R.

Sauersteina
R. Dabrowskib
M. Beckerc
W. Bullmerc

Regulated Two-Stage Turbocharging for

gasoline Engines (2010)

a BorgWarner Turbo Systems

b Hochschule Mannheim, TurboAcademy

c BorgWarner Inc. Engine Systems Group

1 Motivation

On the way to fuel efficient, economic and high-performance engines, turbocharging used in
Diesel engines has also become established with gasoline engines. The demands for gasoline
engines are different though. In the past the utilization of turbocharging to enhance the
performance was subject to compromises in the low end torque, because of the necessity for
larger turbochargers. The customer demand for high-performance engines that are
simultaneously responsive at start-up was accounted for by the ongoing development in
turbocharging and other engine related components, however the limits became increasingly
evident.

Figure 1.1 provides an overview of contemporary turbocharged gasoline engines with respect to
specific power and specific torque at 1500 RPM. Two Diesel engines with VTG and regulated
two-stage turbocharging are also shown.

220
Regulated Two-Stage R2STM
Specific Torque @ 1500 rpm [Nm/l]

200

Diesel
180
R2S

Peak Torque Observed Limit for

Torque Enhanced
160 Diesel Single-Stage
Engines Power Engines
VTG
Turbocharging
140

Peak Power Engines

120

Base Engines
100
60 70 80 90 100 110 120
Specific Power [kW/l]

Figure 1.1: Trade-off on turbocharged gasoline engines between specific power

and specific torque at 1500 RPM

1
From the figure a clustering of the engines into determined characteristic ranges becomes
evident. In the smaller specific values range, preferably cost-effective engines (“Base Engines“)
are to be found that utilize technologies such as port fuel injection, simple mono-scroll turbines
and only one or even none inlet cam phaser. In the higher specific engine power range, a group
of high-power engines (“Peak Power Engines”) can be found that sacrifice low end torque for
high performance targets. The next group can be seen on the left of the image; it features direct
injection and typically one cam phaser to achieve an increased low-end torque at low specific
power (“Peak Torque Engines”). A modern Diesel engine with VTG, shown for the purpose of
comparison, delivers almost equal performance values thus showing the potential of gasoline
engines for higher specific powers.

The next group in the centre of the figure demonstrates an easing of the conflict of goals
between specific low-end torques and specific powers; these engines are equipped with direct
injection, cam phasers on the inlet and outlet valves as well as single-scroll turbines (“Torque
Enhanced Power Engines”). By the combination of the stated technologies a “scavenging”
charge cycle and thus a significant increase in torque at low engine speeds is possible. A further
increase is possible and is design-dependent with 4 and 6 cylinder in-line engines using
separated exhaust gas routing in the exhaust manifold and twin-scroll turbine housings. This
technology enables an improvement in the preconditions for a “scavenging” charge cycle through
avoiding of exhaust-gas pulsation interference of neighbouring cylinders and can significantly
increase the low end torque [1]. The Diesel engine included for comparison with regulated two-
stage turbocharging exhibits a specific low-end torque in the upper range of this group; the
specific power is however limited to a value of approx. 75 kW/l in comparison to the gasoline
engines shown. Further synergy potential by the combination of multi-scroll turbine housings with
variable turbine geometry is possible and the subject of ongoing investigation [2].

Figure 1.1 shows that with the current engines with one-stage turbocharging the specific low-end
torque as well as the specific power is limited to values of about 180 Nm/l (at 75 kW/l) up to 160
Nm/l (at 100 kW/l). Demands for a further increase in the low-end torque, resulting from further
downsizing or an extension of the engine power range to be covered within an engine family, can
be implemented with single-stage turbocharging only to a very limited extend.

The regulated two-stage turbocharging can solve the conflict described above. Concept studies
on different gasoline engines turbocharged in this way have shown that low-end torques
exceeding 200 Nm/l and simultaneous power densities of more than 110 kW/l can be delivered.

2
In the following the design, control strategy and the operational behaviour of this type of 2-stage
turbocharging on a DI gasoline engine are described in detail.

2 Regulated two-stage turbocharging (R2STM) on a DI

gasoline engine

2.1 Thermodynamic aspects of supercharged gasoline engines

In order to realize the specific torque and power objectives of more than 200 Nm/l and 100 kW/l
as described in chapter 1 as well as compliance to future CO2-emission limits with the assistance
of the shift in the load points, gasoline engine turbocharging concepts with more than 20 bar
break mean effective pressure (BMEP) are increasingly becoming the focus of automobile
manufacturers. Single-stage charging systems reach their thermodynamic limits in such
applications.

