You are on page 1of 24

1.

INTRODUCTION
All internal combustion engines generate power by creating explosions using fuel and air. These explosions occur inside the
engine's cylinders and push the pistons down, which turns the crankshaft. Some of the power thus produced is used to prepare the
cylinders for the next explosion by forcing the exhaust gases out of the cylinder, drawing in air (or fuel-air mixture in non-diesel
engines, and compressing the air or fuel-air mixture before the fuel is ignited.
Fig 1. Working of four stroke engine.

!"
There are se#eral differences between diesel engines and non-diesel engines. $on-diesel engines combine a fuel mist with air
before the mixture is taken into the cylinder, while diesel engines in%ect fuel into the cylinder after the air is taken in and compressed.
$on-diesel engines use a spark plug to ignite the fuel-air mixture, while diesel engines use the heat created by compressing the air in the
cylinder to ignite the fuel, which is in%ected into the hot air after compression. &n order to create the high temperatures needed to ignite
diesel fuel, diesel engines ha#e much higher compression ratios than
gasoline engines. 'ecause diesel fuel is made of larger molecules than gasoline, burning diesel fuel produces more energy than burning
the same #olume of gasoline. The higher compression ratio in a diesel engine and the higher energy content of diesel fuel allow diesel
engines to be more efficient than gasoline engines.
1.1. Formation of Nitrogen Oxides (NOx)
The same factors that cause diesel engines to run more efficiently than gasoline engines also cause them to run at a higher
temperature. This leads to a pollution problem, the creation of nitrogen oxides ($(x. )ou see, fuel in any engine is burned with extra
air, which helps eliminate unburned fuel from the exhaust. This air is approximately *+, nitrogen and !-, oxygen.
.hen air is compressed inside the cylinder of the diesel engine, the temperature of the air is increased enough to ignite diesel fuel
after it is ignited in the cylinder. .hen the diesel fuel ignites, the temperature of the air increases to more than -/001 and the air
expands pushing the piston down and rotating the crankshaft.

!"
Fig 2. NOx formation one.
2enerally the higher the temperature, the more efficient is the engine
-. 2ood 3erformance
!. 2ood 4conomy
Some of the oxygen is used to burn the fuel, but the extra is supposed to %ust pass through the engine unreacted. The nitrogen, since it
does not participate in the
!"
combustion reaction, also passes unchanged through the engine. .hen the peak temperatures are high enough for long periods of time,
the nitrogen and oxygen in the air combines to form new compounds, primarily $( and $(!. These are normally collecti#ely referred to
as 5$(x6.
1.2. !ro"#ems of NOx
$itrogen oxides are one of the main pollutants emitted by #ehicle engines. (nce they enter into the atmosphere, they are spread
o#er a large area by the wind. .hen it rains, water then combines with the nitrogen oxides to form acid rain. This has been known to
damage buildings and ha#e an ad#erse effect on ecological systems.
Too much $(x in the atmosphere also contributes to the production of S7(2. .hen the sunrays hit these pollutants S7(2 is
formed. $(x also causes breathing illness to the human lungs.

1.$. %!& %mission 'tandards
Since -+**, $(x emissions from diesel engines ha#e been regulated by the 43A
(4n#ironmental 3rotection Agency. &n (ctober !00!, new $(x standards re8uired the
diesel engine industry to introduce additional technology to meet the new standards
The 43A has regulated hea#y duty diesel engines since the -+*0s. The following chart shows the trend to e#er-lower emissions.
9nderstanding the details of the chart is not of interest to most truckers. 4#en though the emissions standards become increasingly more
difficult to meet, the diesel engine industry has always been able to continue to impro#e engine durability, reliability, performance, and
!"
fuel economy. A 8uick look at the bottom right hand side of the chart also shows that emissions from diesel engines built in !00* and
beyond will approach :ero.
Fig $. %!& (ea)* Dut* %ngine %mission 'tandards
1.+. (o, -an NOx "e redu-ed.
Since higher cylinder temperatures cause $(x, $(x can be reduced by lowering cylinder temperatures. ;harge air coolers are
already commonly used for this reason.
<educed cylinder temperatures can be achie#ed in three ways.
4nriching the air fuel (A=1 mixture.
!"
>owering the compression ratio and retarding ignition timing.
