresponding wavelength are
given by
Moreover, Senecal [40] analytically determined the
relationship of the Levich constant, CL, and the breakup time
constant of the WAVE model, B1. Considering the breakup
length of the WAVE model to be LKH = U ·τKH and assuming
that the viscosity is zero, Eq. (5) reduces to
(7)
(13)
(8)
and therefore
where a is the droplet acceleration, CRT is the RayleighTaylor breakup constant and KRT is the wave number given
by
(14)
Comparing Eq. (11) and (14), the correlation of the
coefficients is given by
(9)
Breakup time in the Rayleigh-Taylor model is defined as
(10)
where Cτ is the breakup time constant and the liquid core
length, LRT, is obtained from Levich theory [38]:
(11)
The radius of the new droplets is calculated as half of the
wavelength obtained above
.
Based on their experiments, Hiroyasu and Arai [39] proposed
the correlation of density and diameter to the Levich constant,
CL, as
(12)
(15)
As this relation shows, the Levich constant is also adjustable
and ranges from 5 to 20. In addition, Patterson and Reitz [35]
investigated the effects of Rayleigh-Taylor breakup constant,
CRT, in the range 1.0 ≤ CRT ≤ 5.33.
GRID INDEPENDENCE
Initially, grid independency tests were conducted to optimize
the grid size in the engine simulation. The simulations were
performed on a 60 mm (D) × 60 mm (W) × 80 mm (H) fixed
chamber with similar conditions as Wang's experiments [7].
Five different sets of grids, 70 × 70 × 105 (0.5 million), 87 ×
87 × 130 (1.0 million), 93 × 93 × 140 (1.2 million), 100 ×
100 × 150 (1.5 million), and 106 × 106 × 160 (1.8 million)
nodes, were generated. Figure 2 shows the results for a 300
MPa spray injection pressure with the five different meshes.
It is seen that 0.5 million cells over-estimates the penetration
significantly. Moreover, the domain with 1.0 million cells
slightly over-predicts the penetration in comparison with the
rest of the grids. From this information, we concluded that, in
the case of ultra-high injection spray model, the domain
should be meshed with 93 × 93 × 140 cells or finer to avoid
any effects of grid size on penetration length. For our
subsequent calculations, the optimal grid size is chosen to be
93 × 93 × 140, with the largest grid size set at 0.65 mm,
which is around 60 mm divided by 93, in the chamber of the
engine model.
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
X and negative Y directions. For convenience, the injectors
are referred to as INJ-0, INJ-90, INJ-180, and INJ-270,
respectively. Nozzle geometry and injection conditions are
summarized in Table 3.
Table 1. Engine operation conditions and initial setups
Figure 2. Spray tip penetration results for the grid
independency tests
Table 2. Diesel fuel properties
COMPUTATIONAL DOMAIN SETUP
The geometry utilized in this study is obtained from the
KIVA3V manual [41]. The engine is sized 82.55 mm in bore
and 92.075 mm in stroke and a high compression ratio of
17.2:1 is reached by the design. Its connecting rod length is
174 mm and it is operated at 1500 rpm. Other engine
operating conditions are listed in Table 1. The engine model
is set at top dead center (TDC) position initially by the
requirement of FLUENT. The geometry is appropriately
meshed with a hybrid mesh and sized by the optimum size
obtained from the previous section. The maximum number of
cells for the engine geometry is 166,500 cells at TDC and
570,000 cells when the piston surface reaches the bottom
dead center (BDC).
All wall boundaries, including the moving boundaries such as
piston and valves, are kept constant at 360 K during the
simulation. A pressure inlet is used at the inlet boundary (one
in positive Y direction) with two inlet pressures (1.0 and 1.5
atm) and fixed temperature at 318 K. The flow direction at
the inlet is set parallel to the intake runner walls. The outlet
boundary (in negative Y direction) is set as a pressure-outlet
where the gauge pressure is zero. The internal interfaces
between zones, as discussed above, do not contribute
physically in the calculation; however, when the valves are
fully closed, the interfaces are treated as walls and the
boundary conditions are determined from the adjacent wall
boundaries.
The diesel fuel which is utilized in this simulation is n-decane
(C10H22) with properties taken according to Wang's
experimental data (Table 2) [7], Four injection points are set
at the center of the cylinder head, offset by 0.5 mm from the
cylinder axis. The nozzle holes are 90 degrees apart from
each other and face towards positive X, positive Y, negative
Table 3. Nozzle configurations and spray injection
parameters
RESULTS AND DISCUSSION
Due to the simplified geometry of the engine used in this
study, no validation with experimental data is available.
