Modeling of Common Rail Fuel Injection System of Four Cylinder Hydrogen Fueled Engine

Modeling of Common Rail Fuel Injection

System of Four Cylinder Hydrogen
Fueled Engine

Mohammed Kamil1, M. M. Rahman2, Rosli A. Bakar3

, , Faculty of Mechanical Engineering,

1 2 3

Universiti Malaysia Pahang, 26600 UMP, Pekan, Pahang

Email: 1mohammedk_22@yahoo.com,
2
mustafizur@ump.edu.my, 3rosli@ump.edu.my

ABSTRACT

Automotive system components, such as injection system, which have

an increasing number of parameters and new design approaches, are
required. It is no longer possible to optimize such a system solely based on
experience or forward optimization. Modeling of hydrogen fuel injection
system is a very important tool that can be used for predicting the effect of
advanced injection strategies on combustion and emissions of a hydrogen-
fueled engine using port fuel injection. This paper addresses the modeling
of common rail fuel injection system for four stroke four cylinder port
injection hydrogen fueled engine. The common rail injection system model
has been developed utilizing GT-FUEL commercial software. Focus is paid
for the wave propagation phenomena arising in this system subsequent
to the injection event and pressure oscillations that influence the injected
fuel quantity. One dimensional flow equations in conservation form
are employed to simulate wave propagation phenomena throughout the
common rail (accumulator). The interaction effect of operational conditions
including the engine speed and rail pressure is illustrated. The required
compromise solutions are discussed in details. It can be seen that the
common rail injection system is a promising enhancement for hydrogen
port injection engine.

KEYWORDS: Hydrogen fueled engine, port injection, common rail, wave

propagation, rail pressure, engine speed.

1.0 INTRODUCTION

Interest in the hydrogen economy has grown rapidly in recent

years. Those countries with long traditions of activity in hydrogen
research and development have now been joined by a large number
of newcomers. The main reason for this surge of interest is that the
hydrogen economy may be an answer to the two main challenges facing

ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010 37

Journal of Engineering and Technology

the world in the years to come: climate change and the need for security
of energy supplies. Both these challenges require the development of
new, highly-efficient energy technologies that are either carbon-neutral
or low emitting technologies. Another reason for the growing interest
in hydrogen is the strong need for alternative fuels, especially in the
transport sector. Alternative fuels could serve as links between the
different energy sectors, especially between the power system and
the transport sector, to facilitate the uptake of emerging technologies
and increase the flexibility and robustness of the energy system as a
whole (Larsen et.al., 2004). The comparison of hydrogen properties
with compressed natural gas (CNG) and gasoline fuels are tabulated
in Table 1.

The unique combustion properties of hydrogen can be beneficial at

certain engine operating conditions and pose technical challenges at
other engine operating conditions. These challenges include abnormal
combustion (pre-ignition, backfire and knocking), and higher heat
transfer losses (MacLean and Lave, 2003). Hydrogen fuel can be
induced to the combustion chamber either by: (i) external injection (e.g.,
using port or manifold fuel injection); or by (ii) direct injection to the
combustion chamber. Injecting hydrogen fuel directly in the cylinder is
not typically accompanied by pre-ignition occurrence and typically has
the potential for the highest power. However, it leads to a reduction in
the thermal efficiency due to non-homogeneity of the mixture inside  29

the cylinder; and the required compact fuel injection system. Then,
hydrogen induction by means of port injection can be considered as
external injection (e.g., using port or manifold fuel injection); or by (ii) direct injection to the
a compromise
combustion chamber. between the sophisticated
Injecting hydrogen fuel directly in the direct injection
cylinder is not typicallysystem
accompaniedand
thepre-ignition
by uncontrollable carbureted
occurrence and typically hassystem.
the potentialAsforathecompromise solution,
highest power. However, the
it leads
to a reduction in the thermal efficiency due to non-homogeneity of the mixture inside the cylinder;
design of the port injection should approach as much as possible
and the required compact fuel injection system. Then, hydrogen induction by means of port
from
the desired
injection featuresasand
can be considered move away
a compromise betweenfrom the undesired.
the sophisticated This system
direct injection suggests and
that
the extensivecarbureted
uncontrollable studiessystem.
and As enhancements mustthebe
a compromise solution, achieved
design of the portwith the
injection
should approach as much as possible from the desired features and move away from the undesired.
port injection system.
This suggests that extensive studies and enhancements must be achieved with the port injection
system.
TABLE 1
Comparison of hydrogen propertiesTABLE 1 with CNG and Gasoline
Comparison of hydrogen properties with CNG and Gasoline (White et al., 2006)
(White et.al., 2006)
Property Gasoline CNG Hydrogen
Flammability limit (Ø) 0.7-4 0.4-1.6 0.1-7.1
Laminar flame velocity (m/s) 0.37-0.43 0.38 1.85
Research octane number 91-99 140 >120
Adiabatic flame temperature (K) 550 723 858
Stoichiometric volume fraction (%) ~2 9.48 29.53
Minimum ignition energy (MJ) 0.24 0.28 0.02
Quenching distance (mm) ~2 2.1 0.64
Lower heating value (MJ/kg) 44.79 45.8 119.7
Heat of combustion (MJ/kg of air) 2.83 2.9 3.37

Common rail (CR) hydrogen injection system for port injection is proposed in this study to improve
the performance of the port injection system. The main key for this enhancement is the flexibility
that can be provided
38 for the
ISSN: most important
2180-3811 injection
Vol. 1 No. parameters: The pressure
1 January-June level generation is
2010
almost independent of the engine speed and of fuel metering; the injection timing and duration can
be optimized for every working conditions. In diesel engines, the CR injection system has met an
 Modeling of Common Rail Fuel Injection System of Four Cylinder Hydrogen Fueled Engine

Common rail (CR) hydrogen injection system for port injection is

proposed in this study to improve the performance of the port injection
system. The main key for this enhancement is the flexibility that can
be provided for the most important injection parameters: The pressure
level generation is almost independent of the engine speed and of fuel
metering; the injection timing and duration can be optimized for every
working conditions. In diesel engines, the CR injection system has
met an extraordinary success and is reasonable of the ever-increasing
share of diesel engines in European automotive market (Catania et.al.,
2008a). The ability of delivering multiple injections, which is one of the
most interesting features of such injection systems, allows an improved
combustion and heat release rate control, resulting in clean and efficient
engine performance. To achieve this result, CR electro-injectors have
thoroughly been investigated by researchers of the automotive field,
in order to obtain fast actuations, retaining, at the same time, a good
precision of the injected fuel volume (Catania et.al., 2008b). However, the
dynamics of CR hydraulic components can cause sensible perturbations
to multiple injections, which hence occur under different conditions
from those expected. The wave propagation phenomena arising in the
system, subsequent to an injection event, lead to pressure oscillations
that influence the injected fuel quantity, particularly when the dwell
time between consecutive injections is changed. The dependence of the
injected quantity on the system dynamics was also pointed by other
authors for diesel engines (Bianchi et.al., 2002; Catalano et.al., 2002;
Henein et.al., 2002; Mulemane et.al., 2004).

The rail pressure is also affected by the system dynamics: Whenever

an injection occurs, the pressure in the rail drops because the rail itself
does not behave as an infinite volume capacity (Bianchi et.al., 2002). A
sensor for continuously monitoring the pressure at the injector inlet
was proposed by (Torkzadeh et.al., 2002) to delivered the information to
ECU and thus adjust the injection parameters, so as to better control the
injected fuel quantity. In fact all these studies and findings are reported
for diesel common rail injection system. It is quite difficult to find a
work reports a detailed analysis of CR hydrogen injection system and
the pressure-wave propagation phenomena in this system. Numerical
models of CR diesel injection system, based on one-dimensional (1D)
partial differential equations, were developed in (Catalano et.al., 2002;
Catania et.al., 2008b; Ferguson and Kirkpatrick, 2001). However, as per
authors’ knowledge, no model has been developed for CR hydrogen
port injection system.

In the present work, CR for hydrogen injected system is suggested and

modeled for port injection hydrogen fueled engine. A one- dimensional

ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010 39

Journal of Engineering and Technology

compressible pipe flow model is developed using GT-FUEL code.

Using this model, the gas dynamic and pressure wave propagation
phenomena of CR injection system were examined briefly; and its
response to the variation in several parameters is clarified. These
parameters include: engine speed, rail pressure, and injector’s holes
diameter. It was shown how such a model is capable of predicting the
complex flow phenomena that occur in the system.

