Analysis of 2014 Formula one hybrid

powertrain
A Preliminary study with focus on its applicability to road
cars

Karthik Upendra, MSC automotive engineering.

Chalmers University of Technology
Department of Signals and Systems

Anders Grauers, Vehicle Analysis group, SHC.

Chalmers University of Technology
Department of Signals and Systems

Swedish Hybrid Vehicle Centre

June-2014
 Abstract

Understanding the 2014 formula one hybrid cars seems interesting for hybrid powertrain engineers as they are
very different from the conventional hybrid powertrain. An analysis is performed through literature review and
discussion, to understand the technicality behind it. Its pros and cons for formula one racing are presented and
then the aim of the report is narrowed down to analyze the waste heat recovery method implemented in formula
one. An attempt has been made to study its efficacy and value for money for road vehicle.

Although the energy recovery potential by using electric machine mounted on common shaft with the
turbocharger is less when compared to other thermal energy recovery methods (such as Rankine cycle), its
compactness, implicity and the ability to reduce turbolag makes it an attractive solution for formula one cars.
Addition of any device in the exhaust path increases the exhaust back pressure which in turn influences the gas
exchange process inside the cylinder. The same phenomenon is seen with recovering the energy from the exhaust
gases using MGU-H in formula one car. It is learnt that, although it deteriorates the engine performance, the
amount of energy recovered is higher and thereby improves the efficiency of the engine. It is confirmed that the
fuel economy can be improved by 5-10% in diesel engines, but its potential to reduce the fuel consumption in
gasoline engine is not clear yet and still needs to be analysed.

Keywords: 2014 formula one powertrain, Exhaust energy recovery method, MGU-H, fuel consumption
reduction, turbolag.

i
 Acknowledgements
We would like extend our gratefulness to Victor Judez for sowing the seed of curiosity about formula one
hybrids and also for the valuable discussion and materials.

ii
 Contents

Abstract i

Acknowledgements ii

Contents iii

1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Aim of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2 Power train layout and system description 2

2.1 Powertrain layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2 Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 System components and the important rule that shapes the power train . . . . . . . . . . . . . . . 3

3 MGU-K energy recovery potential 6

4 Why is it important for engine efficiency in F1 8

5 Energy losses in ICE and different exhaust gas energy recovery methods 9

6 Discussion about exhaust energy recovery via turbine with electric machine mounted on
a common shaft 11
6.1 Discussion about increase in exhaust back pressure and its effect on light duty gasoline engine . . . 12
6.2 Discussion on increase in exhaust back pressure and its effect on heavy duty diesel engine . . . . . 13
6.3 Conclusion from the discussion on increase in exhaust back pressure . . . . . . . . . . . . . . . . . 17

7 Discussion on fuel consumption reduction on standard and real driving cycle 18

8 Summary 20

References 21

iii
iv
1 Introduction
1.1 Background
Downsized turbocharged engine are becoming common in road cars to cut the emission and reduce fuel
consumption. From 2014, formula one has started to see downsized turbocharged engine with somewhat a
strong hybridisation. The regulations imposed by the FIA (Fédération Internationale de l’Automobile) on the
amount of fuel used per race forces the teams to be more energy efficient and at the same time deliver a good
performance fit for racing. These regulations ensures the money spent by the teams are contained and the
performance of the car across all teams are comparable.
The engines are downsized from 2.4 litre naturally aspirated V8 to 1.6 litre V6 single stage turbocharged
engine. Although downsizing reduces the fuel consumption, the overall performance of the car reduces which is
obviously not what we want to see in formula one racing cars. Therefore, the downsized engines are turbocharged
which helps to increase the performance. However, the turbocharged engine still faces poor performance at
low engine speed called the turbolag which makes it slightly unsuitable for racing applications. Turbolag is a
phenomenon where the amount of compressed air that can be sent to the engine is limited by the amount of
exhaust gases required to run the turbocharger during low engine speed regime. One has to be careful with
adding a turbocharger. The addition of the turbocharger on the exhaust side of the engine, ’if not calibrated
accurately’, will cause the back pressure in the exhaust to increase which will in turn result in increased fuel
consumption and also some time causes knocking in gasoline engine. If the turbocharged-engine is calibrated
appropriately then the only downside is the turbo lag. One smart way of reducing the turbolag is by using a
electric machine to speed up the turbocharger during the low engine revs, which is exactly what we see in the
power unit of formula one 2014. The same electric machine can also be used to recover some of the energy
from the exhaust gases which is dealt in detail later in this report.
The new rules also limit the amount of fuel to maximum of 100 kg per race which implicates that the energy
saving has to be accomplished by recovering some of energy dissipated during braking. Therefore, another
electric machine is added to recover the the vehicles kinetic energy during braking.
The implementation of the technology for recovery of vehicles kinetic energy during braking can be seen both
in formula one hybrid electric vehicle (HEV) and road HEV power trains (PTs). The difference in PTs however,
is in the implementation of electric machine mounted on common shaft with the turbocharger. Therefore, there
is a need to analyse the energy recovery potential of this system and list out the pros and cons of this system
and check its application for full/mild/micro hybrid, which forms the basis of this report.

1.2 Aim of the report

Understanding the Formula one hybrids can be important for hybrid power train engineers, as they are very
different than the hybrid powertrains used in conventional cars and heavy vehicles. And therefore can reveal
new interesting insights and ideas.
The aim of this report is to analyse the hybrid powertrain architecture of the formula one, introduced in
2014. Moving further, the objective of the report is narrowed down to analyse the concept of the exhaust
energy recovery potential seen in the formula one cars. This is accomplished by performing literature review
and through discussion with experts in academia.

1
2 Power train layout and system description
In this chapter the power train layout of the 2014 formula one power train is described first. Secondly, the
usage of the power train components are described with a hypothetical driving scenario. Last but not the least,
the important regulations which governs the size and design of the components is discussed.

2.1 Powertrain layout

The power unit of formula one 2014 consists of three main energy converters; two motor generators unit (MGU)
and a internal combustion engine (ICE).1 The architecture resembles the parallel type arrangement and a
typical layout of the formula one power train of 2014 is shown in figure 2.1.

Compressor Turbine

MGU-H

Cooler

Intake Exhaust

Fixed gear
ratio

Transmission Wheels

Single cylinder
MGU-K Of Engine shown
for clarity

Figure 2.1: Typical layout of formula one power train. Single cylinder engine shown for clarity. Electrical
connections are not shown.

One end of the crank shaft is connected to a MGU (MGU-K) via the fixed gear ratio and the other end is
connected to transmission and then to the wheels. A turbine is placed in the path of the exhaust gas, coming
out of the engine, which then spins the compressor. Another electric machine (MGU-H) is mounted on a
common shaft with the turbocharger. An inter cooler is placed downstream of the compressor to cool the
compressed air before passing it to the intake manifold. The ICE and MGU-K are the only energy converters
that are allowed to propel the vehicle. It should be noted here that the layout of the power unit shown in figure
2.1 is not a rule book layout and the teams are allowed to shuffle the components around. A clutch can be
used, for instance, between MGU-H and the turbine. However, there is very little room to deviate from this
arrangement considering the benefits of this arrangement and also respecting the rules.

1 Although battery is a energy converter, the terminology used in this report will be energy storage (ES) and is seldom discussed
in this report.

2
2.2 Strategy
When the car is on the track, it can be propelled by engine or MGU-K or both together. However, the use of
MGU-K is is allowed after the vehicle has for the first time reached 100 kph. When the car is on the track
and under full acceleration, the engine uses the fuel energy to propel the car. During this time, the engine
is operating at full load and the amount of exhaust gas is significant. This enormous energy content in the
exhaust gases turns the turbine to more than 50000 rpm. Consequently, the MGU-H mounted on the common
shaft with the turbine can now be operated in generator mode to recover part of the exhaust energy.
When the car approaches a corner or during braking, the kinetic energy of the vehicle is converted into
electrical energy in the MGU-K and is stored in the battery for later use. Immediately after cornering and
when the driver wants to fully accelerate, the MGU-H operates in motor mode and speeds up the turbine to
overcome the turbolag. When the turbocharger gains sufficient speed the MGU-H can operate in generator
mode to recover part of the energy from the exhaust. And therefore the two MG unit are together rightly
called as the energy recovery system (ERS).
The teams can decide to deploy the energy recovered through ERS to gain positional advantage. For
example, during overtaking, in addition to the fuel energy through the engine, an additional 120kW can be
added through MGU-K continuously for a period of 33.33 seconds (time calculated from back calculation using
the energy in ES) by using the energy from the ES. In addition to this, if the MGU-H is properly calibrated with
the turbocharged-engine, an unlimited amount of energy can be directly sent to the wheels through MGU-K,
(see figure 2.2). Therefore, it is interesting to see the energy content in the exhaust gases and how much of it
can be recovered via the MGU-H.

2.3 System components and the important rule that shapes the
power train
The regulations ensures the money spent by the team are contained and the performance of the cars across all
teams are comparable. These regulations governs the size and design of the power train components. Some of
them which are important for the discussion in this report are listed in this section.

Internal combustion engine (ICE): The primary propulsion unit is a V6, 1.6 Litre, spark ignited, direct
injection, single stage turbocharged engine. The internal combustion unit for the eleven teams competing in
2014 is supplied by Renault, Mercedes and Ferrari. It is said that from 2015 Honda will join this elite group.
The important rules governing the design of the ICE which are interesting for the discussion in this report
taken from [7] are:-
- The maximum speed of the engine is limited to 15000 rpm.
- Below 10500 rpm, the fuel mass flow rate is governed by the equation 2.1. Maximum fuel rate is 100
kg/hr within 10500-15000 rpm.

Q ≤ 0.009 ∗ N + 5.5 (2.1)

Where, Q (kg/hr) is the Fuel mass flow rate and N (RPM) is the Engine speed.This will effectively limit
the torque as a function of engine speed. Therefore, this makes it vital to maximize the engine efficiency,
which is explained in chapter 4.
- The mass of the fuel allowed per race is 100 kg.
- Only single stage turbocharger is allowed. Variable geometry turbine (VGT) or variable nozzle turbine
(VNT) or any device to adjust the gas throat section at the inlet to the turbine wheel is prohibited.
- Two inlet and two exhaust valve per cylinder is allowed.
- The idle speed control target may not exceed 4000 rpm.
- Engine exhaust systems must have only a single tailpipe exit which must be rearward facing and through
which all exhaust gases must pass.

3
 - Variable valve timing and variable valve lift profile systems are not permitted. Variable length intake
trumpets are forbidden in 2014 only.
- The pressure of the fuel supplied to the injectors may not exceed 500 bar.

Energy recovery system: The function of MGU-K is to recover part of the kinetic energy during braking and
also to provide the propulsion power whenever required but within the limits specified by the FIA (Fédération
Internationale de l’Automobile) rules and regulation for 2014 [7]. The function of MGU-H is to recover part of
the energy from the exhaust gases coming out of the engine and also to speed up the turbocharger to overcome
the turbo lag. Since the two MGU units allow for recovery of part of the waste energy, they are called as energy
recovery system (ERS).
The important rules governing the design and utilisation of the two MGU units which are interesting for
the discussion in this report taken from [7] are:-
- The Maximum power, speed and torque of the MGU-K is restricted to 120 kW, 50000 rpm and 200 Nm
respectively. The electrical DC measurements are used to verify the energy and power requirements
are being respected. A fixed efficiency correction of 0.95 is used to monitor the maximum power from
MGU-K, which is 120 kW [7].
- The MGU-H is used to recover part of the exhaust energy and there is no restriction on the size of
MGU-H. However, the maximum speed is restricted to 125000 rpm.

- The engine ancillaries should be driven by the ICE and/or MGU-K.