Therefore, the quality of the model in this study is compared
with other numerical simulation which uses the same
geometry. Jonnalagedda et al. [42] studied an HCCI model
with this simplified vertical port engine geometry and
investigated the grid dependency with RANS and LES
models in KIVA3V. Figure 3 shows the difference of the
volume averaged in-cylinder pressure between the turbulence
models [43]; standard k-ε in our study and Reynolds
Averaged Navier-Stokes (RANS) and large eddy simulation
(LES) models by Jonnalagedda [42]. The pressure difference
at the peak between the standard k-ε and LES model is 8.34
% while the difference between the RANS and LES models is
−5.73 %. The flow structure of the models has also been
examined. Figure 4 is the stream traces on the cut plane
through the middle of the ports (YZ plane). These traces are
captured at 200 CA where the large turbulent kinetic energy
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
is expected at the time when the intake valve is closed.
Unlike the LES model, the standard k-ε model is not able to
depict small eddies in the section; however, it shows the
largest eddies at the bottom right corner, the middle right, and
the left edge of the intake valve pocket. Flow at the time of
injection is also investigated for the validation. Figure 5
captures the stream traces of the natural aspirated and
turbocharged cases in the section of a horizontal (XY) plane
5.7 mm offset from the surface of the cylinder head to piston
surface. In the natural aspirated case, vortices are observed
near the intake valve, which are also captured in the small flat
piston engine [44], In summary, the in-cylinder flow
calculated by the standard k-ε turbulence model does not
duplicate the complex localized details of the flow predicted
by the LES model but is able to reconstruct the larger eddies.
In addition comparison of the flow characteristics before
TDC shows the similar vortex positions. Since our current
interest is to study features of the general flow structure, the
standard k-ε model seems to be adequate for this purpose.
In Fig. 5, it is observed that the sprays emanating from INJ-0
and INJ-180 are exposed to the opposing flow whereas the
spray of INJ- 90 will be injected into the downwind and
INJ-270 will face a much more complex flow compared to
the others. INJ-0 and INJ-180 are sprayed to the upwind flow
and fuel injected from INJ-90 and INJ-270 will experience
catastrophic flow with vortices near the wall in the
turbocharged case. The cores of the vortices that are captured
near the intake port shift to the exhaust port side due to
stronger flow by turbocharging. The effects of these incylinder flows are further discussed in the spray
characteristics section.
Figure 5. Stream traces superimposed by velocity
magnitude contour of naturally aspirated (left) and
turbocharged (right) case at the start of injection (−25.5
ATDC) on XV plane 5.7 mm offset from the cylinder
head surface
Figure 3. Mean cylinder pressure variation for standard
k- ε in present work, and RANS and LES models [42]
Figure 4. Stream traces on YZ cut plane at 200 CA with
standard k-ε in present work (left) and LES model [42]
(right)
Once the fuel is injected into the chamber, strong flows of the
sprays become dominant and build symmetric flow structure
about the axes (Fig. 6). The magnitude of the fuel spray
velocity changes linearly by increasing the injection pressure
and the backflow is also increased relatively; hence, our
simulations predict higher relative velocity and air
entraimnent at the edges of sprays, as observed in the
experiments of Choi et al [19].
The effects of the breakup models, turbocharging, and
injection pressures on spray tip penetration are also
examined. Figure 7 presents the spray tip penetrations for
different injection pressures and densities in the engine
model, using the experiment data by Wang et al [7] and the
simulation in constant volume chamber as a reference. The
difference between simulations and experiments in the
constant volume case indicates that the spray models are
properly set. The reduction of the penetration from the
constant volume case to the engine model confirms the effect
of the flow inside the cylinder. It is also noticed in Fig. 7 that
the KH-RT model cases penetrate slightly less than the
WAVE model cases, as expected [16]. Since the KH-RT
model generally breaks droplets up more quickly into smaller
droplets, evaporation is induced and penetration is reduced.
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
The small droplet size and the effective vaporization
generated by the RT mechanism of KH-RT agree with the
result of Ricart et al. [45] and Xin et al. [46]. The effect of
turbocharging is clearly seen in the slower penetration as
shown in Fig. 7. As stated in the experimental works [7, 24],
high air density induced by turbocharging reduces the spray
penetration significantly.
