Accepted Manuscript

Analysis of low-pressure exhaust gases recirculation transport and control in

transient operation of automotive diesel engines

José Manuel Luján, Héctor Climent, Francisco José Arnau, Julián Miguel-
García

PII: S1359-4311(17)37025-4
DOI: https://doi.org/10.1016/j.applthermaleng.2018.03.085
Reference: ATE 11973

To appear in: Applied Thermal Engineering

Received Date: 3 November 2017

Revised Date: 20 March 2018
Accepted Date: 25 March 2018

Please cite this article as: J. Manuel Luján, H. Climent, F. José Arnau, J. Miguel-García, Analysis of low-pressure
exhaust gases recirculation transport and control in transient operation of automotive diesel engines, Applied
Thermal Engineering (2018), doi: https://doi.org/10.1016/j.applthermaleng.2018.03.085

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 Analysis of low-pressure exhaust gases recirculation transport and
control in transient operation of automotive diesel engines
José Manuel Luján, Héctor Climent*, Francisco José Arnau, Julián Miguel-García

CMT Motores Térmicos, Universitat Politècnica de València, Spain

*Corresponding author: hcliment@mot.upv.es. Telephone: (+34) 96 387 76 50. Postal address:

CMT Motores Térmicos. Universitat Politècnica de València. Camino de Vera s/n. 46022.
Valencia. Spain.

Abstract

The objective of the study is to determine the behavior of the low pressure exhaust gas
recirculation (LP EGR) transport phenomena in the intake manifold during engine
transient operation. The investigation also analyzes the influence of the propagation of
the pressure waves in the intake manifold on the engine performance. In this sense,
there is a clear trade-off: long intake lines improve the engine volumetric efficiency at
low engine speeds but delay the EGR transport in the system.

The experiments were performed on a test bench with a 1.6 liter Euro-5 specification
diesel engine. A CO2 fast tracking measurement device was setup and placed in two
locations in the intake line in order to track the EGR transport in transient operation.
The CO2 concentration is acquired with crank-angle resolution. Three different engine
transients at constant engine speed were studied. They are extreme and worst-case
scenarios in driving situations: (i) from low load to full load, (ii) from full load to low load,
and (iii) from low load to medium load. In this way, it is possible to observe the behavior
of the engine when: (i) leaving the EGR zone, (ii) entering into the EGR zone, and (iii)
changing operating point without leaving the EGR zone.

A consistent methodology that combines experimental results and a 1D model capable

to predict the behavior of the engine was developed. The results obtained in this
investigation show a relevant phenomenon: depending on the synchronization of the
EGR and Exhaust Throttle (ET) valves, an overshoot occurs when the engine enters
into EGR zone. In this study, the results show the importance of the synchronization of
the valves that control the EGR strategy. Comparisons between measured and
modeled CO2 concentrations lead to conclude that the EGR transport during engine
transient operation is correctly predicted within a 1D engine code.

1. Introduction

The development of current diesel engines is focused on lowering fuel consumption

and pollutant exhaust emissions. Since pollutant emission regulations have become
more restrictive, new technologies are being developed. Exhaust gas recirculation
(EGR) strategy is widely used in diesel engines due to the benefit in NOx emissions [1,
2]. As the limit of NOx emissions become more stringent, the EGR rates increase and
strategies become more complex. The introduction of exhaust gases in the combustion

1
chamber inhibits the creation of NOx emissions by reducing the peak combustion
temperature and decreasing the oxygen concentration.

The disadvantages of the EGR have been studied since decades, mainly the soot
increase [3]. It has been studied the effects of the EGR temperature on diesel engines
combustion and emissions [4]. The need to reduce NOx emissions has forced to
increase the EGR rates, develop more complex strategies and create after-treatment
systems.

Nowadays new after-treatment systems such as Selective Catalytic Reduction (SCR)

based in Urea Water Solutions (UWS) have been developed. The UWS is injected in
the exhaust to react with NOx and reduces signally the NOx. Nonetheless this system
presents some disadvantages as the deposits of urea and its byproducts during cold
weather conditions and low exhaust temperatures [5], and the economic spending to
load the tank with UWS. Moreover after-treatment systems entails a penalty of fuel
consumption because of the increase of the back-pressure.

