Application of CFD For Analysis and Design of IC Engines

Application of CFD for Analysis
and Design of IC Engines
Vijayashree and V. Ganesan
Abstract Most of us are not familiar with the concept that an internal combustion
(IC) engine is working on a four-stroke six-event principle. The six events are
suction, compression, combustion, expansion, blow-down and exhaust. However,
the expansion and blow-down happen in the power stroke and they can be clubbed
together. Therefore, we can say that an engine, whether diesel or gasoline, is
working on four-stroke five-event principle. The purpose of this chapter is to make
the reader to familiarize with the complexities involved in the working of a
four-stroke engine. The five events which are completed in four strokes are: suction,
compression, combustion, expansion and exhaust. Application of computational
fluid dynamics (CFD) principles for each process mentioned above is a challenging
job. The difficulty in understanding the working of an IC engine is due to the fact
that we cannot see what is happening inside the cylinder piston arrangement. All
that is described in textbooks is based on our knowledge gained over a period of
time by conducting experiments. There is no doubt in saying that “seeing is
believing”. As it is next to impossible to have a complete experimental flow
visualization, nowadays CFD comes in handy to have theoretical flow visualization.
Well-developed software is available for the simulation in 3D geometries. In this
chapter, we explain step-by-step the details required for the CFD simulation of
various processes in an IC engine. Extensive results obtained over a period of
20 years of research by the application of CFD in analysing the flow in engines are
comprehensively presented and discussed. CFD can be very well applied for ana-
lysing any particular process. It can also be used for the modification of the existing
engine design or can also be employed for a new design of an engine. It is hoped
that readers may be benefitted in understanding the application of CFD for fluid
flow analysis and engine design by reading this chapter. Therefore, the main aim of
this chapter is to make the reader appreciate how exactly CFD can be applied for
Vijayashree V. Ganesan (&)

Department of Mechanical Engineering, Indian Institute of Technology Madras,
Chennai 600 036, India
e-mail: vganesan.iitm@gmail.com
Vijayashree
e-mail: vijaya.shree.iitm@gmail.com
© Springer Nature Singapore Pte Ltd. 2018 251

D. K. Srivastava et al. (eds.), Advances in Internal Combustion Engine Research,
Energy, Environment, and Sustainability,
https://doi.org/10.1007/978-981-10-7575-9_13
252 Vijayashree and V. Ganesan
design of an engine. As it is application oriented, we are not going deep into the
equations, modelling, etc. A number of case studies are presented and discussed.
1 Introduction
Engineers always look for improvement in the design and manufacture of internal
combustion (IC) engines because there is tremendous competition and pressure in
achieving higher performance with lower and lower emissions. The next generation
of engines needs to be compact, flexible, light, and powerful. At the same time, they
should discharge minimum amount of pollutants and use less fuel. This is a
conflicting requirement. To achieve this, innovative engine design is a must to meet
these competing requirements. The ability to accurately and rapidly analyse the
performance of multiple engine designs is very critical. This chapter explains in
detail the various processes in an IC engine, and the results are presented as case
studies at appropriate places. Details regarding these aspects are organized into the
following sections in this chapter.
2 Engine Performance
The performance of an IC engine depends upon complex interactions between

chemical, electronic, fluid, and mechanical systems. However, the major challenge
in design is the complex fluid dynamics of turbulent reacting flows interacting with
moving parts through the intake/exhaust manifolds, valves, cylinder, and piston.
The timescales of the intake air flow, fuel injection, vaporization of fuel, turbulent
mixing process, species transport, chemistry, and pollutant formation all overlap
and need to be considered simultaneously.
Computational fluid dynamics (CFD) has emerged as a useful and powerful tool
in understanding the fluid dynamics of IC engines for design purposes. This is due
to the fact that, compared to analytical, experimental, or lower dimensional com-
putational methods, multidimensional CFD modelling allows designers to simulate
and visualize the complex fluid dynamics involved. This has become possible due
to the availability of superfast digital computers in solving the governing physical
laws for mass, momentum, and energy transport on a 3D geometry, along with
submodels for critical phenomena such as turbulence and fuel chemistry. Insight
provided by CFD analysis helps in design of appropriate geometry, such as ports,
valves, and pistons, as well as engine parameters such as valve timing and fuel
injection.
It should be noted that engine analysis using CFD software has always been
hampered due to the inherent complexity in:
Application of CFD for Analysis and Design of IC Engines 253
• creating a computational mesh in both the moving and non-moving portions of

the domain;
• decomposing the geometry into a topology which should exactly duplicate the
piston motion;
• specifying the motion of the valves;
• solving the unsteady equations for flow, turbulence, energy and chemistry;
• post-processing of results for extracting useful information from very large
number of data sets.
Those who are contemplating to venture on CFD studies on engines should be
aware that this is a time-consuming and error-prone process. Further, it can create a
significant impediment to rapid engine analysis and design feedback. The solution
to this problem is an integrated approach and environment specifically tailored to
the needs of modelling the IC engine. The environment requirements should have:
• minimal inputs from the user;
• the necessary tools to automatically perform a problem set-up;
• the ability to transfer information very rapidly between the different stages of the
CFD analysis;
• the capability to significantly compress the set-up and analysis process;
• minimal loss in the accuracy;
• minimal errors.
The IC Engine Analysis System should provide such an integrated environment
with the capabilities of integrating the setup of most IC engine designs. Further, it
should include:
• bidirectional CAD connectivity to mainstream CAD systems;
• powerful geometry modelling tools in design mode;
• flexible meshing;
• solution using appropriate software;
• powerful post-processing;
• persistent parameterization and design exploration;
• provision to allow users to modify geometry or problem set-up parameters; and
• provision to automatically regenerate analysis results.
Over a period of last one decade, the time taken for geometry adaptation, mesh
generation, and solution convergence has been reduced considerably from several
hours of work to minutes, with minimum errors. The user should specify the engine
parameters and geometry at the beginning of setting up the problem, instead of at
the solution stage.
The next few sections we will introduce the tools in the IC Engine Design and
Analysis System. A deeper examination of the fluid dynamics issues in IC engines
and the CFD modelling process will be carried out first, followed by details of the
IC engine port flow analysis, using STAR-CD platform.
3 Engine Design
Efficient design of an IC engine requires a number of critical decisions to establish

the impact of fluid dynamics on the overall performance of the engine. The major
decisions on the design involve in taking care of the following:
• the type of engine and its specifications;
• peak power to be developed at the design speed;
• the number of cylinders;
• the type of fuel and its emissions characteristics;
• the cubic capacity of the engine;
• overall “packaging” of the engine including all the subsystems.
In future, electronics may play an important role. However, at present,
mechanical systems, especially the port design using different cam configurations
and analysing the flow through them, are quite important. For this, there may be
further specifications, such as the output of the engine at idling speed or at low
RPM.
These design decisions are vital for the evaluation of air–fuel ratio needed by the
engine at that particular operating conditions. This will lead to a number of design
decisions in order to achieve the maximum overall efficiency of the engine. The
power output in terms of various efficiencies is given by the following equation:
gi gc gm gv qi Vd N
pb ¼ ð1Þ
nFQ
where
gi is the indicated thermal efficiency
gc is the combustion efficiency
gm is the mechanical efficiency
gv is the volumetric efficiency of the engine
qi is the density of the air at the intake
Vd is the engine displacement volume
N is the rotational speed
N is the number of revolution per power stroke (n = 2 for 4-stroke and 1 for
2-stroke engine)
F is the fuel air ratio
Q is the calorific value of the fuel per unit mass
• The important point to be taken into account in port design is to achieve

maximum value for each of the efficiency mentioned above. This is required to
obtain maximum power from the minimum fuel input. Of the various efficien-
cies mentioned in Eq. (1), fluid dynamics plays a vital role in the volumetric and
combustion efficiency.
• The secondary goal of engine design is to meet emissions standards. These are
always imposed by regulations. The major pollutants include three oxides, viz.
COx, NOx and SOx, where COx, refers to oxides of carbon [CO, CO2], NOx
refers to oxides of nitrogen [NO, NO2], and SOx, refers to oxides of sulphur
[SO2, SO3]. Further there are unburned hydrocarbons (HC), and polyaromatic
hydrocarbons (PAH or “soot”), which are all products of combustion.
• Pollutants are formed inside the engine due to intense interactions of the
mechanical and chemical processes. They are intimately associated with the
fluid dynamics inside the combustion chamber. With after treatment, the pol-
lutants in the exhaust stream can be considerably reduced. However, it becomes
very costly. Therefore, it is desirable to minimize the pollutant formation at the
source itself. In this, fluid dynamics in the port plays a vital role.
4 Fluid Dynamics During the Four Strokes
Achieving the maximum volumetric efficiency is one of the most important factors
in engine design. It depends on several fluid dynamic phenomena in the intake and
exhaust tracts leading to the combustion chamber. This is the subject matter of this
chapter.
• When the air is sucked into the cylinder during the intake cycle, it passes
through the gap between the intake valve and the valve seat. As it squeezes
through the gap, the flow separation takes place from the walls of the port and
valve surfaces, forming a tangential jet. The jet coming out of the gap may
impinge on the cylinder walls and tumble into the space between the valves and
the piston. This jet imparts angular momentum known as swirl and tumble, to
the fresh charge.
• Swirl is usually defined by a normalized angular momentum value about the
vertical axis through which the piston motion is constrained. Swirl can be
imparted to the incoming air by the proper design of the intake port. This is
called suction swirl.
• The gross motions of the fresh charge in the cylinder form recirculation regions
that enhance mixing. If there is strong swirl, the flow may develop stratification
with regions of high and low velocities. The intake port should be designed to
impart additional angular momentum to the air.
• Squish happens during the last part of the compression stroke where there is a
narrow region between the piston and the cylinder head, and the air may be
squeezed (or “squished”) from the sides of the cylinder into the combustion
chamber. This converts the energy in the swirl into turbulence. Further, it is to
be noted that when the piston travels towards the top during the compression
stroke, most of the energy contained in the tumble (or angular momentum
orthogonal to the swirl axis) of the jet is converted to turbulence as the available
space in the vertical direction is reduced significantly. The swirl will become
stronger as the air is squeezed out to the side. Flow phenomena that affect
volumetric efficiency are:
– flow separation, jet formation and reattachment on the cylinder wall/head;
– mixing due to swirl and tumble in the cylinder volume;
– turbulence production during the compression of air due to squish;
– flow stratification in the cylinder.
Engines that utilize carburetion or port fuel injection (PFI) are known as pre-
mixed engines. If an engine uses a carburettor, then the air and fuel are mixed in the
carburettor and the fuel–air mixture enters the intake manifold. In this type of
engine, at least from the fluid mechanical point of view, a mixture of fresh air and
fuel is inducted into the engine through the intake port. For a port fuel-injected
engine, the fuel is sprayed into the ports. Usually, spraying takes place onto the
back of the intake valve, where it vaporizes and mixes with the intake air.
In all modern direct injection diesel (DI diesel) engines, high-pressure fuel is
injected directly into the combustion chamber as the piston nears the end of the
compression stroke. The liquid fuel spray breaks up into finer droplets and
vaporizes into the surrounding air taking the latent heat of vapourization from air.
High pressure of the fuel spray enhances break-up, and high levels of turbulence in
the cylinder enhance the mixing. In any engine, proper charge motion at the start of
the combustion process is essential for efficient burning of the mixture. However,
often some compromises are needed when an engine operates in the range of speeds
as in the case of automotive engines.
In a spark ignition (SI) engine, a flame front is formed which travels outwards
from the ignition point, consuming the available fuel air mixture. Turbulence again
plays an important role in flame propagation, since the flame travels at the turbulent
flame speed. Hence, if the turbulence levels are high, then the flame front will move
very fast to all parts of the combustion chamber. In case of SI engines, the rapid
flame propagation avoids knock due to autoignition of fuel–air mixture ahead of the
flame. The flame speed depends on the air–fuel ratio of the mixture. If the mixture is
outside of the flammability limits, usually between equivalence ratios of less than
0.5 and greater than 4, the flame will not propagate. The engine starts misfiring.
Similarly, if regions exist inside the cylinder that are outside of the flammability
limits, these regions will not burn. The unburnt mixture will most likely be pushed
out through the exhaust and into the atmosphere.
In compression ignition (CI) engines, due to higher compression ratio, air is
compressed to a high temperature and pressure. Fuel is injected directly into the
combustion chamber. After sometime (physical delay period), spray break-up,
mixing and low-temperature chemical reactions occur. The mixed air and fuel in the
spray plume ignites and starts burning, usually forming a stratified or diffusion
flame. CI engines have no knock limit. However, they are limited by the amount of
mixing in the cylinder and the material strength of the components.
Combustion increases pressure and temperature as the energy contained in the