2.0 MODELING DESCRIPTION

The main components of the suggested CR hydrogen injection system are

shown in Fig. 1. Hydrogen tank is used as a reservoir for hydrogen and
non-corroding and free of leaks. Openings (safety) valves is provided
for excess pressure to escape automatically. Hydrogen tank is situated
far away from the engine to avoid ignition of escaping hydrogen in
case of accident. A pressure regulator is used to required injection
pressure. The injected hydrogen quantity depends upon the injection
period and difference between hydrogen pressure in the rail and port
pressure. However, in case of return-less hydrogen system (which is
suggested in this model because there is neither low pressure tank nor
a compressor for re-compressing hydrogen gas) the pressure regulator
is a part of in-tank unit. Hence hydrogen rail pressure is maintained at
a constant level with reference to the surrounding pressure. This means
that the difference between hydrogen rail pressure and port pressure
is not constant and take into account when the injection duration is
calculated. Hydrogen is supplied to the individual injector with the
help of the common rail. Common rail is used for mounting and
location of the injectors; storage of the hydrogen volume and ensuring
that hydrogen is distributed evenly to all injectors. The electromagnetic
injectors (solenoid-controlled) are injected hydrogen into the intake port
at system pressure. It permits the precise metering of the quantity of
hydrogen required. It is triggered through the engine control unit ECU
driver stages with signal calculated by engine management system.

40 ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010

mounting and location of the injectors; storage of the hydrogen volume and ensuring that hydrogen
is distributed evenly to all injectors. The electromagnetic injectors (solenoid-controlled) are injected
Modeling
hydrogen intoofthe
Common RailatFuel
intake port Injection
system System
pressure. of Fourthe
It permits Cylinder
precise Hydrogen
metering ofFueled Engineof
the quantity
hydrogen required. It is triggered through the engine control unit ECU driver stages with signal
calculated by engine management system.

FIGURE
FIGURE1 1 
CR Hydrogen Injection System
 31

The behavior of the pressureCRwaves is modeled using the
Hydrogen Injection System
equations
 31