The turbine mass flow rate or the performance factor of a turbine in unregulated operation
depends mainly on the engine speed nM [3] (Equation 1):

m T⋅ T3 VH ⋅ i ⎡ 1 ⎤ T3 p2S
= ⋅ ⎢1 + ⎥ ⋅ λa ⋅ ⋅ ⋅ nM [Equation 1]
p3 RL ⎣ Lmin ⋅ λ ⎦ T2S p3

At low engine speeds and consequently low turbine mass flow rates, the representation of an
effective torque build-up of a small turbine with low absorption capacity is necessary, as in
accordance with the first turbocharger main equation (Equation 2), a correspondingly higher
pressure before turbine p3t must be built up for the necessary pressure ratio πV:

κL
⎡ ⎛ κ Abg −1
⎞⎤ κL −1
p2t ⎢ m T c p,Abg T3t ⎜ ⎛p ⎞ κ Abg
⎟⎥
πV = = ⎢1 + ⋅ ⋅ ⋅ ηis,V ⋅ ηis,T ⋅ ηm ⋅ ⎜ 1 − ⎜ 4 ⎟ ⎟⎥ [Equation 2]
p1t c p,L T1t ⎜ ⎝ p3t ⎠ ⎟⎥
⎢ mV
⎣ ⎝ ⎠⎦

In principle, gasoline engine combustion helps providing the necessary turbine power due to
stoichiometric combustion resulting in high exhaust temperatures and high turbine mass flow
rate; however, the turbine pressure ratio πT = p3t / p4 and consequently the turbine size remain the
main influential parameters.

3
At rated power small volute housings with low A/R-ratios lead to undesirably high pressure levels
before the turbine thus negatively affecting the gas exchange of the engine. High residual gas
content, retarded ignition angle and the requirement for enrichment are the result and will not
only lead to an increase in fuel consumption, but to a limitation of the power, should the
enrichment be above a significant level. This power limitation is dependent on the width of the
compressor map, which is a limiting factor for single-stage charging systems due to the high
necessary air mass flow rate for the current power objectives. This factor is exacerbated by the
existing high spread in delivery rate with gasoline engines, requiring the use of larger turbines for
the rated power range. The increasing power density currently observed and the associated
increase in turbine size is at the expense of transient engine operating response with single-
stage turbocharging systems [4].

The transient response depends, according to the principle of conversion of angular momentum
for the turbocharger (Equation 3) on the turbocharger speed nATL, but particularly on the available
surplus exhaust gas enthalpy content for the turbine or the system, as well as the moment of
inertia of the rotating parts ΘATL:

dnATL 1
= 2 ⋅ (PT − PV − PR ) [Equation 3]
dt 4π ⋅ Θ ATL ⋅ nATL

Single-stage turbochargers designed for high specific powers have a noticeable delay in
transient operation due to their relatively large polar moment of inertia and the large turbine
absorption capacity. This is disadvantageous and undesired because of the desirable high
acceleration response from low speeds. The problem described becomes particularly evident on
heavy vehicles as well as in extreme conditions such as high altitudes or pulling out on steep
gradients.

Regulated two-stage turbocharging (R2STM) does not only offer the solution for this conflict of
goals but also utilizes the benefits of the dedicated turbocharger size for a certain operating
range. Because of the small design of the high-pressure stage and the compressors connected
in series, R2S systems can realize the highest charging pressures at low engine speeds, so that
the objective of 200 Nm/l from 1500 RPM can be achieved. As a result of the low polar moment
of inertia and its high bank up ability, the high-pressure stage enables a significant increase of
the low-end torque and the dynamic response.

The design of the low-pressure stage is undertaken for the respective target power (> 100 kW/l)
and the additional reserve requirements. Compared to the Diesel engine, the higher variation in
4
delivery rate of the gasoline engine can be more easily considered, and the low-pressure
compressor at the rated power point can be operated at good levels of efficiency. The high flow
coefficient and low bank up behavior of the low-pressure turbine supports scavenging of the
cylinders, which leads to lower residual gas content and a good fuel economy levels also at
higher loads.

A two-stage regulated turbocharger assembly as a system component of the engine is

dependent on an environment, which can realize the implementation of the generated high
charging pressures in break mean effective pressure (BMEP), and that eases, if possible, the
problem areas of supercharged gasoline engines. The knocking problem requires in addition to a
reduced compression ratio and direct injection special measures that reduce the residual gas
content in the cylinder to a minimum. 4 cylinder concepts with conventional cam timing
configuration and 4-in-1 type exhaust manifolds have a system-related disadvantage. The gas
exchange is affected by the outlet pressure pulse of the cylinder next in the firing order. The high
pressure level before the high-pressure turbine causes high residual gas content in the low-end
torque range, retarded ignition point and late BMF (burnt mass fraction) so that the potential of
the higher charging pressure cannot fully be exploited. As a direct consequence, 3 cylinder
concepts with two-stage regulated charging systems are viewed as a useful alternative [5] to the
4 cylinder engine. Here, active scavenging strategies with the aid of cam phasers at the inlet and
outlet valves can be used to get the associated positive effects on low residual gas content in the
cylinder for good low-end torque. But even on 4 cylinder engines, variable valve timing can
significantly reduce the residual gas content and makes a good combination with both single and
2-stage turbochargers [1].