In spite of the new after-treatment systems, the development of new and more complex
EGR strategies are necessary. The simultaneous use of HP and LP EGR is a recent
strategy to reduce NOx emissions and fuel consumption at low and medium speed and
load conditions. However, LP EGR is especially useful at high loads and in transient
operations [6]. EGR strategies in gasoline engines have been studied to reduce fuel
consumption and NOx emissions. In addition, EGR in gasoline engines can replace
fuel enrichment and avoid the knock [7]. Moreover the comparison between HP and LP
cooled EGR in turbocharged gasoline engines have been studied. HP or LP EGR must
be applied depending on the operation point of the engine [8]. Including the application
of EGR in alternative (liquid and gaseous) fuels like raw oils, processed oils, hydrogen
or natural gas have been studied too [9].

This study is focused in the LP EGR configuration. In one hand, LP EGR presents
some disadvantages versus HP EGR. LP EGR transport takes more time to arrive to
the cylinders due to the length of the intake line, produces higher hydrocarbon (HC)
emissions and, at cold conditions, presents lower efficiency than HP EGR since it
increases the intake temperature [10]. In addition, LP configuration needs more
exhaust energy because the compressor operates under higher amount of gas. In the
other hand, LP EGR shows some advantages compared to HP EGR [11]. HP EGR
reduces NOx emissions but penalizing the fuel consumption and the dispersion of the
EGR among cylinders, which can have consequences for NOx and PM emissions [12].
As to LP EGR loop systems are effective means of simultaneously reducing the NOx
emission and fuel consumption [13]. Moreover, with LP EGR loop systems, the gas
flow through the turbine is unchanged while varying the EGR rate [14]. However, if very
high EGR rates are desired, it is sometimes necessary to close a backpressure valve
placed in the exhaust line, which is usually referred to as exhaust throttle (ET) valve.

The tendency of the new homologation cycles, such as the Worldwide harmonized
Light vehicles Test Cycles (WLTC) and Real Driving Emissions (RDE) cycles, will be
more restrictive with transient operation which are more pollutant. For that reason, in

2
parallel with the new homologation cycles, it is necessary to improve the control of
emissions in transient operations [15].

Experimental and modeling tools were employed to fulfil this study. Traditionally,
experimental tests have been combined with modeling results because there are some
parameters in the engine that are impossible or complex to be measured [16]. The
instantaneous gas concentration or instantaneous mass flow in a certain location
during a given engine transient are good examples.

The paper is structured as follows: In Section 2 the experimental setup is explained.

Section 3 is devoted to the explanation of modeling tools. Section 4 contains main
results and discussion in terms of gases transport in the intake line for different
transients and intake line configurations, synchronization of the EGR and ET valves,
and the overshoot phenomenon. Finally, summary and main conclusions are presented
in Section 5.

2. Experimental setup

The experiments were performed on a test bench with a turbocharged diesel engine,
which is Euro-5 compliant. Table 1 shows the main features of the engine. The engine
includes both LP and HP EGR systems. This study is focused in the LP EGR loop of
the engine, whose schematic layout is depicted in Fig. 1. There are two intake line
configurations depending on the length of the duct between the charge air cooler and
the intake manifold. The first configuration is the original one, while the second includes
an additional 600 mm length pipe. The aim is to analyze the influence of the intake line
acoustics. The transients were performed with an ECU-controlled movement of the
EGR valve. If the desired EGR rate is not achieved with the EGR valve fully open, it is
necessary to regulate with the ET.

Table 1

Engine specifications.

Cylinder number In-line 4

Bore x stroke (mm) 80x79.5
Displacement (cm3) 1600
Compression ratio 15.4:1
Valve number 4/cylinder
Fuel delivery system Common rail. Direct injection.
EGR system HP EGR and LP cooled EGR
Intake boosting Turbocharger with VGT
Intake cooling system Air charge air cooler (ACAC)
Maximum power (kW/rpm) 96/4000
Maximum torque (Nm/rpm) 320/1750

3
Figure 1. Engine schematic layout

A CO2 fast tracking system, based on Non-Dispersive Infra-Red measuring principle,

was used to evaluate the transport of exhaust gases in the intake line [17, 18]. The
CO2 tracking system is able to detect changes in gas concentration with a T90 of 8 ms,
which is a higher sampling frequency than other conventional gas analysis systems,
like the HORIBA MEXA 7170DEGR, often used in engine testing and widely used in
steady operation. Moreover, the probe of the device can be placed in locations where
quick changes in the gas pressure may occur, such as in the intake manifold during
transient operation. The proposed experimental tests have been performed with high
pressure gradients, i.e. from 1 to 2.6 bar (abs) during the load transient.

The fast tracking system is able to measure with two probes at the same time. One of
them was installed upstream the compressor (downstream the LP EGR mixer with the
intake air). Tests were carried out with the other probe placed in different locations: at
charge air cooler outlet and in the intake manifold.