of compressible gas
The behavior of the pressure CRdynamics.
waves CR The governing
Hydrogen
is modeled
Hydrogen
Injectionequations
Injectionusing
System for the CR
Systemthe equations of compressible gas dynami
The
The behavior
hydrogen
The governing of the
injection pressure
equations
behavior of the system
pressure for waves
the
waves are
is CR is modeled
the
hydrogen
modeled unsteadyusing
injection
using the the
mass,equations
equationssystem of
momentum, compressible
are the unsteady
of compressible gas mass,gas
and
dynamics. dynam
momentu
The governing equations for CR
the Hydrogen
CR Injectioninjection
hydrogen System system are the unsteady mass, moment
energy
The
and energyconservation
governing
The behavior the pressure equations,
equations for
conservation
of
the CR hydrogen
equations,
waves which
injection
which
is modeled usingare areexpressed
the expressed
system are the in vector
unsteady mass,
in vector form
momentum,
form (1-2) (Ferguson a
and energyconservation
and energy conservation equations,
equations, which
which are areequations
expressed expressedof compressible
in vector
in vector form
gas dynamics.
form (1-2)
(1-2) (Ferguson and (Ferguson
(1-2) (Ferguson
Kirkpatrick,
The governing 2001): and
equations
Kirkpatrick, 2001):
Kirkpatrick,
for the CR hydrogen2001):
injection system are the unsteady mass, momentum,
Kirkpatrick, 2001): equations, which are expressed in vector form (1-2) (Ferguson and
and energy conservation
Kirkpatrick, 2001): ߩ‫ܣߩ ܣ‬
డ௙
డ௙ డிడி ߩ‫ܣ‬
൅
൅
డ௙ డ௫ ൌ
డிൌ ݂ܶ
݂ܶ ൌ ൌ
൥ߩ‫ܷܣ‬൥ߩ‫ܷܣ‬
ߩ‫ܣ‬ ൩ ൩ (1) (
డ௧
డ௧
డ௙ ൅డி
డ௫ ൌ ݂ܶ ൌߩ‫ܧܣ‬൥ ߩ‫ܷܣ‬
ߩ‫ܧܣ‬ ൩
డ௧൅ డ௫ ൌ ݂ܶ ൌ ൥ߩ‫ܷܣ‬൩ (1)
డ௧ డ௫
ߩ‫ܧܣ‬
ߩ‫ܧܣ‬
ߩ‫ܷܣ‬ Ͳ
ߩ‫ܷܣ‬ Ͳ
ఋ஺
‫ ܨ‬ൌ ቎‫ܣ‬ሺߩܷ ଶ
ߩ‫ܷܣ‬൅ ܲሻ቏ ; ܶ ൌ ቎߬ఠ ξͶߨ‫ܣ‬
Ͳ
൅ ܲ Ͳ቏
ఋ௫ ఋ஺ (2)
‫ ܨ‬ൌ ቎‫ܣ‬ሺߩܷ ߩ‫ܷܣ‬
ߩ‫ܪܷܣ‬
ଶ
൅ ܲሻ቏ ; ߬ఠ ξͶߨ‫ ܣ‬൅ ܲ
ܶ ൌ ‫ݍ‬቎ ξͶߨ‫ ܣ‬ఋ஺ ቏
ఋ஺ (
‫ܨ‬
‫ܨ‬ൌൌ ቎቎‫ܣ‬ሺߩܷ
‫ܣ‬ሺߩܷଶ ଶ
; ቏;
൅ ቏ܲሻ
൅ ܲሻ ߬ఠൌ
ܶ ൌ ቎ܶ ఠ߬ ൅
ξͶߨ‫ܣ‬ ܲ ቏ ൅ ܲఋ௫ ቏
቎ ఠ ξͶߨ‫ܣ‬ (2)
ߩ‫ܪܷܣ‬ ఋ௫ ఋ௫
‫ݍ‬ఠ ξͶߨ‫ܣ‬
ߩ‫ܪܷܣ‬
ߩ‫ܪܷܣ‬ ‫ ݍ‬ξͶߨ‫ܣ‬
where ȡ is the hydrogen density,ఠ A is‫ݍ‬the
ఠ ξͶߨ‫ܣ‬
cross sectional area of the rail, U is the hydrogen velocity,
P is theȡpressure,
where is the E is the total
hydrogen specificAenergy;
density, is H is thesectional
thesectional
cross total specific enthalpy, ߬ఠ is the U iswall theshear
where ȡ is the hydrogen density, A is the cross area of thearea rail, Uofisthe therail,
hydrogen hydrogen veloci
velocity,
the ȡpressure, A
where is‫ݍ‬ఠthe hydrogenwall heatdensity, isarethe thatUnumerically
Hcross sectional area of the߬ఠrail, is the hydrogen veloc
stress, and is the flux. There a number of gas dynamic programs
PPisis the
where
solve pressure,
theρcompressible
is the E isE is
the
hydrogen the
totaltotal
conservation
specific specific
energy;
density,
equations energy;
A is the H
is
toenergy;
predict the is
total
system the
specific
crossflow.total specific
enthalpy,
sectional enthalpy,
is the wall
area of ߬ ఠ is the wall she
shear
the
P is the
stress, and
stress, pressure,
andcode ‫ݍݍ‬ఠ isis thethe E
wall is the
heatheat total
flux. flux.specific
There Thereare a number a ofH is the
gas dynamic total specific
gasprograms enthalpy,
thatprograms
numerically ߬ is the wall sh
ఠ is wall are number
of the CRofhydrogen dynamic ఠthat numerica
GT-FUEL used to develop the gas dynamic model port injection system
solveU
rail,
stress, iscompressible
theand the‫ݍ‬ hydrogen
is the wall
conservation velocity,
heat flux.
equations P
There is
to the
are
predict pressure,
a number
system flow. of E is
gas the total
dynamic specific
programs that numeric
solve
for fourthecylinder
compressible
ఠ engine. The conservation equations
fuel system components to predict
are assembled systemand toflow.fit the desired behavior
energy;
GT-FUEL
solve
of eachthe
GT-FUEL Hcode
fuel is the
system
code
is used
compressible
is total
component.
used specific
to develop
conservation
to
the gas enthalpy,
The underlying
develop the
dynamic
equations
gas physics τωto
model
dynamic isof
of the
the CR
predict
the flow
model wall system shear
hydrogen
isofgoverned
the flow.
CR stress,
by a fullyand
port injection system qωinjection syste
transient,
for four cylinder engine. The fuel system components are assembled and to fit thehydrogen
desired behavior port
is the
GT-FUEL
mean-line
for wall
fourfuel
of each
heat
code
analysis
cylinder is
system engine.
flux.
used
based
component.
There
to
on develop are
one-dimensional
TheThe fuel a
the number
gas
systemphysics
underlying
dynamic of
Navier-Stokes
components gasmodel
of the flow
dynamic of
equation.
areisassembled the Theprograms
CR
governed byand
hydrogen
input
a fully
that
parameters
to transient,port
fit the desiredinjection behavsys
including
for four engine speed
cylinder rpm, rail
engine. The pressure,
fuel and injector’s
system components holes diameter
are were defined
assembled and intothefit model.
the desired behav
numerically
ofmean-line
eacharefuel
There
analysis
several
solve
system based theoncompressible
wayscomponent.
to model the The
one-dimensional underlying
performance
conservation
physics equation.
Navier-Stokes
of a component. ofThe equations
the flow
first
Theisinput
partisin
to
governed predict
parameters
the systems byisa fully transie
of
systemeach fuel system
speed component. Theand underlying physics of the flow governed by a fully transi
the high flow.
including
mean-line engine
analysis rpm,
based rail on
pressure,
one-dimensional injector’s holes diameter
Navier-Stokes were defined
equation. in the model.
pressure hydrogen tank. The temperature of the tank is considered to the The ambient input paramet
mean-line
There are several
including
temperature, analysis
engine ways tobased
300 Kspeed
model on
rpm,
and composition railThe one-dimensional
the performance
pressure,
is pure gas andhydrogen. Navier-Stokes
of a component.
injector’s Theholes
The first part
high diameter equation.
in the systems
pressure inwere the tankThe
defined input
is parame
is in the mod
the high pressure
including engine hydrogen
speed tank.rail
rpm, temperature
pressure, and of the tank holes is considered to were
the ambient
considered
There are as a parameter.
several ways This
to pressure
model the isperformance
reduced to injector’s
a pre-specified
of a component. value diameter
through
The defined
the pressure
first part in the in systems
the mo
GT-FUEL
temperature, code
There
regulator.are The
300 K is
several
latter
andused
ways
is modeled
to
composition
to model
simply
develop
is pure gas
the
as an
thehydrogen.
performance
orifice
gas dynamic
connectionof The high model
a component.
between
pressure of
the tankThe in the
outlet
the
first tank
and
CR
part
next
is
in the system
the high pressure
considered
hydrogen asport
a parameter. hydrogen
injection tank. is
This pressure The reducedtemperature
to a pre-specifiedof thevalue tank isfuel
through considered
the pressureto the ambie
tube. high
the The diameter
pressure of the orificesystem
hydrogen is tank.
considered for
The four cylinder
astemperature
a parameter to engine.
control
of the the The
tank pressure
is system
downstream
considered to the amb
temperature,
regulator. The 300discharge
latter K andcoefficients
is modeled composition
simply as an is pure using
orifice gas the
connection hydrogen.
between The
the tank highoutlet pressure
and
and in
next the tank
components
the regulator. The
temperature,
tube. The diameter areofassembled
300 Kthe and
orifice composition
is and areto
considered fit
calculated
is
as a the
pure desired
gas
parameter hydrogen.
to behavior
geometry
control the
of
Thetheof
pressurehigh each
mating pipefuel
pressure
downstream in the tank
considered
orifice diameter. as aThis parameter.
descriptionThis pressure
is followed for theis check
reduced valves to anda pre-specified
regulator all over value through the pressu
this system.
system
The next component.
considered
the regulator.
regulator. The
part as
Thea discharge
theparameter.
in latter is
hydrogen The underlying
This
coefficients
modeledline apressure
is simply
are calculated
pipe carries as physics
is
an the
usingof
reduced
orifice
hydrogen
thetothe afrom
geometry
connection flow ofisthe
pre-specified governed
between
the tank
mating
to the value
the by
pipethrough
filter.tank
and
The outlet theandpressne
orifice diameter. TheThis description is followed for asthe ancheck valvesconnection
and regulatorbetween all over this system.
regulator.
atube.
fully
conduction
Thetransient,
heat
diameter latter
transfer of isthe
mean-line modeled
between orifice the simply
analysis
pipe
is wallbased
considered orifice
and hydrogen
as
The next part in the hydrogen line is a pipe carries the hydrogen from the tank to the filter. The
on
a one-dimensional
gas is neglected
parameter to control the
andNavier-
the
the tank outlet
initial
pressure and n
downstrea
tube.
the The
temperature
Stokes regulator.
conduction
diameter
of the pipe
equation.
heatThe transfer
of
Thewall
discharge theisinput
between
orifice
considered is considered
the parameters
coefficients to environment
pipe wall
astemperature.
a parameter
including
areandcalculated
hydrogen
The
gas isengine
using the tofilter
controlin thethe
speed
geometry
neglected and the ofrpm,pressure
hydrogen
the mating
initial
downstre
pipe a
circuit
the removes The
regulator. the solid discharge particlescoefficients
which couldare cause wear. It using
calculated is imperativethe that hydrogen
geometry of the is mating pipe
rail
orificepressure,
diameter.
temperature
efficiently
of theand
cleaned This
pipe injector’s
This
in order description
wall is considered
not to damage holes
is to diameter
followedenvironment for
the precision thewerecheck
temperature. defined
valves
The and in
filter the themodel.
regulator
in hydrogenall over this syste
orifice
circuit
The next diameter.
removes
part thethe
in solid description
particles line
hydrogen whichis isfollowed
could
a pipe cause theofcheck
for wear.
carries
the injection
theIt is valvessystem.
imperative
hydrogen and
fromthat
The filter
regulator
hydrogen
the tank
is
allto
isover
the this systT
filter.
There
modeledare in this several
model as ways a pipe similarto model to the deliverythe performance
pipe. However, higher of system.
asurface
component.roughness is
The next
efficiently
conduction
assumed part
cleaned
heat in the
in
totransfer hydrogen
order not to damage
between line theis a
the pipe
pipe carries
precision
inwall the
of the
and hydrogen
injection
hydrogen twogas from
pipesistank. the
The tank
filter
neglected isto the filter. T
and the init
The
modeledfirstininpart
conduction
order
thisheat in
model
consider
the
as a pipe
transfer
the pressure
systems similar is
between to thethe
the
losses high
delivery
pipe
this
pipe.
wall
part. Another
pressure
However,
and hydrogen
hydrogen higher surface
gas is
are added
roughness
neglectedTheto is and the ini
simulate the hydrogen
temperature of the lines.wall
pipe Three is flexible
considered pipes are
to also included
environment intemperature.
the hydrogen line
The upstream
filter
temperature
assumed in orderoftothe tank
consider isispressure
the considered losses in tothisthepart.ambientAnother temperature,
two pipes are added 300 to the hydrog
in
temperature
the rail. removes
circuit
simulate the hydrogen
of the
The purpose theispipe to
solid
lines.
wall
investigate
particles
Three
considered
the effect of the
flexiblewhich pipes are
to
couldenvironment
elasticity
causeof wear.
also included
temperature.
these pipes
It isonimperative
in the hydrogen
The filter
hydrogen
line upstream
wavethatin the hydro
hydrogen
speed. The
circuit function of
removes the thesolidcheck particles
valve in thewhich hydrogen line iscause
could to ensure that hydrogen
wear. It is flow happensthat hydrogen
imperative
efficiently
the rail. Thecleaned
in one direction
purpose isinto order
and no reverse
investigate
flow. not the to effect
damage of thethe precision
elasticity of these of pipes
the injection
on hydrogensystem. wave The filter
efficiently
speed. Theinfunction
modeled cleaned
this of the in checkorder notintothedamage
valve hydrogen the the line isprecision
to ensure
pipe.that of the injection
hydrogen flow happens system. The filte
As stated early, themodelrail hasasthe a pipe
task ofsimilar
storing to hydrogen delivery
and damping However,
the higher
pressure fluctuations surface
as roughness
modeled
in one direction
assumed
effectivelyin in thisandmodel
order
as possible.
no reverse
to2180-3811 as
consider
It is a pipe
flow. similar
theVol.
worth to mention pressure to the
the losses deliveryin pipe.
this part. However,
Another higher
two surface
pipes areroughnes
added
As stated early, ISSN:
the railto has the task of 1thathydrogen
No. expected
1 and fluctuation
January-June for
2010hydrogen rail is lower 41
assumed
than dieselthe
simulate in order
or hydrogen
gasoline consider
fuels. ThisThree
lines. thestoring
is because pressure losses
the fluctuations
flexible pipes are
damping
in
ofalso
these
the
thisincluded
part.
fuels
pressure
Another
are
fluctuations
introduced
in the two
hydrogen
as
by pipes
two line are upstrea
added
effectively as possible. It is worth to mention that the expected fluctuation for hydrogen rail is lower
simulate
sources
the The the purpose
rail. including hydrogen
the pump is lines.
to Three flexible
andinvestigate
injectors. However,
the effect pipes
thereofisareno also
the need included
for a pump
elasticity of theseinin the hydrogen
thepipes
hydrogen line upstre
on hydrogen wa
Journal of Engineering and Technology