- The difference between maximum and minimum state of charge at any instant of time the car is on the
track should not exceed 4 MJ.
- Energy flow from MGU-K to the energy storage is restricted to 2 MJ per lap and the energy released by
the energy storage to MGU-k is restricted to 4 MJ per lap.
- Maximum of 5 kJ of energy can be stored in the control unit. This limits the size of the capacitors in the
power electronics.
- The amount of electrical energy that can be directly transferred from MGU-H (which recovers part of
exhaust energy) to MGU-K is unlimited.

- With the exception of cars starting the race from the pitlane, the MGU-K may only be used during the
race start, once the car has reached 100 kph.
The weight of the energy storage device must be within 20 - 25 kg range. Also, the car must be driven in electric
mode in the pit lane. Summary of the important rules which are interesting for the discussion in this report is
shown in figure 2.2. Although the the addition of ERS may feel like a boon for fuel economy for formula one
cars, several of the energy regulations and the technical limitations hinders the maximum utilization of the
technology. However on the other hand, it can be argued that the energy regulations help to contain the cost
to maintain a comparable performance of car across all the teams.
In the next chapter the kinetic energy available for recovery during braking is analysed and the main
discussion topic of this report is narrowed down.

4
Figure 2.2: Summary of important rules regarding the power unit [7]

5


3 MGU-K energy recovery potential

Figure 3.1 shows the velocity and acceleration profile of formula one car covering one lap of the Monte Carlo
GP taken from [4]. The Kinetic energy recovery system (KERS) is not used in this lap. The weight of the
vehicle is taken as 660 kg, air density is 1.29 kg/m3 , drag coefficient is 0.5, frontal area is 1.5 m2 . From this
the propulsive and braking power is calculated which is shown in figure 3.2. From figure 3.2, it can be seen

that a propulsion unit delivering a maximum power of approximately 550 kW was used to obtain the velocity
and acceleration profile of figure 3.1. From figure 3.2 it can be seen that, approximately 9 MJ of energy is
available for regeneration. This available energy at the wheels will undergo four stages of energy conversion in
case a electrical KERS is used. I.e two stage energy conversion each for charging and discharging. I.e, kinetic
energy at the wheels is converted to electrical energy in the MGU unit and then stored as chemical energy in
the battery during charging and the same reverse process while discharging. Assuming 90 % efficiency each for
MGU and battery, an overall regeneration efficiency of 65% can be obtained. This means, 7.29 MJ of energy
can be recovered during regeneration per lap which is more than the regeneration limit set by the FIA [7]. Note
that out of 7.2 MJ of energy recovered, only 5.8 MJ of energy reaches the wheel due to energy loses in battery
and MGU-K during discharging phase.
The braking time and the braking distance is more in case of KERS when compared to friction braking [4].
Therefore, a proper strategy has to be planned to determine when the KERS has to be deployed and when the
friction brake has to be used. Further, the weight distribution of the car, the friction grip available at the tire,
amount of braking and the integration of KERS with the friction brakes all plays an important role.



(a) Velocity profile (b) Acceleration profile

Figure 3.1: Velocity and acceleration vs distance of a F1 car covering one lap at Monte Carlo

(a) Propulsive and braking power (b) Propulsive and braking energy

Figure 3.2: Propulsive power and energy vs distance of a F1 car covering one lap at Monte Carlo [4]

6
 The Monte Carlo lap is considered the slowest track, meaning that there will be lot of frequent braking
and hence more kinetic energy for regeneration. Assuming that 50% less energy will be available on the other
tracks and assuming 65% regeneration efficiency, there is till 3.6 MJ of energy that can be recovered and out of
that only 2.9 MJ of energy reaches the wheel. Therefore, from above discussion it is clear that the target set
for regeneration of braking energy is nothing special but will require strategy planing and some research and
development involving the longitudinal dynamics of the vehicle.
The addition of the two MGU allows for the reduction of the fuel consumption and also enhances the
performance of the car. To be precise, the amount of fuel consumption reduction and the acceleration boost
that can be obtained from the MGU unit depends mainly on how much energy can be recuperated. We know
from above discussion that the maximum energy that can be recuperated by the MGU-K is limited by the rules.
The only other option for the teams to gain advantage is to use the energy path from MGU-H to MGU-K
(unlimited power flow, see figure 2.2) without using the energy storage. Also, in order to be able to use MGU-H
exhaust gas recovery technology in road cars, its value for money has to realised first. Therefore, it is interesting
for various reasons to analyse the amount of energy that can be recovered from MGU-H units.
Therefore, moving further in this report, it is decided to limit the discussion to analyse the potential of
energy recovery using the MGU-H through a literature review.

7
4 Why is it important for engine efficiency in F1
The amount of fuel that can be used per race is reduced to 100 kg which means that the task has to be
accomplished with less fuel consumption. Therefore, the efficiency of ICE also plays an important role.
The upper limit for maximum power delivered by the engine is not explicitly mentioned in the rules. However,
the constraints on the fuel mass flow rate caps the maximum power delivered by the engine, forcing the teams
to improve the efficiency of the engine to get maximum power out of the engine.

600 800
30% eff
750 35% eff
550
40% eff
700 45% eff
500

450 650

400 600

Torque (Nm)
350 550
Power (kW)

300 500

250 450

200 400

150 350

100 30% eff 300

35% eff
50 40% eff 250
45% eff
0 200
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000
Engine Speed (RPM) Engine Speed (RPM)

(a) Power estimation (b) Torque estimation

Figure 4.1: Power and torque estimation of of ICE of formula one 2014 for different constant ICE efficiency

The teams are however allowed to develop new fuel but these are again governed by the long set of rules
mentioned in the article 19 of the FIA 2014 technical regulations [7]. Hence, the difference in calorific value of
the fuel used by different teams can be assumed to be marginal. Figure 4.1 shows the target power and torque
curve for different constant efficiency of the engine using equation 2.1. Calorific value of the fuel assumed
here is 44.5 M J/kg. Variable geometry turbine is not allowed and only a single stage turbocharger is allowed.
Varying the power level at different engine speed by using variable valve phasing and lift are not allowed as
mentioned in [7].
Figure 4.2, taken from [4], compares the power and torque curve ’estimated/targeted’ for formula one 2014
turbocharged ICE to that of power and torque curves of the turbocharged racing engines of the mid eighties.
The orange and purple curve represents the 1.5 litre turbocharged racing engine of the mid eighties where the
maximum pressure ratio of the turbocharger were limited to 4 and 2 bar respectively. The black curve is an
’estimate’ made from the maximum BMEP of the naturally aspirated racing engine of mid nineties. The other
two curves represents the ’estimate’ made using the fuel flow limit of 2014 F1 rules. Note, the word estimate
is used to emphasise the fact that the derived curve are not the curve currently used in the formula one but
rather an approximation made by the author in [4]. Further, he says that the technologies that develop much
higher torque are already available and therefore the target set by the FIA is nothing special but requires
particular research and development.
From the above discussion it is clear that there is a need to improve the fuel efficiency of the engine. The
maximum efficiency of nowadays GDI engine is at hitting a saturation limit at 38% mark and to get a major
improvement in fuel economy other measures such as recovering the waste energy from exhaust gases has to be
employed. In the next section the amount of useful power and the losses in the engine is explained and the
potential of exhaust energy recovery is explained.

8
 

(a) Power curve (b) Torque curve.

Figure 4.2: Power and torque of some turbocharged racing engines of mid eighties and today’s F1 engine.
The picture also shows the power and torque profile targeted in 1.6 litre F1 engines revving 15000 rpm with
maximum power output of 475 and 525 kW. Taken from [4]

5 Energy losses in ICE and different exhaust gas energy

recovery methods
The engine designs relating to improving the thermodynamic efficiency of the engine is reaching a saturation
limit and other methods needs to be considered to meet the future emission regulations. One such area where
by theto Environmental
there is huge potential Protection
improve the fuel economy Agency
of the vehicle of USA[6].
is by recycling The
the waste energy dissipated 60
by the engine. results showed that 33% of the energy from fuel combustion

Brake Thermal Efficiency/%

was flow
carried away inwith exhaust gas and 29% 55
A typical energy distribution the engine is shown in figure 5.1 was
takencarried
from [13].away
It is seen that, more
than 30% of the with
lossescooling water
are carried awayasby well as heat
the exhaust radiation,
gases and 60%
and more than the isenergy
carried away by both
50
exhaust gases and
leftcooling from the engine.
was converted intoTheses areas provides
indicated work. room
Fig. for
2 isfurther improving the fuel economy
the schematic
of the vehicle.
diagram of test results. 45

40

35
198

Figure 5. The trend

Through ana
referencing the la
that waste heat e
application prosp
Figure 2. Figure
The schematic diagram of flow
engine energy flow distribution saving and emissi
5.1: Engine Energy distribution. [13]
the exhaust gas
Fig. 3 shows the energy flow distribution of a typical exhaust gas energ
gasoline engine under full-load operating conditions. As potential. For this
shown in the figure, indicated9 work, heat loss and exhaust mainly discussed
energy are approximately equal. The waste heat energy of
engine, including heat loss and exhaust gas energy, are over IV. THE CH
60% of the total energy. In conclusion, it is necessary to
 Several methods considered for recycling the energy are shown in 5.2 taken from [16]. It shows the amount
of energy that is available for recycling versus its complexity for automotive applications. An excerpt from
the same reference which explains the applicability of various exhaust energy recovery method for automotive
purpose is given below.
”The differences in the thermal processes in
heat utilization are mainly caused by the ability
to use the temperature difference. Chemical reac- In additio
Gear

Energy
tions have to run within a required temperature process

Mech.
Auxillaries
range (> 400–5000 C) and can therefore only and therm
utilize heat above this temperature level. Due Driving Resistances because
to the process management, the realization of the com
Furtherm

Energy
a high heat input is much more critical for de-
turbochar
vices based on the thermo-acoustic effect or the thermal
Stirling cycle. The heat transfer by gas-to-gas efficiencie
of the Joule process and by gas-to-material of a those of
thermoelectric device makes the heat utilization developm
for a given heat exchanger surface more difficult generate
in comparison to the evaporation procedure of remaining
are well-
a Rankine cycle. Moreover, the latter approach
plants. D
offers the possibility to recover the released heat Kinetic + used for
of both the exhaust gas and the coolant system. Pressure Exhaust Coolant in the ex
In general, system complexity (e.g. size, weight, Exhaust Energy Energy Rankine
automotive application