Three type of engine transient tests were assessed. The first one corresponds to the
classic engine load response at constant engine speed. The engine transient starts at
low load, in an operating condition where the LP EGR strategy is enabled. Suddenly,
the pedal is pushed to its maximum position (full load), where the engine does not work
inside the EGR zone.

The second transient is similar to the first one, but the engine load is increased without
leaving the EGR zone. The third transient is just the opposite of the first one: the
engine starts at full load (outside the EGR zone) and, suddenly, the pedal is released
and let the engine run in low load conditions performing LP EGR.

All these engine transients were run at constant engine speed due to its remarkable
impact on the volumetric efficiency, which is affected by the pressure waves
propagation phenomenon inside the intake manifold. Therefore, in order to avoid
interactions between the intake acoustics and the EGR transport delay it was decided
to remove the engine speed effect by keeping it constant during the transient tests. In
addition, to account for this influence on the torque evolutions, tests were performed
separately at different engine speeds (1250, 1500, 1750 and 2000 rpm) and two
different intake lines.

4
For the sake of repeatability, the instantaneous transients were performed with an ECU
automatic control of EGR and ET valves. The ET valve acts as a backpressure valve. It
is placed in the exhaust line, downstream of the LP EGR inlet. The objective of this
valve is to increase the EGR rate if it is not enough with the EGR valve fully open. It
usually happens at low load and low speed engine conditions. At full load conditions or
during the transient performance, the EGR strategy is usually avoided, since it is
desired to allow the maximum air mass flow to enter into the cylinders. Therefore, the
EGR valve is fully closed and the ET valve is fully open in these conditions. In a fast
engine transient, although the final situation is inside the EGR zone, it is very likely that
the EGR strategy is switched off in the first stage of the transient and activated in the
final part. This leads to a fast operation of both EGR and ET valves in a very short
period of time. The synchronization of these valves will affect the EGR transport
phenomena from the exhaust to the intake manifold.

Several engine parameters were measured to assess the engine performance and
analyze the EGR transport in the intake manifold. The variables together with the
sensors features are presented in Table 2.

Table 2

Sensor Variable Accuracy [%] Range

Thermocouples type K Temperature 1 0 ºC – 1260 ºC

Pressure sensor Pressure 0.3 0 – 6 bar

Gravimetric fuel balance Fuel mass flow 0.2 0 – 150 kg/h

Hot wire meter Air mass flow 1 0 – 720 kg/h

Dynamometer brake Torque 0.1 0 – 480 Nm

NDIR500 CO&CO2 analyzer CO2 2 0 – 20 %

Instrumentation accuracy

The pollutant emissions were measured with specific equipment (Horiba MEXA
7170DEGR), which acquires the NOX, THC, CO, CO2, and O2 concentrations in the
location where the probe is placed. The EGR rate has been obtained experimentally
from CO2 measurement in exhaust and intake manifolds [19] using the following
expression:

(1)

3. Modeling tools

5
The flow behavior inside intake and exhaust systems of internal combustion engines
can be simulated with computer tools. The flow is considered essentially one-
dimensional inside the systems that conform the engine. This situation will be true only
if the length-to-diameter is high enough and the turbulent flow is totally developed. The
governing equations for one-dimensional unsteady compressible non-homentropic
flow, i.e the mass, momentum and energy conservation equations, from a hyperbolic
system of partial differential equations in the vector from of Eq. (2). The vectors are
represented in strong conservative form in Eq. (3):

(2)

Where W is the solution vector, F represents the flux vector (mass, momentum and
energy), and C1 and C2 include the source terms that take into account the effects of
heat transfer, area changes and friction. These vectors are expressed as:

(3)

The flow properties can be obtained at every node of the duct and time instant
considering different numerical methods, time marching and spatial discretization
techniques in the solution of the Eq. (2) and the state equation of the ideal gases [20-
23]. Moreover, it is likely to estimate the inclusion of the chemical species transport
equation to the governing equations system with the same level of precision of the
applied numerical methods and without changes in the solution procedure. It is
required n-1 equations of chemical species conservation in the governing equations
system, where n is the number of the chemical species to be transported, to solve the
transport of the chemical species along the 1D elements. The chemical species
conservation equation in vector form is

, (4)

where Y is a vector including the mass fraction of n−1 different chemical species. The
mass fraction of the chemical species n is given by the compatibility equation

, (5)

6
The effect of the conversion rate and the convective transport have been considered in
Eq. (4) due to chemical reactions. Because of the negligible influence of the term owing
to diffusion among the chemical species compared to the velocity transport in ducts of
internal combustion engines, it is no considered. Bearing in mind the chemical species
transport in 1D elements, the governing equations system in vector and strong
conservative form is formulated as [24]:

(6)

The adaptation of the numerical methods is needed, due to the complexity of the
equations system with chemical species. In this work it is used the two-step Lax-
Wendroff method case [25] because it offers fast and good results to this type of study
which does not analyze the internal part of the engine like combustion chamber or
elements close to the cylinders.