<educing the amount of (xygen in the cylinder
4nriching the air fuel (A=1 mixture to reduce combustion temperatures. ?owe#er, this increases ?; and carbon monoxide (;(
emissions. Also >owering the compression ratio and <etarded &gnition Timing make the combustion process start at a less than the
optimum point and reduces the efficiency of combustion.
Fig +. NOx redu-tion "* #o,ering t/e tem0erature
These techni8ues lowers the cylinder temperature, reducing $(x, but it also reduces fuel economy and performance, and creates
excess soot, which results in more fre8uent oil changes. So, the best way is to limit the amount of (xygen in the cylinder. <educed
oxygen results in lower cylinder temperatures. This is done by circulating some exhaust gas and mixing it into the engine inlet air. This
process is known as 4xhaust 2as <ecirculation.
!"
!. 4@?A9ST 2AS <4;&<;9>AT&($
4xhaust 2as <ecirculation is an efficient method to reduce $(x emissions from the engine. &t works by recirculating a 8uantity of
exhaust gas back to the engine cylinders. &ntermixing the recirculated gas with incoming air reduces the amount of a#ailable (! to the
combustion and lowers the peak temperature of combustion. <ecirculation is usually achie#ed by piping a route from the exhaust
manifold to the intake manifold. A control #al#e within the circuit regulates and times the gas flow.
2.1. Uses of %x/aust 1as Re-ir-u#ation
1irst, exhaust gas recirculation reduces the concentration of oxygen in the fuel-air mixture. 'y replacing some of the oxygen-rich
inlet air with relati#ely oxygen-poor exhaust gas, there is less oxygen a#ailable for the combustion reaction to proceed. Since the rate of
a reaction is always dependent to some degree on the concentration of its reactants in the pre- reaction mix, the $(x-producing reactions
proceed more slowly, which means that less $(x is formed.
&n addition, since there is less oxygen a#ailable, the engine must be ad%usted to in%ect less fuel before each power stroke. Since we
are now burning less fuel, there is less heat a#ailable to heat the fluids taking place in the reaction. The combustion reaction therefore
occurs at lower temperature. Since the temperature is lower, and since the rate of the $(x-forming reaction is lower at lower
temperatures, less $(x is formed.
2.2. 2asi- !arts Of %1R
There are A basic parts of 42<
42< Bal#e
!"
42< ;ooler
42< Transfer 3ipe
Typical 1our Stroke Ciesel 4ngine with 'asic 3arts of 42<
Figure 3
!"
.hen 42< is re8uired engine electronic controls open the 42< #al#e. The exhaust gas then flows through the pipe to the cooler. The
exhaust gases are cooled by water from the truck cooling system. The cooled exhaust gas then flow through the 42< transfer pipe to the
intake manifold.
Figure 4

2.$. %1R O0erating Conditions
There are three operating conditions. The 42< flow should match the conditions
-. ?igh 42< flow is necessary during cruising and midrange acceleration
!. >ow 42< flow is needed during low speed and light load.
!"
A. $o 42< flow should occur during conditions when 42< flow could ad#ersely affect the engine operating efficiency or #ehicle
dri#ability. ie, during engine warm up, idle, wide open throttle, etc.
2.+. %1R Im0a-t on %C'
The 4;7 (4lectronic ;ontrol 7achine considers the 42< system as an integral part of the entire 4;S. Therefore the 4;7 is
capable of neutrali:ing the negati#e aspects of 42< by programming additional spark ad#ance and decreased fuel in%ection duration
during periods of high 42< flow. 'y integrating the fuel and spark control with the 42< metering system, engine performance and the
fuel economy can actually be enhanced when the 42< system is functioning as designed.
2.3. %1R T/eor* of O0eration
The purpose of the 42< system is to precisely regulate the flow under different operating conditions. The precise amount of
exhaust gas must be metered into the intake manifold and it #aries significantly as the engine load changes. 'y integrating the fuel and
spark control with the 42< metering system, engine performance and the fuel economy can be enhanced. 1or this an 4;7 (4lectronic
;ontrol 7achine is used to regulate the 42< flow. .hen 42< is re8uired 4;7 opens the 42< #al#e.The 4;7 is capable of
neutrali:ing the negati#e aspects of 42< by programming additional spark ad#ance and decreased fuel in%ection duration during periods
42< flowThe exhaust gas then flows through the pipe to the cooler. The exhaust gases are cooled by water from the #ehicleDs cooling
system. The cooled exhaust gas then flow through the 42< transfer pipe to the intake manifold.