Figure 6. Velocity vector field of INJ-0 superimposed by
velocity magnitude contour at 2.5 CA after start of
injection (−23.0 ATDC) on XY plane 5.7 mm offset from
the cylinder head surface
Figure 7. Spray tip penetration variations of INJ-0 at
different pressures (100, 200 and 300 MPa) and
conditions associated with the comparison between the
experiment and simulation in the constant volume vessel
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
Figure 8 presents the variation of each injector's spray
penetration in the turbo and non-turbocharging cases. In the
naturally aspirated case, as discussed previously in Fig. 5,
INJ-0 and INJ-180 have upwind flow towards the injection
points and hence their penetrations are similar. On the other
hand, the penetration of INJ-270 is greatly reduced from the
other sprays due to complexity of the flow and the existence
of a vortex near the wall. In the case of turbocharging, the
difference is not large compared to the naturally aspirated
case; however, the penetrations of INJ-90 and INJ-270 are
reduced since both of them are exposed to complicated flow.
The effect of the vortices on vaporization is consistent with
the discussion by Jagus [27]. Also note that the spray
penetration at 300 MPa injection pressure is lower than 100
and 200 MPa in the experiment and the computation. Fast
atomization induced by higher injection pressure results in
droplet vaporization and the penetration is reduced.
Figure 8. Deviation of spray tip penetration among
injectors at 300 MPa injection pressure with naturally
aspirated (top) and turbocharged (bottom) cases
The results for Sauter mean diameter (Fig. 9) display
transition of droplet size during the injection. Within 1.0 CA
after the start of injection each model shows rapid change of
droplet size. The maximum difference of SMD between
WAVE and KH-RT model is found to be 19.5% in the 200
MPa turbocharging case. Additionally, the difference
between naturally aspirated and turbocharged case is not
significant since the effect of the air density on SMD is
factored by the power 0.06 as defined by the modified SMD
correlation of Ejim et al. [47],
(16)
where SMD is given in µm, kinematic viscosity is in m2/s,
surface tension is in N/m, ρl and pg (density of the fuel and
air) are in kg/m3, and ΔP (the pressure difference between
injection and ambient pressure) is in bar. Within 1.0 CA after
the start of injection, droplet sizes of all models converge to
similar diameter and remain constant, as discussed by
Nishida et al. [5]. The higher the injection pressure, the
higher the velocity, as discussed in the previous section. Only
the 300 MPa injection pressure case is atomized the droplets
break into smaller size than SMDs obtained from the
correlation (red-dashed line in Fig. 9). Since the correlation
of SMD is developed within 0.5 ms [47], the larger SMDs in
lower injection pressures are due to the short time period of
the breakup process. Still, the correlation between high spray
velocity and small SMD is observed in the results and match
the study of Post and Abraham [26]. The SMD limitation
after 300 MPa injection pressure reported by Lee et al. [11] is
not reproduced in this study; however, the reduction of the
rate of change in SMD is observed where the rate of
reduction from 100 MPa to 200 MPa is 48.9% whereas the
change rate from 200 MPa to 300 MPa is 41.8%.
The dispersion of sprays is analyzed by the spray cone angles
in different conditions. The correlation of the cone angle with
the injection pressure is observed in Fig. 10. The results of
larger cone angle at higher injection pressure matches the
discussion of Park et al. [48] that small droplets are found
downstream of the cross-flow. In this study the small droplets
on the edge of the spray are entrained by the opposing flow,
captured by the flow, and enhance the size of the cone angle.
The effect of turbocharging is also shown for the 300 MPa
case (Fig. 10) which predicts wider cone angle, as found by
Naber and Siebers [24] whereas no change or even narrower
cone angles are captured in the 100 and 200 MPa cases. It is
assumed that the sprays are not yet fully developed at 2.5 CA
after the start of injection for those cases. The change of cone
angle in time shows oscillation during the injection but it is
likely to maintain the constant value right after the start of
injection. The constant but oscillating spray cone angle
during the injection is also found at different location of spray
in Naber and Siebers' work [24].
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
Figure 10. Variations of spray cone angle in different
cases and injectors at 2.5 CA after start of injection
(−23.0 ATDC) and transition of spray cone angle of 300
MPa injection pressure until sprays impinge to the walls
Figure 9. Sauter mean diameter; INJ-0 at different
injection pressures (100, 200, and 300 MPa)
Finally, the fuel mass fractions for each case are examined.