fuel is liberated and the chemical reaction is completed. The combustion of fuel
produces a spike in pressure and temperature due to sudden energy release taking
place with the production of exhaust gases. Some of the energy is radiated and
convected to the cylinder walls, cylinder head, piston and the valves, and some of it
is lost. One-third of the energy goes into the power stroke, where the exhaust gases
expand under high pressure and push the piston down towards bottom centre
position. A thermodynamic energy balance will show that the energy produced due
to combustion is used for work done due to expansion. The thermal losses include
heat losses through the walls and the enthalpy loss in the exhaust gases at high
temperature.
During the exhaust stroke, the exhaust gases are pushed out through the exhaust
valves, which start opening towards the end of the expansion stroke. Like intake
process, this process also involves formation of a high-speed, high-temperature jet
in the gap between the exhaust valves and ports.
During the process of combustion, the fuel, which is a long chain hydrocarbon,
breaks up into smaller molecules. The carbon and hydrogen contained in these
molecules combine with the oxygen in the air in exothermic reactions. If the fuel–
air ratio is stoichiometric at each location in the combustion chamber, carbon
dioxide and water vapour will be formed. However, at some regions, where the
fuel–air ratio is rich due to inadequate mixing, there will be insufficient oxygen
molecules and the combustion will not be complete. Because of this, carbon
monoxide (CO) and unburnt hydrocarbon molecules will be produced.
Some of the unburnt hydrocarbons will be polycyclic or polyaromatic as the
carbon chains wrap around each other and form solid particles. This is called soot.
If the carbon monoxide and unburnt hydrocarbons are then transported to a region
with adequate oxygen, then there is a possibility of combustion to reach comple-
tion. Otherwise, they will leave with the exhaust gases and cause loss of energy.
Due to high temperatures in cylinder, the nitrogen molecules contained in the air
break up under intense heat. The nitrogen ions combine with the available oxygen
radicals to form nitrous oxides or NOx. If the fuel contains nitrogen or sulphur
atoms, they will also form NOx and sulphur oxides (SOx).
Now, it must be clear that the combustion efficiency of the engine and pollutant
formation depends upon:
1. fluid dynamics of swirl;
2. tumble;
3. mixing;
4. turbulence production during the intake and compression strokes.
Energy loss is mainly due to incomplete combustion, the heat transfer losses to
the wall, and the exhaust losses. Engine emissions include COx, NOx, SOx, soot,
unburnt hydrocarbons and water.
5 Design of Engines with High Efficiency
The design of engines with high efficiency has to take into account the complex
fluid dynamics that occurs in the manifolds and cylinders. Several design issues
come to the forefront here:
• port flow design;
• combustion chamber and piston shape;
• squish;
• compression ratio;
• design for low speed and idle;
• spark and injection timing.
5.1 Port Flow Design
The air flow rate through the intake manifold ports depends on the pressure dif-
ference between the cylinder and the manifold. Further, it also depends on the
throttle position in case of SI engines. A critical factor here is the packaging. That is
to say, the engine and its supporting systems have to fit in a certain amount of
space. Further, it should allow easy access for future maintenance. This means that
the intake manifolds and engine ports might be routed around other parts. This
introduces an additional resistance to the air flow and affects the swirl and tumble in
the cylinder. Port flow design to achieve a given air flow rate and desired levels of
swirl/tumble within a certain packaging layout to maximize volumetric efficiency
thus becomes a critical fluid dynamics design problem. This chapter is devoted to
understanding the fluid dynamics of port flow right from the straight port and ports
with bend angle
5.2 Combustion Chamber and Piston Shape
Apart from port flow, another critical design issue is the size and shape of the
combustion chamber, the piston crown shape, and the layout of the valves. Here,
the chamber can be flat, a hemispheric dome or a pent roof, while the piston crown
can be flat, domed, inclined or a bowl.
The valves can be positioned as “straight”, i.e. the valves are aligned with the
cylinder axis as shown in Fig. 1, or they can be “canted”, i.e. they are at an angle to
the cylinder axis and normal to the surfaces of the combustion chamber (Fig. 2).
It is a known fact that the volumetric efficiency and the amount of air intake into
the cylinder are dependent on the ratio of the intake valve area relative to the
cross-sectional area of the cylinder. Hence, it is desirable to have the intake valves
Fig. 1 Straight valve engine [1]
Fig. 2 Canted valve engine [1]
be as large as possible relative to the bore. Therefore, modern engines are designed
with four valves (two intakes and two exhausts) instead of two. However, if the
combustion chamber is flat, the surface area available gets limited for the valve
layout to just the cross section. If the combustion chamber is hemispheric or pent
roof, it opens up more surface area for the intake and exhaust valves, allowing them
to be larger and more efficient. The deficiency in such a design is that the com-
bustion chamber volume and surface area are large. This implies that the flame front
for combustion has to travel a longer distance, increasing the chance of incomplete
combustion. Further, the compression ratio will decrease since there is a larger
volume at the top centre. A larger wall surface area increases the heat losses during
combustion. This will affect combustion efficiency.
This deficiency can be overcome by changing the piston shape from the flat to a
domed or any desirable shape to reduce the volume. This means that the flame front
has to travel around the piston dome to reach all parts of the combustion chamber
volume. Again, this will increase the time taken for complete combustion which in
turn increases the possibility of knocking in SI engines. To overcome this defect,
the piston could be incorporated with a bowl in the centre, which would reduce the
flame travel time. However, this will reduce the compression ratio.
5.3 Squish
In addition, geometric design consideration is to have a “squish” region. Squish is

nothing but the region around the perimeter of the combustion chamber with as
minimum clearance volume as possible between the piston and the cylinder head at
top centre (refer Fig. 3). As mentioned earlier, a small or low squish region causes
the swirling air flow to get squeezed out into the combustion chamber. This causes a
turbulent jet to form, which converts the mean flow energy in the swirl into tur-
bulence, increasing the turbulent flame speed and the compression ratio and the
combustion efficiency. However, if the squish clearance is too low, there is a chance
of collision between the piston and the cylinder head when the engine material
expands due to high temperature. Further, mass manufacturing of engines requires
that there should be room for production and manufacturing tolerances. Therefore,
the squish needs to be low enough to allow higher turbulence production, but at the
same time high enough to allow room for variability due to thermal expansion and
manufacturing tolerances.
Fig. 3 Illustration of swirl and squish

5.4 Compression Ratio
The compression ratio is defined as the ratio of the total cylinder volume to the
clearance volume (refer Fig. 4). It is a critical factor in combustion efficiency and
pollutant formation. It may be noted that a high compression ratio enhances the
combustion efficiency. At the same time, the higher temperatures due to high
compression ratio cause more NOx to form, thus increasing the emissions.
In the 1970s, automotive engines had much higher compression ratios, since the
emissions norms were not stringent. Over the period with stricter environmental
regulations on emissions, the compression ratios were reduced to meet the new
standards. However, in the 1990s, technological advances in catalytic converters
and increase in combustion efficiency allowed higher compression ratios again and
improved fuel economy. An additional consideration, especially for diesel engines,
one should take care of the materials used for the piston and combustion chamber to
withstand the peak temperatures and pressures encountered with high compression
ratios and high boost levels.
Thus, intake/exhaust ports designs, along with heads, valves and pistons, involve
interplay and trade-offs between the volumetric and combustion efficiency, pollu-
tant formation and packaging considerations, materials choices and manufacturing
tolerances. In this connection, the ability to accurately analyse the engine fluid
dynamics plays a key role in optimizing the engine to effectively deliver the peak
power, while meeting emissions standards and further taking into consideration
packaging and manufacturing constraints.
Fig. 4 Illustration of total volume and clearance volume

5.5 Design for Low Speed and Idle
In many cases, the performance of engines at idle or low speed is a critical design
consideration. For automobile engines, the engine is designed for peak power at a
specific speed. This speed is typically high. Further, the engine will still have to
perform well at lower speeds and at idle speed also. Variable valve timing, which
allows the valves to have different lift profiles and opening and closing events for
different engine speeds, is being tried in many modern engines. Here, the goal is to
maximize the volumetric and combustion efficiency. This is achieved by producing
optimal levels of swirl, tumble, and turbulence at both low and high speeds. This
will enable the combustion charge, i.e. the air/fuel mixture, to get well mixed, and
the turbulent flame speed is high enough for complete combustion. There could be
additional geometric design changes to the ports and valves for low speed or idle
conditions. For example, valve masking and valve shrouding are used to impart
additional swirl to the air flow and increase the jet velocity (refer Fig. 5).
5.6 Spark and Injection Timing
During the engine cycle, the spark timing in SI engine and start of injection in CI
are to be optimized to provide the desired power or torque with minimum pollutant
formation. Strategies, such as exhaust gas recirculation (EGR), are used to mini-
mize the peak temperature in the engine by increasing the thermal mass of the
intake air. This reduces the NOx production, which is strongly a function of high
temperatures. For diesel engines, there is a trade-off between soot production and
Fig. 5 Inlet valves used to induce high swirl at low engine speeds and low valve lift
NOx. Note that soot production decreases with temperature, whereas NOx pro-
duction increases with temperature. Thus, designing high efficiency engines to meet
performance and emissions standards requires trade-offs which should take engine
fluid dynamics into account.
6 Application of CFD in IC Engine Design
This section emphasizes the role of CFD modelling in IC engine design and per-
formance analysis. The information in this section is divided into the following
subsections:
• the role of CFD in engine design;
• types of CFD in engine performance.
6.1 The Role of CFD in Engine Design
As already described, IC engines involve complex fluid dynamic interactions

between air flow and fuel injection. Further, moving geometries and combustion
play a vital role. Fluid dynamics phenomena such as fuel jet formation and wall
impingement are to be considered. Further, swirl and tumble and turbulence are
equally crucial for obtaining high efficiency and meeting emissions norms. The
design problems that are encountered include port flow design, combustion
chamber shape design, variable valve timing, injection and ignition timing, and
design for low or idle speeds.
There are several tools available in practice during the design process. These
include:
1. experimental investigation using test or flow bench set-ups;
2. one D codes;
3. analytical models;
4. empirical/historical data;
5. computational fluid dynamics (CFD).
Of these, CFD has very high potential for providing useful and detailed infor-
mation. It can provide insights that can be fed back into the design process. This is
because in CFD we solve the fundamental equations that describe fluid flow directly
on a mesh which describes the three-dimensional geometry. Further, incorporation
of submodels for fuel injection, chemistry, combustion and turbulence enhances the
capability of predictions. A number of techniques are available for modelling
motion of the moving geometry and its effect on fluid flow.
From the CFD results, it is possible to visualize the flow phenomena on 3D