Additionally to 1D elements, 0D elements are employed too. 0D elements are used as

much as possible because the resolution of their equations consume less resources
and are faster to solve. The characteristic of the 0D elements is that they can
accumulate mass so that the flow conditions are constant throughout its volume at
each calculation step. Such as the case of the turbine, cylinders, after-treatment
devices, etc. These kinds of elements are solved by means of a filling and emptying
model [26] that includes the mass and energy conservation equations for open systems
combined with the ideal gas state equation.

The chemical species transport across 0D elements involves the addition of n−1 mass
conservation equations to calculate the mass fraction of n − 1 chemical species,

(7)

where is the mass of the chemical species j inside the 0D element and is the
mass fraction of the chemical species j entering to or exiting from the 0D element

7
through the boundary condition i. Finally, the mass fraction of the chemical species j at
time instant will be

(8)

As in the case of the 1D elements, the mass fraction of the chemical species n is given
by the compatibility Eq. (5). The species considered in the model of the present study
are air, burned gas and fuel.

Finally, other submodels may be used to account for the EGR, VGT and ET valves
movement [27] although, in this study, the time evolution of every valve position was
considered as an input to the model. This is a suitable approach since one of the
objectives of the study is to assess the 1D model capability to reproduce the EGR
transport in the intake line and not to develop a complex model of the valves actuation
system.

4. Results and discussion

Once experimental and modeling tools have been explained, it is possible to analyze
and discuss the results. It is important to differentiate three cases:

Influence of the intake line length

Fig. 2 presents the EGR valve position on the left, the ET valve position on the middle
and the CO2 concentration measured in the intake manifold on the right. Fig. 2 shows a
comparison during the tip-in engine operation at 1250 rpm between the original
configuration and the additional 600 mm length pipe. Since the transient starts in an
engine running condition where the EGR strategy is active, the EGR valve is
completely open to achieve the desired EGR rate. Once the engine is running steadily
in low load situation the pedal is pushed at 0.5 seconds and a quick engine load
transient is requested. Hence, the EGR valve closes and the ET valve fully opens
because the ECU detects the transient situation. It is observed the immediate effect in
CO2 levels. That response verifies the capacity of the fast tracking system at detecting
rapid changes in gas concentration. Later, that response will be compared with model
results.

From the EGR emptying point of view there is not much difference between the original
duct and the configuration with an additional 600 mm pipe. A delay of nearly 20 ms is
detected when the additional duct is used. The contribution of the additional duct to the
overall intake line volume leads to an increase of 10% which explains the limited
influence in the EGR emptying process.

8
Figure 2. EGR (left) and ET (middle) valves movement and CO2 evolution (right) in a
tip-in transient operation at 1250 rpm with original intake line configuration and with the
additional 600 mm duct.

Nevertheless, the intake manifold pressure shows differences during transient

operation depending on the intake line length. Fig. 3 presents, on the left plot, higher
intake pressure values when the additional 600 mm length duct is used (dashed line).
Wave propagation phenomena is better tuned at 1250 rpm with the longest line. Engine
tests without EGR have been performed to isolate both phenomena: EGR transport
delay and pressure pulses propagation. It is observed in purple lines (without EGR)
that the wave propagation presents a better transient response with the longest intake
line (purple dashed line). Similar results are depicted in red (with EGR).

The smoke limiter strategy remained invariable during the testing campaign. However,
since the objective of the study is to assess the influence of pressure wave propagation
and EGR transport in the engine performance, the injected fuel gets modified if the air
mass flow evolution changes from one configuration to another, and so the engine
torque. Fig. 3 also shows the engine torque evolution on the right. The initial peak in
the torque evolution does not have a relation with the engine behavior but with the
control of the brake. Since a sudden change in the pedal position is requested and the
transients have to be performed at constant speed, the system that controls the brake
overreacts leading to a peak in the brake torque, which is released immediately after.

The analysis is focused in the torque evolution after 1 s. At first, it is possible to detect
in red lines (with EGR) that the torque increases faster in case of the shortest intake
line (solid line) than in case of longest intake line (dashed line). However, once the
emptying of the EGR effect has finished, the longest intake line surpasses the shortest
one due to the wave propagation phenomenon. In purple lines (without EGR) the effect
in the early stage of the transient is similar with both intake line lengths. It happens
because the EGR is not activated. However, it is observed how the wave propagation
phenomenon appears as the transient evolves, promoting a better transient with the
longest intake line.