Figure 11 shows the sections of XY, YZ, and ZX plane at two
different times and is placed right, top, and left, respectively,
in each group of contours. Larger portion of the fuel mass
contour in the 300 MPa cases at 2.5 CA after start of injection
indicates fast injection process due to high injection pressure.
Fast penetration of spray at high injection pressure was also
observed by Tao and Bergstrand [12]. The completion of
injection at an early stage results in extra time for fuel-air
mixing and contributes to a better mixture formation at
combustion as reported by Yokota [3]. The uniform color
contour of the high injection pressure cases at TDC indicates
an even and lean mixture of fuel in the chamber.
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
strong swirl motion but generates a complicated flow at the
time of injection. From this study, in the ultra-high injection
pressure cases, the following conclusions regarding spray and
flow characteristics can be made, and are in agreement with
the results reported by other researchers.
• KH-RT hybrid breakup model estimates smaller droplets
and larger dispersion than the WAVE model. However, the
spray characteristics calculated by the WAVE model are not
significantly different from KH-RT since the breakup
constant is found by using the correlation to RT constants.
• Four sprays injected in the chamber are exposed to different
flow structures and found to be in good agreement with other
works on spray characteristics such as spray tip penetration
Sauter mean diameter and spray cone angle. The ultra-high
injection pressure breaks droplets to smaller size; thus,
reduced spray penetration and higher dispersion rate is
predicted. Also, the droplet vaporization is accelerated by the
existence of vortices in the direction of the sprays.
• Turbocharging is found to reduce spray tip penetration and
to dissipate the angle of spray. On the other hand, the effect
of turbocharging on SMD is not predicted as indicated in the
correlation stated by Ejim [47].
• The ultra-high injection pressure promotes a faster and
effective mixture process and allows extending the time to
form the mixture of air-fuel.
• The rate of SMD change is reduced at higher injection
pressure, as reported by Lee et al [11]. However, the SMD
limitation which Lee et al. has observed is not detected in this
study.
REFERENCES
1. Heywood, J.B., “Internal Combustion Engine
Fundamentals,” McGraw-Hill, ISBN-0-07-028637-X:
512-554, 1988.
2. Kato, T., Tsujimura, K., Shintani, M., Minami, T. et al.,
“Spray Characteristics and Combustion Improvement of D.I.
Diesel Engine with High Pressure Fuel Injection,” SAE
Technical Paper 890265, 1989, doi: 10.4271/890265.
Figure 11. Contours of fuel mass fraction at two
different times
CONCLUSIONS
The effects of ultra-high injection pressure sprays are studied
in a simplified engine geometry. The engine model is
properly set up and different conditions are examined by
means of injection pressures, spray models and inlet
pressures. The simple engine geometry does not induce
3. Yokota, H., Kamimoto, T., Kosaka, H., and Tsujimura, K.,
“Fast Burning and Reduced Soot Formation via Ultra-High
Pressure Diesel Fuel Injection,” SAE Technical Paper
910225, 1991, doi: 10.4271/910225.
4. Kuti, O.A., Xiangang, W.G., Zhang, W., Nishida, K., and
Huang, Z.H., “Characteristics of the Ignition and Combustion
of Biodiesel Fuel Spray Injected by a Common-Rail Injection
System for a Direct-Injection Diesel Engine,” Proceeding of
the Institution of Mechanical Engineers, Part D: Journal of
Automobile Engineering 224:1581-1596, 2010, doi:
10.1243/09544070JAUTO 1503.
5. NISHIDA, K., ZHANG, W., and MANABE, T., “Effects
of Micro-Hole and Ultra-High Injection Pressure on Mixture
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
Properties of D.I. Diesel Spray,” SAE Technical Paper
2007-01-1890, 2007, doi:10.4271/2007-01-1890.
6. Wang, X.G., Kuti, O. A., Zhang, W., Nishida, K., and
Huang, Z.H., “Effect of Injection Pressure on Flame and Soot
Characteristics of the Biodiesel Fuel Spray,” Combustion
Science and Technology 182:1369-1390, 2010, doi:
10.1080/00102201003789139.
7. Wang, X.G., Huang, Z.H., Kuti, O.A., Zhang, W., and
Nishida, K., “Experimental and Analytical Study on
Biodiesel and Diesel Spray Characteristics under Ultra-High
Injection Pressure,” International Journal of Heat and Fluid
Flow 31:659-666, 2010, doi: 10.1016/j.ijheatfluidflow.