geometry and can be analysed numerically. This can provide excellent insight into
the complex interactions that take place inside the engine. This will enable the
designer to compare different designs and quantify the trade-offs. Various
trade-offs, such as soot versus NOx, swirl versus tumble and impact on turbulence
production, combustion efficiency versus pollutant formation, provide a platform
for optimal designs. Hence, nowadays, CFD analysis is used extensively as part of
the design process in power generation, in transportation and especially in auto-
motive engineering. With the rise of modern and inexpensive computing power due
to the availability of superfast digital computers and 3D CAD systems, it has
become much easier for analysts to perform CFD analysis. In increasing impor-
tance, the CFD analyses performed can be classified into:
• Port flow analysis: it provides quantification of swirl, tumble and flow rate, with
static engine geometry by placing the piston at different locations during the
suction stroke of the cycle.
• Cold flow analysis: here, moving geometry is brought into picture and the air
flow can be evaluated. However, no fuel injection or reactions are considered.
• In-cylinder combustion simulation: in this expansion and exhaust strokes with
fuel injection, mixing, reactions, ignition and pollutant formation with moving
geometry are considered.
• Full cycle simulation: entire engine cycle with air flow, fuel injection, com-
bustion, and reactions intake and exhaust are predicted.
However, the CFD analysis of various processes for engines is highly complex,
time-consuming and error-prone. Typically, the researcher has to go through several
steps to set up the problem. Any minor error can lead to failure of the simulation.
Once the analysis has been set up, it takes many hours or even days of computation
to get the solution depending upon the grid density and evaluation of the results.
The results coming out of simulation are fairly complex, with large data sets. It
requires time and effort to analyse and get useful and meaningful information,
which can be fed back to the design stage.
Automation and process compression thus become a critical need. In the next
section, we will further evaluate the practical issues facing engineers in conducting
a CFD analysis on IC engines. Following that, we will explain the results obtained
from an integrated environment such as STAR-CD workbench and analyse some
typical results obtained for the IC engine analysis.
6.2 Port Flow Analysis
In port flow analysis, the geometry of the ports valves and cylinders is “frozen” at
critical points during the engine cycle. The air flow through the ports is analysed
using CFD for the various case studies undertaken. The flow rate through the engine
volume, swirl and tumble in the cylinder and turbulence levels are evaluated. Fluid
dynamics phenomena such as separation, jet formation, valve choking, wall
impingement and reattachment, as well as the secondary motions, can be visualized
and analysed.
The results provide snapshots of the fluid dynamics throughout the engine cycle
and can be used to modify the port geometry to produce desired behaviour of the air
flow. Simulation validation wherever possible has to be performed using the data
available in the literature. The results do not capture dynamic phenomena such as
expansion and compression of air due to piston movement and turbulence pro-
duction from swirl and tumble.
In practice, conducting port flow analysis at a single point is fairly straightfor-
ward because the geometry is static, which fits well with the workflow and capa-
bilities in CFD software. One can start with the port, valve and cylinder geometry at
a particular position, create a mesh, specify the mass flow rate or pressure drop for
the compressible flow and a turbulence model and compute the results. The
RANS-based turbulence models are used to compute the effect of turbulence. Since
the turbulent flow interactions with the walls are critical, mesh refinement in the
near wall region is necessary using inflation or boundary layers. However, when the
number of critical positions and hence the number of cases increase, the problem
complexity increases significantly. Setting up large numbers of static cases with
identical mesh and flow settings is time-consuming, with scope for error. In the
following sections, we will present three case studies on port flow.
6.2.1 Case Study 1: Steady Flow CFD Analysis for Improving

Volumetric Efficiency
An industrial project was undertaken to analyse the flow through the inlet and
exhaust manifolds of a popular two wheeler engine in India and then, if necessary,
arrive at a modified design based on the CFD analysis. The objective of this project
was to estimate the volumetric efficiency of the base engine and if necessary modify
the design to arrive at better volumetric efficiency. The results presented below
establish how usefully CFD can be applied for improving the engine port design.
Figure 6 shows the geometry of the intake port. A steady flow analysis at three
different valve lifts was carried out to estimate the volumetric efficiency and also to
examine the flow through the ports of the existing design. As transient analysis
takes time in making fixed and moving meshes, steady flow analysis was thought of
to get quick results and make design modifications if necessary for improving the
flow and volumetric efficiency. Initially, a CAD model of the inlet port was
developed and was meshed with one lakh and seventy-five thousand grid points
(refer Fig. 6).
The meshed geometry for different valve lifts for the inlet manifold is shown in
Fig. 7. Prescribing appropriate boundary conditions, viz. pressure boundary con-
dition at inlet and exit boundary being the cylinder bottom centre, the flow for the
Fig. 6 CAD model of inlet port geometry with grid
2mm Valve lift 4mm Valve lift 5.6mm Valve lift

(Maximum)
Fig. 7 Meshed geometry of the inlet port at three-valve lift position
three-valve lifts has been studied. Figure 8 shows the velocity vector along with
velocity contours for the three-valve lift positions.
As can be seen that the engine has pent-roof cylinder head and the piston being
stationary as the flow pass through the inlet port initially even with 2-mm valve lift,
there was a slight flow separation near the bend. As the vale lift increases from 2 to
4 mm, there is more separation and, at maximum lift of 5.6 mm, there is large flow
separation at the bend region. Further, it causes recirculation behind the vale head.
This can hamper volumetric efficiency of the engine. The pressure contours shown
in Fig. 9 for the three valves were analysed and found that it is favouring separation
in the bend region of the inlet manifold. There is a distinct pressure gradient at the
bend region which helps to separate the flow (refer Fig. 9).
2 mm valve lift (IVO)
5.6 mm valve lift (Max. valve lift)
Fig. 8 Velocity vector and velocity contours for three-valve lifts during valve opening process
To reduce the effect of separation in the inlet manifold design, changes were
made and the bend angle was changed from 60° to 45° and the flow analysis was
carried out again. This helped considerably in reducing the recirculation in the inlet
manifold, and the final design was arrived at. The volumetric efficiency improved
from 82.5 to 86%. Thus, this case study has established that steady flow analysis
using CFD can be successfully applied for the design changes of the inlet manifold
5.6 mm valve lift (Max. valve lift)
Fig. 9 Total pressure contours for the three-valve lifts
by which volumetric efficiency could be considerably improved and a final

geometry was arrived at. Similar analysis was carried out for the exhaust manifold
also, and the design changes helped to increase the scavenging efficiency. The
meshed geometry of the exhaust manifold is shown in Fig. 10. The velocity vectors
in the exhaust manifold are shown in Fig. 11.
Exhaust Port and Valve
Hemispherical combustion chamber

1,75,000 cells
Fig. 10 Meshed geometry of the exhaust manifold
Based on the results obtained, similar modification was carried out in the exhaust
manifold also and final design of the cylinder head with inlet and exhaust manifold
is shown in Fig. 12. This case study has clearly established that CFD can be used as
design tool along with flow analysis.
6.2.2 Case Study 2: Steady Flow Analysis on Multi-cylinder Inlet

Manifold
Having got the confidence by means of the case study in the previous section in
improving the design by CFD analysis in a single-cylinder engine, a multi-cylinder
manifold analysis was carried out to improve the design and flow characteristics.
This is also a typical industrial project. In this, a four-cylinder inlet manifold along
with throttle body was taken for the analysis. The purpose of this study was to
understand the flow behaviour of a multi-cylinder throttle body assembly where
there is one inlet and multiple outlets. The main aim of this case study was to
establish equal flow rate to all the cylinders. The main problem in a multi-cylinder
SI engine is the maldistribution of fuel–air mixture in various cylinders. This can be
easily analysed using CFD under steady flow conditions. Initially, a CAD model
was developed and the corresponding meshed model is shown in Fig. 13.
Fortunately for the engine under consideration, the valve timing was such that at
any time only one cylinder valve was in the open position, whereas for other three
cylinders it is in closed position. Flow studies have been carried out for different
valve lift conditions (25, 50 and 100%). The flow results are presented for all the
three-valve lift condition. Compared to the manifold in the case study 1, this
manifold was more complex and meshing was a little more challenging.
Exhaust Valve lift = 2 mm
Exhaust Valve lift = 6 mm
Fig. 11 Velocity vectors in exhaust manifold for two-valve lifts

Fig. 12 Final design of intake and exhaust manifold
Fig. 13 Meshed geometry of the multi-cylinder inlet manifold
Because the manifold geometry is complex, it was decided to carry out the
analysis dividing the flow domain in five regions as indicated in Fig. 14. The
various regions were carefully chosen so that one can understand the complete flow
characteristics. Region 1 is the one where the possibility of recirculation is there.
Region 2 is the region where flow passes over the vale stem. Flow region 3 is one
where the flow starts entering the cylinder. Region 4 is the one where flow sepa-
ration can take place. Region 5 is the one where the bulk flow takes place. Thus,
this configuration provides the possibility for all the flow complications and
Flow in Port
Fig. 14 Division of flow domain into number regions
Fig. 15 Flow velocity vectors at various lift conditions in manifold 3
therefore it was a challenging job to undertake. The necessary boundary conditions

were specified, appropriate equations were solved, and the flow velocity vectors
were obtained for all three-valve lift conditions. The details are shown in Fig. 15.
We will explain the results obtained for full-valve lift condition only.
Flow Analysis in Various Regions
As shown in Fig. 15, in region 1, there is a formation of strong recirculation zone.

This is due to the projection in the inlet manifold. This will cause flow losses, and
this is to be avoided. On analysing the region 2, there is a restriction imposed by
valve stem. Because of the stem, the flow starts separating behind the stem. It may
be noted that this cannot be avoided. Coming to region 3, as can be seen, there is
possibility for flow separation because of the slight bend and due to the stepped
configuration as shown in Fig. 15a. Even though the step increases the flow area if
the separation takes place, it will nullify the positive effect. When region 4 is
analysed, it is clearly seen that there is a strong separation. This is to be avoided. As
shown in Fig. 15d up to 50% of the manifold length, the bulk flow is normal.
Beyond that because of the manifold bend, valve stem as well as valve head effect,
there is back pressure which decelerates the flow. The deceleration is minimum for
full-valve lift conditions. Hence, this region should be looked into more carefully.
Thus, the steady flow analysis could throw some useful flow structure and the
design was modified taking into consideration all the details and drawbacks
described, by which uniform flow distribution could achieved in all the four
cylinders. Again, this study has established that CFD can be used as a tool for
design modifications. As this was a steady flow analysis, the time required to obtain
the solution was comparatively very much less.
From this study, we could find that the fourth manifold was getting less air since
the air has to travel a longer distance. In order to get uniform flow, the plenum
chamber design was slightly modified based on the CFD results and by which
uniform distribution could be achieved. It should be emphasized here that because
of the CAD model availability, the geometry could meshed accurately and flow
vectors are shown in Fig. 15. This is the second case study by which we could
establish that CFD can be applied for the design. The CAD model of the modified
design is shown in Fig. 16.
6.2.3 Case Study 3: Steady Flow Analysis on Six-Cylinder V-Engine

Inlet Manifold
Based on the confidence obtained through the above two case studies, we went on
to evaluate a highly complex intake manifold with power valve of a six-cylinder V
engine. Three manifolds are in the right bank, and three others are in the left bank.
The manifold was very carefully designed, and the main purpose of this study was
to evaluate the flow coefficients for all the six manifolds to see whether the flow is
uniform. It is to be emphasized here that the manifold was designed after many
iterations and was expected to give uniform flow distribution. Again, it was sub-
jected to steady flow analysis and the flow coefficient values are evaluated.
Figure 17 shows the meshed geometry under consideration. Meshing was really
a very challenging job, and it took considerable time to arrive at appropriate mesh
size. Again, appropriate boundary conditions were prescribed and equations were
Fig. 16 CAD model of the modified design of the inlet manifold
Fig. 17 Six-cylinder inlet manifold versus V type engine
solved. It may be noted there was a power valve which can be kept either in closed
position or in open position. When it is kept in closed position, flow will pass
through the left bank to all the six cylinders, whereas when it is in open position
flow will take place through the right bank to all and the six cylinders will get the
Fig. 18 Grid arrangement when the power valve is in open position
flow. As it is a complex design, it will be difficult to visualize the flow when one
sees this for the first time.
Meshed geometry is shown in Fig. 18 when the power valve is open and flow
velocity vectors for all the six cylinders are shown in Fig. 19. From the velocity
vectors, flow coefficients were calculated and are given in Table 1. It is surprising
to see that the flow coefficient K is same to first decimal accuracy. Establishing this
by means of experiments would have been time-consuming and costly. Thus, by
means of this CFD study, we could validate the design parameters. As this was a
steady flow analysis, the results obtained were very fast and less costly.
Table 1 gives the flow coefficient based on the quantity of flow, and the pressure
drop shows clearly that the flow coefficient is the same to the first decimal accuracy.
Therefore, there was no necessity to make any changes in the design. This further
establishes the use of CFD in confirming the design
6.3 Cold Flow Analysis
Modelling of cold flow takes into account mainly the airflow and possibly the fuel
injection without involving reactions. The main aim is to capture the mixture for-
mation process. For this, accurate accounting of the interaction of moving piston
with the fluid dynamics of the induction process. In this analysis, the changing
Manifold 1
Manifold 2
Manifold 3
Manifold 4
Fig. 19 Flow velocity vector for all the six manifolds