K and composition is pure gas hydrogen. The high pressure in the tank
is considered as a parameter. This pressure is reduced to a pre-specified
value through the pressure regulator. The latter is modeled simply as an
orifice connection between the tank outlet and next tube. The diameter
of the orifice is considered as a parameter to control the pressure
downstream the regulator. The discharge coefficients are calculated
using the geometry of the mating pipe and orifice diameter. This
description is followed for the check valves and regulator all over this
system. The next part in the hydrogen line is a pipe carries the hydrogen
from the tank to the filter. The conduction heat transfer between the
pipe wall and hydrogen gas is neglected and the initial temperature
of the pipe wall is considered to environment temperature. The filter
in the hydrogen circuit removes the solid particles which could cause
wear. It is imperative that hydrogen is efficiently cleaned in order not
to damage the precision of the injection system. The filter is modeled
in this model as a pipe similar to the delivery pipe. However, higher
surface roughness is assumed in order to consider the pressure losses in
this part. Another two pipes are added to simulate the hydrogen lines.
Three flexible pipes are also included in the hydrogen line upstream
the rail. The purpose is to investigate the effect of the elasticity of these
pipes on hydrogen wave speed. The function of the check valve in the
hydrogen line is to ensure that hydrogen flow happens in one direction
and no reverse flow.

As stated early, the rail has the task of storing hydrogen and damping
the pressure fluctuations as effectively as possible. It is worth to
mention that the expected fluctuation for hydrogen rail is lower than
diesel or gasoline fuels. This is because the fluctuations of these fuels
are introduced by two sources including the pump and injectors.
However, there is no need for a pump in the hydrogen system because
hydrogen is stored in a high pressure and consequently no pumping
pulsations. The flow inside the rail is considered as 1D isentropic flow.
The isentropic flow is an adiabatic a flow in which there is no heat
exchange) frictionless flow. This means that the effects of viscosity
and heat transfer are restricted to thin layers adjacent to the walls, i.e.,
only important in the wall boundary layers. This assumption becomes
more realistic because of gas hydrogen density is very low (0.0824 kg/
m3) compared with liquid gasoline (730 kg/m3) and that hydrogen is
injected in the ambient temperature. Figure 2 shows the configuration
of the common rail. The CR is modeled using a series of pipes and flow-
splits. Four flow-splits, connected by three pipes, are composing the
rail. The flow-splits are composing from an intake and two discharges.
The intake draws hydrogen fuel from the preceding flow component
(pipe). One discharge supplies hydrogen to the next pipe and other

42 ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010

 Modeling of Common Rail Fuel Injection System of Four Cylinder Hydrogen Fueled Engine
 32
 32
supplies hydrogen to the injector. The discharge of the last flow-split
whichis there
closed is nowith cup tofrictionless
heat exchange) preventflow. anyThis flow
means through it because
that the effects of viscosity there
and heat are
no more
transfer
which are
there is no flow
restricted to components.
thin
heat exchange)layers adjacent
frictionless Conservation
to the walls, i.e.,
flow. This means that onlyof themomentum
important in the wall
effects of viscosity is solved
boundary
and heat
layers. This
transfer areinassumption
restricted tobecomes
thin layers more realistic
adjacent to thebecause
walls, of
i.e.,gas
onlyhydrogen
importantdensity is very
in the wall low
boundary
based
(0.0824 3 3-dimentional flow-splits even3 though the flow in GT-FUEL
layers.kg/m This ) assumption
compared with becomesliquidmoregasoline (730because
realistic kg/m ) and of gasthathydrogen
hydrogendensityis injected in the
is very low
is otherwise
ambient
(0.0824 temperature. based
kg/m3) comparedFigure with2on aliquid
shows 1D version
thegasoline
configuration of kg/m
(730 the
of the3Navier-Stokes
common
) and rail. Theequation.
that hydrogen CR is modeled
is injected inThe the
usingcharacteristic
a series
ambient of pipeslength
temperature. and
Figure 2and
flow-splits.
shows expansion
Four flow-splits,
the configuration diameter
connected are
of the common defined
by three pipes,
rail. The forCReach
are composing port
is modeled
theusing
rail. aThe flow-splits are flow-splits.
composing Four from flow-splits,
an intake and
in
hydrogen
theseries of pipes and
flow-split. The characteristic length istwo
thedischarges.
connected by three pipes,
distance The intake
from are the draws
composing
port
the rail.fuel Thefrom the preceding
flow-splits flow component
are composing from an(pipe). intakeOneanddischarge supplies The
two discharges. hydrogen
intaketodraws
the
hydrogen fuel from the preceding flow component (pipe). One discharge supplies hydrogen tothe
next plane
pipe andto the
other opposite
supplies hydrogen sideto of
the the
injector. flow-split
The discharge (used
of the in
last to calculate
flow-split is closed the
with propagation
nextcup
pipe to prevent
and other of
any pressure
flow
supplies through
hydrogen waves
to the through
it because there areThe
injector. nothe moreflow-split).
discharge the lastThe
flowofcomponents. expansion
Conservation
flow-split is closed
ofwith
momentum is solved anybased
otherwise
cup to prevent
diameter based is
is used
onsolved
a 1D based
in inthrough
flow 3-dimentional
calculating
versionin of
it because flow-splits
the
the Navier-Stokes
therekinetic even
are no
equation.
though
more
energyflowthe flow in GT-FUEL
components.
losses
The characteristic
due Conservation
to area is
of momentum 3-dimentional flow-splits even though the flow in lengthGT-FUEL and is
expansion
expansion
otherwise based inside
diameter 1Dthe
onarea defined flow-split.
for each
version of theport Figure
in 3 equation.
shows
the flow-split.
Navier-Stokes Thethe geometry
characteristic
The characteristic lengthof isflow-
length the
and
splitfrom
distance
expansion no.diameter
1. The
the sensors
portareplane
definedto the and
for monitors
opposite
each portsidein theofare
the equipped
flow-split
flow-split. The(used in thein torail
characteristic to sense
calculate
length isthethe
propagation
distance of pressure
from the port waves
plane through
to the flow-split).
the opposite side of the Theflow-split
expansion(used diameter
in towithis used inthe
calculate
and monitor
calculating the kinetic
the variations
energywaveslosses through
in
due to area
pressure and temperature crank
propagation of pressure theexpansion
flow-split).insideThetheexpansion
flow-split. diameter
Figure 3 showsis used the in
angle
geometry of throughout
flow-split no. 1. the
The run
sensors session.
and monitors These
are are
equipped fixed
in
calculating the kinetic energy losses due to area expansion inside the flow-split. Figure 3 shows the in
rail flow-split
to sense and no.the
monitor 4
and
thegeometry pipe
variations of in no. 2 of
pressure
flow-split no. the
and rail
1. temperature
The respectively.
sensors with crank angle
and monitors throughout
are equipped in the
the run session.
rail to sense andThese are
monitor
fixed
the in flow-split
variations inno. 4 and pipe
pressure no. 2 of the rail
and temperature withrespectively.
crank angle throughout the run session. These are
fixed in flow-split no. 4 and pipe no. 2 of the rail respectively.

FIGURE 2
FIGURE 2
CommonFIGURE
rail configuration
2
Common rail configuration
Common rail configuration

FIGURE 3
FIGURE
Geometry 33
of flow-split
FIGURE
Geometry ofofflow-split
Geometry flow-split

In the present model, the fuel injector motion (opening and closing)
is prescribed instead of being calculated. The opening and closing
profiles of the injectors are imposed using pipes and orifices to model
the injectors and orifice represents the injector holes. The opening

ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010 43

  33
Journal of Engineering and Technology 

and closing of the injector is controlled through the orifice diameter.