9
Figure 3. Intake pressure (left) and torque evolution (right) during a tip-in transient
operation at 1250 rpm with original intake line configuration and with the additional 600
mm duct, with EGR strategy and without EGR strategy.

At a higher engine speed, 1750 rpm, Fig. 4 shows the intake manifold pressure during
the engine load transient on the left. Concerning the tests without EGR (in purple), the
original duct provides a faster response. The original line is better tuned in terms of
pressure waves propagation at this engine speed than the longest line. Similar
comments can be stated for the transients with EGR (in red), where a faster pressure
rise is achieved with the shortest line. In the right plot of Fig. 4 the torque evolution is
presented too. In this case, it is still observed that the beneficial effect of the wave
propagation becomes relevant with the shortest line, as it has been demonstrated in
the tests without EGR (purple lines). Engine torque evolution in tests with EGR is
fastest with the shortest line too. In addition to the effect of the wave acoustics, there is
also the quickest EGR emptying with the shortest intake line, which leads to a more
rapid torque recovery in the early stages of the transient.

Figure 4. Intake pressure (left) and torque evolution (right) during a tip-in transient
operation at 1750 rpm with original intake line configuration and with the additional 600
mm duct, with EGR strategy and without EGR strategy.

Fig. 5 presents a comparison between measurements and predicted results by the 1D

engine model at 2000 rpm during the tip-in load transient. The plots correspond to the
EGR valve position on the left, the CO2 concentration at the compressor inlet in the
middle and the CO2 concentration in the intake manifold on the right. As commented
previously, the valve position for the 1D model is directly imposed from the

10
experimental data. The highly-scattered signal in the compressor inlet is due to the
mixing of the exhaust gas coming from the exhaust line with the fresh intake air coming
from the air filter. The mixture is far from being homogeneous despite the presence of
an EGR mixing device and this explains its high variability captured by the gas
analyzer. Results show a delay in the CO2 concentration measurements between
compressor inlet and intake manifold because of the length of the intake line. In this
transient maneuver, the CO2 takes about 0.4 seconds to disappear completely from the
intake line since the EGR valve closes.

Secondly, it is observed that the calculated results are very close to the experimental
results. These results show that the model can predict the reality and it is concluded
that, for this type of EGR transport, the model is valid. The initial increase in the CO2
concentration measured in the intake line is not realistic so an issue related to the
device behavior due to the large intake pressure variations might be happening.
Anyway, a good model response to the EGR transport phenomenon is found when
leaving the EGR zone.

Figure 5. EGR valve movement (left) and comparison between measurement and
predicted results by 1D model of the CO 2 evolution at the compressor inlet (middle) and
in the intake manifold (right) in transient operation from 2 bar BMEP to full load at 2000
rpm.

Influence of the EGR and ET control

Fig. 6 presents, on the left, the valves position (the EGR on top and the ET at the
bottom) and, on the right, the CO2 concentration at the compressor inlet (top) and at
the intake manifold (bottom). In this case it is observed the effect of entering in the
EGR zone, from full load to 2 bar BMEP, at 2000 rpm. Regarding the measurement of
the movement of the valves, in this tip-out operation the control strategy makes the
EGR to open completely and enables the EGR strategy control to the ET valve. This
valve, in a first phase, starts to close but an immediate opening peak is observed,
followed again by a closing evolution up to the final position. If the ET valve would have
moved directly to the final position, a remarkable CO2 concentration overshoot would
have occurred, as described in the following paragraphs. It is the opening movement of
the ET valve in the middle of the transient, the responsible of reducing the overshoot
effect shown at the beginning of the transient and depicted in the graphs on the right.

11
Figure 6. EGR (top left) and ET (bottom left) valves movement and comparison
between measurement and predicted results by 1D model of the CO 2 evolution at the
compressor inlet (top right) and in the intake manifold (bottom right) in transient
operation from full load to 2 bar BMEP at 2000 rpm.

Fig. 7 shows, as previously, the valves position on the left, EGR on top and ET at the
bottom. While on the right, it presents the CO2 concentration at the compressor inlet on
top and the CO2 concentration in the intake manifold at the bottom. In this case the
transient occurs inside the EGR zone, from 2 bar to 11 bar BMEP at 2000 rpm. As in
the transient to full load, it is also observed that the ECU commands a closing of the
EGR valve at the initial part of the transient in the top left plot and an opening of the ET
valve at the bottom left plot. Later, since the engine remains inside the EGR zone, the
EGR valve is again open and the ET performs the control of the air mass flow.