Manifold 5
Manifold 6
Fig. 19 (continued)
behaviour of the air flow which tumbles into the cylinder can be studied. Further, it
can take into account swirl generated by intake valves and the exhaust jet coming
out of the exhaust valves. Further, the turbulence generated from swirl and tumble
due to compression and squish can be conveniently analysed.
This information will be very useful to decide whether the conditions in the
cylinder at the end of the compression stroke are conducive for combustion and
flame propagation. It is a known fact that turbulence enhances flame propagation.
Further, it aids complete combustion during the power stroke. For the best power
output, homogeneity of mixture with right air/fuel ratio throughout the combustion
process is a must. CFD analysis can provide the degree of charge stratification.
However, cold flow simulations do not include the significant thermodynamic
changes that accompany combustion. This is because the flow characteristics during
the expansion and exhaust strokes do not reflect actual conditions. In terms of
validation, experimental Particle Image Velocimetry (PIV) or LDV data in cycling
engines is not easy to obtain as with port flow analysis. However, transparent
pistons and cylinders can be used to gather velocity information for simple engine
configurations.
CFD modelling of cold flow involves additional work. First of all, one should
incorporate the valve and piston motion. Further, boundary conditions, turbulence
models and other required submodels should be specified. Details of valve and
piston geometry should be given, along with the lift curves. Engine geometric
characteristics, such as bore, stroke, connecting rod length and compression ratio,
should be given as input data in order to calculate piston position as a function of
Table 1 Flow coefficient details with power valve open and closed position
Power Mass flux Density Q (m3/ Inlet Outlet (P2 (P2 – K
valve (kg/s) (kg/m3) min) “P2” “P1” –P1) P1)
closed (Pa) (Pa) (Pa) (mm of
H2O)
Manifold 1 0.0908265 1.16203 4.69040 99889 97058 2831 288.583 3.6218
Manifold 2 0.0901336 1.16209 4.65489 99855 97058 2797 285.117 3.6274
Manifold 3 0.0917929 1.16195 4.73994 99879 97,058 2821 287.564 3.5776
Manifold 4 0.0905110 1.16182 4.67427 99,843 97,058 2785 283.894 3.6046
Manifold 5 0.0935049 1.16208 4.82780 99,870 97,058 2812 286.646 3.5069
Manifold 6 0.0883281 1.16175 4.56181 99,850 97,058 2792 284.606 3.6981
Power Mass flux Density Q (m3/ Inlet Outlet (P2 (P2 – K
valve open (kg/s) (kg/m3) min) “P2” “P1” –P1) P1)
(Pa) (Pa) (Pa) (mm of
H2O)
Manifold 1 0.0937697 1.16200 4.84181 99,911 97,058 2853 290.826 3.4221
Manifold 2 0.0961538 1.16214 4.96431 99,898 97,058 2840 289.501 3.4274
Manifold 3 0.0952559 1.16212 4.91804 99,914 97,058 2856 291.131 3.4693
Manifold 4 0.0970425 1.16222 5.00985 99,916 97,058 2858 291.335 3.4069
Manifold 5 0.0961706 1.16223 4.96479 99,915 97,058 2857 291.233 3.4373
Manifold 6 0.0958784 1.16217 4.94996 99,921 97,058 2863 291.845 3.4512
crank angle. Since the volume in the cylinder is changing throughout the engine
cycle, the mesh change should take place accordingly. Methods to automatically
modify the mesh during piston motion also need to be incorporated. The CFD
calculation is an inherently transient computational problem. It involves moving
deforming or dynamic mesh. All the geometric motion is a function of a single
parameter, viz. the position of the crankshaft in its rotation, or crank angle.
The preprocessing from geometry to solver set-up is typically time-consuming
and challenging. Here, we should separate or decompose the geometry into moving
and stationary parts. Typically, the intake ports are split off from the cylinder and
valves. The region between the valve margin and valve seat, which opens
and closes during valve motion, should be separated. The combustion chamber and
piston region may be also decomposed or separated into smaller parts. Then, each
part can be meshed accordingly for the solver set-up. Any errors at this stage can
lead to failures downstream during the solution process.
The run times for solver runs can be fairly long since the motion is typically
resolved with small time steps (approximately 0.25 crank angle) to get accurate
results, and the simulation is run for two or three cycles to remove the initial
transients. Finally, the large volume of transient data that results from the CFD
solution needs to be post-processed to obtain useful insight and information. Thus,
cold flow analysis would also benefit from design automation and process
compression.
6.3.1 Case Study 4: Cold Flow Transient Analysis

on a Single-Cylinder Engine
As is evident that in the previous case studies, only steady flow analysis has been
carried out. It will provide first-cut information for the design modification
requirements. However, true picture will emerge only if a transient flow analysis
with moving piston is carried out. In this case study, a CFD analysis has been
carried out to investigate the effect of intake port configurations such as intake port
cross-sectional area, bend angle, engine speed, and intake pressure on the flow field
inside the cylinder of a DI diesel engine under cold flow condition. Further, to
evaluate consistency of in-cylinder flow structure results, a grid independence study
has been conducted to arrive at the optimum grid density that can be used for
subsequent simulation studies.
In-cylinder fluid dynamics in direct injection CI (diesel) engines play a vital role
during combustion process. In particular, in-cylinder fluid dynamics contribute to
the fuel-air mixing, which is one of the most important factors for the control of the
fuel burning rate. It significantly affects the ignition delay, magnitude of the pre-
mixed burn, magnitude and timing of the diffusion burn, and the emissions of nitric
oxide and soot. In order to have better mixing, swirl is generated during the intake
stroke as a result of intake port shape and its orientation. In IC engines, the fuel
evaporation and mixing processes are strongly influenced by the turbulent nature of
the in-cylinder flows. The velocity gradient in the mean flow is one of the major
reasons for the turbulent fluctuations. The air-jet created by flow during the intake
process interacts with the cylinder wall and moving piston to generate large-scale
rotating flow, both in the vertical and in the horizontal planes. The behaviour of the
in-cylinder turbulent flow can be characterized by monitoring the kinetic energy
and the integral length scale variation of turbulent eddies that contribute to turbu-
lence production during intake and compression processes. In-cylinder flow char-
acteristics at the time of fuel injection and subsequent interactions with fuel sprays
and combustion are the fundamental considerations for the engine performance and
exhaust emissions of a diesel engine.
7 Engine and Operating Conditions
A single-cylinder DI diesel engine having a toroidal combustion chamber with two

direct intake ports whose outlet is tangential to the wall of the cylinder and two
exhaust ports to simulate the flow field inside the cylinder has been chosen. The
geometry with different intake port configurations such as constant cross-sectional
area (CCS), varying cross-sectional area (VCS), combination of both (CVCS) and
varying cross-sectional area intake ports with different bend angles (20°, 30° and
40°) have been investigated. However, the results pertaining to varying
cross-sectional area (VCS) with 20° bend angle only will be presented and dis-
cussed. Geometrical details and operating conditions of the engine are given as
Table 2 Geometrical details of the engine

Bore 130.0 mm Static bumping clearance 1.55 mm
Stroke 150.0 mm Compression ratio 15.5:1
Connecting rod length 275.0 mm Intake valve diameter 44.4 mm
Displacement 1991 cc Intake valve opening 9° bTDC
Rated speed 2000 rpm Intake valve closure 62° aBDC
Bowl diameter 76.0 mm Exhaust valve opening 43° bBDC
Bowl depth 29.0 mm Exhaust valve closure 10° aTDC
Bowl volume 113.5 cc
input, and appropriate fuel injection strategy has been adopted. The geometrical
details of the engine are given in Table 2.
8 Methodology
The CFD code, STAR-CD, has been used as solver. The code solves the discretized
Navier–Stokes equations. For physical modelling, the RNG k–e turbulence model
with standard wall function has been employed. The solution algorithm is based on
the pressure-correction method. It uses the PISO algorithm with the first-order
upwind differencing scheme (UD). For the solution of the momentum, energy and
turbulence equations, the implicit temporal discretization method is employed.
Spatial discretization is done using the UD. For solving momentum, turbulent
kinetic energy/dissipation and temperature, the appropriate equations with central
differencing scheme are used.
The calculations begin at TDC of the intake stroke and complete at 30° after
TDC (aTDC) of compression. This means a cold flow analysis has been carried out.
The results of different variables plotted are presented in non-dimensional form.
Normalization is done with respect to mean piston speed which facilitates the
comparison. The various equations are represented through Eqs. (2–6).
v ð hÞ
Swirl ratio; SRðhÞ ¼ 2pN ð2Þ
60 r
uð hÞ
Radial velocity ¼ ð3Þ
Vp
where v = tangential velocity in m/s, h = crank angle in degrees, N = engine speed

in rpm, r = distance from cylinder axis in m, u = radial velocity in m/s, VP = mean
piston speed in m/s.
2 1=2
u0 ðhÞ 3k
Turbulence intensity ¼ ¼ ð4Þ
Vp Vp
where u0 ¼ fluctuating velocity in m/s and k = turbulent kinetic energy in m2/s2.

Swirl is defined as the organized rotation of the charge in the cylinder with
respect to the cylinder axis. With an initial angular momentum in the incoming
charge, swirl is created. Swirl ratio is defined as the ratio of total angular
momentum of the in-cylinder fluid about the cylinder axis to the total moment of
inertia of the fluid about the cylinder axis:
X
N Cells X
N Cells
where HZ ¼ fvi ðxi xÞ ui ðyi yÞgmi ¼ fvi xi ui yi gmi

i¼1 i¼1
ð5Þ
X
N Cells
2 2
X
N Cells
and MZ ¼ fðxi xÞ þ ðyi yÞ gmi ¼ fx2i þ y2i gmi

i¼1 i¼1
Pn 2 1=2
u0 ðhÞ i 3 ki VðhÞi qðhÞi
Mass averaged turbulence intensity ¼ ¼ Pn : ð6Þ
VP VP i VðhÞi qðhÞi
where Hz represents the total angular momentum of the in-cylinder fluid about the
cylinder axis (z-axis), Mz is the total moment of inertia of the fluid about
the cylinder axis, N is the angular speed of the crankshaft in rpm, ui and vi represent
the local components of velocity in x and y directions, respectively, mi is the mass
accumulated in each cell, V is the cell volume, q is the density, h is crank angle, and
xi and yi represent axes in radial and tangential direction.
9 Modelling Procedure
A hexahedral block structure mesh is employed for the entire computational domain
of the engine with 505,542 cells. The preprocessor GAMBIT is used to create the
entire computational domain of the engine including intake and exhaust ports.
STAR-CD is used for the solution of governing equations and post-processing the
results. The meshed intake and exhaust port geometry is shown in Fig. 20.
10 Boundary and Initial Conditions
The boundary and initial conditions (Table 3) employed are as follows: by

neglecting dynamic effects, constant pressure boundary conditions are prescribed to
both intake and exhaust ports. Smooth wall option is used in order to take it into
o o o
20 bend angle 30 bend angle 40 bend angle
Fig. 20 Varying cross-sectional area with different bend angles
Table 3 Boundary and initial conditions