InThe orifice
the present diameter
model, is defined
the fuel injector motionas a function
(opening of the
and closing) main drive
is prescribed instead (crank
of being
calculated. The opening and closing profiles of the injectors are imposed
shaft) angle. This simplification is done because the focus of this study using pipes and orifices to
model the injectors and orifice represents the injector holes. The opening and closing of the injector
isis on thethrough
controlled gas dynamic phenomena
the orifice diameter. The orificeindiameter
the common
is defined asrail. Eachofinjector
a function the main
is considered
drive fourThis
(crank shaft) angle. holes. Figureis4done
simplification shows
becausethe
thedeveloped common
focus of this study is on therail
gas
dynamic phenomena in the common rail. Each injector is
model for hydrogen fueled engine utilizing the GT-FUEL codes.considered four holes. Figure 4 shows the
developed common rail model for hydrogen fueled engine utilizing the GT-FUEL codes.

FIGURE 44
FIGURE
GT-FUEL Model approach setup for the CR hydrogen injection system
GT-FUEL Model approach setup for the CR hydrogen injection system

3.0 RESULTS AND DISCUSSION

3.0
In this
RESULTS AND DISCUSSION
section the results obtained from the developed model will be analyzed. The results are
focusing on the wave propagation and gas dynamic behavior of hydrogen gas flow in four strokes
In this
four section
cylinders the results
port injection engine.obtained
The effectsfrom the speed,
of engine developed model
rail pressure willholes
injector be
diameter will be investigated. The present section is subdivided into three parts to highlight the
analyzed. The results are focusing on the wave propagation
trends with each of the above stated parameters. Young modulus for flexible pipes is 200 GPa,
and gas
dynamic
single injectionbehavior
event, and theofresults
hydrogen gasno.
are for cycle flow in fourarestrokes
5 conditions adopted infour cylinders
the model.
port
The flowinjection
in commonengine. The due
rail is unsteady effects
to the of engine
periodic speed,
injectors railand
opening pressure injector
closing. The opening
and closing of the injector creates finite amplitude compression and rarefaction pressure waves that
holes diameter will be investigated. The present section is
propagate through the common rail hydrogen flow. These pressure waves may aid or inhibit the
subdivided
into three
hydrogen parts
exchange to highlight
processes. Therefore thethesize
trends with each
of the common of thetoabove
rail necessity stated
tune properly to
parameters.
obtained Young
the preeminent gas modulus
exchange. Theforeffect
flexible pipes
of engine ison
speed 200gasGPa, single
dynamic injection
of hydrogen flow
event, and the results are for cycle no. 5 conditions are adopted in the
model.

The flow in common rail is unsteady due to the periodic injectors

opening and closing. The opening and closing of the injector creates
finite amplitude compression and rarefaction pressure waves that
propagate through the common rail hydrogen flow. These pressure
waves may aid or inhibit the hydrogen exchange processes. Therefore

44 ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010

 Modeling of Common Rail Fuel Injection System of Four Cylinder Hydrogen Fueled Engine

the size of the common rail necessity to tune properly to obtained the
preeminent gas exchange. The effect of engine speed on gas dynamic of
hydrogen flow is investigated. The engine speed of 3000, 4000 and 5000
rpm were considered in this investigation. The results presented are 0.5
mm injector holes diameter and 3 bar rail pressure.

Figure 5 shows the pressure variation in the common rail with crank  34
angle for different engine speed. The complexity of the phenomena that 
occur is apparent. The amplitude of the pressure fluctuations increases
substantially
is investigated. with
The engine decreasing
speed of 3000, 4000engine speed.
and 5000 Maximum
rpm were consideredamplitude of
in this investigation.
0.047presented
The results bar was arereported at 3000
0.5 mm injector rpm,
holes while
diameter andthe maximum
3 bar rail pressure.amplitude of
Figure0.032
5 showsbarthe pressureatvariation
happens 5000 rpm.in theHowever,
common rail as with crank angle
the engine speedfor different
increase, engine
speed. The complexity of the phenomena that occur is apparent. The amplitude of the pressure
the frequency of the pressure waves increases. The rapid closing of
fluctuations increases substantially with decreasing engine speed. Maximum amplitude of 0.047 bar
the injectors
was reported at 3000 results
rpm, while in the
high oscillations
maximum waves
amplitude known
of 0.032 as the at
bar happens “water
5000 rpm.
hammer
However, effect”speed
as the engine which can be
increase, the seen clearly
frequency of theinpressure
figurewaves
5. Higher speeds
increases. The rapid
closinglead
of theto higher number of closing and opening times for each injector
injectors results in high oscillations waves known as the “water hammer effect”
which and
can be seen clearly in figure 5. Higher speeds lead to higher number
this is reflecting in higher frequency for the pressure pulsations. of closing and opening
times for each injector and this is reflecting in higher frequency for the pressure pulsations. Each
Each injection triggers a pressure wave in the rail. This wave influences
injection triggers a pressure wave in the rail. This wave influences the fuel quantity of the next
injections. Flowquantity
the fuel of the next
rate differences amonginjections. Flow
the injectors causerate differences
differences among
of fuel the from
quantity
injectors
cylinder cause differences of fuel quantity from cylinder to cylinder.
to cylinder.

 Waterhammer

FIGURE 55
FIGURE
Rail Pressure profile for different
Rail Pressure profile for differentengine
engine
The speed of sound is an important parameter considered in this study due to the pressure waves
Thealong
propagate speedtheofcommon
sound is railanwith
important
velocity, parameter considered
which depends on the speedin this studyFor a
of sound.
due to the pressure waves propagate along the common rail
substance that is not perfect gas, it is desirable to express the speed of sound in terms ofwith
physical
velocity, which depends on the speed of sound. For a substance that
property of the substance: the compressibility (John and Keith, 2006). Thus, the sound speed can be
calculated using the van der Waals equation, which can be expressed as in Eq.
is not perfect gas, it is desirable to express the speed of sound in terms (3):

డ௣
ܽ ൌ ටߛሺ ሻ ் (3)
డఘ
ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010 45
where Ȗ is the specific heat ratio (1.407 for hydrogen at 0 0C) and ሺ߲‫݌‬Τ߲ߩሻ ் is the partial
 FIGURE FIGURE
5 5
Journal of Engineering Rail
and Technology
Pressure FIGURE
Rail profile
Pressure 5 for engine
forprofile
different different engine
Rail Pressure profile for different engine
The speedThe of speed
sound of is an
soundimportant parameterparameter
is an important considered in this study
considered duestudy
in this to thedue
pressure
to the waves
pressure waves
Thephysical
of speed of sound is an important
property of the parameter considered
substance: the in compressibility
this study due to the pressure waves
(JohnofForand
propagate along
propagate thealong
common the rail with
common velocity,
rail which
with velocity, depends
which on the speed
depends on the of speed
sound. a For a
sound.
propagate along the common rail with velocity, which depends on the speed of sound. For a
substanceKeith,
that is
substance 2006).
notthat Thus,
perfect
is notgas, the
it
perfectissound
desirable
gas, it speed
is to can
express
desirable to be
the calculated
speed
express of
the sound
speed using
in
of the
terms
sound
substance that is not perfect gas, it is desirable to express the speed of sound in terms of physical van
of
in der
physical
terms of physical
propertyWaals
of the substance:
property
property of the the compressibility
substance: the (John
compressibility
equation, which can be expressed as in Eq. (3):
of the substance: the compressibility and
(John Keith,
(John
and 2006).
and
Keith, Keith,Thus,
2006). the
2006).
Thus, sound
Thus,
the speed
the
sound can
sound
speed can be
speed
be can be
calculatedcalculated
using theusing
calculated van the
usingderthe Waals
vanvan
derderequation,
Waals
Waals whichwhich
canwhich
equation,
equation, be
canexpressed
becan as in
as Eq.
be expressed
expressed (3):
as(3):
in Eq. in Eq. (3):