system integration) is increasing with growing approach

heat utilization. But the complexity of some
Complexity for

Acoustic 2-Loop- POSSIB

technologies (Stirling, thermoacoustic) is dis- Rankine
Stirling RANKIN
proportionally high in comparison to the heat 1-Loop-
recovery achievable for automotive applications. Joule Rankine (B)
Catalytic 1-Loop- As shown
For this reason these systems are excluded in the Rankine (A) cycle ca
following evaluation.In addition to the heat uti- Combined
associate
lization, the efficiency of the process itself plays Turbines Seebeck
• Proc
an important role. Turbo-machines and thermo- liquid
chemical technologies are not considered, because Turbine
• Proc
their efficiency benefits are strongly linked to the Heat utilization
Turbo-machines supe
combustion engine and, hence, hard to assess. supp
Furthermore, turbo-machines, and in particu- Thermo-dynamic Thermo-chemical
• Proc
lar turbochargers, can be applied in combination Thermo-electric Thermo-acoustic
vapo
with the thermal processes. Current efficiencies gene
Figure 1 Energy utilization vs. complexity of different
of thermo-electrical devices are lower than those Figure 5.2: Energy utilization Vs complexity of different heat
heat recovery systems • Proc
of thermo-dynamic processes, but future material recovery systems. [16] med
development could offer the chance to efficiently diss
generate electrical power directly on-board. The two remaining thermo-dynamic
The differences in the processes (Joule, Rankine)
thermal processes in heatare
well-established in gas and steam cogeneration plants. utilization
Due to itsare mainly
better caused the
efficiency by steam
the ability
cycleto isuse thefor
used
temperature
lower temperature levels (300 − 9000 C) as found in the exhaust gas ofdifference. Chemical
an IC engine. reactions have
To summarize to run
the Rankine
within a required temperature range (>400–500°C) and
steam cycle can be identified as a favorable approach for the recuperation of waste heat.”
can therefore only utilize heat above this temperature
The main advantage of energy recovery from exhaust level. Due
gases to using
the process
rankinemanagement,
cycle is thatthe realization
the system of is
a high heat input is much more critical
independent of the engine cycle, i.e. this method do not greatly influence the back pressure in the exhaustfor devices based
manifold and hence doesn’t influence the gas exchangeonprocess
the thermo-acoustic effect or the Stirling cycle. The
inside the cylinder significantly. However, the
heat transfer by gas-to-gas of the Joule process and by
turbosteamer’s complexity, size, weight and cost has still
gas-to-material of a[8].
to be addressed On the otherdevice
thermoelectric hand,makes
the exhaust
the
energy recovery by using a turbine with the electric heat
machine mounted
utilization for a on
givena common shaft issurface
heat exchanger less complex
more
and occupies less space. Although, it has some disadvantages such as poor
difficult in comparison efficiency
to the due procedure
evaporation its influence of aon
the gas exchange process of the cylinder as explainedRankine
above, its cycle. Moreover,
compact size the
andlatter
less approach
complexity offers
makesthe it
more attractive solution for future research. Therefore,possibility
the exhaust to energy
recover recovery
the released
by turbineheat with
of both the
a electric Wt,pump
machine mounted on a common shaft is considered hereexhaust gas study.
for further and theThis
coolant system.
method In general,
is same as the system
one seen
complexity (e.g. size, weight, system integration) is
in formula one which uses MGU-H mounted on a common shaft with the turbocharger.
increasing with growing heat utilization. But the
complexity of some technologies (Stirling, thermo-
acoustic) is disproportionally high in comparison to the
heat recovery achievable for automotive applications. For
this reason these systems are excluded in the following
10
evaluation.

Figure 2
6 Discussion about exhaust energy recovery via turbine
with electric machine mounted on a common shaft
As apparent from figure 5.2, the exhaust gases consists of energy which can be classified into pressure, kinetic
and thermal energy. This is also confirmed in [6]. Apart from these three energy form, there exists traces of
unburnt fuel which is negligible and is ignored in the future discussion. The energy recovery through MGU-H
is mainly driven by the pressure gradients and/or kinetic energy, whereas, the energy recovery through rankine
cycle is determined mainly by thermal energy [5].
The authors in [6] performs a study on a 75 kW two stroke engine and concludes that the exhaust energy
increases with raising speed and load. A maximum of 92 kW is present at the highest engine speed-load
operating point from a 75 kW two stroke engine. Further they mention that the thermal energy is the major
form and occupies more than 90% at full speed range, more than 95% at high speed operating point. The
pressure energy takes the second place followed by the kinetic energy. It illustrates that the amount of energy
available for recovery through MGU-H (which is the pressure and/or kinetic energy) is comparatively lesser
than the amount of energy available for recovery through thermal method such as rankine cycle. On the
positive side, the exhaust energy recovery using MGU-H is less complex and it doesn’t rule out the use of
thermal energy recovery methods.
Recovering part of the exhaust energy by using an electric machine mounted on common shaft with the
turbocharger is refereed by different names in the literature. US based corporation, caterpillar refers to it as
’Electrical turbocompound (ETC)’in [10] [1]. They predict a fuel consumption improvement of 5-10 % on a
heavy duty 14.6 litre diesel engine. Mitsubushi refers to it as hybrid turbo in [12]. Garrett (taken over by
Honeywell) refers to it as e-turbo and also some times electrically assisted turbocharger (EAT) in [17] [3]. UK
based, micro-hybrid specialist, controlled power technologies (CPT) refers toDownloaded from SAE International
it as ’Turbine-generator by Chalmers University of T
integrated
Gas energy recovery system (TIGERS)’ in [9].
The amount of power available at the turbine
depends on the exhaust mass flow rate, tempera-
ELECTR
ture of the exhaust gases and the pressure of the
exhaust gases. Theses three parameters vary de- The sel
pending on the type of engine (two/four stroke driven b
engine or petrol/diesel engine), size of engine With the
(heavy/light duty engine) and last but not the motor/ge
least the combustion environment and gas ex- synchron
change process inside the cylinder. In [10], the magnet
energy available for recovery on a heavy duty point, a
14.6 litre diesel engine is estimated and the same centrifug
is shown in figure 6.1. A maximum of approx- inertia h
imately 40 kW is available for recovery at rated High ma
engine power of 354 kW. Figure 6.1 is very useful Figure 6.1: Compressor and turbine power Vs engine power. all turbo
and an estimate of energy available for recovery Taken from [10] results o
can be made for light duty engines and formula Fig. 2 Compressor and turbine power in turbo machine
one car engines which is of the order of 5-10 kW compound engine machine
and 50-60 kW respectively. But, it must be noted that, recovering all of this energy has a direct influence on advantag
In the case that the power requirement of the
the gas exchange process of the cylinder which in turn affects the performance of the engine.
compressor cannot be met, the electrical machine can
Recovering energy with the electric machine mountedbe usedon as a motor
common to accelerate
shaft the turbo shaft.
with the turbocharger hasThen,
direct Stator di
influence on the back pressure in the exhaust manifold.the electricity neededcauses
This sometimes to runtheasexhaust
a motor would
gases be
(residual Stack len
generated
gases) to flow back into the cylinder during the last part by the electric
of the exhaust stroke.machine mounted onis the
This phenomenon crankto
refereed Length o
shaft, to
as internal EGR. A small amount of internal EGR is found or be
it advantageous
could be supplied by an the
as it reduces on-board
pumpingenergy
losses. Rotor ine
However, careful consideration must be taken as thestorage device.
hot residual gases may create a condition favourable for Air gap l
knocking in gasoline engine. Also, high amount of residual gasses deteriorate the performance of the engine at Rotor sti
The primary
high loads. A brief discussion on the effect of increasing objective
the exhaust of pressure
back the ETC insystem,
a lightasduty
presented
gasoline
in this paper, is to improve fuel economy. Thus, the Machine
engine and heavy duty diesel engine is presented below in section 6.1 and 6.2 respectively. System e
principal mode of operation is when the electric machine
on the turbocharger shaft acts as a generator, and the
electric machine on the engine crankshaft works as a
motor. The power electronics shown in Figure 1 regulate Table 1
the
11 operation of the generator, and it maintains the
electrical bus at the desired voltage of 340 V. A second
electric machine, acting as a motor, draws power from Further
the electrical bus and assists the engine by injecting phase 6
mechanical power into the crankshaft. In addition to the rotor lam
crankshaft-mounted motor, there could be other vehicle power r
 

1734

6.1 Discussion about increase in exhaust back pressure and its ef-
fect on light duty gasoline engine
The authors in [5] performs a study on naturally aspirated 2 litre four stroke gasoline engine to investigate
the potential of exhaust waste energy recovery using power turbine technology for light duty application.
Restrictions with decreasing diameter were mounted in the exhaust to simulate different vane positions of a
VGT (Variable Geometry Turbine) and in-cylinder pressure measurements were performed to evaluate the effect
of increased exhaust back pressure on intake- and exhaust pumping losses and on engine performance. Four
engine operating points around 2300-2750 rpm and torque ranging from 20-75 Nm were selected based on the
high residence time during NEDC cycle.



(a) Individual pumping power loss contribution at low (b) Individual pumping power loss contribution at high
load load

(c) Additional available power verses additional pump- (d) Additional available power verses additional pump-
ing power at low load ing power at high load

Figure 6.2: Effect of increase in back pressure on pumping losses taken from [5]

Figure 6.2 taken from [5] shows the effect of increasing exhaust back pressure on the pumping losses of the
engine. In figure 6.2a and 6.2b, the individual pumping power loss contributions for intake (intake valve plus
throttle valve), exhaust (exhaust valve plus restriction), throttle valve and exhaust restriction are depicted for
low (20 Nm torque) and high load (75 Nm torque) respectively. The pumping power loss contribution for the
restriction can also be interpreted as the potentially recoverable power. At low load, as the exhaust back pressure
is increased, the throttle is opened more to maintain the same IMEP and hence the intake pumping losses
decreases slightly. The increase in exhaust pumping losses due to a restriction is only partially compensated by
the decrease in intake pumping losses and therefore inserting a restriction in the exhaust is not PMEP neutral.
And also the recoverable power is also very low. At high load, the intake pumping losses are low. In this case

12
the decrease in intake pumping losses does not have much influence anymore while exhaust pumping losses are
at least one order of magnitude higher.
In figure 6.2c and 6.2d, the additional power available together with the additional pumping power due to
increasing exhaust back pressure is plotted. It is obvious that the additional pumping power increases faster
both for low and high load. However, for high load the two lines diverge much slower and for low exhaust back
pressure they keep pace up to about 1.5 bar.
In figure 6.3, the combined ISFC1 (Indicated
specific fuel consumption) is plotted. The term
combined means that turbine shaft power is re-
covered by an electric generator and used to drive
auxiliary electrics. For slightly higher exhaust
back pressure, fuel savings are observed for an
engine net IMEP of 1.9 and 2.4 bar due to the
thermal efficiency increase of the engine. The
thermal efficiency is increased due to the hot
residual gases in the cylinder which decreases the
need for throttling. However, for a net IMEP of
4.7 bar fuel savings up to 2.7 % are observed in
which the exhaust back pressure is 1.5 bar. The
recovered power at this point is nearly 1800 W,
which suggests that in this case the fuel savings
can be ascribed to the recovered power by the
turbine. From figure 6.3, it is clear that there is Figure 6.3: Combined ISFC
an optimum exhaust back pressure which gives
the maximum reduction in fuel consumption.
The author in [5], goes on to say that, there is no significant fuel savings obtained with the exhaust waste
heat recovery using power turbine technology on NA aspirated engine. He also mentions that to prevent
large negative pumping losses, the intake pressure should ideally be maintained higher than exhaust pressure.
However, the above paper doesn’t take into account the influence of the compressed air from the turbocharger
on the scavenging/gas exchange process of the cylinder. The influence of volumetric efficiency due to turbo
charging is neglected which is known to have positive influence on the fuel economy. In order to obtain more
insight, a similar analysis preformed in [11] on a downsized version of six cylinder heavy duty engine is studied.
and is explained below.

6.2 Discussion on increase in exhaust back pressure and its effect

on heavy duty diesel engine
The author in [11] performs a study on a downsized version of a six cylinder heavy duty turbocharged truck
engine. The engine has a rated power of 360 kW at 1700 rpm and the maximum BMEP is 33 bar. The engine
is already equipped with turbocharger. The fuelling rate, boost air pressure and temperature are maintained
the same as for non turbo-compound case. Here, the author performs a study on mechanical and electrical
turbo-compounding. In this report, the results pertaining to electrical turbo-compounding (which is same as
the MGU-H technology) is discussed. The engine simulation code here is based on a three dimensional multi
zone combustion model. For more information about the simulation model, the reader is referred to the original
paper [11].
The investigation is conducted for increase in exhaust back pressure and three different turbocharger
efficiency (49 %, 54 % and 59 %). The 59 % efficiency of the turbocharger amounts to approximately 80 %
efficiency each for both turbine and compressor. This is close to what is achieved by caterpillar (approximately
1 ISFC (Indicated specific fuel consumption) is a measure of fuel consumed with respect to the power delivered by expansion of

the gasses inside the cylinder. The power delivered is measured by the P-V (indicator) diagram. Whereas, BSFC (brake specific
fuel consumption) is the measure of fuel consumed with respect to the power delivered to the crankshaft. The power at the
crankshaft is measured by a dynamo and hence includes friction losses.