Boundary conditions Initial conditions
Intake and exhaust ports Intake and exhaust ports
Type Constant pressure 1.01 bar Pressure 1.01 bar
Temperature 303 K Temperature 303 K
Turbulent intensity 5%a Turbulent intensity 5%a
Length scale 1 mma Length scale 1 mma
Cylinder–valve interface Cylinder
Type Attach Pressure 1.0 bar
Wall Temperature 303 K
Turbulent intensity 5%
Type Smooth wall, no slip
Length scale 1 mm
Wall heat Adiabatic
a
Source www.adapco-online.com
account frictional effects at the walls. Walls are considered as adiabatic. The initial
values of pressure and temperature prescribed are assumed to be in the whole
domain. The residual swirl of the flow in the cylinder at the end of the exhaust
stroke is considered as negligible. This means that the flow is supposed to be
quiescent initially. The initial turbulent intensity is assumed as 5% of the mean
flow. The integral length scale is calculated using Prandtl’s mixing length model.
The walls in the entire are considered adiabatic.
In-cylinder fluid dynamics in DI diesel engines plays a vital role during com-
bustion process. In particular, in-cylinder fluid dynamics plays a vital role in
controlling the fuel burning rate. Burning rate in turn affects the ignition delay, rate
of the premixed as well as the diffusion burning. Further, it controls the emissions
of NOx and soot. For better mixing, during the intake, stroke swirl is generated.
This is achieved by designing appropriate intake port shape and its orientation.
Fuel evaporation and mixing process in an in IC engines are strongly influenced
by the in-cylinder turbulence. The turbulent fluctuations are created by the velocity
gradient in the mean flow. The jet action by flow during the intake process interacts
with the cylinder wall and moving piston. This generates large-scale rotating flow,
both in horizontal and in the vertical planes. In-cylinder turbulent flow behaviour
can be characterized by the kinetic energy and the integral length scale variation of
turbulent eddies. These turbulent eddies contribute to turbulence production during
both intake and compression processes. Engine performance and exhaust emissions
of a diesel engine are controlled by in-cylinder air motion at the time of fuel
injection and subsequent interactions with fuel sprays and combustion.
The objectives of this case study are to investigate the effect of inlet port con-
figuration on the in-cylinder flow dynamics. The results obtained through simula-
tions to meet the objective are presented in this section in the following order:
• validation of the code used and the grid independence study;
• study of the effect of variable cross-sectional area intake port with different bend
angles under cold flow motoring condition.
11 Validation of the Code
For the validation of the CFD results, a 20° bend angle having varying
cross-sectional area intake ports is considered. The predicted results are compared
with the experimental results of Payri et al. [57, 58]. This configuration has been
chosen since experimental results are available for such a configuration. The
operating speed of the engine, intake pressure and temperature are 1000 rpm,
1.01 bar and 303 K, respectively. The results obtained for swirl ratio, turbulent
intensity and radial velocity variations at different crank angles are compared with
Laser Doppler Velocimetry (LDV) measurements of the same engine. The engine
under consideration is equipped with toroidal combustion chamber in which
measurements have been made at different locations at a radial distance of 34 mm
from the axis of the cylinder and at a distance of 4, 8, 12 and 16 mm (Z1, Z2, Z3,
Z4), respectively, from the cylinder head the details of which are shown in Fig. 21.
One of the exhaust valves of the cylinder under consideration is made inoperative.
A thick quartz window is inserted in the space for measurement purpose.
1 : 4 mm from the cylinder head

r = 34 mm
Fig. 21 Measuring locations

Fig. 22 Swirl ratio versus Location Z1 = 4 mm, Expt. Payri et al.

crank angle at various
locations (Z1 = 4 mm and
r = 34 mm Predicted
4
Z2 = 8 mm from the cylinder
head) compared with
experimental LDV data for
3
varying cross-sectional area
SR
with 20° bend angle
1
280 300 320 340 360 380 400
Crank Angle, deg.
Figures 22 shows the distribution of swirl ratio at various crank angles (70°
during compression stroke and 30° during expansion stroke) around compression
TDC for the two measuring locations. It is shown that swirl ratio decrease is
observed during compression stroke. It may be due to swirl getting converted into
squish slowly. The wall friction also may play a part. As can be seen from the figure
as the piston approaches TDC, swirl is enhanced. It is due to the flow acceleration
which causes the reversal of its angular momentum within the smaller diameter
piston bowl. During the expansion stroke as the flow starts exiting, reverse squish
action takes place. The sudden fall in the swirl ratio is caused by the piston bowl
and wall friction. As shown in Fig. 22, the predicted results are in good agreement
with experimental results at all locations. During the expansion stroke, there is a
faster decay of swirl at locations especially near the cylinder head. This is due to
reverse squish–swirl interaction. As the piston moves away from TDC, this inter-
action becomes less and less.
Figure 23 shows the distribution of non-dimensional turbulent intensity, which
decays almost linearly as it approaches TDC at 4- and 8-mm location. This is due to
squish and swirl interaction. As the bowl has an open geometry at the top, the radial
motion gets weakened and the squish effect becomes small. It is true particularly at
the measurement points. As can be seen from the figure, the measuring points are
located relatively far away from the walls. It is evident that turbulence dissipation
rate is higher compared to the turbulence generation rate. The predicted results are
in good agreement with experiments.
12 Grid Independence Study
Results presented here has been carried out with varying cross-sectional area intake
ports with a 20° bend angle which is used for validation, to study the effect of grid
density on the simulated results. For this purpose, the computational domain
Fig. 23 Non-dimensional Location Z1 = 4 mm, Expt. Payri et al.

turbulent intensity versus r = 34 mm
crank angle at various 0.8 Predicted
locations (Z1 = 4 mm and
Z2 = 8 mm from the cylinder
head) compared with 0.6
experimental LDV data for
u'/Vp
varying cross-sectional area
with 20° bend angle
0.4
0.2
280 300 320 340 360 380 400
Crank Angle, deg.
Location Z2 = 8 mm, Expt. Payri et al.

r = 34 mm
0.8 Predicted
0.6
u'/Vp
0.4
0.2
280 300 320 340 360 380 400
Crank Angle, deg.
consisting of 337,978 (grid A), 505,542 (grid B), 691,354 (grid C) and 912,680
(grid D) cells has been used. Figure 24 shows mass averaged swirl ratio (SR) and
non-dimensional turbulent intensity during suction and compression strokes for
various grid densities. These mass averaged variables are considered to be the most
appropriate to characterize the flow within the cylinder. The swirl in the cylinder is
generated during the intake stroke, and the swirl increases with increasing piston
speed. After reaching the maximum piston speed, the piston starts decelerating
towards BDC, the mass averaged cylinder velocity decreases, and swirl starts
decreasing slowly during the rest of the intake stroke (say up to 180°). This
reducing trend continues in the first part of the compression stroke due to slow
acceleration and compression effect. However, while approaching TDC, swirl is
enhanced as the flow accelerates in preserving its angular momentum within the
smaller diameter piston bowl. During the expansion stroke, there is reverse squish,
as the flow exits from the piston bowl which is mainly responsible for the sudden
fall in swirl velocity. A similar behaviour can be observed in case of the turbulent
intensity also. There is a rapid increase at the beginning of the intake stroke caused
Fig. 24 Temporal distribution of mass averaged swirl ratio for different grids
by the shear stresses associated with the intake jet entering the cylinder and piston
acceleration. This is followed by a reduction of turbulent intensity up to a point just
before TDC and a slight rise up to TDC and drop during early expansion. With
higher and higher cell density, there is slight increase in swirl ratio near com-
pression TDC and slight decrease between maximum valve lift and earlier to TDC.
However, a small increase in turbulent intensity is observed throughout the stroke.
The comparison of flow quantities such as the mass averaged swirl ratio and
non-dimensional turbulence intensity obtained at 115 CAD corresponds to maxi-
mum valve lift, 345 CAD near the point of fuel injection and at compression TDC.
Grids B and D show almost same maximum swirl ratio at all the three crank angles
considered, and grid B shows very small reduction in turbulent intensity compared
to higher grid densities. Considering the computational expense and time, the grid
B with 505,542 cells is used for further computations.
13 Parametric Studies
Based on the confidence gained by the validation, simulation has been carried out to
study the effects of intake port cross-sectional area, bend angle, engine speed and
intake pressure on in-cylinder flow structure. Three intake port cross sections CCS,
VCS and CVCS are considered. To start with a 20° bend angle has been taken for
constant cross-sectional area intake ports. Three engine speeds (1000, 1500 and
2000 rpm) and different intake pressures (1.01, 1.1, 1.2, 1.4, 1.6 and 1.71 bar) have
been studied. The main aim is to present the results of a comprehensive CFD study
on the flow characteristics inside the cylinder of a heavy-duty DI diesel engine and
provide an insight to the influence of above-said parameters on flow characteristics
near TDC just before the start of combustion. However, only VCS configuration
with 20° bend angle will be presented.
13.1 Effect of Intake Port Cross-Sectional Area
Simulation is carried out to study the effect of intake port cross-sectional area on
flow structure with initial values of pressure and temperature of 1.01 bar and
303 K, respectively, at an engine speed of 1000 rpm. The flow in the cylinder
during the intake and compression stroke is analysed. The volumetric efficiency
estimated from the flow is 88.79%. Figures 25 and 26 show the temporal variation
of mass averaged swirl ratio and non-dimensional turbulent intensity during suction
and compression strokes. The mass averaged swirl ratio and non-dimensional tur-
bulence intensity are analysed for three crank angles, viz. 115 CAD corresponds to
maximum valve lift, 345 CAD near the point of fuel injection and at compression
TDC. Consider the swirl ratio at 115 CAD where the intake valves are fully opened.
The swirl ratio at this crank angle is higher due to maximum valve lift condition.
The reason for the increase in swirl ratio is pressure drop due to convergent
cross-sectional area, which results in increased flow velocity. Hence, the flow
accelerates and because of the 20° bend angle there is flow deflection. Due to
acceleration, the momentum increases, which manifests itself as swirl. Further, it is
interesting to note that with respect to 115 CAD (full-valve lift position), the swirl
ratio correspondingly at 345 CAD and 360 CAD is 23.13 and 51.88% more for
varying cross sections. This is due to enhancement of swirl by squish effect and also
due to high initial velocity. Turbulence intensity will fall with commencement of
intake valve closure, and turbulence fluctuations again increase beyond the intake
valve closure due to increased fluid movement. Both swirl ratio and turbulent
intensity fall after compression TDC due to reverse squish–swirl interaction.
Figure 27 shows the distribution of non-dimensional velocity and turbulent
intensity, respectively, at a cross section passing through the axis of intake valves at
35 CAD (35° aTDC). It is the initial stage of intake stroke at which intake valves
have just started opening. As can be seen, two toroidal vortices appear near the wall
of the cylinder as well as at the centre of the bowl because strong interaction of
Fig. 25 Temporal
distribution of mass averaged
swirl ratio in variable area
intake port cross sections with
20° bend angle
Fig. 26 Temporal distribution of mass averaged non-dimensional turbulent intensity in different

intake port cross sections with 20° bend angle
Fig. 27 Non-dimensional velocity field and turbulent intensity in axial midsection at 35 CAD
(35° aTDC) in a computational domain with varying cross-sectional area intake ports with 20°
bend angle
intake jets impinges on the walls of the cylinder and the piston during the induction.
In the initial stage of intake stroke, the space in the small volume contained between
the cylinder head and the piston bowl is not sufficient for a vertical flow to develop
clearly in any direction. During the induction, the intake jets impinge on the walls
of the cylinder and the piston. Due to this strong interaction of intake jets, toroidal
vortices appear near the wall of the cylinder. The two induction jets collide and
impinge on the bottom surface of the combustion chamber creating two more
toroidal vortices (recirculation zones) inside the combustion chamber bowl. In all
the cases, maximum value of turbulent intensity observed near the cylinder wall and
at the centre of the combustion chamber where the jet impinges. The turbulent
intensities are highly non-uniform with maximum values in the strong recirculation
zone where the jets collide.
Fig. 28 Non-dimensional velocity and turbulent intensity field at 165 CAD (165° aTDC) in a
computational domain with varying cross-sectional area intake ports with 20° bend angle
Figure 28 shows the distribution of non-dimensional velocity and turbulent

intensity at 165 CAD (165° aTDC). It is the final stage of intake stroke at which
intake valves are starting to close. This crank angle has been chosen to study the
flow structure development near the intake valve closing. The vortex at the centre of
the cylinder seems to have stabilized and elongated at the same location, and there
is no other vortex observed below the intake and exhaust valves have been dis-
appeared. The vortex, nearer to left-side cylinder wall and above the piston top
surface, is closer to cylinder wall, and the other one is closer to the corner adjacent
to cylinder wall and piston top surface. One more additional vortex is observed in
the centre of the combustion chamber. When the piston is approaching BDC, the
piston speed is low and the inlet velocity of the flow entering the cylinder is also
low. This makes the vortex observed earlier gets stabilized at the same location.
The turbulent intensity distribution is almost uniform. The maximum value of
turbulent intensity still appears around the valves. A small zone of strong turbulence
observed near the intake and exhaust valves. After the closing of intake valves, the
turbulence becomes more uniform during remaining part of the compression stroke
because the discharge velocities and the shear stresses induced in the top portion of
the cylinder gradually decay.
Figure 29a, b shows the distribution of non-dimensional velocity and turbulent
intensity, respectively, at 345 CAD (near the point of fuel injection). There are two
vortices with opposite rotational directions observed near the centre of the com-
bustion chamber. The maximum value of turbulent intensity is observed in the
intake side, centre and exhaust side of the combustion chamber. Maximum values
of velocity and turbulent intensity are observed in this configuration.
Figure 30a, b shows the distribution of non-dimensional velocity and turbulent
intensity at compression TDC (360 CAD). For cases a and c, two toroidal vortices with
opposite rotational directions near the combustion chamber wall and one almost at the
Fig. 29 a Non-dimensional velocity field at 345 CAD in a computational domain with varying
cross-sectional area intake ports with 20° bend angle. b Non-dimensional turbulent intensity field
at 345 CAD in a computational domain with varying cross-sectional area intake ports with 20°
bend angle
centre of the combustion chamber are observed. For the configuration under consid-
eration, there are four vortices with opposite rotational directions observed in the
combustion chamber. The appearance of toroidal vortices with opposite rotational
directions is due to strong interaction between squish and swirl near TDC and also due
to central projection in the bowl. The centrifugal force caused by the tangential vortex
impedes the flow from entering radially towards the central zone of the cylinder. The
maximum value of turbulent intensity is observed in the centre of the combustion
chamber. From the results so far presented, it may be concluded that the varying
cross-sectional area intake port configuration provides plausible results.
Fig. 30 a Non-dimensional velocity field at TDC (360 CAD) in a computational domain with
varying cross-sectional area intake ports with 20° bend angle. b Non-dimensional turbulent
intensity field at TDC (360 CAD) in a computational domain with varying cross-sectional area
intake ports with 20° bend angle
13.2 In-Cylinder Combustion Simulation
Combustion process involves simulation of the power stroke during the engine
cycle, starting from closing of valves to the end of the expansion stroke. Since the
valves are closed, the combustion chamber is the main flow domain and the piston
is the sole moving part. These simulations are also known as “in-cylinder com-
bustion”. Combustion simulation in multidimensional modelling is less complicated
geometrically than a port flow simulation. If the geometry is rotationally symmetric,
the entire domain can be modelled as a sector to speed up the calculation.
Typically, the initial flow field at this stage can be obtained from
• a cold flow simulation if 3D geometry is used;

