With increasing concern about energy shortage and environmental protection,
research on improving engine fuel economy and reducing exhaust emissions has
become the major researching aspect in combustion and engine development. Due
to limited reserves of crude oil, development of alternative fuel engines has
attracted more and more concern in the engine community. Alternative fuels usually
belong to clean fuels compared to diesel fuel and gasoline fuel in the combustion
process of engines. The introduction of these alternative fuels is beneficial
to slowing down the fuel shortage and reducing engine exhaust emissions. Hydrogen
fuel is regarded as one of the most promising alternative fuels for automobiles
in future. Technology of the optimum control on hydrogen-fueled engines is a
key to improve its performances in every respect. An important issue with energy
usage is the associated undesirable emissions. High flame speed leading to good
thermal efficiency, wide flammability limits, absence of carbon based emissions,
qualitative mixture control and high diffusivity leading to good mixing are
some of advantages of hydrogen. Hydrogen induction techniques play a very dominant
and sensitive role in determining the performance characteristics of the hydrogen
fueled internal combustion engine (Suwanchotchoung, 2003).
Hydrogen fuel delivery system can be broken down into three main types including
the carbureted injection, Port Fuel Injection (PFI) and Direct Injection (DI)
(COD, 2001). In direct injection, the intake valve is
closed when the fuel is injected into the combustion cylinder during the compression
stroke (COD, 2001). Like PFI, direct injection has long
been viewed as one of the most attractive choices for supplying hydrogen fuel
to combustion chamber (White et al., 2006; Verhelst
et al., 2006; Zhenzhong et al., 2002;
Mohammadi et al., 2007; Guo
et al., 1999; Jorach et al., 1997;
Kim et al., 1995). This view is based on: its
prevention for abnormal combustion: pre-ignition, backfire and knock; and the
high volumetric efficiency, (since hydrogen is injected after intake valve closing).
The improved volumetric efficiency and the higher heat of combustion of hydrogen
compared to gasoline, provides the potential for power density to be approximately
115% that of the identical engine operated on gasoline (White
et al., 2006). However, it is worthy to emphasize that while direct
injection solves the problem of pre-ignition in the intake manifold, it does
not necessarily prevent pre-ignition within the combustion chamber (COD,
2001). In fact the difficulties and limitations accompanied with DI are
more serious and severe than those of PFI. Direct injection during the compression
stroke needs high pressure hydrogen and thus effectively requires liquid hydrogen
storage. Metal hydrides can only provide low pressure hydrogen, compressed hydrogen
could be used but this limits the effective tank contents as the tank can only
be emptied down to the fuel injection pressure. Compressing gaseous hydrogen
on board would mean an extra compressor and a substantial energy demand (Verhelst,
2005). Furthermore, a high-pressure, high flow-rate hydrogen injector is
required for operation at high engine speeds and to overcome the in-cylinder
pressure for injection late in the compression stroke. The high pressure was
defined by White et al. (2006) as greater than
80 bar to ensure sonic injection velocities and high enough mass flow rates
for Start of Injection (SOI) throughout the compression stroke. The need for
rapid mixing necessitates the use of critical flow injectors and the short time
duration with late injection requires high mass flow rates. The valve leakage
at the valve seat and the losses associated with the injection system are another
issues (Kim et al., 1995; Tsujimura
et al., 2003; Kim et al., 2006). Guo
et al. (1999) have kept the injector in a status such that it is
always not under a high pressure, so pre-ignition caused by the injector's leakage
at initial stage of starting the engine was avoided. While, a seal made of an
elastomer material has been used with success to prevent valve leakage at the
valve seat (Homan et al., 1983; Green
and Glasson, 1992). It is apparent that the structure of DI system is more
sophisticated, expensive and attend great durability problem (COD,
2001; Stockhausen et al., 2002; Yi
et al., 1996).