13
85 % each for both turbine and compressor) in their ETC [10]. The simulation results for engine speed of 1700
Downloaded from SAE International by Chalmers University of Technology, Thursday, June 05, 2014 06:22:08 PM
rpm is discussed.
The exhaust pressure is increased 1 bar above
the value corresponding to the non turbocom-
4.0
pound case as shown 0in figure 6.4. It is inter- 1700 rpm Engine Speed

Exhaust Pressure (bar)

esting to note that the -5 maximum exhaust back
1700 rpm, Kis=80% 3.5
NOx Variation (%)

pressure for the non-turbocompound

Downloaded from SAE International by case (i.e.University of Technology, Thursday, June 05, 2014 06:22:08 PM
Chalmers
-10 3.0
the standard turbocharged downsized diesel en-
gine) was approximately -15 2.8 bar and 1.7 bar for 2.5

100 % and 25 % load -20 respectively. Load Also caterpil-

2.0
100%
lar’scurve
heavy duty
being dieselas
reduced engine
exhaust with the
75% electrical
pressure increases. 2 Load
-25 1.5
turbocompund
The maximumoperates with exhaust
bsfc improvement 50% back
attributed pres-
to electric 25 %
25% 3.8 % for the 100 %
sureturbocompounding
of 3.5 bar [10]. alone
The is approximately
-30 reason for maintaining

bsfc Variation (%)

0 1.0
highly efficient T/C. On the other hand the total overall
1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 0 0.2 0.4 0.6 0.8 1 1.2
suchbsfchigh improvement
back pressure could
due
Power tobe because
Turbine both the
Pressureelectric
Ratio (-) Exhaust Pressure Increase(bar)
exhaust gases in theand
turbocompounding diesel
increasedengine T/C are at a islower
efficiency 6.5%. -2
Therefore a considerable bsfc
Fig 7a: Variation improvement
of specific especially
NOx emissions at pressure
vs. power turbine Fig 8: Exhaust manifold pressure variation vs. exhaust pressure
temperature of the order
low and medium load is ratio
of 600 degree
at 1700 rpm for
experienced various
only
celsiushigh Figure 6.4: Exhaust
enginealoads.
using increasemanifold
for 1700 rpm andpressure variation
25%, 100% engine load vs exhaust
andefficiency
internal T/C.
EGRAtcould -4
25% help
250 loadreduce the NOx.
the maximum
Loa d
bsfc pressure increase for 1700 rpm and 25 %, 100 % engine load
improvement is attributed
225 by 70% to100%
T/C efficiency 1700 rpm, 100% Load
The effect
increase and of increase
by 30% of exhaust
to turbocompounding. 75%back pres-
At high load EFFECT ON BSFC-In this case,
Standard T/C the
Eff. effect of electrical
Soot Variation (%)

200 -6
turbocompounding on10% engine bsfc isT/C
given
Eff. as function of
surethe situation is to
(compared reversed and the overall50%
the175non-turbocompound improvement
case)is Increased
turbine inlet pressure increase (compared to the non
attributed by 42% to T/C efficiency
150 in figure 6.5a
increase
25% and 58% to 20% Increased T/C Eff.
on the BSFC
electric is shown
turbocompounding. Therefore, the andprevious
6.5b. It must beturbocompound
noted herecase).
-8 thatAsthe fuelling
already rateforisbsfc
mentioned maintained
125 improvement two different approaches are adopted, one
the analysis
same as that
reveals of
that the
electricalnon-turbocompounding
turbocompounding can be case. Therefore 0
comparing to estimate
0.2 to
bsfc the 0.4 the
0.6 Bsfc,
non-turbocompound
Downloaded from SAE International by Chalmers University of Technology, Thursday, June 05, 2014 06:22:08 PM the
0.8 case 1 overall
for each power
beneficial
output if we use a100
is considered, highly
i.e. efficient
the powerT/C.generated
Consequently from the engine T/CExhaust
efficiency
and the Pressure
considered
mechanical andIncrease
a second
power (bar)
comparing
generated bsfc from the
the need for increasing75T/C efficiency is clearly revealed. to the value corresponding to the non-turbocompound
turbo-generator. 50 casevariation
Fig 9c: bsfc for the due
standard T/Cturbocompounding
to electric efficiency (existing
vs. one). In the
exhaust
1700 rpm, Kis=80%
25 first increase
pressure case thefor effect of electrical
1700 rpm, 100% load turbocompounding
and various T/C is
revealed while in efficiencies
the second the result is the combined
0
curve being reduced as1.4exhaust pressure effect
22 of both electrical turbocompounding and increased
2 1.5 1.6 1.7 1.8 1.9 increases.
2 2.1 2.2 2.3 2.4 T/C overall efficiency.
The maximum bsfc improvement Power Turbine attributed to electric
Pressure Ratio (-)
turbocompounding alone is approximately 3.8 % for the In00Figs 9a,b and 9c,d is given the overall bsfc variation
(%)
bsfc Variation (%)

0
Variation(%)

highly efficient T/C.

Fig 7b: On theofother
Variation exhausthand the
soot vs. withtotal
power overall
turbine pressure ratio
vs. exhaust pressure increase for 25% and 100% load
at 1700 rpm for various engine loads
bsfc improvement due to both electric respectively for the two approaches adopted. As
-2
bsfcVariation

turbocompounding
-2 and increased T/C efficiency is 6.5%. -2
observed, a different behavior is revealed at 25% and
Therefore a considerable bsfc improvement especially at 100% load, Figs 9a,b and 9c,d respectively. At low
-4
low and medium load is FOR
RESULTS ELECTRICAL
experienced onlyTURBOCOMPOUNDING
using a high engine load, the bsfc reduction is significantly lower
-4
compared to the one at full engine load.
efficiency-4 T/C. At 25% load the maximum bsfc
The theoretical investigation for electrical -6
improvement isturbocompounding
attributed by concentrates
70% to T/Con efficiency 1700 rpm, 100% Load
bsfc

1700 rpm, 25% Load the effect of exhaust In Fig 9a is depicted 1700 rpm, the100%bsfc variation
Load compared to the
increase-6and bypressure
30% to increase
turbocompounding.
Standard before the T/C
T/C Eff.
At high load
turbine and on the -6
one without turbocompounding
Standard T/C Eff.
forEff.
each T/C efficiency
Standard T/C
the situation is effect
reversed of T/Cand the overallT/C
efficiency. improvement is -8 10%As Increased T/Cbsfc
Eff.
10% IncreasedAs already
Eff. mentioned, in the considered at 25% load. 10% Increased T/C Eff. improvement
revealed,
attributed by 42% to T/Cinvestigation
present efficiency increase
exhaust and 58% tohas been
pressure starts to deteriorate 20% afterIncreased
a certain T/Cvalue
Eff. of exhaust
20% Increased T/C Eff. -8 20% Increased T/C Eff.
electric -8turbocompounding.
increased approximately Therefore,1 the bar previous
above the value -10
pressure increase, which is shifted slightly towards
0 corresponding
analysis reveals that 0.4to turbocompounding
0.2electrical the non
0.6 turbocompound
0.8 can1 case.
be In Fig.8 higher 00 values 0.2
0.2as T/C 0.4
efficiency
0.4 0.6increases.
0.6 0.8
0.8 On the 11 other
we isuse
beneficial ifExhaust given the variation
a Pressure
highly efficient of exhaust
T/C.
Increase pressure for the
Consequently
(bar) hand Exhaust
in Fig.9b Pressure
Exhaust is given the Increase
Pressure bsfc variation
Increase (bar)
compared to
(bar)
standard efficiency T/C vs. exhaust pressure increase the non-turbocompound case for the standard T/C
the need for increasing T/C efficiency is clearly revealed.
for 25% and 100% load. From this graph it is obvious efficiency. Obviously in turbocompounding
this case bsfc vs. reduction is
(a) Fig 9c: bsfc variation due due
to variation
electric exhaust
thatBsfc
Fig 9a: bsfc variation due to variation
electric
the maximum for 25
turbocompounding
pressure %across
ratio load the turbine is
vs. exhaust Fig 9d: Total bsfc (b)
significantly Bsfc
variation higher to both forturbocompounding
electric
because 100%
itand isvariousload
included and
pressure increase for 1700 rpm, 25% load and various T/C efficiencies T/C pressure
efficiency increase
increase for 1700
vs. rpm,
exhaust 100% load
pressure increase for T/Crpm the
1700
4.0 which is rather high and explains the reason for contribution of and both electric turbocompounding and
efficiencies
100% load.
increasing exhaust pressure only by 1 bar. This is
Figure 6.5:22 Bsfcobviously variation due to electric turbocompounding
a disadvantage compared to mechanicalEFFECTimprovement
increased T/C efficiency. The observed maximum bsfc
vs exhaust
2ON EXHAUST
pressure
attributed
increase for 1700
to electric turbocompounding
TEMPERATURE-Because of
rpm for
low/high load and various turbocharger
turbocompounding because the reduction efficiency. Taken
of bsfc isthe increasealone of from
is exhaust
1% for [11].
the Standard
highly
pressure efficient
before T/C turbocharger
the efficiency is efficiency
turbine, itwhile
(%)

0 expected to be lower. On the other hand, the absolute the total bsfcof turbine
improvement due to withboth
corresponds0 to 49 %, the 10 %value turbocharger corresponds examined the efficiency
variation inlet temperature
mechanical to 54
0% and the last one corresponds to 59
Variation(%)

bsfc Variation (%)

exhaust pressure compared to

exhaust turbocompounding
pressure increase and increased T/C efficiency is 3.3%.
as shown in Figs 10a-b for
% efficiency. turbocompounding, Fig.3, is significantly lower affecting
less the net engine power output, as shown latter on. 25% and 100%load, loadFigsrespectively. Since exhaust
Variation

-2 -2 100%
For 9c,d, bsfc reduction continues to
-2 temperature improve depends only on
with exhaust exhaust
pressure pressure
increase levelofitsthe
the slope
variation -4 is the same for all T/C efficiencies examined.
Figure -4 6.5a and 6.5b shows the effect of electrical turbocompounding with
As shown exhaust temperature at the turbine inlet
different turbochargers having
-4
different efficiency. It can be seen that at low load, the increases bsfc reduction is lower compared to the one at full load.
bsfc

1700 rpm, 25% Load -6

almost linearly with exhaust pressure the
bsfc

For low load, -6 after a certain

1700 rpm, 25%point
Load
Standard T/C Eff. of exhaust pressure slope being
increase, slightly
the lower
bsfc
1700 at
starts
rpm, high
100% to exhaust
deteriorate.
Load pressure. A The
maximum of
-6 Standard T/C Eff. maximum temperature increase Standard ranges from 600C at low
approximately 1% bsfc reduction
10% Increased can be seen at low load,
T/C Eff. whereas,
-8 a0 maximum
engine load to 50 C at full engine
ofT/C 3.8Eff.
load. This
% is seen at high loads.
reveals that
20% Increased T/C Eff. 10% Increased T/C Eff. 10% Increased T/C Eff.
Note that,-8Standard turbocharger efficiency
20% Increased T/C Eff. correspondsno to 49 %, the
significant 10 %exists
problem turbocharger
for theT/C
20% Increased T/Ccorresponds to 54 %
Eff. from the
-8 -10 of electric turbocompounding. The increase
application
efficiency and00 the last
0.2 one
0.2
0.4 corresponds
0.4
0.6
0.6
0.8 to 591 % efficiency
0.8 1
as explained earlier.
Exhaust Pressure Increase (bar) compared to0 mechanical
0.2 0.4 0.6 0.8 is slightly
turbocompounding 1
Figure 6.6a Exhaust
and 6.6bPressure shows Increase
the impact (bar) on bsfc variation
lower due due toExhaust
thetolower
bothPressure
absolute Increase
exhaust
electrical (bar)
pressure value.
turbo-compounding and
Fig 9b: Total bsfc variation due to both electric turbocompounding and
improved
Fig
T/C9a: turbocharger
bsfc variation
efficiency increasedue efficiency.
vs.toexhaust
electric Allforvs.
turbocompounding
pressure increase the curves
exhaust
1700 rpm in the graph
Fig 9d: Total bsfccorresponds
variation due to bothto percentage
electric of bsfc
turbocompounding and variation
pressure increase for 1700 rpm, load and various T/C efficiencies
and 25% load.
compared to the non turbo-compound case with standardT/C efficiency increase vs. exhaust pressure increase for 1700 rpm
turbocharger efficiency of 49 %. Therefore, the dashed
and 100% load.
2
line and thin solid line corresponds to bsfc improvement due to the addition of
EFFECT ON EXHAUST TEMPERATURE-Because of
electrical turbo-compound and
also due to increasing the turbocharger efficiency to 54% and 59%
the increase respectively.
of exhaust Whereas,
pressure before the thick
the turbine, it is solid line
bsfc Variation (%)