• patching-in can be done based on a cold flow analysis;
• running the piston putting off the combustion to obtain charge compression.
As with cold flow, a moving deforming mesh model can be used for the piston
motion. Since only the piston motion is included, geometric decomposition is not
required in the simulation. The modelling of the fluid dynamics in the valve port
region and their effect on combustion need not be included in the in-cylinder
combustion simulations.
Models pertaining to the fuel spray, combustion and pollutant formation should
be included. In direct injection engines, the fuel spray from the tip of the injector
nozzle should be included at the specific crank angle and duration using appropriate
spray model. For port fuelled engines, it may be assumed that the combustion
charge is well mixed. A chemical mechanism describing the reaction of fuel vapour
with air is used to describe combustion. Models for turbulence–chemistry inter-
action should be specified. Submodels for NOx and soot formation must be
incorporated to calculate pollutant formation, which can be coupled with the
combustion calculation or calculated as a post-processing step.
The main challenge in modelling in-cylinder combustion process lies in the
physics for spray modelling and combustion. The spray can be considered as a
column of liquid entering the domain at high speed which subsequently breaks into
droplets due to aerodynamic forces. These droplets can coalesce into larger droplets
or break into even smaller droplets. During this process, droplets exchange mass
with the surrounding gases. Submodels for coalescence and break-up, as well as
heat and mass transfer, are used to capture spray dynamics. The most important
point is that the CFD mesh should be sufficiently fine to capture the coupling
between the liquid droplets and the gases in the cylinder accurately. If liquid spray
impinges upon the cylinder walls, it is possible to form a thin liquid film. This will
undergo its own processes of movement and vaporization and will require a sep-
arate treatment.
To evaluate combustion, detailed chemical mechanisms for pure fuels that
constitute the components of diesel fuel and gasoline involve hundreds of species
and thousands of reactions. These reactions are coupled with the fluid dynamics due
to the similar timescales of fluid mechanical motions and chemical reactions. The
energy release during the combustion process increases pressures and temperatures
for the working medium. This in turn affects the fluid motions inside the cylinder.
A direct computation of this coupled interaction, while including detailed chemistry
is staggeringly expensive in terms of computation time and is impractical for
complex geometries. Therefore, submodels should be included to take care of this.
Reduced order mechanisms can capture most of the essential chemistry in a
narrower range of temperature and equivalence ratio. They should be used along
with a submodel for turbulence–chemistry interaction. One such model is the
probability density function (PDF) approach which allows an efficient computation
of turbulence–chemistry interaction. Flame propagation is modelled using a pro-
gress variable-based approach which calculates the transient flame front speed and
location. These approaches enable computation of the combustion process on large

meshes in complex geometries with a reasonable computational power.
Simplified mechanisms can be used to compute the NOx formation due to
• high temperatures (thermal NOx);
• nitrogen in the fuel (fuel NOx);
• fuel reactions in the flame front (prompt NOx);
• sulphur oxides in the fuel (SOx);
• soot formation.
These pollutants are generally a very small percentage of the total mass in the
cylinder. Therefore, these calculations can be decoupled from the calculation of the
main energy release. In some cases, this is done during the post-processing oper-
ation at the end of the simulation. However, it is more accurate to include the
pollutant formation stage in the simulation. This is true especially for the pollutants
arising from incomplete combustion that oxidize later in the cycle.
13.2.1 Case Study 5: Combustion Analysis in a DI Diesel Engine
In the previous case study, we investigated the effect of port bend angle in cylinder
flow field. I found that a variable area port with 20° bend angle was giving better in
cylinder air motion. In this case study, we will move one step forward to simulate in
cylinder combustion process on a single-cylinder DI diesel engine. First, we will
explain the combustion model used and also give models for NOx and soot for-
mation, and then, the predicted results will be presented and explained.
13.3 Modelling of Combustion Process
The process of combustion is modelled using extended flame coherent model. It is

named as ECFM-3Z model which is a general-purpose combustion model capable
of simulating the complex mechanisms of turbulent mixing, flame propagation,
diffusion combustion and pollutant formation. The flame propagation phase is
modelled by the flame surface density (FSD) transport equation which is given by

@R l R
þ r:ðuRÞ r: D þ t r ¼
@t Sct q
X 2 e 2 qu X l c X 1 2 1 @p

Cdivu r:u þ Ca C þ Ul bUl þ Sconv
3 k 3 qb c 1 c 3 cp @t
where D is the molecular diffusivity, Cdivu is an empirical parameter with default

value of 1.0, C is a correction factor that takes into account the flame chemical
timescale and the flame interaction with the walls, C is the ITNFS (Net Flame
Fig. 31 ECFM 3Z combustion model
Stretch) function, qu and qb the density of the unburnt and burnt gases, Ul the
effective laminar flame speed, c is the isentropic coefficient, Sconv is an additional
contribution to the FSD from convection at the spark plug, a and b are empirical
coefficients for the production and destruction terms, lt is the turbulent viscosity,
p is the thermodynamic pressure, and c is the Reynolds-averaged progress variable.
It can also be used for in-cylinder analysis in a multi-injection environment and
for multi-cycle simulations. ECFM-3Z model is recommended for general com-
bustion simulation of both gasoline and diesel engines with homogeneous or
non-homogeneous fuel–air mixtures. “3Z” stands for three zones of mixing, namely
the unmixed fuel zone, mixed gases zone and the unmixed air plus EGR zone which
is shown in Fig. 31. The three zones are too small to be resolved by the mesh and
are therefore modelled as subgrid quantities. The mixed zone is the result of tur-
bulent and molecular mixing between gases in the other two zones and is where
combustion takes place.
13.4 NOx Model
The principal reactions for the formation of thermal NOx is recognized and pro-
posed by Zeldovich mechanism. The rate of formation of NOx is significant only at
higher temperature since the thermal fixation of nitrogen requires breaking of strong
N2 bonds. This effect is represented by the high activation energy of reaction, which
makes the reaction as the rate-limiting step of Zeldovich mechanism.
13.5 Soot Model
The formation and emission of carbonaceous particles is a process that is often

observed during the combustion of hydrocarbons. In general, the soot formations in
IC engines are due to incomplete combustion of fuel. Apart from the loss of fuel and
combustion efficiency, it affects the health of human life and environment. Hence,
there are strict legislative demands to produce cleaner engines. There are many
models proposed for soot formation. In Chapter “Characterization of Ringing
Operation in Ethanol Fueled HCCI Engine Using”, the details of all models, viz.
RNG k–e, droplet break-up, ECFM-3Z combustion model for diesel combustion,
NOx model, soot model, are explained.
Initially, a cold flow analysis was carried out and, from that analysis, it was
found that out of eight-piston top configurations studied, flat with central bowl and
pent roof with offset bowl pistons exhibits higher velocity, TKE, TR, turbulence
intensity and length scale as compared to those of other piston top configurations.
Hence, CFD analysis is extended for combustion studies with these piston top
profiles (called as optimized piston top profiles). However, only results pertaining
to flat piston-with-centre-bowl will be presented and discussed.
13.6 Spray Structure
The most significant feature of the diesel engine is the use of liquid hydrocarbon as
a fuel. The liquid fuel injected through the nozzle breaks up, atomizes and evap-
orates in a high temperature surrounding air and burns as being mixed with air. The
behaviour of fuel spray during the break-up, atomization and vaporization mainly
depends on the air entrainment with a higher velocity and the higher temperature
inside the engine cylinder [16].
Figure 32 depicts the plot of fuel injection and spray evaporation for the flat
piston-with-centre-bowl at the end of fuel injection (717 CAD). Figure 32a shows
the interaction of flow field with diesel fuel injection. Figure 32b shows the
exploded view of diesel fuel spray evaporation having larger droplets in head region
and a very less number of evaporated droplets near the tail region inside the engine
cylinder.
Fig. 32 Plot of fuel injection on flow field and spray evaporation for flat piston-with-centre-bowl
13.7 Diameter and Droplet Velocity
Figure 33 shows the variation of fuel droplet diameter with crank angles for the flat
piston top profiles with central bowl during the period of injection. From Fig. 33, it
is clear that fuel droplet diameter is around 1.8 lm during the start of injection. As
more fuel gets accumulated between 705° and 715° crank angle and because of
adhesion, the diameter of droplets gets increased. Near the end of injection as the
fuel nozzle getting closed, the fuel break-up is not proper and therefore the diameter
of the droplets shoots up.
Figure 34 depicts the variation of fuel droplet velocity with crank angles for the
flat piston top profiles with central bowl during the period of injection. From
Fig. 34, it is noted that the flat piston-with-centre-bowl gives a maximum fuel
droplet velocity of about 75 m/s.
Fig. 33 Variation of droplet

diameter with crank angle
with flat
piston-with-centre-bowl
Fig. 34 Variation of droplet

velocity with crank angle for
flat piston-with-centre-bowl
(a) At 710 CAD (b) 720 CAD
Fig. 35 Pressure contours for flat piston-with-centre-bowl
13.8 Peak Pressure
Figure 35 depicts the pressure contours for flat piston-with-centre-bowl under firing
condition at an engine speed of 1500 rpm with a conventional start of fuel injection
timing at 23° bTDC. It is observed that flat piston-with-centre-bowl develops peak
pressure of 67.9 and 77 bar at 710 and 720 CAD under full load condition which is
reasonably a good value for such a configuration.
13.9 Heat Release Rate
Figure 36 shows the variation of heat release rate with crank angles for flat
piston-with-centre-bowl at an engine speed of 1500 rpm. From Fig. 36, it shows the
early start of heat release during premixed combustion period and slower rate of
combustion during diffusion combustion period. The maximum heat release rate of
78 J/°C A is obtained with the flat piston.
Fig. 36 Heat release rate

versus crank angle
13.10 Peak Temperature
Knowledge of the temperature distribution inside a combustion chamber of an

engine is the fundamental step to understand the combustion process and flame
propagation. As well, if the peak temperature is too high it can disrupt the engine
functioning, since the engine is made of material that dilates, and as a consequence,
the dimension of the different parts of the engine gets distorted. Figure 37 illustrates
the computed temperature contour plot for two different piston top profiles on the
midplane passing between the centrelines of intake and exhaust valve through axis
for different crank angle degrees (say 720–1080 CAD in steps of 30 CAD). From
these figures, it is observed that flat piston-with-centre-bowl gives peak an about
2369 K. Further, it beautifully depicts the process of combustion during the
expansion stroke. As expressed earlier, even though such a process to visualize
720 CAD 750 CAD 780 CAD
810 CAD 840 CAD 870 CAD
Fig. 37 Temperature contours for flat piston-with-centre-bowl

experimentally is extremely difficult, CFD analysis throws a good light on theo-

retical visualization in the process of combustion.
14 Full Cycle Simulations
As the name indicates, full cycle simulations essentially involve all the five pro-
cesses, viz. suction, compression, combustion, expansion and exhaust. This means
the prediction of all details right from cold flow analysis, in-cylinder combustion
and complete simulations of the entire engine cycle. This is nothing but a transient
computation of turbulent airflow, spray and combustion, and exhaust with moving
valves and pistons. The initial flow field is obtained from a cold flow simulation or
by running the engine without combustion for a cycle before turning on spray and
combustion.
The advantage of full cycle simulations is that they provide the full picture of
engine performance, taking into account intake and exhaust valve fluid dynamics,
mixing, turbulence production, spray, combustion and flame propagation, and
pollutant formation. It is to be emphasized here that they are extremely complex to
set up and expensive to run.
The geometry preparation is a complex exercise. It includes geometry from the
throttle body, ports, valves, combustion chamber, cylinder and the piston, making it
difficult to perform clean-up, decomposition and meshing. The solver set-up has to
include moving mesh, airflow, turbulence, spray, turbulence–chemistry and flame
propagation, and pollutant formation.
From the above, the need is for process compression and automation all the way
from geometry to post-processing. If properly done, it will considerably reduce the
time needed for problem set-up and post-processing. Further, accurate and efficient
models for chemistry, spray and combustion, as well as efficient solver techniques,
are required to get the solution in the shortest time possible.
14.1 Case Study 7: Expansion Process with NOx and Soot