Another important challenge for DI is the extremely short time for hydrogen-air
mixing. For early injection (i.e., coincident with Inlet Valve Closure (IVC)
maximum available mixing times range from approximately 20-4 msec across the
speed range 1000-5000 rpm, respectively (White et al.,
2006). This insufficient time leads to unstable engine operation at low
hydrogen-air equivalence ratios due to insufficient mixing between hydrogen
and air (Rottengruber et al., 2004). As an attempt
to fix this problem, Guo et al. (1999) used a
fast response high pressure solenoid valve to improve hydrogen jet penetration
and mixture formation in the combustion chamber and to prevent backfire occurring
in the hydrogen supply pipe between the valve and the combustion chamber. Jorach
et al. (1997) suggested early injection to provide more time for
mixing process. However, in practice, to avoid preignition, SOI is retarded
with respect to IVC and mixing times are further reduced (White
et al., 2006).
Among the subsequent problems of the inadequate mixing time for DI system,
is the unacceptable high level of NOx emissions. The low grade of homogenization
is responsible for forming rich areas in the combustion chamber. The reaction
temperatures in these rich areas may rise up more than 2300 K (Jorach
et al., 1997). Several researchers have tried to surmount this problem
via proper adjusting for injection time. Late injection, as a solution, was
investigated by Mohammadi et al. (2007) and
Jorach et al. (1997). However, this measure is
insufficient and the system will be susceptible for pre-ignition as stated above.
Therefore, additional transactions like utilization of other techniques such
as EGR and after-treatment methods are required to bring the NOx emission to
acceptable level (Mohammadi et al., 2007).
As a whole, both PFI and DI have their advantages and disadvantages. DI is
better for full load performance (maximum power output), PFI is better at part
load (maximum engine efficiency) (Verhelst, 2005; Verhelst
et al., 2006). Some designs proposed utilizing dual-injection (both
of PFI and DI) in the same engine (Kim et al., 2006;
Yi et al., 2000; Blair, 1999).
The dual-injection strategy was suggested to take advantage of the high thermal
efficiencies at low and medium loads with PFI system and the high power output
with DI system. (White et al., 2006). Kim
et al. (2006) introduced the following strategy: using PFI only under
idling and low load because no backfire occurs. For the case of high load, most
of the fuel is injected directly into the cylinder during the compression process
and the rest, which guarantees that the intake mixture is lean enough so, that
no backfire occurs, is supplied into the intake pipe to increase the mixing
rate. Excellent results were reported, such that the maximum torque of the dual-injection
was increased by about 60% compared to a hydrogen engine using external mixture
preparation and the brake thermal efficiency was higher by about 22% at low
load compared with direct-cylinder injection hydrogen engine. The objectives
of this study are to investigate the effect of air fuel ratio on engine performance
and instantaneous behavior of intake, exhaust port pressure and cylinder pressure
on the crank angle of the direct injection hydrogen fueled engine.
MATERIALS AND METHODS
This study was conducted at high computing laboratory, Automotive Excellence Centre, Faculty of Mechanical Engineering, Universiti Malaysia Pahang, Kuantan in 2008.
Hydrogen engine modeling
Engine performance parameters: The Brake Mean Effective Pressure (BMEP)
can be defined as the ratio of the brake work per cycle Wb to the
cylinder volume displaced per cycle Vd and it can be expressed as
in Eq. 1 (Heywood, 1988):
Equation 1 can be rewrite for the four stroke engine as in Eq. 2:
where, Pb is the brake power and N is the rotational speed.
Brake efficiency (ηb) can be defined as the ratio of the brake power Pb to the engine fuel energy as in Eq. 3:
is the fuel mass flow rate and LHV is the lower heating value of hydrogen.
The Brake Specific Fuel Consumption (BSFC) represents the fuel flow rate
per unit brake power output and can be expressed as in Eq. 4
The volumetric efficiency (ηv) of the engine defines as the
mass of air supplied through the intake valve during the intake period ()
by comparison with a reference mass, which is that mass required to perfectly
fill the swept volume under the prevailing atmospheric conditions and can be
expressed as in Eq. 5:
where, ρai is the inlet air density.
The burning rate (Xb) of combustion process was modeled using Wiebe function, which can be expressed as Eq. 6:
where, θ is the crank angle, θi is the start of combustion, Δθ is the combustion period and a and n are adjustable constants.