2010.03.006.
8. Wang, X.G., Huang, Z.H., Zhang, W., Kuti, O.A., and
Nishida, K., “Effects of Ultra-High Injection Pressure and
Micro-Hole Nozzle on Flame Structure and Soot Formation
of Impinging Diesel Spray,” Applied Energy 88:1620-1628,
2011, doi: 10.1016/j.apenergy.2010.11.035.
9. Zhang, W., Nishida, K., Gao, J., and Miura, D., “An
Experimental Study on Flat-Wall-Impinging Spray of
Microhole Nozzles under Ultra-High Injection Pressures,”
Proceeding of the Institution of Mechanical Engineers, Part
D: Journal of Automobile Engineering 222:1731-1741, 2008,
doi: 10.1243/09544070JAUTO858.
17. Karrholm, F. and Nordin, N., “Numerical Investigation
of Mesh/Turbulence/Spray Interaction for Diesel
applications,” SAE Technical Paper 2005-01-2115, 2005,
doi:10.4271/2005-01-2115.
18. Ghasemi, A., and Djavareshkian, M.H., “Investigation of
the Effect of Natural Gas Equivalence Ratio and Piston Bowl
Flow Field on Combustion and Pollutant Formation of a DI
Dual Fuel Engine.” Journal of Applied Sciences 10:1-11,
2010, doi: 10.3923/jas.2010.1369.1379.
19. Choi, W., and Choi, B.C., “Estimation of the Air
Entraimnent Characteristics of a Transient High-Pressure
Diesel Spray,” Proceedings of the Institution of Mechanical
Engineers, Part D: Journal of Automobile Engineering
219:1025-1036, 2005, doi: 10.1243/095440705X34630.
20. Desantes, J.M., Arregle, J., Lopez, J.J., and Garcia, J.M.,
“Turbulent Gas Jets and Diesel-Like Sprays in a Crossflow:
A Study on Axis Deflection and Air Entraimnent,” Fuel
86:2120-2132, 2006, doi: 10.1016/j.fuel.2006.03.025.
21. McCracken, M.E., and Abraham, J., “Characterization of
Mixing Enhancement in Swirl-Spray Interactions in Diesel
Engines,” Combustion Science and Technology 174(10):
93-124, 2002, doi: 10.1080/00102200290021452.
10. Zhang, W., Nishida, K., and Tian, J.P., “Spray-Spray and
Spray-Wall Interactions in Diesel Sprays from Micro-Hole
Nozzles under Ultra-High Injection Pressures,” presented at
ICLASS 2009, 2009, Vail, CO, July, 2009.
22. Park, S.W., and Lee, C.S., “Investigation of Atomization
and Evaporation Characteristics of High-Pressure Injection
Diesel Spray Using Kelvin-Helmtholtz Instability/Droplet
Deformation and Break-Up Competition Model,” Proceeding
of the Institution of Mechanical Engineers, Part D: Journal
of Automobile Engineering 218:767-777, 2004.
11. Lee, S.H., Jeong, D.Y., Lee, J.T., Ryou, H.S., and Hong,
K., “Investigation on Spray Characteristics under Ultra-High
Injection Pressure Conditions,” International Journal of
Automotive Technology 6(2): 125-131, 2005.
23. Iyer, V., and Abraham, J., “Penetration and Dispersion of
Transient Gas Jets and Sprays,” Combustion Science and
Technology 130: 315-334, 1997, doi:
10.1080/00102209708935747.
12. Tao, F. and Bergstrand, P., “Effect of Ultra-High
Injection Pressure on Diesel Ignition and Flame under HighBoost Conditions,” SAE Technical Paper 2008-01-1603,
2008, doi:10.4271/2008-01-1603.
24. Naber, J. and Siebers, D., “Effects of Gas Density and
Vaporization on Penetration and Dispersion of Diesel
Sprays,” SAE Technical Paper 960034, 1996, doi:
10.4271/960034.
13. Lin, S.P., and Reitz, R.D., “Drop and Spray Formation
from a Liquid Jet,” Annual Review of Fluid Mechanics
30:85-105, 1998, doi: 10.1146/annurev.fluid.30.1.85.