Prediction
So far, we have illustrated up to expansion process in the previous case studies. In

this section, we will present the results for the exhaust process so that we will have
the satisfaction of presenting all the processes of an IC engine by different case
studies, which is the aim of this chapter. Figure 38 shows the complete exhaust
process starting from 900 to 1080 CAD in terms of 30 CAD. As the piston moves
from bottom centre to top centre, it pushes the product of combustion. During the
process, the temperature of the working fluid in the cylinder drops which is well
presented in this figure.
(s) 900 CAD (t) 930 CAD (u) 960 CAD
(v) 990 CAD (w) 1020 CAD (x) 1080 CAD
Fig. 38 Temperature contours for flat piston-with-centre-bowl during exhaust
14.2 NOx and Soot Emissions
Comparison of Figs. 39 and 40 shows a trade-off between NOx and soot emissions
for the flat piston-with-centre-bowl. NOx formation at the beginning of injection is
higher. This is due to the rate of temperature rise is higher during the early part of
combustion.
In order to lower pollutant emissions of diesel engines, enormous efforts have
been taken to overcome the trade-off between NOx and soot emissions. As it is well
known, due to the NOx and soot trade-off relation, it is difficult to reduce these two
pollutants simultaneously. For example, changing the combustion chamber shape to
achieve a better in-cylinder air motion is an effective way to reduce soot formation
by increasing the peak cylinder temperature and pressure. However, this usually
results in an increase in NOx production.
Fig. 39 NOx emission versus

crank angle
Fig. 40 Soot emission versus

crank angle
15 Conclusion
As mentioned in the early part of this chapter, the main aim of this chapter is to
introduce to the readers the application of CFD in IC engine analysis and design has
been achieved by means of different case studies. Consciously, equations have been
avoided. Different case studies have been presented purposely to establish that it
can be applied to different geometries with different input conditions. From the case
studies presented, it can be concluded that mentioned aim has been achieved.
Acknowledgements The results presented here are from the theses of our Punctilious, hard
working and Dedicated (PhD) scholars. We gratefully acknowledge their efforts in producing the
results. Authors are thankful to CD-ADAPCO for providing licence free of cost for performing
CFD studies.
Bibliography
1. Amin M, Saray RK, Shafee S, Ghafouri J (2013) Numerical study of combustion and
emission characteristics of dual-fuel engines using 3D-CFD models coupled with chemical
kinetics. Fuel 106:98–105
2. Andreas WB, Josefsson G, Ekdahl R, Ogink R, Grandin B (2011) The effect of tumble flow
on efficiency for a direct injected turbocharged downsized gasoline engine, SAE Paper
No. 2011-24-0054
3. Andreassi L, Cordiner S, Rocco V (2003) Modelling the early stage of spark ignition engine
combustion using the KIVA-3V code incorporating an ignition model. Int J Engine Res 4(3):
179–192
4. Assanis DN, Hong SJ, Nishimura A, Papageorgakis G, Vanzieleghem B (2000) Stududies of
spray break-up and mixture stratification in a gasoline direct injection engine using KIVA-3V.
J Gas Turbine Power 122:485–492
5. Auriemma M, Corcione FE, Diana S, Police G, Valentino G (2001) Investigation of the intake
tumble flow in a prototype GDI engine using a steady state test rig, SAE Paper No. 2001-01-022
6. Auriemma M, Corcione FE, Martino UD, Valentino G (2001) Analysis of the intake flow in a
diesel engine head using dynamic steady flow conditions, SAE Paper No. 2001-01-1307
7. Auriemma M, Caputo G, Corcione FE, Valentino G, Riganti G (2003) Fluid-dynamics
analysis of the intake system for a HDDI diesel engine by STAR-CD code and LDA
technique, SAE Paper No. 2003-01-0002
8. Barths H, Hasse C, Peters N (2000) Computational fluid dynamics modeling of non-premixed
combustion in direct injection diesel engines. Proc IMechE Vol 1 J Engine Res
9. Beard OC, Miche M (2003) Improved modeling of DI diesel engines using sub grid
description of spray and combustion, SAE Paper No. 2003-01-0008
10. Bedford Frederick C (2001) Modeling production and abatement of nitrogen oxide pollutants
in direct injection diesel engines. University of Wisconsin-Madison
11. Benny P (2008) Effect of inlet manifold configuration on in-cylinder air motion and
combustion in a DI diesel engine, Ph.D. thesis, IIT Madras, India
12. Bertrand HD (2002) Practical diesel engine combustion analysis. SAE international,
Warrendale, PA, USA
13. Bianchi GM, Cantore G, Fontanesi S (2001) Turbulence modelling in CFD simulation of ICE
intake flows, SAE Paper No. 2001-01-049
14. Cipolat D (2007) Analysis of energy release and NOx emissions of a CI engine fuelled on
diesel and DME. J Appl Therm Eng
15. Corgard DD, Reitz RD (2001) Effects of alternative fuels and intake port geometry on HSDI
diesel engine performance and emissions, SAE Paper No. 2001-01-0647
16. Cyril C (2002) Combustion process in diesel engine. Ph.D. thesis, University of Brighton
17. Franz XT, Srinivasan S (2009) CFD-based optimization of fuel injection strategies in a diesel
engine using an adaptive gradient method. Appl Math Model 33:1366–1385
18. Ganesan V (2000) Computer simulation of compression—ignition engine processes.
Universities Press, Hydrabad
19. Ghandhi JB, Herold RE, Shakal JS, Strand TE (2005) Time resolved particle image
velocimetry measurements in an internal combustion engine, SAE Paper No. 2005-01-3868
20. Gnana Sagaya Raj A (2014) In-cylinder air motion and combustion in diesel engines, Ph.D.
thesis, IIT Madras, India
21. Gnana Sagaya Raj A, Mallikarjuna JM, Ganesan V (2012) Flow investigation on different
combustion chamber configuration in a DI diesel engine—a CFD approach. Int J Adv Mater
Res 505:420–423
22. Gnana Sagaya Raj A, Mallikarjuna JM, Ganesan V (2013) Analysis of in-cylinder air motion
in a DI diesel engine with four different piston bowl configuration—a CFD and PIV
comparison, SAE Paper No. 2013-01-2786
23. Gnana Sagaya Raj A, Mallikarjuna JM, Ganesan V (2013) CFD prediction of combustion on
direct injection diesel engine with two different combustion chamber configurations, SAE
Paper No. 2013-01-2804
24. Gnana Sagaya Raj A, Mallikarjuna JM, Ganesan V (2013) Energy efficient piston
configuration for effective air motion—a CFD study. Appl Energy 102:347–354
25. Goldsworthy L (2006) Computational fluid dynamics modeling of residual fuel oil
combustion in the context of marine diesel. IMechE J Engine Res 7:181–199
26. Heywood JB (1988) Engines: an introduction. McGraw-Hill Inc, NY
27. Hossainpour S, Binesh AR (2009) Investigation of fuel spray atomization in a DI heavy-duty
diesel engine and comparison of various spray breakup models. Fuel 88:799–805
28. Hountalas DT, Papagiannakis RG (2000) Development of a simulation model for direct
injection dual fuel diesel-natural gas engines, SAE Paper No. 2000-01-0286. https://www.
sharcnet.ca/Software/Ansys/17.0/en-us/help/wb_icom/icom_intro_fd.html
29. Huan YC, Zhou L, Li B, Jiang DM (2001) Experimental study on fuel spray development
history near the spark plug gap in the direct injection circularly stratified charge combustion
system. Trans Chin Soc Int Combust Engines 19(6):497–501
30. Huang RF, Huanga CW, Changa SB, Yanga HS, Lina TW, Hsub WY (2004) Topological
flow evolutions in cylinder of a motored engine during intake and compression strokes.
J Fluids Struct 20:105–127
31. Jaichandar S, Annamalai K (2012) Influences of re-entrant combustion chamber geometry on
the performance of Pongamia biodiesel in a DI diesel engine. Energy 44:633–640
32. James Gunasekaran E (2008) Modeling of flow and combustion in a four stroke DISI engine,
Ph.D. thesis, IIT Madras, India
33. Jayashankara B (2009) Multi-dimensional modeling of CI engine processes, Ph.D. thesis, IIT
Madras, India
34. Jayashankara B, Ganesan V (2010) Effect of fuel injection timing and intake pressure on the
performance of a DI diesel engine—a parametric study using CFD. Energy Convers Manag
51:1835–1848
35. Kar MP, Kiat H, Gan S (2012) In-cylinder diesel spray combustion simulations using parallel
computation: a performance benchmarking study. Appl Energy 93:466–478
36. Kawahara N, Tomital F, Kasaharal D, Sumida M (2004) Liquid sheet break-up of
high-pressure swirl injector for DISI engine. In: Comodia 2004
37. Kulkarni Y, Aghav Y, Sohi NS, Dani AD (2005) Optimization of inlet port performance on
emission compliance of naturally aspirated DI diesel engine, SAE 2005-26-010
38. Kyungnam K, Arai M (2002) Diesel spray and adhering fuel on an impingement wall, SAE
Paper No. 2002-01-1628
39. Lappas P, Evans RL (2006) A numerical and experimental study of the squish-jet combustion
chamber. ImechE Int J Engine Res 7:471–487
40. Lee BH, Song YJ, Chang JH, Jeon CH (2010) Effect of the number of fuel injector holes on
characteristics of combustion and emissions in a diesel engine. Int J Autom Technol 11(6):
783–791
41. Lee D, Rutland CJ (2001) Multidimensional modeling of a six mode diesel test cycle using a
PDF combustion model, SAE Paper No. 2001-01-0585
42. Li Y, Zaho H, Leach B, Ma T (2004) Development of a fuel stratification spark ignition
engine. Proc Inst Mech Eng J Autom Eng 219(7):923–934
43. Li Y, Zaho H, Leach B, Ma T, Ladommatos N (2004) Characterization of an in-cylinder flow
structure in a high-tumble spark ignition engine. Int J Engine Res 5(5):375–400
44. Li Y, Zhao H, Peng Z, Ladommatos N (2002) Particle image velocimetry measurement of
in-cylinder flow in internal combustion engines—experiment and flow structure analysis. Proc
Inst Mech Eng 216(Part D):65–81
45. Lin L, Shulin D, Jin X, Jinxiang W, Xiaohong G (2000) Effects of combustion chamber
geometry on in-cylinder air motion and performance in DI diesel engine, SAE Paper
No. 2000-01-0510
46. Liu Y, Lu F, Reitz RD (2005) The use of non-parametric regression to investigate the
sensitivities of high-speed direct injection diesel emissions and fuel consumption to engine
parameters. IMechE Vol 7 J Engine Res
47. Lu L, Shulin D, Jin X, Jinxiang W, Xiaohong G (2000) Effects of combustion chamber
geometry on in-cylinder air motion and performance in DI diesel engine, SAE Paper
No. 2000-01-0510
48. Lumley JL (1999) Engines: an introduction. Cambridge University Press
49. Micklow GJ, Gong W (2001) Combustion modeling for direct injection diesel engines. Proc
Inst Mech Eng 215(Part D):191–205
50. Micklow GJ, Gong WD (2007) Intake and in-cylinder flowfield modelling of a four-valve
diesel engine. Proc IMechE J Autom Eng 221(Part D):1425–1440
51. Mikalsen R, Roskilly AP (2009) A computational study of free-piston diesel engine
combustion. Appl Energy 86:1136–1143
52. Miles PC (2000) The influence of swirl on HSDI diesel combustion at moderate speed and
load, SAE paper No. 2000-01-1829
53. Mohamed AJ, Kantchev G, Abid MS (2011) Influence of intake manifold design on
in-cylinder flow and engine performances in a bus diesel engine converted to LPG gas fuelled,
using CFD analyses and experimental investigations. Energy 36:2701–2715
54. Muralikrishna B, Mallikarjuna JM (2008) Effect of engine speed on intake valve flow
characteristics of a diesel engine—an analysis using particle image velocimetry. In: First
international conference on emerging trends in engineering and technology, IEEE, New York.
https://doi.org/10.1109/ICETET.2008.97
55. Muralikrishna B, Mallikarjuna JM (2009) Tumble flow analysis in an unfired engine using
particle image velocimetry. World Acad Sci Eng Technol 54:430–435
56. Paul CM, Collin R, Hildingsson L, Hultgvist A, Andersson OI (2007) Combined
measurements of flow structure, partially oxidized fuel, and soot in a high-speed,
direct-injection diesel engine. In: Proceedings of the combustion institute, vol 31, Issue 2
57. Payri F, Benajes J, Margot X, Gil A (2003) CFD modeling of the in-cylinder flow in
direct-injection diesel engines. Comput Fluids 33:995–1021
58. Payri F, Desantes JM, Pastor VJ (1996) LDV measurements of the flow inside the combustion
chamber of a 4-valve D.I. diesel engine with axisymmetric piston-bowls. Exp Fluids
59. Prasad BVVSU, Sharma CS, Anand TNC, Ravikrishna RV (2011) High swirl-inducing piston
bowls in small diesel engines for emission reduction. Appl Energy 88:2355–2367
60. Ra Y, Hruby EJ, Reitz RD (2005) Parametric study of combustion characteristics in a
direct-injection diesel homogeneous charge compression ignition engine with a low-pressure.
Proc IMechE J Engine Res 6:215–230
61. Rakopoulos CD, Hountalas DT, Rakopoulos DC, Giakoumis EG (2005) Experimental heat
release rate analysis in both chambers of an indirect injection turbocharged diesel engine at
various load and speed conditions, SAE Paper No. 2005-01-0926
62. Rakopoulos CD, Kosmadakis GM, Pariotis EG (2010) Investigation of piston bowl geometry
and speed effects in a motored HSDI diesel engine using a CFD against a quasi-dimensional
model. Energy Convers Manag 51:470–484
63. Ramdasi SS, Marathe NV, Moorthi S, Balasubramanian AK (2005) An integrated design
approach for robust engine development: a case study, SAE Paper No. 2005-26-002
64. Sagaya Raj A, Mallikarjuna J, Venkitachalam G (2013) CFD prediction of combustion on
direct injection diesel engine with two different combustion chamber configurations, SAE
Technical Paper 2013-01-2804. https://doi.org/10.4271/2013-01-2804
65. Saijo K, Kojima T, Nishiwaki K (2005) Computational fluid dynamics analysis of the effect of
mixture heterogeneity on combustion process in a premixed charge compression ignition
Engine. IMechE J Engine Res 6(5)
66. Sangjin H, Wooldridge MS, Hong GI, Assanis DN, Pitsch H (2005) Development and
application of a comprehensive soot model for 3D CFD reacting flow studies in a diesel
engine. Combust Flame 143:11–26
67. Shabbir S, Gokhale N, Thatte V, Gandhi N, Deshmukh B, Aghav Y, Kumar M,