12
Figure 7. EGR (top left) and ET (bottom left) valves movement and comparison
between measurement and predicted results by 1D model of the CO 2 evolution at the
compressor inlet (top right) and in the intake manifold (bottom right) in transient
operation from 2 bar to 11 bar BMEP at 2000 rpm

The calculated results by the engine model are presented in Fig. 6. As in the
experimental data, the overshoot phenomenon when entering in the EGR zone is
properly captured. On the left plots of Fig. 6 and 7, the position of the EGR and ET
valves in the model are the same as in the tests, since the movement of the valves in
the model is imposed from the experiments. The model performance when predicting
the CO2 concentration along the intake line is observed in the plots on the right of Fig.
6 and 7. Both concentrations, at the compressor inlet and in the intake manifold, are
very similar to measured data during the early stages of the transient (between the
period of 0.5 s and 2.5 s), which is the relevant phase of the present study.

The left plot in Fig. 8 shows the pressure at the inlet and outlet of the LP EGR system
in the transient from full load to 2 bar BMEP at 2000 rpm. Fig. 8 demonstrates that the
overshoot is caused because in the first stage of the transient, in full load steady
operation, the exhaust pressure is much higher than the inlet compressor pressure. In
this situation the EGR valve is completely closed. The pressure at the inlet of the LP
EGR system (exhaust line) is high due to the pressure loss in the exhaust line when
the mass flow through the engine is very high. On the contrary, the pressure at the
outlet of the LP EGR system (compressor inlet) is low due to the high pressure loss in
the air filter as a result of the high air mass flow through the engine. In the final part of
the transient, in low load conditions, both pressure traces are closely related since the
EGR valve is fully open. In fact, the pressure difference is the pressure loss in the EGR
line due to the EGR mass flow. During the transient, there is an initial phase where
both pressures approach to each other very rapidly, indicating that the EGR mass flow
has an important value.

The plot of the right shows the mass flow evolutions in three locations: at the outlet of
the air filter (green), through the compressor (purple) and through the EGR valve (red).
The large difference between exhaust and intake pressure together with the rapid
opening of the EGR valve promotes a large amount of exhaust gases through the EGR
line, which is the root cause of the EGR overshoot in the intake manifold. There is no
need to close the ET at the same time as the EGR valve opens because of the initial
EGR overshoot. In the last part of the transient, once the exhaust and intake pressures
are closer, EGR mass flow goes down and it is necessary to close the ET to recover
the EGR rate.

13
Figure 8. Pressure inlet and outlet of LP EGR system (left) and mass flow rate
evolutions (right) in the transient from full load to 2 bar BMEP at 2000 rpm.

In order to check the impact of the synchronization between the EGR and the ET
valves, a simulation, where the ET moves one second later than the opening of the
EGR valve, is performed. Simulation results are shown in Fig. 9 comparing the original
synchronization (in red) with one second delay (in purple). Fig. 9 presents on the left
the movement of the valves (EGR valve on top and different synchronization of the ET
valve at the bottom) and, on the right, the burned gas fraction (inlet of the compressor
on top and in the intake manifold at the bottom). It is observed that when the ET closes
at the same time as the EGR opens, an important EGR overshoot is created at the
compressor inlet and is transported through the intake line up to the intake manifold
and intake valves. In fact, there is no need to close the ET at the same time as the
EGR valve opens because two effects are accumulated: (a) initial EGR increase due to
the high value of exhaust-intake pressure difference and (b) the increase in the exhaust
pressure due to ET valve partial closing. Fig. 9 shows that the delay in the opening of
the ET valve reduces considerably the overshoot. The results of the combination of
these two phenomena show that the synchronization of the EGR and ET valves is
essential to reduce the overshoot phenomenon.

14
Figure 9. EGR (top left) and ET (bottom left) valves movement and comparison of the
effect of different valve synchronization results given by 1D model of burned gas
fraction at the compressor inlet (top right) and in the intake manifold (bottom right) in
transient operation from full load to 2 bar BMEP at 2000 rpm

5. Summary and Conclusions

The tradeoff between wave propagation and the emptying of the EGR in engine load
transients leaving the EGR zone was evaluated. Engine performance was assessed
with two intake lines in transient operation at different engine speeds starting with and
without EGR. Three different transients were tested at 1250, 1500, 1750 and 2000 rpm.
It has been remarkable the role of the CO2 fast tracking system, which has allowed to
carry out the study because of its fast response and capability to measure under
pressure variation conditions during transient operations.