0
examined the variation of turbine inlet temperature with
exhaust pressure increase as shown in Figs 10a-b for
-2 25% and 100% load respectively. Since exhaust
14
temperature depends only on exhaust pressure level its
variation is the same for all T/C efficiencies examined.
-4 As shown exhaust temperature at the turbine inlet
1700 rpm, 25% Load
increases almost linearly with exhaust pressure the
-6 Standard T/C Eff. slope being slightly lower at high exhaust pressure. The
10% Increased T/C Eff.
maximum temperature increase ranges from 600C at low
 increase and by 30% to1700 turbocompounding. At high load -6 1700Standard T/C Eff.

b
bsf
rpm, 25% Load

bsf
rpm, 100% Load
the situation-6 is reversed and Standard
the overall improvement is
T/C Eff. 10%Standard
Increased T/C
T/C Eff.Eff.
-8
attributed by 42% to T/C efficiency increase
10% Increased T/Cand
Eff. 58% to 20%10%
Increased T/CT/C
Increased Eff.Eff.
electric turbocompounding. 20% Therefore, theEff.previous -8
Increased T/C 20% Increased T/C Eff.
-8 -10
0 0.2 0.4 0.6 0.8 1
analysis reveals that electrical turbocompounding can be
beneficial if we0 use a0.2 0.4
highly efficient 0.6 0.8
T/C. Consequently 1 Exhaust
0 Pressure
0.2 0.4 Increase
0.6 (bar)
0.8 1
ExhaustT/C
the need for increasing Pressure
efficiencyIncrease (bar)
is clearly revealed. Exhaust Pressure Increase (bar)
Fig 9c: bsfc variation due to electric turbocompounding vs. exhaust
Fig 9a: bsfc variation due to electric turbocompounding vs. exhaust pressure
Fig increase
9d: Total for 1700
bsfc variation duerpm, 100%
to both load and
electric various T/C
turbocompounding and
pressure increase for 1700 rpm, 25% load and various T/C efficiencies T/C efficiency increase vs. exhaust pressure increase for 1700 rpm
efficiencies
and 100% load.
2 2 2
EFFECT ON EXHAUST TEMPERATURE-Because of
the increase
0 of exhaust pressure before the turbine, it is
bsfc Variation (%)

0
bsfc Variation (%)

bsfc Variation (%)

examined the variation of turbine inlet temperature with
exhaust pressure increase as shown in Figs 10a-b for
-2
-2 -2 25% and 100% load respectively. Since exhaust
temperature depends only on exhaust pressure level its
-4 is the same for all T/C efficiencies examined.
variation
-4 -4 As shown exhaust temperature at the turbine inlet
-6 almost linearly with exhaust pressure the
increases
Downloaded
1700
1700 from
rpm,
rpm, SAE
25% International
25% Load
Load by Chalmers University of Technology, Thursday,1700
Junerpm,
05, 2014 06:22:08 PM
-6 -6 Standard
Standard T/CT/C
Eff.Eff.
slope being slightly lower at100%
highLoad
exhaust pressure. The
Standard T/C Eff.
10% Increased
maximum temperature increase ranges from 600C at low
-8
10% Increased T/CT/CEff.Eff.
20% Increased
engine load to 500C at full10% Increased T/C Eff.
engine load. This reveals that
20% Increased T/CT/CEff.Eff.
-8 -8 no significant
-10 problem 20% Increased T/C Eff.
exists for the T/C from the
0 0 0.20.2 0.40.4 0.60.6 0.80.8 1 1 application
60 0
of electric
0.2 turbocompounding.
0.4 0.6 0.8The increase
1
Turbine Inlet Temp. Variation (oC)

Exhaust
Exhaust
120
Pressure
Pressure Increase
Increase (bar)
(bar) comparedExhaust
to mechanical
Pressure turbocompounding
1700 rpm, 25% Load
Increase is slightly
(bar)

Generated Power (kW)

1700 rpm, 25% Load Standard T/C Eff.
lower due to the lower absolute exhaust pressure value.
50
FigFig
9a:9b: Total
bsfc bsfc(a)
100 variation
variation dueBsfc to Standard
duevariation
to electric both T/C Eff.
electric
for turbocompounding
turbocompounding 25Eff.vs.
%forexhaust
load and Fig 9d: Total bsfc (b) Bsfc
variation
10% Increased T/C Eff.
duevariation for
to both electric 100% load and
turbocompounding
T/C increase
pressure efficiencyfor
increase
1700 rpm, 25% 10%
vs. exhaustload Increased
pressure
and variousT/C T/C
increase 1700 rpm
efficiencies 20%pressure
T/C efficiency increase vs. exhaust Increased T/C Eff.for 1700 rpm
increase
80 and 25% 20%load.
Increased T/C Eff. 40 and 100% load.
2
Figure 6.6: Total Bsfc variation due to both electric turbocompounding and improved turbocharger
60 EFFECT30ON EXHAUST TEMPERATURE-Because of
efficiency vs exhaust pressure increase for 1700 rpm for
the low/high
increase of load. Taken
exhaust from
pressure [11].theStandard
before turbocharger
turbine, it is
bsfc Variation (%)

0 40
efficiency corresponds to 49 %, the 10 % turbocharger corresponds
examined20the to 54 % efficiency
variation andtemperature
of turbine inlet the last one
withcorresponds
to 59 % efficiency.
20 exhaust pressure increase as shown in Figs 10a-b for
-2 25% and 10 100% load respectively. Since exhaust
0
temperature depends only on exhaust pressure level its
variation is0 the same for all T/C efficiencies examined.
-4 the-20
represents bsfc
0
variation
0.2 0.4
due 0.6
to electrical
0.8 1
turbo-compounding
As shown exhaust
0
and0.2standard
temperature
0.4
turbocharger
0.6
at the turbine
0.8
efficiency
1
inlet of 49%.
The author goes on to say
25%that ”At 25% increases almost linearly with exhaust pressure the
Exhaust
1700 Pressure
rpm, Increase
Load (bar) load the maximum bsfc Exhaust improvement
Pressure Increase is attributed(bar) by 70% to T/C
al by Chalmers University of Technology,
-6 Thursday, June 05, 2014T/C 06:22:08 PM slope being slightly lower at high exhaust pressure. The
efficiency increase andStandard by 30% Eff. to turbo-compounding. At high
maximum load theincrease
temperature situation rangesis from
reversed 0 and low the overall
Fig 10a: Variation of turbine 10%
inlet temperature
Increased T/Cvs. exhaust
Eff. pressure Fig 11a Variation of generated
0
power vs. exhaust pressure60 C at
increase at
improvementincrease for is
1700attributed
rpm and 25% load by for42% toT/CT/C
various
20% Increased T/C Eff.
efficiency increase1700
efficiencies. engine loadandrpm 58%
to and 25%to
50 C at electric
full
load engine turbo-compounding”.
load.
for various T/C This reveals
efficiencies that
-8 no significant problem exists for the T/C from the
0 0.2 0.4 0.6 0.8 1 application of electric turbocompounding. The increase
Exhaust Pressure Increase (bar) compared 60 to mechanical turbocompounding is slightly
Turbine Inlet Temp. Variation ( C)

60
1700 rpm, 25% Load lower due to the lower 1700 rpm, 100%
absolute Load pressure value.
exhaust
o

120
Generated Power (kW)

Generated Power (kW)

Fig 9b: Total bsfc Standard

1700 rpm, T/C
100% Load Eff. Standard T/C Eff.
50 variation due to both electric
Standard
turbocompounding
T/C Eff.
and 50
increase vs. exhaust10%
T/C efficiency100 Increased
pressure T/C for
increase Eff.1700 rpm 10% Increased T/C Eff.
Eff. 10%
and 25%20% Increased T/C
load.Increased Eff.
T/C Eff. 20% Increased T/C Eff.
Eff. 4080 20% Increased T/C Eff. 40

3060 30
40
20 20
20
10 10
0

0-20 0
0.8 1 00 0.2
0.2 0.4
0.4 0.6 0.8
0.8 11 0 0.2 0.4 0.6 0.8 1
e (bar) ExhaustPressure
Exhaust Pressure Increase
Increase (bar)
(bar) Exhaust Pressure Increase (bar)

xhaust pressure FigFig

11a10b: Variation
Variation of turbine inlet
of generated
(a) Generated powertemperature
vs. exhaust
power vs.
for exhaustincrease
pressure%pressure
load at
25efficiencies. Fig 11b Variation of generated power vs. exhaust pressure increase at
(b)andGenerated
/C efficiencies. increase for rpm
1700 1700 rpm
and andload
25% 100% forload for various
various T/C
T/C efficiencies 1700 rpm 100% load for power for
various T/C 100% load
efficiencies

EFFECT ON TURBINE POWER PERCENTAGE

Figure 6.7: Generated power Vs exhaust pressure increase at
CONVERTED TO1700 rpm for low and
ELECTRIC-Another high loads
important
EFFECT 60 ON GENERATED POWER- In the case of parameter is the part of turbine power which is
1700 rpm,generated
electric turbocompounding, 100% Load electric power is converted to electric power and its variation with, engine
Figure 6.7a and 6.7b Standard
shows theEff. generated electric power vs increase in exhaust pressure. The maximum
Generated Power (kW)

T/C
significantly
50 important because it defines the size of the load and exhaust gas pressure increase. This variation
power generated
required is 15 kW
electric generator.10% Increased
Asand 33 kW
witnessed T/Cfrom
Eff.
for 25%
Figs 11a-and 100% load respectively.
is illustrated in Figs 12a-b for 25% and 100% load
Eff.
b corresponding to 25% 20%
and Increased T/C Eff.
100% load respectively respectively. As shown, the percentage of turbine power
Eff. 40
Figure 6.8aelectric
generated and 6.8b shows
power the percentage
increases of turbine
with exhaust powertowhich
converted electric is converted
power increasestowith electric
exhaustpower
gas versus the
pressure at a decreasing slope. Generated electric pressure and T/C efficiency. The relative percentage
increase in30exhaust back pressure. At low load approximately
power increases with the increase of T/C efficiency as
55% of the turbine power
presents a maximum value of 55% at low engine load
is converted to electric
and expected.
at high load approximately
For the 30% the
present application is converted
maximum to electric.
while theThis shows that
corresponding one atatfullhigh
load load,
is 30%.most of the turbine
This is
20 electric power ranges from 15 kW at 25%
generated reasonable given that the turbocharger compressor
power is utilised ny the compressor to provide the required boost pressure.
engine load to 33 kW at 100% engine load for the highly absorbs more power at full engine load to provide the
efficient10
Figure T/C. required
6.9a and 6.9b shows the relative decrease in engineboost pressure.
power versus the increase in exhaust pressure.
The maximum 0
decrease in engine power is 16% and 5% for 25% and 100% load respectively.
0 0.2 0.4 0.6
0.8 1
Figure 6.10a and 6.10b shows the 0.8
variation 1
of total system power versus the increase in exhaust pressure. A
e (bar) Exhaust Pressure Increase (bar)
maximum of 94 kW and 393 kW is achieved for low and high load respectively. Further, the author concludes
xhaust pressure
T/C efficiencies.
thatFigboth turbo-compounding
11b Variation (mechanical
of generated power vs. exhaust and
pressure increase
1700 rpm and 100% load for various T/C efficiencies
at electrical turbo-compounding) results in reduction in primary

engine power output but the overall power output increases due to the power generated from the exhaust gas
EFFECT ON TURBINE POWER PERCENTAGE
CONVERTED TO ELECTRIC-Another important
n the case of parameter is the part of turbine power which is
ectric power is converted to electric power and its variation with, engine
the size of the load and exhaust gas pressure increase. This variation
15
from Figs 11a- is illustrated in Figs 12a-b for 25% and 100% load
d respectively respectively. As shown, the percentage of turbine power
with exhaust converted to electric power increases with exhaust gas
erated electric pressure and T/C efficiency. The relative percentage
C efficiency as presents a maximum value of 55% at low engine load
 1700 rp

Net Engi
Turbine
20
-20
10
0 -24
0 0.2 0.4 0.6 0.8 1 0 0.2
Exhaust
Downloaded from SAE International by Chalmers University of Technology, Thursday, JunePressure Increase
05, 2014 06:22:08 PM (bar) Exhaust Pres
Fig 12a: Variation of turbine power percentage converted to electric vs. Fig 13a: Relative decrease of net
exhaust pressure increase at 1700rpm and 25% load for various T/C increase at 1700 rpm and 25%
efficiencies.