Nandgaonkar M, Babu MK (2011) Efficient approach for optimization of piston bowl shape,
compression ratio and EGR for DI diesel engine, SAE Paper No. 2011-24-0013
68. Shi Y, Reitz RD (2008) Optimization study of the effects of bowl geometry, spray targeting,
and swirl ratio for a heavy-duty diesel engine operated at low and high load. ImechE Int J
Engine Res 9:325–346
69. STAR-CD Version 324 user guide, Computational Dynamics ltd. London
70. Star Methodology 3.2, CD-ADAPCO Ltd
71. STAR-CD Version 324 Methodology guide, Computational Dynamics ltd. London
72. Steeper RR, Stevens EJ (2000) Characterisation of combustion, piston temperatures, fuel
sprays and fuel air mixing in a DISI optical engine, SAE paper 2000-01-2900
73. Thornhill D, Li H, Fleck R, Cuningham G (2006) Modeling of unsteady gas-dynamic flow in
a pipe and at its exit using CFD. Proc IMechE J Autom Eng 220(Part D)
74. Tonini S, Gavaises M, Theodorakakos A (2008) The role of droplet fragmentation in
high-pressure evaporating diesel sprays. Int J Therm Sci 48:554–572
75. Towers DP, Towers CE (2004) Cyclic Variability measurements of in-cylinder engine flows
using high speed particle image velocimetry. Meas Sci Technol, 1917–1925
76. Treea DR, Svenssonb KI (2006) Soot processes in compression ignition engines. Prog Energy
Combust Sci
77. Vanderwedge VA, Lounsberry TH, Hochgreb S (2000) Numerical modelling of fuel sprays in
DISI engines under early operating conditions, SAE paper 2000-01-0273
78. Versteeg H, Malalasekara W (2008) An introduction to computational fluid dynamics—the
finite volume method, 2nd edn. Pearson Publication
79. Vincent M, Barton I, Angelberger C, Poinsot T (2004) CFD evaluation of effects of split
injection on combustion and emissions in a DI diesel engine, SAE No. 11PFL-0029
80. Vincent M, Barton I, Angelberger C, Poinsot T (2004) Towards large eddy simulation in
internal-combustion engines: simulation of a compressed tumble flow, SAE Paper
No. 2004-01-1995
81. Vitek O, Polasek M (2002) Tuned manifold systems-application of 1-D pipe model, SAE
Paper No. 2002-01-0004
82. Wagner GJ, Liu WK (2000) Turbulence simulation and multiple scale subgrid models.
Comput Mech 25:117–136
83. Willy D, Dupont A, Baby X, Charnay G, Boree J (2003) PIV measurements of internal
aerodynamic of diesel combustion chamber, SAE Paper No. 2003-01-3083
84. Willy D, Dumont P, Dupont A, Baby X, Charnay G, Boree J (2004) Airflow cyclic variations
analysis in diesel combustion chamber by PIV measurements, SAE 2004-01-1410
85. Wu HW, Peng SW (2004) Numerical analysis of thermal turbulent flow in the bowl-in-piston
combustion chamber of a motored engine. Int J Therm Sci 43:1011–1023
86. Xu H (2001) Some critical technical issues on the steady flow testing of cylinder heads, SAE
Paper No. 2001-01-1308
87. Young J, Munoz RH, Anderson RW, Lavoie G (2000) Study of stratified charge DISI engine
with an air force fuel injection system, SAE paper 2000-01-2901
88. Yun H, Reitz RD (2005) Combustion optimization in the low-temperature diesel combustion
regime. Proc IMechE Int J Engine Res 6(Part D)
89. Zhang Y, Tomoaki I, Nishida K (2001) Characterisation of mixture formation in
split-injection diesel sprays via laser absorption-scattering (LAS) technique, SAE Paper
No. 2001-01-3498
90. Zhao H, Li Y, Peng Z, Ladommatos N (2002) Tumbling flow analysis in a four valve spark
ignition engine using particle image velocimetry. Int J Engine Res 3(3):139–155
91. Zhen L, Wang T, Li X, Li L, Zhang G (2012) Parametric design of the tangential intake port
in diesel engines. Proc IMechE J Autom Eng 221(Part D):1–13
92. Zheng OP, Zhang HM, Zhang DF (2005) A computational study of combustion in
compression ignition natural gas engine with separated chamber. Fuels 84
93. Zhu Y, Zhao H, Ladomatos N (2005) Computational fluid dynamics study of the effects of the
RE_Entrant lip shape and toroidal radii of piston bowl on a high-speed direct injection diesel
engine’s performance and emissions. Proc IMechE J Autom Eng 219(Part D)
94. Zhu Y, Zhao H, Melas DA, Ladommatos N (2004) Computational study of the effects of the
re-entrant lip shape and toroidal radii of piston bowl on a HSDI diesel engine’s performance
and emissions, SAE Paper No. 2004-01-0118
95. Zhu Y, Zhao M, Ladommatos N (2003) Computational study of the effects of injection
timing, EGR and swirl ratio on a HSDI multi-injection diesel engine emission and
performance, SAE Paper No. 2003-01-0346

Application of CFD For Analysis and Design of IC Engines

Uploaded by

Copyright:

Available Formats

Application of CFD For Analysis and Design of IC Engines

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Application of CFD For Analysis and Design of IC Engines

Uploaded by

Copyright:

Available Formats

Application of CFD for Analysis

and Design of IC Engines

Vijayashree and V. Ganesan

Vijayashree V. Ganesan (&)

© Springer Nature Singapore Pte Ltd. 2018 251

The performance of an IC engine depends upon complex interactions between

• creating a computational mesh in both the moving and non-moving portions of

Efﬁcient design of an IC engine requires a number of critical decisions to establish

• The important point to be taken into account in port design is to achieve

4 Fluid Dynamics During the Four Strokes

Combustion increases pressure and temperature as the energy contained in the

5 Design of Engines with High Efﬁciency

5.1 Port Flow Design

5.2 Combustion Chamber and Piston Shape

Fig. 1 Straight valve engine [1]

Fig. 2 Canted valve engine [1]

In addition, geometric design consideration is to have a “squish” region. Squish is

Fig. 3 Illustration of swirl and squish

5.4 Compression Ratio

Fig. 4 Illustration of total volume and clearance volume

5.5 Design for Low Speed and Idle

5.6 Spark and Injection Timing

6 Application of CFD in IC Engine Design

6.1 The Role of CFD in Engine Design

As already described, IC engines involve complex fluid dynamic interactions

From the CFD results, it is possible to visualize the flow phenomena on 3D

6.2 Port Flow Analysis

6.2.1 Case Study 1: Steady Flow CFD Analysis for Improving

Fig. 6 CAD model of inlet port geometry with grid

2mm Valve lift 4mm Valve lift 5.6mm Valve lift

Fig. 7 Meshed geometry of the inlet port at three-valve lift position

2 mm valve lift (IVO)

4 mm valve lift (IVO)

5.6 mm valve lift (Max. valve lift)

2 mm valve lift (IVO)

4 mm valve lift (IVO)

5.6 mm valve lift (Max. valve lift)

Fig. 9 Total pressure contours for the three-valve lifts

by which volumetric efﬁciency could be considerably improved and a ﬁnal

Exhaust Port and Valve

Hemispherical combustion chamber

Fig. 10 Meshed geometry of the exhaust manifold

6.2.2 Case Study 2: Steady Flow Analysis on Multi-cylinder Inlet

Exhaust Valve lift = 2 mm

Exhaust Valve lift = 6 mm

Fig. 11 Velocity vectors in exhaust manifold for two-valve lifts

Fig. 12 Final design of intake and exhaust manifold

Fig. 13 Meshed geometry of the multi-cylinder inlet manifold

Fig. 15 Flow velocity vectors at various lift conditions in manifold 3

therefore it was a challenging job to undertake. The necessary boundary conditions

Flow Analysis in Various Regions

As shown in Fig. 15, in region 1, there is a formation of strong recirculation zone.

6.2.3 Case Study 3: Steady Flow Analysis on Six-Cylinder V-Engine

Fig. 16 CAD model of the modiﬁed design of the inlet manifold

Fig. 17 Six-cylinder inlet manifold versus V type engine

Fig. 18 Grid arrangement when the power valve is in open position

6.3 Cold Flow Analysis

Fig. 19 Flow velocity vector for all the six manifolds

6.3.1 Case Study 4: Cold Flow Transient Analysis