డ௣ డ௣
డ௣
 35
ܽ ൌ ටߛሺ ܽܽ ൌ
ሻൌ்ටߛሺ
ටߛሺడఘሻ்ሻ ் (3)
(3)
 (3)
డఘ డఘ

0
where Ȗ where
is theȖȖ
where ோ் isisthe
specificthespecific
ோ்ఘఉheat
specificheatheat
ratio ratioratio
(1.407 (1.407 for hydrogen
for(1.407
hydrogen
for hydrogen 0 C)
at 0at0C) and0andሺ߲‫݌‬
at 0 ሺ߲‫݌‬Τ߲ߩሻ is the partial
C) Τand ்isΤthe
߲ߩሻ ்ሺ߲‫݌‬ ߲ߩሻ ்partial
is the partial
ܽ derivative
ൌ ටߛሾ of൅pressure െ ʹߙߩሿ
మ with respect to density for a constant temperature. Finally the speed (4)
where
derivativederivative γ of is
of pressure
ଵିఉఘ withthe specific
respect
pressure
ሺଵିఉఘሻ withtorespect
densityheattofor ratio
a constant
density (1.407
for for temperature.
hydrogen
a temperature.
constant Finally the at
speed
Finally ofofsound
00the
C) sound
and
speed of sound
can be expressed as in Eq. (4) (John and Keith, 2006):
can be expressed
(∂p⁄∂ρ) asisinthe
can be expressed Eq. (4) (John
as in
partial and(John
Keith,and
Eq. derivative
(4) 2006):
Keith, 2006): with respect to density for
of pressure
T ߚ are the constants used in van der Waals’s equation of state, T is the absolute
where ߙ and
a constant
temperature, andtemperature. Finally
ܴ is the gas constant thebespeed
which can ofassound
expressed can be expressed as
in Eq. (5):
35
 35
 35

in Eq. (4) (John and Keith, 2006):  
௨௡௜௩௘௥௦௔௟௚௔௦௖௢௡௦௧௔௡௧ሺ࣬ሻ
ܴൌ (5)
௠௢௟௔௥௠௔௦௦ሺ௠ሻ  35  35
ܽ ൌ ටߛሾ
ோ்
൅
ோ்ఘఉ
ோ்ோ் ோ்ఘఉ
െ of ோ்ఘఉ
ʹߙߩሿ 
(4)
35 
Theܽܽ ൌ
ଵିఉఘ ൌටߛሾ
speedටߛሾ ofଵିఉఘ
ሺଵିఉఘሻ ൅మ൅
sound మ െమ ʹߙߩሿ
hydrogen െ ʹߙߩሿ at 0 0C is equals to 1270 m/s compared with 331 m/s for air at (4) (4)
ଵିఉఘ ሺଵିఉఘሻ
ሺଵିఉఘሻ 
same condition. This comes from the difference in the molar mass for hydrogen (2.016) compared
withோ் air
where (28.97). This implies that theinderpressure waves inequation
hydrogen gasT propagate
T state, faster in the
where
ܽ ൌ ටߛሾ ߙ and ൌߚ൅
ܽwhere
ߙ ߙand
areோ்ఘఉthe
andோ்ߚ ߚ are
െare
the
constants
ோ்ఘఉthe constants
used
constants inused
van usedvan der
vanWaals’s
inWaals’s derequation
Waals’s of of state,
state,
equation ofis is
thetheabsolute
absolute
T is the absolute (4) (4)
common ටߛሾrail.
temperature, ோ்Figure
andమ൅ ʹߙߩሿ
6is demonstrates
ோ்ఘఉ
the మgasെ ʹߙߩሿ
constant thewhich
behavior can of expressed
be the sound as of in speed
Eq. in rail for selected speed
(5):
where
temperature,
ଵିఉఘ ටߛሾα
ൌ and
ܽtemperature, ܴ is
ሺଵିఉఘሻ
ଵିఉఘ
and ܴ βܴare
theሺଵିఉఘሻ
൅ gas
ሺଵିఉఘሻ
isconstant
మ െ the
ʹߙߩሿ constants
which can be used expressed in expressed
van
as in Eq. der(5):asWaals’s equation of
(4)
range. It can ଵିఉఘ be andobserved the
that gas
this constant
figure iswhich similarcan be
trends with Figure in This
5. Eq. (5):indicates that the
where
state,
pressure
and
where
Twave
are
is
and
the
theߚlower
absolute
propagation
௨௡௜௩௘௥௦௔௟௚௔௦௖௢௡௦௧௔௡௧ሺ࣬ሻ
constants
ߚarearetheamplitudes.
the constants
temperature,
is strongly
used inused van
depends on and
usedder inWaals’s
van
the sound
der
R is speed.
equation
Waals’s
the
of
gas
equation
Higher
state,
constant
speed leads
TofisTrail
state,
the absolute
which
Tabsolute
isof
to higher
thesound
can
absolute
ܴ ൌ ߙfrequencies
ܴ ൌߚ ௨௡௜௩௘௥௦௔௟௚௔௦௖௢௡௦௧௔௡௧ሺ࣬ሻ
ߙand (5)
௨௡௜௩௘௥௦௔௟௚௔௦௖௢௡௦௧௔௡௧ሺ࣬ሻ
bewhere
ܴ expressed
ൌ ߙ and
௠௢௟௔௥௠௔௦௦ሺ௠ሻ
asܴ in constants
Eq.gas(5): At in
engine van der
speed Waals’s
3000 equation
rpm, the of
maximumstate, is the
speed (5)
of (5)
temperature,
temperature,
and isnoticed
the
andܴgas
m/sܴis௠௢௟௔௥௠௔௦௦ሺ௠ሻ
௠௢௟௔௥௠௔௦௦ሺ௠ሻ
1320.4
temperature, and is
constant
iswhile
the thegasit is which
constant
1319.69
constant canm/s
which which
becan
atexpressed
can
5000
be be expressed
rpm.
expressed as
Inin asEq.
fact inthe (5):
asdifference
Eq. in Eq. (5):
(5): between is very
0
The compared
small speed of sound with sound of hydrogen speed.0 The C is0 equalsoftorail
at 0 fluctuation 1270speed m/sofcompared
sound is with 0.053% 331from m/s for engineair at
The ௨௡௜௩௘௥௦௔௟௚௔௦௖௢௡௦௧௔௡௧ሺ࣬ሻ
speed The
sameofat௨௡௜௩௘௥௦௔௟௚௔௦௖௢௡௦௧௔௡௧ሺ࣬ሻ
sound
speed
condition. of sound
of hydrogen
௨௡௜௩௘௥௦௔௟௚௔௦௖௢௡௦௧௔௡௧ሺ࣬ሻ
This comes atfrom
of hydrogen 0thatCthesound
isat equals
0 speed
difference C is toin 1270
equals
the m/s
to 1270
molar compared
mass m/s
for withspeed.
compared
hydrogen 331 m/s
with
(2.016) for
331 air
compared m/s at for air at
ܴ ൌ speed ܴ
ܴ ൌൌ 3000 rpm. This implies less depends on the engine This change (5)
(5) (5)
same condition.
may same This
condition.
௠௢௟௔௥௠௔௦௦ሺ௠ሻ
withcome air from comes
௠௢௟௔௥௠௔௦௦ሺ௠ሻ
(28.97). This
This
the hydrogen
௠௢௟௔௥௠௔௦௦ሺ௠ሻ from comes
implies the from
difference
that the
temperature the pressure
and in
differencethe waves
density. molar in the mass
in molarfor hydrogen
hydrogen massgas for (2.016)faster
hydrogen
propagate compared
(2.016)
in thecompared
with air common
(28.97).
with air rail. This
(28.97). implies
Figure This that
implies
6 demonstrates the pressure
that
the thebehavior wavesof in
pressure hydrogen
thewaves
sound in gas propagate
of hydrogen
speed in rail gas faster inspeed
forpropagate
selected the in the
faster
The speedThe
common Theof speed
rail.
range.
common sound
speed
Figure
It can ofof6
rail. sound
be sound
hydrogen of of
demonstrates
observed
Figure hydrogen
hydrogen
at 0the
that
6 demonstrates
0
this is0at0equals
Catbehavior
figureC0theis0isCequals
is
ofto equals
similar 1270
the
behavior tosound
1270
m/s
trendsto the
of m/s
1270
of
with compared
compared m/s of
speed
Figure
sound compared
in with with
5.rail
This
speed for331
331 with
m/sm/sfor
inselected
indicates
rail for
331
for air
air
that m/s
speed atat for speed
the
selected air at
same same
condition.
same
range. It pressure condition.
condition.
This
can be Itobserved
wave comesThis
This comes
from comes the
that thisis figure from the
difference
from difference
the difference
isdepends in
similarison the in the
molar
trends inmolar
themass
with mass
molarfor
Figure for hydrogen
hydrogen
mass for
5. Figure (2.016)
hydrogen
(2.016)
Thisspeed indicates compared
compared
(2.016)
that thecompared
range. can propagation
be observed strongly
that this figure the sound
similar speed.
trends with Higher 5. leads
This to higher
indicates that the
with
pressure with
air (28.97).
with air
air
frequencies
wave (28.97).
This
(28.97).
andimplies
propagation This
lower This implies
that
implies
is amplitudes.
strongly thethatpressure
At the
that pressure
the
engine thewaves
waves
pressure
speed in
3000 inspeed.
waves
hydrogen
rpm, hydrogen
in maximum
the gas
hydrogen
gas propagate
propagate gasspeed
rail faster
propagate
faster in
oftosound in faster
the
the in the
of to higher
pressure
common wave
rail. propagation
Figure 6 demonstrates is depends
strongly
the
on
depends
behavior
sound
of on
the the
sound soundof
Higher
speed.
speed in
speed
Higher
rail for
leadsspeed
selected
higher
leads
speed
common 1320.4
rail.
common
frequencies and
frequenciesm/s
Figure
lower is
rail. noticed
6anddemonstrates
Figure
amplitudes.
lower while it
6 amplitudes. is
demonstrates
Atthe 1319.69
engine behavior
At the
speed m/s
engine ofat
behavior
3000 5000
thespeedrpm,rpm.
soundof3000theIn fact
of maximum
the sound
speedthe
rpm, the in difference
of maximum
rail
speed
railfor between
inselected
speed rail
of for
rail soundis
speed very
speed
selected
of speed
sound of
range.
small It can be with
compared observed sound that this The
speed. figure is similaroftrends
fluctuation rail with Figure
speed of sound 5. isThis
0.053%indicatesfrom that
enginethe
range.
1320.4Itm/s
can
range.
is
1320.4be Itobserved
noticed
m/s caniswhile
be observed
that
noticed it this
is figure
that
1319.69
while it this
is is
m/s similar
figure
1319.69 at 5000 istrends
m/s similar
rpm.
at with
5000In trends
Figure
fact
rpm. with
theIn 5. Figure
This
difference
fact the indicates
5. This
between
difference indicates
that
is the that
very
between is the
very
pressure
speed wave rpm.
at 3000 propagation
This is strongly
implies that sound depends speed on lessthe sound
depends speed.
onHigherHigher
the engine speed
speed. leadsThis to change
higher
pressure wave
pressure
small compared
smallcomepropagation
wave
with
compared propagation
sound withis strongly
speed.sound isspeed.
The depends
strongly
fluctuation
The ondepends
the sound
of rail onspeed
thespeed.
sound
of soundspeed. isspeed
Higher
0.053% leads
isspeed
oftosound
from higher
leads
engine ofto engine
higher
frequencies
may and
from lower
the amplitudes.
hydrogen At engine
temperature andfluctuation
speed
density. 3000 of
rpm, rail
thespeed
maximum of sound
rail speed 0.053% from
frequenciesfrequencies
andrpm.
speed at 1320.4
3000
speed lower
at
m/s 3000 and
amplitudes.
isThis lower
implies
rpm.
noticed This
whileamplitudes.
At
that engine
implies
it is sound
1319.69 At
that engine
speed m/s 3000
sound less speed
rpm,
depends
speed
at 5000 3000
theIn
less
rpm. onmaximum
rpm,thethe
depends
fact the
engine
onmaximum
rail speed
thespeed.
difference engine rail
ofspeed.
This
between sound
speed
change of sound
This
is very changeof
1320.4
may comem/s
1320.4
is noticed
from
may
small m/shydrogen
the
come
compared iswhile
from noticed
theitsound
with iswhile
1319.69
temperature
hydrogen ittemperature
speed. isThe
m/s
1319.69
and at 5000
density.
fluctuation m/srpm.
and at
of 5000
density. In speed
rail fact
rpm. the
ofInsound
difference
fact the difference
between
is 0.053% fromis between
very is very
engine
small compared
small atcompared
speed with
3000sound rpm.with speed.
This sound
implies The speed.
fluctuation
that The fluctuation
sound of rail
speed lessspeed of rail
depends of on sound
speed of
is 0.053%
the engine soundspeed. isfrom
0.053%
This engine
changefrom engine
speed at 3000
speed
may rpm.
comeat 3000 Thisthe
from rpm.
implies Thisthat
hydrogen implies
soundthat
temperature speed sound
and less speed
density. depends less on depends
the engine on the speed.
engine Thisspeed.change This change
may comemay
fromcome
the hydrogen
from the hydrogen
temperature
temperature
and density.
and density.