Furthermore, the heat transfer in-side the cylinder was modeled using a formula
which is closely emulates the classical Woschni correlation. Based on this correlation,
the heat transfer coefficient hc can be expressed as Eq.
hc = 3.26B-0.2 p0.8
where, B is the bore in meters, p is the pressure in kPa, T is temperature in K and w is the average cylinder gas velocity in m sec-1.
The hydrogen gas fuel was injected directly in-side the cylinders using the
four sequential pulse fuel injectors. The AFR was imposed for the injectors.
Then, the injected fuel rate was estimated using the Eq. 8
(Ferguson and Kirkpatrick, 2001):
is the injector delivery rate (g sec-1),
the reference density used to calculate volumetric efficiency (kg m-3),
FAR is the fuel air ratio and Pw is the injection duration (°CA).
The four cylinders were then connected together through the engine part which translates the force acting on each piston into the crankshaft (brake) power. In the engine model: engine type was 4-stroke type; the number of cylinders is set to four; the configuration inline had been chosen; and simulation with prescribed engine speed was specified rather than engine load. Furthermore, engine friction model was imposed to model friction in the engine. The Friction Mean Effective Pressure (FMEP) was modeled based on Eq. 9:
where, Speedmp represents the mean piston speed and Pmax is the peak cylinder pressure.
Engine model: The engine model for an in-line 4-cylinder direct injection
engine was developed for this study. Engine specifications for the base engine
are tabulated in Table 1. The specific values of input parameters
including the AFR, engine speed and injection timing were defined in the model.
The boundary condition of the intake air was defined first in the entrance of
the engine. The air enters through a bell-mouth orifice to the pipe. The discharge
coefficients of the bell-mouth orifice were set to 1 to ensure the smooth transition
as in the real engine. The pipe of bell-mouth orifice with 0.07 m of diameter
and 0.1 m of length are used in this model. The pipe connects in the intake
to the air cleaner with 0.16 m of diameter and 0.25 m of length was modeled.
The air cleaner pipe identical to the bell-mouth orifice connects to the manifold.
A log style manifold was developed from a series of pipes and flow-splits. The
intake system of the present study model is shown in Fig. 1.
The total volume for each flow-split was 256 cm3. The flow-splits
compose from an intake and two discharges. The intake draws air from the preceding
flow-split. One discharge supplies air to adjacent intake runner and the other
supplies air to the next flow-split. The last discharge pipe was closed with
a cup to prevent any flow through it because there is no more flow-split. The
flow-splits are connected with each other via pipes with 0.09 m diameter and
0.92 m length. The junctions between the flow-splits and the intake runners
were modeled with bell-mouth orifices. The discharge coefficients were also
set to 1 to assure smooth transition, because in most manifolds the transition
from the manifold to the runners is very smooth. The intake runners for the
four cylinders were modeled as four identical pipes with. 04 m diameter and
0.1 m length. Finally the intake runners were linked to the intake ports which
were modeled as pipes with 0.04 m diameter and 0.08 length. The air mass flow
rate in e intake port was used for hydrogen flow rate based on the imposed AFR.
The second major part of the engine model is the powertrain model which is
shown in Fig. 2. In the powertrain, the induced air passes
through the intake cam-driven type valves with 45.5 mm of diameter to the cylinders.
|| Engine specification
|| Intake system model
The valve lash (mechanical clearance between the cam lobe and the valve stem)
was set to 0.1 mm. The overall temperature of the head, piston and cylinder
for the engine parts are listed in Table 2.
|| Powertrain model
of the piston is higher than the cylinder head and cylinder block wall temperature
because this part is not directly cooled by the cooling liquid or oil.
The last major part in the present model is the exhaust system which is shown
in Fig. 3. The exhaust runners were modeled as rounded pipes
with 0.03 m inlet diameter and 800 bending angle for runners 1 and
4; and 40° bending angle of runners 2 and 3. Runners 1 and 4 and runners
2 and 3 are connected before enter in a flow-split with 169.646 cm3 volume.