At low engine speeds (i.e. 1250 rpm) the longest intake line was tuned and the wave
propagation phenomenon is more effective in terms of engine torque than the effect of
fast EGR emptying inside the intake line. When the engine speed increases, the
longest intake line loses benefits due to both: not being acoustically tuned and higher
volume to be emptied. Therefore, faster transients are achieved at higher engine
speeds with the short intake line. Needless to say that the results may change with the
specific application (mainly due to the dimensions and layout of the intake line), so
special care should be paid when extrapolating the results. However, it is possible to
say that, if the objective is to increase the performance of the engine in terms of torque
evolution at low engine speeds, then, it is necessary to take into account wave
phenomena probably with a longer intake line. In the other hand, if the objective is to
increase the engine performance at high speeds, then, long intake line is not needed
anymore and EGR transport is faster with a small sized intake manifold. In any case,
the methodology is consistent and can be properly used in other situations.

An interesting result of this study has been able to quantify the delay of the exhaust
gases transport in the intake manifold during the transient where the engine leaves the
EGR zone, from 2 bar BMEP operation point to full load. The delay was quantified in
0.4 seconds at 2000 rpm since the EGR valve closing.

In case of entering into the EGR zone, it has been shown that the synchronization of
the EGR and ET valves is very important to avoid or reduce the overshoot effect. It was

15
demonstrated that a slower opening of the EGR valve or a delay in the closing of the
ET valve helps to reduce significantly the overshoot.

A 1D model approach is valid for capturing the transport phenomena inside the intake
line during transient operation in all performed cases because the evolution of the
predicted CO2 concentration is very similar to measured data. Moreover, it is very clear
that the model is valid for any case as long as the position of the valves are correctly
defined from the engine tests.

References

1. Ming Zheng, Graham T. Reader, J. Gary Hawley. “Diesel engine exhaust gas
recirculation – a review on advanced and novel concepts.” Energy Conversion and
Management 45 (2004) 883-900.

2. Deepak Agarwal, Shrawan Kumar Singh, Avinash Kumar Agarwal. “Effect of

Exhaust Gas Recirculation (EGR) on performance, emissions, deposits, and
durability of a constant speed compression ignition engine.” Applied Energy 88
(2011) 2900-2907.

3. Tokura, N., Terasaka, K., and Yasuhara, S., "Process through which Soot
Intermixes into Lubricating Oil of a Diesel Engine with Exhaust Gas Recirculation,"
SAE Technical Paper 820082, 1982

4. Ladommatos, N., Abdelhalim, S.M., Zhao, H., “Effects of exhaust gas recirculation
temperature on diesel engine combustion and emissions” Automobile Engineering
212 (1998) 479-500.

5. Sadashiva Prabhu S, Nagaraj S Nayak, N Kapilan, Vijaykumar Hindasageri. “An

experimental and numerical study on effects of exhaust gas temperature and flow
rate on deposit formation in Urea-Selective Catalytic Reduction (SCR) system of
modern automobiles.” Applied Thermal Engineering 111 (2017) 1211-1231

6. Zamboni, G., Simone, M., Capobianco, M., “Hybrid EGR and turbocharging
systems control of low NOx and fuel consumption in an automotive diesel engine.”
Applied Energy 165 (2016) 839-848.

7. Wei, H., Zhu, T., Shu, G., Tan, L., Wang, Y., “Gasoline engine exhaust gas
recirculation – A review.” Applied Energy 99 (2012) 534-544.

8. Shen, K., Li, F., Zhang, Z.,Sun, Y.,Yin, C., “Effects of LP and HP cooled EGR on
performance and emissions in turbocharged GDI engine.” Applied Thermal
Engineering 125 (2017) 746-755.

9. Thangaraja, J., Kannan, C., “Effect of exhaust gas recirculation on advanced

diesel combustion and alternate fuels - A review.” Applied Energy 180 (2016) 169-
184.

10. José Manuel Luján, Carlos Guardiola, Benjamín Pla, Alberto Reig. “Switching
strategy between HP (high pressure) – and LPEGR (low pressure exhaust gas

16
 recirculation) systems for reduced fuel consumption and emissions.” Energy 90
(2015) 1790-1798.

11. Giorgio Zamboni, Massimo Capobianco. “Experimental Study of the effects of HP

and LP EGR in an automotive turbocharged diesel engine.” Applied Energy 94
(2012) 117-128.

12. Alain Maiboom, Xavier Tauzia, Jean-François Hétet. “Influence of EGR unequal
distribution from cylinder to cylinder on NOx-PM trade-off of a HSDI automotive
Diesel engine.” Applied Thermal Engineering 29 (2009) 2043-2050.