Net Engine Power Change (%)

Downloaded from SAE International by Chalmers University of Technology, Thursday, June 05, 2014 06:22:08 PM

(%)Electric (%)
100 0

Net Engine Power Change (%)

Turbine Power to Electric (%)
100 0 1700 rpm, 100% Load
1700 rpm, 25% Load
90 90 Standard T/C Eff.
Standard T/C Eff.
-4
80 10% Increased T/C Eff. 80
-4 10% Increased T/C Eff.
20% Increased T/C Eff. 70 20% Increased T/C Eff.
70 -8
-8
Turbine Power to Electric (%)

100
60 60 0

Power to
1700 rpm, 25% Load
5090 Standard T/C Eff. 50 -12

Power Change
-12
4080
10% Increased T/C Eff. 40-4
20% Increased T/C Eff. -16
3070 -16
30
-8 1700 rpm
1700 rpm, 25% Load

Net EngineTurbine
2060 20 Standard T/C Eff. -20
-20
1050 -12
10 10% Increased T/C Eff.

040
20% Increased T/C Eff. -24
-240
300 -16 0 0.2
0.2 0.4 0.6 0.8 1 00 0.21700 0.4
0.2 0.4 25%0.6
rpm, 0.6 0.8
Load 0.8 1 1
Exhaust Pressure Increase (bar)
20 ExhaustPressure
Exhaust Pressure Increase
Increase
Standard (bar)
T/C Eff. (bar) Exhaust Pres
-20
10 10% Increased T/C Eff.
Fig 13b Relative decrease of net
(a) 0Turbine
Fig 12a: Variation power
of turbine power to electric
percentage convertedfor 25 %
to electric vs. load Fig13a:
Fig (b)
12b Relative
VariationTurbine
decrease
-241700increase power
of turbineofpower
net 20% to
percentage
engine powerelectric
converted
vs.
Increased exhaustfor
T/C Eff. 100%
topressure
electric vs. load
increase at 1700 rpm and 100%
exhaust pressure increase at 1700rpm and 25% load for various T/C increase
exhaust at
pressure rpm and at 25% load forand
1700rpm various
100%T/C efficiencies
load for various T/C
0 efficiencies.
0.2 0.4 0.6 0.8 1 0 0.2 efficiencies.
0.4 0.6 0.8 1
nal by Chalmers University of Technology, Thursday, June 05, 2014 06:22:08 PM
Figure 6.8: 100 Turbine
Exhaustpower percentage
Increaseconverted to electric Vs exhaustPressure
pressureIncrease
increase at 1700 rpm for low

Net Engine Power Change (%)

Pressure (bar) 0 Exhaust (bar)
Electric (%)

1700 rpm, 100% Load

and high loads
90 Standard
Fig 12a: Variation of turbine power T/C Eff.
percentage converted to electric vs.
EFFECT ON TOTAL POWE
ofFig
Downloaded from SAE International by Chalmers University EFFECT13a: Relative
Technology, decreaseJune
-4ON Thursday,
NET ENGINE
of net05,
engine
2014power
POWER-
vs. exhaust
06:22:08 PM pressure
In Figures 13a-b is in Figures 14a-b is given th
exhaust
80pressure increase at 1700rpm and 25%
10% Increased load
T/C Eff.for various T/C increase at 1700 rpm and 25% load for various T/C efficiencies
efficiencies.
20% Increased T/C Eff.
given the relative decrease of net diesel engine power for 100% and 25% engine
70 for 25% -8and 100% respectively as function of exhaust of exhaust pressure increa
Net Engine Power Change (%)
Turbine Power to Electric (%)
(%)

100 0
60 0 1700 rpm, 100% Load pressure increase. The relative decrease has a system obtains a maximum
Power to

ff. 90 Standard T/C Eff. maximum -12 value of 16% at low engine load and 5% at full of the standard T/C efficien
Change

50
-4 -4
T/C Eff. 410 80 10% Increased T/C Eff. engine load. The relative
The contribution reductionisofhigher
of the second net engine powerto
compared the increased one, as it is s
40 1700 rpm,20%100% Load T/C Eff.
Power(kW)

T/C Eff. 70 Increased does not

-16
mechanical depend on T/C efficiency and varies linearly in Fig 14b the benefit in tota
30-8
405 Standard T/C Eff. -8 turbocompounding.
1700 rpm, 100% Load engine load is rather limited
with exhaust pressure increase. Compared to
60
Turbine

20 10% Increased T/C Eff. mechanical Standard

turbocompounding T/Crelative
the Eff. reduction is
Finally -20
Total Power

40050 20% Increased T/C Eff.

the
-12 variation of soot
10% (variation
T/C Eff.expressed
Increased pressure for
-12 lower
10 FSN) since
with aexhaust
lower exhaust manifold
pressure
20% increase
Increased T/C
has beenin
is illustrated
Eff.
395 40 used
0
-16
Figs as already
-24
16a-b for shown.
25% and 100% load respectively. The -16
Net Engine

390300 0.21700 rpm,

0.4 25% 0.6
Load 0.8 1 effect of T/C 0 efficiency0.21700isrpm, 0.4extremely
100% 0.6Loadsmall0.8 because 1 soot
Exhaust Pressure
20 Increase
Standard T/C Eff. (bar) is affected Exhaust
-20 from thePressure exhaust StandardIncrease
T/C Eff.level
pressure (bar) as already
-20
385
10 10% Increased T/C Eff. mentioned which results to 10% Increased
internal EGR. T/CForEff. this reason
Fig 13b Relative decreaseonly of netforengine thepower vs. exhaust T/Cpressure
System

Fig 12b Variation 20% Increased

of turbine power percentage T/C Eff.to electric vs.
converted results are given 20% Increased
standard T/C Eff. efficiency
380 -240 increase -24
at 1700 rpm and 100% load for various T/C efficiencies
exhaust pressure increase at 1700rpm and 100% load for various T/C since for the 0two others they are practically identical. As
0.8 1 00 0.2 0.4
0.2efficiencies.
0.4 0.6
0.6 0.80.8 1 1 0.2 0.4 0.6 0.8 1
375 observed we have anPressure increase Increaseof soot emissions with
ase (bar) Exhaust
ExhaustPressure PressureIncrease Increase(bar) (bar) Exhaust (bar)
exhaust pressure increase ranging from 30% at low load
370 EFFECT
verted to electric vs. Fig
Fig13a: Relative
12b(a) Net
Variation decrease
of engine
turbineofpower
netpower
engine power
percentage vs.
change exhaust
convertedfor topressure
25 %
electric load
vs.
toFig50%13b ON
(b)
TOTAL
at full
Relative load.
NetdecreasePOWER
Obviously
engine
OFthe
of net power
engine THE
power SYSTEM-Finally,
percentage
vs. exhaust
change for increase
100% of
pressure load
EFFECT ON 0 NETrpm 0.2
ENGINE 0.4
POWER- 0.6In Figures
0.8 1 insoot
Figures
increase14a-b isrpm
given the total power atT/C
the flywheel
oad for various T/C increase
exhaust at 1700
pressure and at
increase 25% load forand
1700rpm various
100%T/C load for 13a-b
various is
efficiencies T/C emissions
Downloaded
at 1700 and fromcompared
100% SAE to
load International
for various by mechanical
Chalmers University of Technology, Thursday, June 05
efficiencies
Exhaust
given the relative Pressure
decrease of netIncrease(bar)
efficiencies. diesel engine power forturbocompounding
100% and 25% engine load respectively
is lower due to the lower maximum as function
Change (%)

Figure
for 25% 6.9:and Relative
0 100% respectivelydecrease as of function
net engine of exhaustpower vs exhaust
ofexhaust
exhaust pressure increase
pressurepressure.
manifold increase.The at percentage
The 1700 rpm increase
total power for
of thelowofand high load
Fig 14a Totalincrease.
pressure power of theThe systemrelative
vs. exhaust pressure increase
decrease has at a system obtains a maximum value of 378 kWbeforexplained
the case by
1700 rpm and 100% load for various T/C efficiencies
soot
EFFECT emissions
ON TOTAL is rather
POWER high and
OFa THE can SYSTEM-Finally,
f. maximum -4 value of 16% at low engine load and 5% at full of the standard T/C efficiency and value of 393 kW for
the low absolute
in increased
Figures 14a-b values
is given ofthethetotal
non-turbocompound
power case.
T/C Eff. EFFECT
engine load.ON
100 The NET ENGINE
relative POWER-
reduction In Figures
of net engine power 13a-b is the
In any case one,
theseas itresults
is shown in Fig
should 14a.atAs
only
the
beas
flywheel
depicted
considered
given the relative 1700 rpm,of25%
decrease net Load
diesel engine power for 100% 410 and 25% engine load respectively function The contribution of the se
Power(kW)

T/C Eff. inqualitatively

Fig 14b the benefit
does not depend on T/C efficiency and varies linearly due toin1700 totalrpm, power 100% ofexperimental
the
Loadsystem at data 25% for
(%) Power (kW)

the lack of
for 25%
with 98-8and 100%
exhaust pressurerespectively
Standardas
increase. function
T/C Eff.
Compared of exhaustto
of exhaust
engine load
405 is pressure
rather increase.
limited compared The
Standard T/C
total
to 100%
Eff.
powerload. of the mechanical turbocompound
pressure turbocompounding
increase. The10% relative
Increas decrease
ed Eff. hasis a
T/Creduction
model
systemcalibration,
obtains a maximumas already mentioned.
value of 378 kW for the case
mechanical the relative 10% Increased T/C Eff.
Power

maximum 96-12 value of 16% at 20% low engine

Increas load
ed T/CandEffic. 5%been
at full of the standard
400 T/C efficiency and a value of 393 kW for Finally the variation of so
lower since a lower exhaust manifold pressure has 2 one, as it is shown 20% Increased T/C Eff.
engine load. The relative reduction of net engine power the increased in Fig 14a. As depicted FSN) with exhaust pressu
used as already shown. 395the benefit in total power of the system at 25%
does not -16 depend on T/C efficiency and varies linearly
94 in Fig 14b
Engine