!"
Fig 5. Re#ations/i0 "et,een %1R Ratio and 6oad
+. %1R 6I7IT'
This is based on an experiment conducted. The research ob%ecti#e is to de#elop fundamental information about the relationship
between 42< parameters and diesel combustion instability and particulate formulation so that options can be explored for maximi:ing
the practical 42< limit, thereby further reducing nitrogen oxide emissions while minimi:ing particulate formation. A wide range of
instrumentation was used to
ac8uire time-a#eraged emissions and particulate data as well as time-resol#ed combustion, emissions, and particulate data. The
results of this in#estigation gi#e insight into the effect of 42< le#el on the de#elopment of gaseous emissions as well as mechanisms
responsible for increased particle density and si:e in the exhaust. A sharp increase in hydrocarbon emissions and particle si:e and density
was obser#ed at higher 42< conditions while only slight changes were obser#ed in con#entional combustion parameters such as heat
release and work. Analysis of the time-resol#ed data is ongoing.

!"
The ob%ecti#e of this work is to characteri:e the effect of 42< on the de#elopment of combustion instability and particulate
formation so that options can be explored for maximi:ing the practical 42< limit. .e are specifically interested in the dynamic details
of the combustion transition with 42< and how the transition might be altered by appropriate high-speed ad%ustments to the engine. &n
the long run, we con%ecture that it may be possible to alter the effecti#e 42< limit (and thus $(x performance by using ad#anced engine
control strategies.

4xperiments were performed on a -.+ liter, four-cylinder Bolkswagen turbo-charged direct in%ection engine under steady state, low
load conditions. 4ngine speed was maintained constant at -!00 rpm using an absorbing dynamometer and fuel flow was set to obtain
A0, full load at the 0, 42< condition. A system was de#ised to #ary 42< by
manually deflecting the 42< di#erter #al#e. The precise 42< le#el was monitored by comparing $(x concentrations in the exhaust and
intake. $(x concentrations were used because of the high accuracy of the analy:ers at low concentrations found in the intake o#er a wide
range of 42< le#els.
+.1. Com"ustion C/ara-teriation ,it/ (C and NOx %missions
Steady state measurements were made of ;(, ;(!, ?;, $(x, and (! concentrations in the raw engine-out exhaust using
<osemount and ;alifornia Analytical analy:ers. ;rank angle resol#ed measurements were also made of ?; concentration in the exhaust
using a 1ast 1lame &oni:ation Cetector. The ?; sampling probe was located in the exhaust manifold and the data were recorded.
!"
Fig 8. Trade9off "et,een (C and NOx -on-entration as a fun-tion of %1R 6e)e#
Time-a#eraged ?; and $(x concentrations in the raw engine-out exhaust are shown in the 1igure #ersus 42< le#el. This figure
shows $(x concentration decreasing and ?; increasing with increasing 42< as would be expected. $ote the sudden increase in ?; and
le#eling-off in $(x at approximately "/, 42<, where there appears to be a significant shift in combustion chemistry. This ma%or
transition is in sharp contrast to the slight changes obser#ed in the integrated pressure parameters, ?< and &743. 'ecause of the
suddenness of the emissions change at "/, 42<, it is clear that dynamic engine beha#ior at or abo#e this operating point will be highly
nonlinear. Thus it is imperati#e that any control strategies being considered should be able accommodate such beha#ior.
!"
+.2. Com"ustion C/ara-teriation ,it/ !7
(ur measurements ha#e identified significant changes in 37 emissions with 42< le#el as was expected. Similar to the gaseous
emissions (e.g., ?; and $(x, there was a sharp increase in 37 at a critical 42< le#el. This critical le#el corresponding to a sharp
increase in 37 was obser#ed in mass concentration, particle si:e, and particle density.
a) 7ass Con-entration
A Tapered 4lement (scillating 7icrobalance (T4(7 was used to measure particulate mass concentration and total mass
accumulation as a function of time. A sample of diluted exhaust is pulled through a -! mm filter to the end of a tapered 8uart: element.
The fre8uency of the element changes with mass accumulation. The instrument has approximately A sec resolution on mass
concentration.