FIGURE
FIGURE 66
Rail speed of ofsound
Rail speed fordifferent
sound for different engine
engine

Figure 7 shows the instantaneous hydrogen mass flow rate in the crank angle domain through
Figurerail.
common 7 The
shows the between
interaction instantaneous hydrogen
the speeds curves mass flow
make distinguishing rate
between inquite
them the
crank However,
difficult. angle domain through
higher hydrogen common
flow rated rail. The
can be observed interaction
at 5000 between
rpm engine speed. The
rapid closing and opening of the injectors cause high fluctuation in mass flow rate throughout the
the speeds curves make distinguishing between them quite difficult.
However, higher hydrogen flowFIGURE
rated6 can be observed at 5000 rpm
Rail speed of sound for different engine

Figure 7 shows the instantaneous hydrogen mass flow rate in the crank angle domain through
46common rail. The interaction
ISSN: betweenFIGURE
2180-3811 the FIGURE
1 FIGURE
speeds
Vol. No.6 1 make
6 curves 6January-June
distinguishing between them quite
2010
difficult. However, higher RailRail
Railhydrogen
speed speed
speed of sounddifferent
of flowof
sound sound
for
rated for
different
can be engine
engine
observed
for at engine
different 5000 rpm engine speed. The
rapid closing and opening of the injectors cause high fluctuation in mass flow rate throughout the
 Modeling of Common Rail Fuel Injection System of Four Cylinder Hydrogen Fueled Engine

engine speed. The rapid closing and opening of the injectors cause
high fluctuation in mass flow rate throughout the rail. Reverse flow
(negative flow out) can be noticed in some parts of the cycle. The gas
dynamic effects distort the common rail and lead to the reverse flow
that pressure imbalance occurs between the accumulator (rail) and the
pipes of injector. A maximum mass flow of 60 mg/s rate is reported at
5000 rpm; while a minimum value of 29 mg/s (reverse flow) is obtained
at 3000 rpm.

Figure 8 shows the variation of injector mass flow rate through the
injector orifice with crank angle and speed. It is clearly shown that  36
the influence of the oscillations in rail pressure depends on injection
rate. The injected mass flow rate influence the several parameters
rail. Reverse flow
including (negative
the flow out)coefficient,
discharge can be noticed orifice
in some parts of the cycle.
diameter, The gas of
density dynamic
the
effects distort the common rail and lead to the reverse flow that pressure imbalance occurs between
fuel and difference between hydrogen pressure at orifice
the accumulator (rail) and the pipes of injector. A maximum mass flow of 60 mg/s rate is reported exit and
atport pressure.
5000 rpm; Hence value
while a minimum the offluctuations
29 mg/s (reverse inflow)
theisinjected
obtained at mass are due to
3000 rpm.
Figure 8 shows the variation
the pressure of injector
fluctuations mass flow
across therateinjector
through the injector The
orifice. orificevalue
with crank
of angle
the
and speed. It is clearly shown that the influence of the oscillations in rail pressure depends on
injector hole discharge coefficient is essential for accurate modeling
injection rate. The injected mass flow rate influence the several parameters including the discharge of
the injected mass flow rate. A discharge coefficient for injector hole of
coefficient, orifice diameter, density of the fuel and difference between hydrogen pressure at orifice
exit and is
0.765 portconsidered
pressure. Hence the fluctuations
throughout theinanalysis.
the injectedHigher
mass aremass due toflow
the pressure
rates
fluctuations across the injector orifice. The value of the injector hole discharge coefficient is
can be observed for higher speed. The mass flow rate fluctuations
essential for accurate modeling of the injected mass flow rate. A discharge coefficient for injector occur
in very
hole small
of 0.765 amplitudes
is considered (maxima
throughout of Higher
the analysis. 0.5 mg/s massat 5000
flow ratesrpm;
can bewhile
observedit for
is
higher speed. The mass flow rate fluctuations
0.344 mg/s at 3000 rpm with high frequencies. occur in very small amplitudes (maxima of 0.5 mg/s
at 5000 rpm; while it is 0.344 mg/s at 3000 rpm with high frequencies.