Figs 16a-b for 25% and 1

1700 rpm, 100% Load 0 is rather limited compared to 100% load.
Net Total

with exhaust pressure increase. Compared to engine load

390 effect of T/C efficiency is e
Total

92 Standard T/C Eff.

mechanical -20 turbocompounding the relative reduction is is affected from the exhau
10% Increased T/C Eff. -2
lower since a lower exhaust manifold pressure has been 385
Variation

mentioned which results to i

90
System

20% Increased T/C Eff.

used as-24 already shown.
NOx System

380-4 results are given only for

0.8 1 88 0 0.2 0.4 0.6 0.8 1 since for the two others the
ase (bar) Exhaust Pressure Increase (bar) 375-6 observed we have an incre
86 exhaust pressure increase r
Fig 13b Relative decrease of net engine power vs. exhaust pressure
370-8
verted to electric vs. 1700 rpm, 25% Load to 50% at full load. Obvious
increase at01700 rpm0.2 and 100% 0.4 0.6 T/C0.8
load for various efficiencies1 0 0.2 0.4 0.6 0.8 1
oad for various T/C Standard T/C Eff. soot emissions com
Exhaust Pressure Increase(bar) -10Exhaust Pressure Increase(bar)
10% Increased T/C Eff. turbocompounding is lower
-12 (b)
20% Increased T/C Eff. exhaust manifold pressure.
Fig 14b Total (a)powerSystem total OFpower for 25increase % load
of the system vs. exhaust pressure at Fig 14a Total powerSystem
of the system total powerpressure
vs. exhaust for 100% increaseloadat
EFFECT 1700 ONrpm TOTAL
and 25% POWER
load for various THET/CSYSTEM-Finally,
efficiencies 1700 rpm and 100% load for various T/C efficiencies
soot emissions is rather hig
in Figures 14a-b is given the total power at the flywheel 0 0.2 0.4 0.6 0.8 1 the low absolute values of th
Figures 13a-b is
for 100%6.10: and 25% engine load respectively as function 100Variation of Exhaust Pressure (bar)
el engine power Figure Total power of the system vs exhaust pressure increase 1700 atrpm, 1700 25%rpm Load for low and high load In any case these results
Total Power (kW)

ction of exhaust EFFECT OF E-TURBOCOMPOUNDING

of exhaust pressure increase. The total power of the ON ENGINE Fig 15a Variation of specific NOx emissions qualitatively due to the la
98 Standard T/Cvs.Eff.
exhaust pressure
crease has a OUT
system EMISSIONS-
obtains a maximum As invalue the ofcase
378 kW of formechanical
the case increase for 1700 rpm and 25% load for various T/C efficiencies. model calibration, as alread
10% Increas ed T/C Eff.
ad and 5% at full turbocompounding,
of the standard T/C efficiency electricalandturbocompounding
a value of 393 kW foris
96 20% Increas ed T/C Effic.
et engine power accompanied
the increased
energy. The author by
one, anasincrease
it is shown
also of inexhaust
mentions Fig 14a. pressure.
that,As depicted
maximum For bsfc reduction of 0.2-2% from 25% to 100% load and 1 bar 2
d varies linearly in Fig
this 14b theitbenefit
reason, in total power
is expected of the system
an effect on engine at 25% out 942
Compared to exhaust
tailpipe back
engine emissions, pressure
load is ratherNOx limited increase
and compared
Soot, whichusing
to 100% a
is load.conventional
examined turbocharger is possible. However, with the increased 0
NOx Variation (%)

ive reduction is turbocharger

herein. efficiency, approximately 3.3-6.5 % reduction in
920 bsfc is possible when increasing exhaust back
essure has been -2
(%)

pressure by 1 bar. This is in agreement with the caterpillars90prediction of 5-10% improvement in fuel economy
System

The impact of electric turbocompounding on diesel -2 -4

on engine
a 14.6specific
litre diesel engine is[10].
NOx Variation

NOx emissions depicted in Figs 15a-b

88
-4
for 25% and 100% load respectively. As revealed the -6
NOx variation for both engine loads examined i.e. low 86
-6 -8
and high is quite similar. The maximum reduction of 00.2 0.4 0.6 0.8 1
1700 rp
specific NOx emissions is 7% for the maximum value of S
16 -8Exhaust1700
Pressure Increase(bar)
rpm, 100% Load -10
exhaust pressure increase and maximum T/C efficiency. 1
Standard T/C Eff.
The observed higher reduction of NOx with increasing -10 power of the system10% 2
Fig 14b Total vs. exhaust pressure
Increased T/C Eff.increase at -12
T/C efficiency is mainly attributed to the increase of 1700 rpm and 25% load for various T/C efficiencies
0 0.2 0
20% Increased T/C Eff.
overall power output because specific values are used. -12
Variation of Ex
The overall reduction of NOx emissions in the case of
EFFECT OF0 E-TURBOCOMPOUNDING
0.2 0.4 0.6 0.8 ENGINE
ON 1
Fig 15a Variation of specific NO
electric turbocompounding is partially attributed to the Variation As
of Exhaust Pressure (bar)
OUT EMISSIONS- in the case of mechanical increase for 1700 rpm and 25%
6.3 Conclusion from the discussion on increase in exhaust back
pressure
The above two section discussed the effect of increasing the exhaust back pressure on a light duty naturally
aspirated gasoline engine and heavy duty turbocharged diesel engine. It is clear that the addition of any system
in exhaust stream of the engine increases the back pressure which in turn increases the exhaust pumping losses
irrespective of whether it is gasoline or diesel engine.
Addition of energy recovery system along the exhaust gas path of the engine has two main effects
- The exhaust back pressure increases which causes the exhaust gases (or hot residual gases) to flow back
into the cylinder during the last phase of the exhaust stroke. This phenomenon is referred to as internal
EGR which is known to deteriorate the engine power. It also creates condition favourable for knocking in
gasoline engine.

- The efficiency of the engine increases because the amount of power recovered from the exhaust gases
through MGU-H is higher than the power lost by the engine. This is proved correct for heavy duty diesel
engine, but, it is still not clear in case of light duty gasoline engine.
In gasoline engine the exhaust gases are at higher temperature of the order of 9000 C when compared to
diesel engine whose exhaust gases are of the order of 6000 C. Therefore, gasoline engines are more susceptible
to knocking problems and hence the exhaust back pressure has to be maintained low when compared to diesel
engine. This could change in years to come because the combustion in gasoline engine are becoming more or less
similar to diesel engine. Therefore, for the current engine technology, the MGU-H seems more suitable for diesel
engine than gasoline engine. However, the UK based, micro-hybrid specialist, controlled power technologies
(CPT) has developed ’Turbine-generator integrated Gas energy recovery system (TIGERS)’ which is currently
in validation phase for proof-of-concept design [9]. It is said that 1.2 kW can be recovered with turbocharger
speed of 45000 rpm. An earlier paper from the same company in 2007 estimated that the 6kW can be recycled
at 80000rpm turbine speed from a 2 L gasoline engine [14].
It is seen that there exists an optimum exhaust back pressure for various speed-load operating points of the
engine that gives maximum reduction in fuel consumption. Therefore, by controlling the waste gate along with
the MGU-H speed, an optimum control can be developed for diesel engine which gives optimum fuel economy.
In literature there already exists novel methods for controlling the VGT and EGR for heavy duty diesel engine,
for example see [20].
The FIA rules allows a maximum of 4 MJ of energy in the ES at any time the car is on the track, out of
which a maximum of 2 MJ of energy per lap can be recovered while braking by using MGU-K. This means, a
maximum of 2 MJ of energy per lap may be recovered from MGU-H to get maximum out of imposed rules but
this comes at the cost of reduced engine power. One lap is estimated to be around for 90-120 seconds, which
means that the energy has to be recovered at the rate of 22-16 kW from MGU-H continuously. The energy
content in exhaust gases are higher in heavy duty engines which is approximately around 40 kW for engine
whose rated power is 360 kW. The rated engine power for formula one engine is around 450 kW which means
nearly 50 kW of energy is available for recovery using the electrical turbo compound technology. But, practical
recoverable energy might be limited due to the increase in exhaust back pressure. Hence it is estimated that
the size of the MGU-H is to be around 40-50 kW.
At first it may seem like MGU-H technology is not appropriate for formula one as it decreases the engine
output power. But, the total system power increases as explained in section 6.2 thereby improving the fuel
economy of the vehicle and in turn staying within 100 kg per race limit. Although, the exhaust pumping losses
increases due to increase in exhaust back pressure, it can be said that this pumping loses is recovered by the
MGU-H along with some additional power regenerated from the exhaust gases. This energy generated form
MGU-H can be sent directly to MGU-K instead of storing it in the ES, thereby avoiding two stage of energy
conversion losses (to and fro from the battery). This energy flow path has no restriction imposed by FIA and
if exploited will improve the fuel economy as well as performance of the formula one car. Due to the direct
energy flow from MGU-H to MGU-K, the size of the battery can be reduced to 2MJ (energy recovery limit of
MGU-K) or an electro mechanical flywheel can be implemented instead of a battery which is more efficient.

17
 Implementing the MGU-H technology in road cars seems advantageous as more aggressive downsizing can
be adopted. An analysis of assisted turbo-charging with light hybrid power train is presented in [18]. The cost
benefit analysis of assisted turbo-charging is compared with full hybrid.
Heavy duty truck engines uses engine braking during deceleration to remove excessive load on the friction
brake pads. The MGU-H can be used with the engine braking thereby improving the braking performance/time.
For formula one, this means, energy can be recovered during braking not only from MGU-K but also from
MGU-H. This is a vague idea and could lead to less energy recovered or also might lead to surge problems in
compressor.

7 Discussion on fuel consumption reduction on stan-

dard and real driving cycle
The author in [15] has performed a study to analyse the potential of electric assisted turbocharger for heavy
duty diesel engine with 6 cylinder and eight litre displacement which used a variable geometry turbine (VGT).
The vehicle used is a urban bus with laden mass of 16.5 tons. A 1-D simulation is carried out for the base line
model and validated experimentally. The baseline engine model used a VGT which was then replaced with a
fixed geometry turbine for simulation purpose. The energy storage system is a super capacitor. The electric
machine used here has constant torque of 1 Nm upto 60000 rpm and a constant power of 6.3 kW upto 12000
rpm. It is capable of generating 7.6 Kw power in generator mode. The function of the electric machine in the
electric assisted turbocharger is to reduce the turbo lag during acceleration (from engine idle speed) and to
recover part of exhaust energy to power up the engine auxiliaries thereby reducing the fuel consumption.