3article mass concentration and total mass accumulation were measured on dilute exhaust using the T4(7. 7ass accumulation
rates were calculated based on o#er -00 mass data points and are shown in the figure as a function of 42< le#el. 7ass accumulation
rates begin to increase significantly at A0, 42< and continue to increase rapidly until the maximum 42< le#el. The intersection of the
particulate mass and $(x cur#es represents a region where the engine out particulate mass and $(x concentration are minimi:ed for this
engine condition.
!"
Fig :. Re#ation of !7 &--umu#ation Rate and NOx emission ,it/ %1R.
") !arti-#e 'ie
A Scanning 7obility 3article Si:er (S73S was used to measure the steady state si:e distribution of the particulates in the exhaust
stream. The particles are neutrali:ed and then sorted based on their electrical mobility diameter. The range of the S73S was set at -- nm
E /0/ nm.
!"
3article si:ing was performed on dilute exhaust using the S73S. $umber concentration #s. particle diameter is shown in the figure
for se#eral 42< le#els. Two aspects of the data stand out. The first is the increasing number concentration with le#el of 42<. The
second is the increasing particle si:e. $ote that the particle si:e at the peak concentration increases by a factor of approximately two
between A0, and /A, 42<.
1ig -0. Time-a#eraged si:e distributions as measured by the S73S.
The likely mechanism for particle growth is the reintroduction of particle nuclei into the cylinder during 42<. The recirculating
exhaust particles ser#e as sites for further condensation and accumulation leading to larger particles. A significant fraction of the
!"
measured si:e distribution appears larger than the /00 nm upper bound of the S73S for the highest 42< rates. This is significant
because these particles contain much of the exhaust particulate mass.
The fre8uency plot in the figure illustrates the disappearance of small particles and the growth of much larger particles. The
di#ergence between the cur#es for particles F -00 nm and particles G0--00 nm increases significantly at A0, 42< and continues to
increase. The figure does appear to show that the smallest particles are contributing to the growth of the largest ones. The increase in
larger particles is less steep than the increase in particle mass in the figure.
1ig --. 1re8uency of occurrence of particle si:e classes as a function of 42<.
!"
+.$. NOx redu-tion effe-t of %1R
Fig. 12 shows the typical $(x reduction effect of 42< at the mid-speed range of the test engine.9nder all load conditions, the
amount of $(x decreases as the 42< rate increases. The graph also shows that the $(x reduction cur#es with the 0 , 42< point as
the origin slope downward at different angles according to the loadH the higher the load, the steeper the angle. &n other words, the $(x
reduction effect at the same 42< rate
increases as the engine load becomes higher.
!"
Fig.12. Relationship between EGR rate and NOx
&t is generally known that there are two reasons to reduce $(x by 42<. The first of them is the reduction of combustion
temperature. The addition of exhaust gases to the intake air increases the amount of combustion- accompanying gases (mainly ;(!,
which in turn increases the heat capacity and lowers the combustion temperature. The second effect is the reduction of oxygen
concentration in the intake air, which restrains the generation of $(x. Fig. 1$ shows the $(x emission test results as a function of the
concentration of oxygen in the intake air=42< gas mixture. This graph shows that the $(x reduction rate depends mostly on oxygen
concentration, and not on the engine load or 42< rate.
!"
1ig ! <elationship between oxygen concentration and $(x reduction
Fig.1$ shows the results of $(x emission tests conducted while #arying both the engine operating conditions and 42< rate, in
which the test results shown in Fig. 1$ are merged. As in Fig.1$, almost all the data are on or in a single cur#e, indicating that there is a
strong correlation between the oxygen concentration and $(x reduction rate. The reason for this is thought to be as followsI &n Fig.12,
the $(x reduction rate under a certainload is different from that under another load e#en when the 42< rate remains the same because
the difference in load causes a difference in the amount of combustion-accompanying gases and oxygen concentration in 42< gas,
which in turn changes the oxygen concentration in the intake gas (mixture of intake air and EGR gas).
3. INT%RN&6 %1R
.hen a fraction of the combustion products is still present in the cylinder at the moment that the exhaust #al#es close, the mixture
at the beginning of the next engine cycle will consist of air and fuel, as well as combustion products. These products are called internal
!"
42< (in contrast to external 42<, which means that exhaust gases are recycled to the intake system, after which they mix with the air
and fuel. The fraction of internal 42< that is present in the cylinder at the beginning of the compression stroke is mainly dependent on
the timing of the intake and exhaust #al#es.