25. Kennaird, D., Crua, C., Lacoste, J., Heikal, M. et al., “InCylinder Penetration and Break-Up of Diesel Sprays Using a
Common-Rail Injection System,” SAE Technical Paper
2002-01-1626, 2002, doi:10.4271/2002-01-1626.
14. Jiang, X., Siamas, G.A., Jagus, K., and Karayiannis,
T.G., “Physical Modelling and Advanced Simulations of Gasliquid Two- phase Jet Flows in Atomization and Sprays.”
Progress in Energy and Combustion Science 36:131-167,
2010, doi: 10.1016/j.pecs.2009.09.002.
15. Djavareshkian, M.H., and Ghasemi, A., “Investigation of
Jet Break-Up Process in Diesel Engine Spray Modelling,”
Journal of Applied Sciences 9(11):2078-2087, 2009.
16. Hossainpour, S., and Binesh, A.R., “Investigation of Fuel
Spray Atomization in a DI Heavy-Duty Diesel Engine and
Comparison of Various Spray Breakup Models,” Fuel
88:799-805, 2009, doi: 10.1016/j.fuel.2008.10.036.
26. Post, S., and Abraham, J., “Modeling the Outcome of
Drop-Drop Collisions in Diesel Sprays,” International
Journal of Multiphase Flow 28:997-1019, 2002.
27. Jagus, K., Jiang, X., Dober, G., Greeves, G., Milanovic,
N., and Zhao, H., “Assessment of Large-Eddy Simulation
Feasibility in Modelling the Unsteady Diesel Fuel Injection
and Mixing in a High-Speed Direct-Injection Engine,”
Proceedings of the Institution of Mechanical Engineers, Part
D: Journal of Automobile Engineering 223:1033-1048, 2009,
doi: 10.1243/09544070JAUTO1052.
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
28. “ANSYS FLUENT Theory Guide,” ANSYS, Inc.,
Canonsburg, PA. 2010.
29. Nijdam, J.J., Guo, B.Y., Fletcher, D.F., and Langrish,
T.A.G., “Lagrangian and Eulerian Models for Simulating
Turbulent Dispersion and Coalescence of Droplets within a
Spray,” Applied Mathematical Modelling 30:1196-1211,
2006, doi: 10.1016/j.apm.2006.02.001.
30. Schmidt, D.P., and Rutland, C. J., “A New Droplet
Collision Algorithm,” Journal of Computational Physics
164:62-80, 2000, doi: 10.1006/jcph.2000.6568.
31. Sazhin, S.S., “Advanced Models of Fuel Droplet Heating
and Evaporation,” Progress in Energy and Combustion
Science 32:162-214, 2006, doi: 10.1016/j.pecs.2005.11.001.
32. Reitz, R.D., “Modeling Atomization Processes in HighPressure Vaporizing Sprays,” Atomization and Spray
Technology 3:309-337, 1987.
33. Hwang, S.S., Liu, Z., and Reitz, R.D., “Breakup
Mechanisms and Drag Coefficients of High-Speed
Vaporizing Liquid Drops,” Atomization and Spray 6:353-376,
1996.
34. Lee, C.S., Kim, H.J., and Park, S.W., “Atomization
Characteristics and Prediction Accuracies of Hybrid BreakUp Models for a Gasoline Direct Injection Spray,”
Proceedings of the Institution of Mechanical Engineers, Part
D: Journal of Automobile Engineering 218:1041-1053, 2004.
35. Patterson, M. and Reitz, R., “Modeling the Effects of
Fuel Spray Characteristics on Diesel Engine Combustion and
Emission,” SAE Technical Paper 980131, 1998, doi:
10.4271/980131.
36. Liu, A., Mather, D., and Reitz, R., “Modeling the Effects
of Drop Drag and Breakup on Fuel Sprays,” SAE Technical
Paper 930072, 1993, doi: 10.4271/930072.
37. Liu, F.S., Zhou, L., Sun, B.G., Li, Z.J., and Harold, J.S.,
“Validation and Modification of WAVE Spray Model for
Diesel Combustion Simulation,” Fuel 87:3420-3427, 2008,
doi: 10.1016/j.fuel.2008.05.001.
42. Jonnalagedda, S., Nguyen, A.T., Zhou, B., and
Sobiesiak, A., “Numerical Studies on HCCI Engine
Combustion” Proceedings of the 18th Annual Conference of
the CFD Society of Canada, CFD 2010, London ON,
Canada, May, 2010.