An extract from the same reference [15] is shown
in figure 7.1. The electric machine is used as a mo- 


   
 
tor during the acceleration period to reduce turbo  





 
lag. Once the required turbo pressure is reached the     
electric machine is operated as generator (however     
   
discontinuously). ’8 seconds after the beginning of the  


cycle the upper threshold level (0.65) is reached for the   



SOC, and one of the two engine-driven alternators  

can be switched off, with a power saving of about 4-5     
            
kW, which can continue until the SOC falls below a 
lower threshold level (0.6) at about 19 seconds from  

 
  
the start of the cycle’. Not much about the control 



is explained in the paper, however, the above excerpt 




suggests that the control here is designed such that   


the engine-driven alternator is switched off once the 


upper SOC (state of charge) threshold is reached and 
 
the simulation itself is stopped when the lower SOC      

(state of charge) threshold is reached. He also per-         
 
forms a simulation on five different driving cycles and 
  
concludes that 6 to 1 % of fuel consumption reduction 
 
  
is achievable with the lower value corresponding to 






congested traffic condition. The reason for lower value


in congested traffic is given to be that the the electric 


machine consumes most of the power to overcome  
turbo lag and the operating conditions to generate 
 
electricity are very less and are of small duration. He    

 
also suggests that the turbo lag can be reduced as
 
much as 30 % but careful selection of compressor is 


required in order to avoid the compressor operating

above surge limit. Figure 7.1: The operating mode of MGU-H. Taken from
[15]
More about the surge problem and the optimum       
 


        
18         



         
      
        
        
        
In the United States, the federal urban driving cycle (FUDS)
represents a typical city driving cycle, while the federal highway

The reference vehicle is the FIAT Punto equipped with a 4

cylinders, 1.3 liters EURO 5 Diesel turbocharged engine. The Figure 6 shows the power of the two machines for the baseline
main technical data of vehicle and driveline are reported in case (i.e. the PTC is zero in eq. 6);
6 grey and black lines
Table II. represent the compressor power and the turbine power,
respectively. From the graph the two curves are overlapped,
The simulations were carried out assuming a constant except in the transient phases (see an example in Figure 7).
7
size of electric machine for reducing turbolag is dis-
electrical load of 370 W, though the electrical load is not When a difference between the turbine and the compressor
cussed in [19]. Turbolag reduction is also discussed in [3] and [12].
constant along the real driving cycle and exhibits greater powers occurs the turbochargers accelerate or decelerate
The
values upauthor
to 1 kW in [2] initial
in the caries out The
phase. a study
poweron 1.3 W
of 370 Litre
was four according to the power balance of equation (6).
(6)
calculated considering the contribution of the auxiliaries
cylinder EURO 5 Diesel turbocharged engine with a vehicle
indicatedof
weight in Table III. It is worth 1200
approximately noting kg.
that the
Anelectric load assisted
electric is
slightly higher than
turbocharger 350 W
is used that isfunction
whose the valueandproposed by thelocation
mounting
guidelines
are same of asthe European
that Commission
of a formula one [1].
MGU-HTherefore,
unit described
additional electric devices (e.g. steering, lightning, air
earlier. In this work a simple control strategy with two
modes (running separately), one motor and one generator
behaviour, were not considered. In the baseline case the
mode, is used. There is no energy storage system used
electrical power is supplied to the auxiliaries by a conventional
in this simulation. A constant 1]. electrical load (due to the
auxiliaries) of 370 W is maintained in two available modes
(Generator and motor mode). When in generator mode
Figure 6. Comparison between turbine and compressor power during a
the power generated (maximum 370 W) is provided to the Figure 7.2:
NEDC for theExtract from reference [2] showing the
baseline case.
auxiliaries thereby removing the need for the engine to energy content in exhaust gasses over NEDC cycle
generate power from the fuel.
Figure 7.2 taken from the same reference [2] shows the
power of turbine and compressor for the model without the electric assist over a NEDC cycle. This shows
the energy content of the exhaust gases. The author reports 4% to 6% reduction of fuel consumption based
on simulation results on three standard driving cycle. He concludes that the possibility of exploiting ETC to
supply power to the electric loads is higher in extra urban cycles then in urban cycles due to higher engine load.

19
8 Summary
From 2014, formula one has started to see downsized turbocharged engine with somewhat a strong hybridisation.
Understanding the formula one hybrids is interesting as they are different than other hybrid powertrain used in
conventional cars and heavy vehicles. In this report an attempt has been made to analyse the power train
architecture of formula one 2014 and analyse the possibility of its application on a road vehicles by studying its
potential to reduce the fuel consumption and also improve the drive-ability.
A typical layout of the formula one 2014 is presented. Quick response and power is of at-most priority in
formula one. On the other hand, fuel consumption and cost is of at-most priority for road cars. This difference
in requirement reflects the PT architecture. For example, in road HEVs, MGU-K is typically located in between
the engine and the transmission. A clutch is used in between the engine and MGU-K to allow for pure electric
mode and engine stop during idling thereby reducing the fuel consumption. But, in formula one PTs, there is
no clutch which allows for decoupling of the engine from the MGU-K. Although, technically speaking, it can be
argued that the formula one PTs allow electric mode operation by deactivating the cylinders, but it comes with
added losses such as engine inertia and engine air pumping losses and therefore unsuitable for fuel economy
optimisation.
It is shown that Fuel restriction of 100 kg per race and the maximum fuel flow rate at various speed forces
the team to be energy efficient. The kinetic energy that can be recuperated during braking using MGU-K in
formula one driving is high but is limited by the FIA regulations. One of the reason might be to contain the
money spendings of each team and thus have a comparable performance of the car across all the teams.
The implementation of the technology for recovery of vehicles kinetic energy (i.e technology referring to
MGU-K) during braking can be seen both in formula one HEV and road HEV power trains (PTs). The
difference in PTs however, is in the implementation of MGU-H and therefore a detailed analysis about its
energy recovery potential (ERP) is carried out with the help of literature review.
A brief overview about the applicability of various exhaust energy recovery methods for automotive purpose
is presented. Although the ERP of ETC1 method is less when compared to thermal methods (such as rankine
cycle), its compactness, simplicity and its ability to reduce the turbo lag makes it more attractive for formula
one cars and road cars. And also it doesn’t rule out the use of other methods.
It is clear that addition of any system in the exhaust path will increase the back pressure which in turn
influences the gas exchange process inside the cylinder. Although the back pressure deteriorates the engine
power, the amount of power recovered from the exhaust gases by the ETC is higher and thereby improving the
efficiency of the formula one car engine. If the recovered energy from the MGU-H can be directly transferred to
the MGU-K to propel the vehicle in addition to the engine power, then the acceleration performance of the car
increases. Hence, it is a very attractive solution for formula one car. But, similar efficacy and the value for
money of this technology for road cars still needs to be researched. However, from various research papers in
literature it is seen that this solution is cost effective for heavy duty vehicles for both drive-ability and fuel
economy.
It is seen that there is an optimum back pressure and hence a coordinated control of the waste gate,
turbocharger speed and also perhaps a EGR valve is required for reduced fuel consumption and emission for
heavy/light duty gasoline/diesel engine. By studying the published paper from the OEMs such as Caterpillar,
Garret, CPT, Ricardo and Magneti Marelli Powertrain, it can be concluded that 5-10% reduction in fuel
consumption can be obtained from diesel engine. The fuel consumption benefit on a gasoline engine is not clear
as there are very few published papers. The main problem in gasoline engine seems to be the high temperature
of the exhaust gases and hence the back pressure has to be maintained low in order to avoid knocking. But,
there seems to be a smart or a work around to overcome this problem as can be seen in formula one cars which
is not clear yet.
1 The term ETC (Electric turbo compound) is used here for explanation which refers to the MGU-H technology in formula one.

20
References
[1] M. Algrain. “Controlling an electric turbo compound system for exhaust gas energy recovery in a diesel
engine”. In: Electro Information Technology, 2005 IEEE International Conference on. 2005, 6 pp.–6. doi:
10.1109/EIT.2005.1627004.
[2] Ivan Arsie et al. A Comprehensive Powertrain Model to Evaluate the Benefits of Electric Turbo Compound
(ETC) in Reducing CO2 Emissions from Small Diesel Passenger Cars. Apr. 2014. doi: 10.4271/2014-
01-1650. url: http://dx.doi.org/10.4271/2014-01-1650.
[3] Middlemass C. Balis C. and Shahed S. M. “Design & Development of e-Turbo for SUV and Light Truck
Applications”. in 9th Diesel Engine Emissions Reduction (DEER) Workshop, 2003, pp. 1-6.
[4] Alberto Boretti. KERS Braking for 2014 F1 Cars. Sept. 2012. doi: 10.4271/2012- 01- 1802. url:
http://dx.doi.org/10.4271/2012-01-1802.
[5] Rutger Dijkstra et al. Experimental Analysis of Engine Exhaust Waste Energy Recovery Using Power
Turbine Technology for Light Duty Application. Sept. 2012. doi: 10.4271/2012-01-1749. url: http:
//dx.doi.org/10.4271/2012-01-1749.
[6] F. R. Du F. Z. Ji and X. B. Zhang. Research on the characteristics of exhaust gas energy flow for piston
engine. Nov. 2012. doi: 10.1201/b13718-157. url: http://dx.doi.org/10.1201/b13718-157.
[7] Fédération Internationale de l’Automobile. http://www.fia.com/sites/default/files/regulation/
file/1-2014TECHNICALREGULATIONS2014-01-23_0.pdf. Accessed: 2014-06-03.
[8] Raymond Freymann, Wolfgang Strobl, and Andreas Obieglo. “The turbosteamer: A system introducing
the principle of cogeneration in automotive applications”. English. In: MTZ worldwide 69.5 (2008),
pp. 20–27. doi: 10.1007/BF03226909. url: http://dx.doi.org/10.1007/BF03226909.
[9] Andrew Haughton and Andy Dickinson. Development of an Exhaust Driven Turbine-Generator Integrated
Gas Energy Recovery System (TIGERS ). R Apr. 2014. doi: 10 . 4271 / 2014 - 01 - 1873. url: http :
//dx.doi.org/10.4271/2014-01-1873.
[10] Ulrich Hopmann and Marcelo C. Algrain. Diesel Engine Electric Turbo Compound Technology. June 2003.
doi: 10.4271/2003-01-2294. url: http://dx.doi.org/10.4271/2003-01-2294.
[11] D.T. Hountalas, C.O. Katsanos, and V.T. Lamaris. Recovering Energy from the Diesel Engine Exhaust
Using Mechanical and Electrical Turbocompounding. Apr. 2007. doi: 10.4271/2007- 01- 1563. url:
http://dx.doi.org/10.4271/2007-01-1563.
[12] Yamashita Y. Sumida K. Ogita H. Jinnai Y. Ibaraki S. “Development of the ‘hybrid turbo,’ an electrically
assisted turbocharger”. Mitsubishi Heavy Industries, Technical Review Vol. 43 No. 3, Sep.2006.
[13] Fu Jianqin et al. “A study on the prospect of engine exhaust gas energy recovery”. In: Electric Information
and Control Engineering (ICEICE), 2011 International Conference on. Apr. 2011, pp. 1960–1963. doi:
10.1109/ICEICE.2011.5777049.
[14] M. Michon et al. “Modelling and Testing of a Turbo-generator System for Exhaust Gas Energy Recovery”.
In: Vehicle Power and Propulsion Conference, 2007. VPPC 2007. IEEE. 2007, pp. 544–550. doi: 10.
1109/VPPC.2007.4544184.
[15] F. Millo et al. The Potential of Electric Exhaust Gas Turbocharging for HD Diesel Engines. Apr. 2006.
doi: 10.4271/2006-01-0437. url: http://dx.doi.org/10.4271/2006-01-0437.
[16] J. Ringler et al. Rankine Cycle for Waste Heat Recovery of IC Engines. Apr. 2009. doi: 10.4271/2009-
01-0174. url: http://dx.doi.org/10.4271/2009-01-0174.
[17] S. M. Shahed. An Analysis of Assisted Turbocharging with Light Hybrid Powertrain. Apr. 2006. doi:
10.4271/2006-01-0019. url: http://dx.doi.org/10.4271/2006-01-0019.
[18] S. M. Shahed. An Analysis of Assisted Turbocharging with Light Hybrid Powertrain. Apr. 2006. doi:
10.4271/2006-01-0019. url: http://dx.doi.org/10.4271/2006-01-0019.
[19] Nicola Terdich et al. Off-Road Diesel Engine Transient Response Improvement by Electrically Assisted
Turbocharging. Sept. 2011. doi: 10.4271/2011-24-0127. url: http://dx.doi.org/10.4271/2011-24-
0127.
[20] J. Wahlstrom, Lars Eriksson, and L. Nielsen. “EGR-VGT Control and Tuning for Pumping Work
Minimization and Emission Control”. In: Control Systems Technology, IEEE Transactions on 18.4 (2010),
pp. 993–1003. issn: 1063-6536. doi: 10.1109/TCST.2009.2031473.

21
Swedish Hybrid Vehicle Centre
Chalmers University of Technology
Hörsalsvägen 11, level 5
SE-412 96 Göteborg

Phone: +46 (0) 31 772 10 00

www.hybridfordonscentrum.se