The )a#)e timing of traditional engines, such as the Ciesel and (tto engines, is such that the fraction of exhaust gases (or
residuals at the start of the cycle is as small as possible. Traditional engines ha#e <esidual 2as 1ractions (<21 in the range /--/ mass
,.
4. T%C(NIC&6 I''U%'
4.1. Com"ustion Contamination
4xhaust gas from any combustion process may ha#e certain contaminants, including acid forming compounds, unburned and
partially burned hydrocarbons, air pollutants, and li8uid water. These contaminants can be successfully reintroduced into the combustion
chamber but may lead, o#er time, to serious combustion degradation and instability, and shorter component life. Such effects need to be
fully understood and documented, and appropriate impro#ements made to the combustion process to protect the customerDs in#estment
and maintain true long-term emissions compliance. This acti#ity would be a key element of any ma%or engine manufacturerDs
de#elopment process.
4.2. Contro# '*stem 'ta"i#it*
!"
;ontrol systems for modern engines ha#e been de#eloped o#er two decades and in#ol#e integrated strategies to ad%ust air=fuel
ratio, ignition timing, and air flow rates to maintain emissions control at #arying loads, speeds, and fuel conditions. These systems are at
the heart of successful engine operation today and are #ital to satisfactory long term operation. Adding 42< into the combustion process
introduces further complexity that must be carefully integrated into the entire engine control system approach for successful operation
o#er a wide range of conditions. 1or instance, if fuel 8uality changes o#er time, the air=fuel ratio, ignition timing, air system rates, and
the 42< rate must be ad%usted accordingly to keep the combustion system stable and emissions in compliance. (n the other hand, if the
engineDs load changes rapidly from part load to full load and back to part load, the 42< system dynamics must be included in the o#erall
control strategy response to make sure the engine operates smoothly during this transition.
4.$. 7ateria#s and Dura"i#it*
42< systems may decrease long-term life of the components affected, including the 42< coolers and control #al#es, the pistons
and cylinder heads, exhaust manifolds and sensors, as well as the post engine catalyst. (perating a few hundred hours per year may not
lead to any significant materials degradation in the o#erall lifespan of an engine. ?owe#er, continuous duty applications at J/00 hours
per year may cause near term emissions noncompliance and longer term materials breakdown, shorter component life, and e#en
unexpected, catastrophic engine failures. To minimi:e or eliminate the potentially negati#e impacts of 42< on engine components,
compatible components and designs must be used that often re8uire thousands of hours of lab and field test operation for #alidation.
Although both expensi#e and time consuming, such efforts are a necessary part of pro#ing any new combustion design including 42<
systems. Therefore, ma%or engine manufacturers worldwide need to plan for and execute these tests in order to de#elop the materials
needed for successful 42< applications.
!"
4.+. 6i;uid Dro0out
Curing exhaust gas recirculation, the gasses must be cooled with an external cooler before being reintroduced into the cool inlet
manifold of an engine. The cooling process for the 42< may result in li8uids being formed in the return lines, depending on
temperatures and local humidity, much as li8uids are formed in the tailpipe of an automobile at certain conditions. This li8uid dropout
could be a continuous stream that needs to be carefully understood and managed with the needs of the local en#ironment in mind. .hile
there may be ways to reintroduce this li8uid into the combustion process, doing so may create further problems with combustion and
lead to other emissions complications and instability. As such, managing li8uid dropout needs careful study and de#elopment in an
integrated de#elopment program.
5. CONC6U'ION
Thus, as seen that using 4xhaust 2as <ecirculation Techni8ue in engines, the emissions are #ary much controlled due to lesser
amounts of $(x entering the atmosphere. Thus the emission le#els to be maintained are attained by the engines. As seen, 4xhaust 2as
<ecirculation is a #ery simple method. &t has pro#en to be #ery useful and it is being modified further to attain better standards. This
method is #ery reliable in terms of fuel consumption and highly reliable. Thus 42< is the most effecti#e method for reducing the nitrous
oxide emissions from the engine exhaust. 7any of the four wheeler manufacturers used this techni8ue like 1ord ;ompany, 'en: 7otors
etc to impro#e the engine performance and reduce the amount of pollutants in the exhaust of the engine.
!"
!"