Conservation of momentum is solved in 3-dimentional flow-splits even though
the flow in GT-Power is otherwise based on a one-dimensional version of the
Navier-Stokes equation. Finally, a pipe with 0.06 m diameter and 0.15 m length
connects the last flow-split to the environment. Exhaust system walls temperature
was calculated using a model embodied in each pipe and flow-split. Table
3 are listed the parameters used in the exhaust environment of the model.
|| Temperature of the mail engine parts
|| Parameters used in the exhaust environment
|| Exhaust system model
It is worthy to mention that one of the most attractive combustive features for hydrogen fuel is its wide range of flammability. A lean mixture is one in which the amount of fuel is less than stoichiometric mixture. This leads to fairly easy to get an engine start. Furthermore, the combustion reaction will be more complete. Additionally, the final combustion temperature is lower reducing the amount of pollutants. The air-fuel ratio AFR was varied from rich limit (AFR = 27.464:1 based on mass where the equivalence ratio (φ = 1.2) to a very lean limit (AFR =171.65 where (φ = 0.2) and engine speed varied from 2500 to 4500 rpm. BMEP is a good parameter for comparing engines with regard to design due to its independent on the engine size and speed.
Variation in air fuel ratio on engine performance: Figure 4 shows the effect of air-fuel ratio on the brake mean effective pressure. It can be seen that BMEP decreases with increases of AFR and speed. This decrease happens with two different behaviors. For rich mixtures (low AFR), BMEF decreases almost linearly, then BMEP falls with a non-linear behavior. Higher linear range can be recognized for higher speeds. For 4500 rpm, the linear range is continuing until AFR of 42.9125 (φ =0.8). The non-linear region becomes more predominant at lower speeds and the linear region cannot be specified there. The total drop of BMEP within the studied range of AFR was 8.08 bar for 4500 rpm compared with 10.91 bar for 2500 rpm. At lean operating conditions (AFR = 171.65, (φ = 0.2 the engine gives maximum power (BMEP = 1.635 bar) at lower speed 2500 rpm) compared with the power (BMEP = 0.24 bar) at speed 4500 rpm.
Figure 5 shows the variation of the brake thermal efficiency with the air fuel ratio for the selected speeds. Brake power is the useful part as a percentage from the intake fuel energy. The fuel energy is also covered the friction losses and heat losses (heat loss to surroundings, exhaust, enthalpy and coolant load). Therefore, lower values of ηb can be seen in Fig. 5. It can be observed that the brake thermal efficiency is increases nearby the richest condition (AFR ≅ 35) and then decreases with increases of AFR and speed. The operation within a range of AFR from 38.144 to 42.91250 (φ = 0.9-0.8) gives the maximum values for ηb for all speeds. Maximum ηb of 35.4% at speed 2500 rpm can be seen compared with 26.3% at speed 4500 rpm. Unaccepted efficiency ηb of 3.7% can be seen at very lean conditions with AFR of 171.65 (φ = 0.2 for speed of 4500 rpm, while a value of 23.86% was recorded at the same conditions with speed of 2500 rpm. Clearly, rotational speed has a major effect in the behavior of ηb with AFR. Higher speeds lead to higher friction losses.
Figure 6 depicts the behavior of the brake specific fuel consumption BSFC with AFR. It is easy to perceive from the Fig. 6 that there is an optimum minimum value of BSFC occurred within a range of AFR from 38.144 (φ = 0.9) to 49.0428 (φ = 0.7) for the selected range of speed. At very lean conditions, higher fuel consumption can be noticed. After AFR of 114.433 (φ = 0.3) the BSFC rises up rapidly, especially for high speeds. At very lean conditions with AFR of 171.65 (φ = 0.2), a BSFC of 125.87 g kW h-1 was observed for the speed of 2500 rpm; while it was 809 g kW h-1 for 4500 rpm.