FIGURE
FIGURE 77
Rail Rail
Mass Mass
FlowFlowRate
Ratefor
for Different
Different Engine
Engine

ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010 47

  37
Journal of Engineering and Technology 

FIGURE
GURE 88
FIG
Injector
Injector Mass Mass
M Flow
FlowRate
R for
Rate forDiff
fferent Engin
Different ne
Engine
Figure
F 9 deppicts the varriation in preessure drop aacross the in njector orificce with crankk angle. Thiss is a
r Figure
relevant andd9 flexible
depicts thetovariation
facctor in pressure
improove hydrogen n penetrationdrop across
n and allows the injector
s effective m
mixing orifice
durinng the
diffusion
d with crankbuurning phase e. Again, the e pressure fl
luctuation is
s present
angle. This is a relevant and flexible factor to improve in the behavior r of the presssure
drop
d across the injector orifice. The similarity bbetween the two t curves (pressure
( waaves and presssure
d hydrogen
drop) emphaasizes penetration
that the
t behaviorand r ofallows
the latteereffective mixing
is a subseequent to theeduring
former. Lothe
owerdiffusion
speed 3000
burning
rpm
r phase.
gives higher
h Again, the
amplittude and low pressure fluctuation is present
wer frequenccy. This flucctuation musst be damped in the
d outbehavior
as mucch as
p of thetopressure
possible avoid suppldrop across
lying the thewith
cyylinders injector
h differentorifice.
q
quantitiesThe similarity
off hydrogen between
g Becaus
gas. se, as
stated earlieer, the injecteed hydrogenn mass depennds mainly on o pressure ddifference accross the injector
o the two
orifice. Thuuscurves (pressure
the fluctua vwavesam
ation gives variable and pressure
ounts of hyddrogen.drop) emphasizes
Thee maximum pressurethat
droopthe
of
2 behavior
2.632 of theatlatter
bar is obtained is compared
3000 rpm ca subsequent
wiith 2.626 tobar
the former.
r at 5000 rpmm. AsLower
a whoole,speed 3000
the puls ation
o rpm
of gives
hydrogen higher
n pressure w amplitude
waves has im andim
mportant loweronfrequency.
mpact injeection param This
meters.fluctuation
The effect is must
reelated
substantiallyy with enginne speed. Thiis imposes thhe engine sppeed parameeter take intoo account forr any
t be damped
tuning proceess foroutcomm as much
mon as possible
rail injeection systemm. Thetoeffec
avoid supplying
ct of engine the
speed variaationcylinders
seems to t be
u with different
unimportant t for somequantities
s
studied of hydrogen
paraameters. Thiis maygas. belon Because,
ngs as stated
to the iinteraction w earlier,
with variatioonthe
of
o injected
other parametershydrogen massinndepends
like thee changes mainly
hydrogen ddensity, onthe
so th
hat pressure
total iimpactdifference
b small.across
will be
the injector orifice. Thus the fluctuation gives variable amounts of
hydrogen. The maximum pressure drop of 2.632 bar is obtained at 3000
rpm compared with 2.626 bar at 5000 rpm. As a whole, the pulsation
of hydrogen pressure waves has important impact on injection
parameters. The effect is related substantially with engine speed. This
imposes the engine speed parameter take into account for any tuning
process for common rail injection system. The effect of engine speed
variation seems to be unimportant for some studied parameters. This
may belongs to the interaction with variation of other parameters like
the changes in hydrogen density, so that the total impact will be small.

48 ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010

 Modeling of Common Rail Fuel Injection System of Four Cylinder Hydrogen Fueled Engine
 38


FIGURE
FIGURE 99
PressurePressure
drop across injector
drop across forfordifferent
injector engine
different engine speeds
speeds

4.0 CONCLUSIONS

4.0 CONCLUSIONS
A numerical investigation of single injection system with CR hydrogen fueled engine has been
performed. The highly unsteady wave propagation phenomena taking place in the system play a
major role in the proper understanding of important injection-system characteristics. From the
Aanalysis
numerical investigation
of system of single
behavior with different injection
operation system with
and geometrical CR hydrogen
parameters, the following
fueled engine
conclusions are drawn:has been performed. The highly unsteady wave
propagation phenomena taking place in the system play a major role in
i. The proposed common rail injection system for hydrogen engine can play dramatic role in
the proper understanding of important injection-system characteristics.
developing port injection hydrogen engine; exactly similar to the extraordinary success for
From the
theCRanalysis of system
injection system behavior
in diesel engine. Thewith different
CR system operation
is responsible and
for the ever-
geometrical parameters,
increasing the
share (� 40%) of following
diesel engines in conclusions are drawn:
automotive market.

ii. Although low pressures (2-4 bars) have been investigated for the port injection system, the
i. The
gas proposed
dynamic common
and pressure rail
propagation injection
phenomena systemgasfor
for hydrogen fuelhydrogen
is still effective and
engine
has caninplay
major effect dramatic
the performance role
of the CRin developing
injection system. port injection

iii.
hydrogen engine; exactly similar to the extraordinary success
Hydrogen gas fuel CR injection system is considerably simpler than standard diesel fuel
for thesystem.
injection CR injection system
This is because inisdiesel
the fuel stored atengine. The inCR
high pressures system
a cylinder fuel tank
is no
and responsible
need for highfor the pump.
pressure ever-increasing share
However, the high (≅ velocity
sound 40%) of fordiesel
hydrogen fuel
leads to higher propagation speed pressure oscillations.
engines in automotive market.

ii. Although low pressures (2-4 bars) have been investigated

for the port injection system, the gas dynamic and pressure
propagation phenomena for hydrogen gas fuel is still
effective and has major effect in the performance of the CR
injection system.

iii. Hydrogen gas fuel CR injection system is considerably

simpler than standard diesel fuel injection system. This is
because the fuel is stored at high pressures in a cylinder
fuel tank and no need for high pressure pump. However,

ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010 49

Journal of Engineering and Technology

the high sound velocity for hydrogen fuel leads to higher

propagation speed pressure oscillations.

5.0 ACKNOWLEDGMENT

The authors would like to thank Universiti Malaysia Pahang for

provides laboratory facilities and financial support.

6.0 REFERENCES

A. Catalano, V.A. Tondolo, and A. Dadone. 2002. Dynamic rise of pressure in

the common-rail fuel injection system. SAE Paper No. 2002-01-0210.

A. Mulemane, J.S. Han, P.H. Lu, S.J. Yonn and M.C. Lai. 2004. Modeling
dynamic behavior of diesel fuel injection systems. SAE Paper No.
2004-01-0536.

A.E. Catania, A. Ferrari, M. Manno and E. Spessa. 2008a. Experimental

investigation of dynamics effects on multiple-injection common rail
system performance. Journal of Engineering for Gas Turbines and
Power, ASME. Volume 130. pp. 032806 (13 pages).

A.E. Catania, A. Ferrari, M. Manno. 2008b. Development and application of a

complete multijet common-rail injection-system mathematical model
for hydrodynamic analysis and diagnostics. Journal of Engineering for
Gas Turbines and Power, ASME. Volume 130. pp. 062809.

C.M. White, R.R. Steeper, and A.E. Lutz. 2006. The hydrogen-fueled internal
combustion engine: a technical review. International Journal of Hydrogen
Energy. Volume 31. pp. 1292–1305.

D.D. Torkzadeh, U. Kiencke and M. Keppler. 2002. Introduction of a new

non-invasive pressure sensor for common-rail systems. SAE Paper No.
2002-01-0842.

Ferguson, C.R. and Kirkpatrick, A.T. 2001. International combustion engines.

Second edition. New York: John Wiley and Sons, Inc.

G.M. Bianchi, S. Falfari, P. Pelloni, F. Filicori, and M. Milani. 2002. A numerical

and experimental study towards possible improvements of common
rail injectors. SAE Paper No. 2002-01-0500.

H. Larsen, R. Feidenhans’l and L. Sønderberg. 2004. Hydrogen and its

competitors. Report prepared for Risø National Laboratory. Risø National
Report 3, 2004.

50 ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010

 Modeling of Common Rail Fuel Injection System of Four Cylinder Hydrogen Fueled Engine

H.L. MacLean and L.B. Lave. 2003. Evaluating automobile fuel/propulsion

system technologies. Progress in Energy and Combustion Science. Volume
29. pp. 1-69.

J. John and T. Keith. 2006. Gas Dynamics. Third edition. UK: Pearson Education,
Inc.

N.A. Henein, M.C. Lai, I.P. Singh, L. Zhong and J. Han. 2002. Characteristics
of a common rail diesel injection system under pilot and post injection
modes. SAE Paper No. 2002-01-0218.

ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010 51

Journal of Engineering and Technology

52 ISSN: 2180-3811 Vol. 1 No. 1 January-June 2010