43. Versteeg, H.K., and Malalasekera, W., “An Introduction
to Computational Fluid Dynamics,” Pearson Education
Limited, Essex, ISBN-978-0-13-127498-3:72-110, 2007.
44. Bianchi, G.M., Cantore, G., Parmeggiani, P., and
Michelassi, V., “On Application of Nonlinear k-ε Models for
Internal Combustion Engine,” Journal of Engineering for
Gas Turbines and Power 124:668-677, 2002, doi:
10.1115/1.1454115.
45. Ricart, L.M., Reitz, R.D., and Dec, J.E., “Comparisons of
Diesel Spray Liquid Penetration and Vapor Fuel Distributions
with In-Cylinder Optical Measurements,” Journal of
Engineering for Gas Turbines and Power 122:588-595, 2000.
46. Xin, J., Ricart, L., and Reitz, R.D., “Computer Modeling
of Diesel Spray Atomization and Combustion,” Combustion
Science and Technology 137:171-194, 1998, doi:
10.1080/00102209808952050.
47. Ejim, C.E., Fleck, B.A., and Amirfazli, A., “Analytical
Study for Atomization of Biodiesels and Their Blends in a
Typical Injector: Surface Tension and Viscosity Effects,”
Fuel 86:1534-1544, 2007, doi: 10.1016/j.fuel.2006.11.006.
48. Park, S.W., Kim, S., and Lee, C.S., “Breakup and
Atomization Characteristics of Mono-Dispersed Diesel
Droplets in a Cross- Flow Air Stream” International Journal
of Multiphase Flow 32:807-822, 2006, doi: 10.1016/
j.ijmultiphaseflow.2006.02.019.
CONTACT INFORMATION
Kohei Fukuda
fukuda@uwindsor.ca
Mechanical, Automotive and Material Engineering
University of Windsor
401 Sunset Ave.
Windsor, ON, CANADA N9B 3P4
38. Levich, V.G., “Physicochemical Hydrodynamics,”
Prentice Hall, New Jersey, ISBN-0-13-674440-0, 1962.
ACKNOWLEDGEMENTS
39. Hiroyasu, H. and Arai, M., “Structures of Fuel Sprays in
Diesel Engines,” SAE Technical Paper 900475, 1990, doi:
10.4271/900475.
This research is funded by Ontario Ministry of Research and
Innovation through the Ontario Research Fund under the
Green Auto Power Train project.
40. Senecal, P.K., “Development of a Methodology for
Internal Combustion Engine Design Using MultiDimensional Modeling with Validation through
Experiments,” Ph.D. thesis. Department of Mechanical
Engineering, University of Wisconsin-Madison, Madison
2000.
41. Amsden, A.A., “KIVA-3V: A Block-Structured KIVA
Program for Engines with Vertical or Canted Valves,” Los
Alamos National Laboratory, New Mexico, 1997.
DEFINITIONS/ABBREVIATIONS
U
σ
Velocity
Surface Tension
Gratis copy for Kohei Fukuda
Copyright 2012 SAE International
E-mailing, copying and internet posting are prohibited
Downloaded Thursday, March 15, 2012 10:26:55 AM
v
ρ
ASOI
ATDC
BDC
CA
CFD
HCCI
Kinemetic Viscosity
Density
After Start of Injection
SMD
TAB
TDC
Sauter Mean Diameter
Taylor Analogy Breakup
Top Dead Center
After Top Dead Center
Bottom Dead Center
Crank Angle
Computational Fluid Dynamics
Homogeneous Charge Compression Ignition
KH-RT
Kelvin-Helmholtz Rayleigh-Taylor
LES
NA
RANS
Large Eddy Simulation
Natural Aspirated
Reynolds Averaged Navier-Stokes
The Engineering Meetings Board has approved this paper for publication. It has
successfully completed SAE's peer review process under the supervision of the session
organizer. This process requires a minimum of three (3) reviews by industry experts.
All rights reserved. No part of this publication may be reproduced, stored in a
retrieval system, or transmitted, in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior written permission of SAE.
ISSN 0148-7191
Positions and opinions advanced in this paper are those of the author(s) and not
necessarily those of SAE. The author is solely responsible for the content of the paper.
SAE Customer Service:
Tel: 877-606-7323 (inside USA and Canada)
Tel: 724-776-4970 (outside USA)
Fax: 724-776-0790
Email: CustomerService@sae.org
SAE Web Address: http://www.sae.org
Printed in USA