Figure 7 shows how the AFR can affect the maximum temperature
inside the cylinder. In general, lower temperatures are required due to the
reduction of pollutants. It is clearly demonstrated how the increase in the
AFR can decrease the maximum cylinder temperature with a severe steeped curve.
||Variation of brake mean effective pressure with air fuel ratio
for various engine speeds
||Variation of brake thermal efficiency with air fuel ratio
||Variation of brake specific fuel consumption with air fuel
ratio for different engine speed
||Variation of maximum cylinder temperature with air fuel ratio
But for rich mixtures, the maximum cylinder temperature drops down with a linear
manner. The effect of the engine speed on the relationship between maximum cylinder
temperatures with AFR seems to be minor. At rich operating conditions (AFR =
27.464, φ = 1.2) and a speed of 3000 rpm, a maximum cylinder temperature
of 2767 K was recorded. This temperature dropped down to 1345 K at AFR of 171.65
(φ = 0.2). This lower temperature inhibits the formation of NOx
pollutants. In fact this feature is one of the major motivations toward hydrogen
Instantaneous behaviour on crank angle: The intake port and exhaust
port pressures in terms of crank angle are shown in Fig. 8
and 9, respectively. The gas dynamic effects play a very important
rule here. It distorts the exhaust flow which is shown in Fig.
||Instantaneous intake port pressure distributions with crank
angle for different speed
||Instantaneous exhaust port pressure distributions with crank
angle for various engine speeds
The rise of the pressure at the end of the exhaust stroke can lead to
reverse flow into the cylinder past the exhaust valve; however, the high vacuum
in the beginning of the first stroke is highly desired to banish the burnt gases
out of the cylinder. At speed of 3000 rpm, a maximum pressure of 1.64 bar and maximum vacuum of
0.72 bar were recorded. The response of fluctuation of the amplitude to the
engine speed in case of exhaust pressure seems to be less than the intake pressure.
But the fluctuation is also increasing with the increase of the engine speed.
Figure 10 shows the behavior of the cylinder pressure at
the last cycle (12th cycle) for WOT and stoichiometric operation conditions.
The behavior of the pressure follows the combustion phenomenon that occurs.
The effect of the rotational speed on the instantaneous behavior of the cylinder
pressure is minor.
||Instantaneous cylinder pressure distributions with crank angle
for various engine speed
The maximum pressure has been observed at engine speed of 2500 rpm however the minimum pressure was obtained at 4500 rpm.
It has been adequately emphasized that hydrogen fuel possesses some properties
which are uniquely different from the corresponding properties of conventional
hydrocarbon fuels. This was primarily the reason why initially the research
and development work on. The symptoms of unsteady combustion the most pronounced
effect in an internal combustion engine. Hydrogen has long since been attempted
as a fuel for the internal combustion engine. In general, it is desirable to
have maximum volumetric efficiency for engine. The importance of efficiency
is more critical for hydrogen engines because of the hydrogen fuel displaces
large amount of incoming air due to its low density (0.0824 kg m-3
at 25°C and 1 atm.). This reason reduces the efficiency to high extent.
A stoichiometric mixture of hydrogen and air consists of approximately 30% hydrogen
by volume, whereas a stoichiometric mixture of fully vaporized gasoline and
air consists of approximately 2% gasoline by volume (White
et al., 2006). Therefore, the low efficiency for hydrogen engine
is expected compared to gasoline engine works with same operating conditions
and physical dimension. However, the higher efficiencies can be gained with
direct injection of hydrogen, which can be shown in Fig. 5.
The maximum value of efficiency for the selected range of speed was around 85%.
At further higher engine speed beyond these values, the flow into the engine
during at least part of the intake process becomes chocked. Once this condition
occurs, further increases in engine speed decrease the flow rate significantly.
Thus, the efficiency decreases sharply because of the higher speed is accompanied
by some phenomenon that have negative influence on efficiency. These phenomenon
include the charge heating in the manifold and higher friction flow losses which
increase as the square of engine speed. Due to dissociation at high temperatures
following combustion, molecular oxygen is present in the burned gases under
stoichiometric conditions. Thus, some additional fuel can be added and partially
burned. This increases the temperature and the number of moles of the burned
gases in the cylinder. These effects increases the pressure were given increase
power and mean effective pressure (Ferguson and Kirkpatrick,
The AFR for optimum fuel consumption at a given load depends on the details
of chamber design (including compression ratio) and mixture preparation quality.
It varies for a given chamber with the part of throttle load and speed range
(Ferguson and Kirkpatrick, 2001). It is clearly seen (Fig.