13. José Manuel Luján, Héctor Climent, Benjamín Pla, Manuel Eduardo Rivas-Perea,
Nicolás-Yoan François, José Borges-Alejo, Zoulikha Soukeur. “Exhaust gas
recirculation dispersion analysis using in-cylinder pressure measurements in
automotive diesel engines.” Applied Thermal Engineering 89 (2015) 459-468.

14. Alain Maiboom, Xavier Tauzia, Samiur Rahman Shah, Jean-François Hétet.
“Experimental Study of an LP EGR System on an Automotive Diesel Engine,
compared to HP EGR with respect to PM and NOx Emissions and Specific Fuel
Consumption” SAE Int. J. Engines 2(2):597-610, 2010, doi:10.4271/2009-24-0138.

15. David Heuwetter, William Glewen, David Foster, Roger Krieger, and Michael
Andrine. “Experimental Investigation of Transient Response and Turbocharger
Coupling for High and Low Pressure EGR Systems.” SAE Int. J. Engines7(2): 977-
985, 2014, doi:10.4271/2014-01-1367

16. J. Galindo, J. R. Serrano, F. J. Arnau, and P. Piqueras. “Description of a Semi-

Independent Time Discretization Methodology for a One-Dimensional Gas
Dynamics Model.” J. Eng. Gas Turbines Power 131(3), 034504 (Feb 10, 2009) (5
pages), doi:10.1115/1.2983015

17. Sutela, C., Collings, N., and Hands, T., “Fast Response CO2 Sensor for
Automotive Exhaust Gas Analysis,” SAE Technical Paper 1999-01-3477, 1999,
doi:10.4271/1999-01-3477.

18. Sutela, C., Collings, N., and Hands, T. “Real Time CO2 Measurement to
Determine Transient Intake Gas Composition under EGR Conditions.” SAE
Technical Paper 2000-01-2953, 2000, doi:10.4271/2000-01-2953.

19. J. M. Luján, H. Climent, L. M. García-Cuevas, A. Moratal, “Volumetric efficiency

modelling of internal combustion engines based on a novel adaptative learning
algorithm of artificial neural networks” Applied Thermal Engineering 123 (2017)
625-634.

20. J.R. Serrano, F.J. Arnau, P. Piqueras, A. Onorati, G. Montenegro. “1D gas
dynamic modelling of mass conservation in engine duct systems with thermal
contact discontinuities.” Mathematical and Computer Modelling 49 (5–6) (2009)
1078–1088.

17
21. J. Galindo, J.R. Serrano, F.J. Arnau, P. Piqueras. “High-frequency response of a
calculation methodology for gas dynamics based on Independent Time
Discretization.” Mathematical and Computer Modelling 50 (5–6) (2009) 812–822.

22. J. Galindo, J.R. Serrano, F.J. Arnau, P. Piqueras. “Description of a semi-

Independent Time Discretization methodology for a one-dimensional gas dynamics
model.” Journal of Engineering for Gas Turbines and Power 131 (2009) 034504.

23. A.J. Torregrosa, J.R. Serrano, F.J. Arnau, P. Piqueras. “A fluid dynamic model for
unsteady compressible flow in wall-flow diesel particulate filters.” Energy 36 (2011)
671–684.

24. J.R. Serrano, H.Climent, P.Piqueras, O.García-Alfonso, “Analysis of shock

capturing methods for chemical species transport in unsteady compressible flow.”
Mathematical and Computer Modelling 57 (7-8) (2013) 1751-1759.

25. P.D. Lax, B. Wendroff. “Systems of conservation laws.” Communications on Pure

and Applied Mathematics 17 (1964) 381–398.

26. J. Galindo, J.R. Serrano, F.J. Arnau, P. Piqueras. “Description and analysis of a
one-dimensional gas-dynamic model with Independent Time Discretization.” in:
Proceedings of the Spring Technical Conference of the ASME Internal Combustion
Engine Division, 2008, pp. 187–197.

27. J Galindo, H Climent, C Guardiola and J Domenech, T. Modeling the vacuum

circuit of a pneumatic valve system. Journal of Dynamic Systems, Measurement,
and Control. Vol. 131(3), 2009.

18
HIGHLIGHTS

A methodology to study the LP-EGR in diesel engine transient operations is proposed

Intake line acoustics and LP-EGR transport influence on the tip-in operation

Even with EGR, long lines may provide better load response at low engine speeds

Transient operation of the valves modifies remarkably the in-cylinder EGR evolution

A 1D model is suitable to analyze the LP-EGR transport and acoustics phenomena

19