6) that the higher fuel is consumed at higher speeds due to the greater
friction losses that can occur at high speeds. The value BSFC at speed of 2500
rpm was doubled around two times at speed of 4000 rpm; however the same value
was doubled around five times at speed of 4500 rpm. This is because of very
lean operation conditions can lead to unstable combustion and more lost power
due to a reduction in the volumetric heating value of the air/hydrogen mixture.
The instantaneous behavior is at the 12th cycle for Wide Open Throttle (WOT)
and stoichiometric operation. These Fig. 8 and 9
are very important to investigate the backfire or pre-ignition occurrence in
details. However, for the present case there is neither backfire nor pre-ignition
and this is the case of normal combustion and shows typical results of pressure
variation. The crank angle axis is divided into four parts to indicate the four
strokes which take two cycles (720 degrees). The pressure seems to be a series
of pulses. Each pulse is approximately sinusoidal in shape. The complexity of
the phenomena that occur is apparent. Back flow from the cylinder into the intake
manifold can be recognized during the early part of the intake process until
the cylinder pressure falls below the manifold pressure. This happens within
about 40crank angle degrees and stops when the angle crank reaches 400 degree
from the life cycle. Backflow also occur early in the compression stroke before
the inlet valve closed due to rising cylinder pressure. The amplitude of the
pressure fluctuations increases substantially with increasing engine speed.
From Fig. 8, the maximum intake pressure was recorded 1.1
bar at speed 4500 rpm during the compression stroke, while it was 1.093 bar
at speed of 2500 rpm. At the intake stroke, when high intake vacuum is occurred,
the flow is continuously inward and flow pulsation is small. For high speed,
larger pulses can be seen. At high speeds more fuel is required and consequently
more vacuum in the intake port. A vacuum of 0.792 bar was calculated in 4500
rpm compared with 0.925 bar at 2500 rpm.
The behavior of the pressure follows the combustion phenomenon that occurs. The effect of the rotational speed on the instantaneous behavior of the cylinder pressure is minor. This curve can be divided into three parts for discussion purpose. The first part corresponds the flame development period which consumes about 5% of the air fuel mixture. Very little pressure rise is noticeable and little or no useful work is produced. The second part corresponds the flame propagation period which consumes about 90% of the mixture. During this time, pressure in the cylinder is greatly increased, providing the force to produce work in the expansion stroke. The third part corresponds to flame termination period which consumes about the rest of the mixture (5%). In general this behavior is like the behavior of the traditional gasoline fuel, however it is necessary to keep in mind that during the hydrogen combustion, the flame velocity is rapid and the main changes of cylinder pressure (the second part) occur in a shorter time. Whilst experimental data are not available to verify these predictions, the authors are presented here to illustrate some of the insights that this type of simulation tool may provide to future engine systems designers.
The present study considered the performance characteristics of a four cylinders hydrogen fueled internal combustion engine with hydrogen being injected directly in the cylinder. The following conclusions are drawn:
||At very lean conditions with low engine speeds, acceptable
BMEP can be reached, while it is unacceptable for higher speeds. Lean operation
leads to small values of BMEP compared with rich conditions
||Maximum brake thermal efficiency can be reached at mixture
composition in the range of (φ = 0.9 to 0.8) and it decreases dramatically
at leaner conditions
||The desired minimum BSFC occurs within a mixture composition
range of (φ = 0.7 to 0.9). The operation with very lean condition (φ<0.2)
and high engine speeds (>4500) consumes unacceptable amounts of fuel
||Lean operation conditions results in lower maximum cylinder
temperature. A reduction of around 1400 K can be gained if the engine works
properly at (φ<0.2) instead of stoichiometric operation
||Hydrogen combustion results in moderate pressures in the cylinder.
This reduces the compactness required in the construction of the engine.
But, if abnormal combustion like pre-ignition or backfire happens, higher
pressures may destroy the connecting rod and piston rings. Therefore, much
care should be paid for this point
The authors would like to express their deep gratitude to Universiti Malaysia Pahang (UMP) for provided the laboratory facilities and financial support under project No. RDU 08-05-074.