2.2 Single-stage turbocharging – design and limitations

The design and adaptation of a one-stage turbocharger is frequently undertaken with the
demands that the target torque is reached as quickly as possible at a defined rated power, see
Fig. 1.1. Transferred to the compressor map on the one hand means a very steep rate at the
surge line that limits the compressor performance characteristic to low mass flow rates. On the
other hand, the choke line as the limitation for high mass flow rates should be far enough away
from the surge line. With the required rated power, the maximum throughput of the compressor is
defined and according for this size the design-dependent available characteristic map width. With
the use of adequate valve overlaps and an existing positive pressure drop between the inlet and
outlet side, using a “scavenging” charge cycle, the engine operating curve can be shifted,
particularly for low engine speeds to higher mass flow rates, and thus the torque can be
significantly increased [1]. By corresponding adaptation of the turbine, the effective torque range
5
can be shifted to lower engine speeds, and thus the achievable air mass flow variation between
maximum torque and rated power can be increased a little.

On the turbine side, with the series industrialisation of variable turbine geometry (VTG) [6], the
prerequisite for a solution suitable for gasoline engines for increasing the mass flow rate variation
is provided. Studies indicate that this technology facilitates considerable increases in the startup
torque [1].

On the compressor side, measures for extending the characteristic map width as well as for
increasing the achievable pressure ratios were examined, where the focus of the measures
concentrated on enhancing the pressure ratios and should be discussed in more detail.

In order to achieve the necessary charging pressures for supercharged gasoline engines with
one-stage charging, modified technology compared to today’s state-of-the-art is necessary. An
increase of the charging pressure can generally be achieved in three ways [7]:

• Increase of the circumferential speed by increasing the turbocharger speed

• Increasing the impact of the rotor outlet flow through a larger blade exit angle
• Generation of swirl on the rotor inlet by using an inlet guide or outlet guide vanes in the
diffusor
An increase of the boost pressure by increasing the rotor speed causes greater mechanical
stress because of the higher centrifugal force. In particular, the increased stresses on the
compressor impeller can act unfavorably on the lifetime of the compressor impellers, based on
cast aluminum alloys used up to this point. This could be partly counteracted by series
introduction of milled compressor impellers, which can bear higher loads because of the
homogeneous microstructure of the raw parts used (forged circular blanks).

A further increase in the service life is possible by the use of higher-quality materials, such as
titanium alloys, but this enhancement is subject to a more complex manufacturing process and
higher costs. Increased material densities, such as those with Titanium alloys, lead to increased
moments of inertia with an impaired transient response, should the existing geometry be
retained. Furthermore, an increase of the pressure ratio can cause outlet temperatures that
exceed the admissible material temperatures of the materials currently used in the compressor
and the downstream components and require the use of more suitable materials.

Measures with a modified blade exit angle, generation of swirl in the rotor inlet as well as outlet
guide vanes in the diffusor are described and have been evaluated in [7] and should therefore
not be studied further.

6
To conclude, after careful considerations of the above mentioned modifications the one-stage
compressors do not appear to be suitable for providing large air mass variations in combination
with high pressure ratios that are required by highly downsized turbocharges gasoline engines.

2.3 Regulated two-stage turbocharging (R2STM) – design and regulation

The limitations of single-stage turbocharging shown in the previous chapter lead to deliberations
concerning the introduction of regulated two-stage turbocharging, which can significantly
increase the charging pressure over the entire engine speed range. An advantage with this
charging process is that in the two-stage regulated turbocharging range, the respective stage
loading compared to one-stage turbocharging can be reduced.

Hereby, a second charging stage designed for a lower flow rate ensures the required increase in
charging pressure with low engine speeds, hereinafter referred to as a high-pressure
turbocharger. This is situated upstream (compressor side) or downstream (turbine side) of the
main charging stage, so that regulated two-stage turbocharging can be employed at low flow
rates. With this arrangement, the startup torque is represented decisively by the high-pressure
stage, which in addition to the lower space requirement also offers a considerably reduced
moment of inertia for the rotating parts and thus provides the best preconditions for transient
states [7]. So for the low-pressure turbocharger a compressor with a higher maximum flow rate
can be used.

The associated design reserves can be used either for representation of a higher power or for
compliance to extended application reserves for critical applications (operation at high-altitude,
see Chapter 3.4). In figure 2.3.1, the set-up of the turbocharging system and its actuators can be
seen.

The control of the charging group is operating point-dependent controlled using the turbine
bypass, waste-gate and compressor bypass control elements. The turbine bypass determines
the share of the required total turbine power for the turbocharger, which the high-pressure
turbine generates. At the lowest speed and at full load, the turbine bypass is practically closed,
so that the exhaust gas passes both turbocharger stages and the air is compressed in two-
stages on the compressor side. As the engine speed increases, the low-pressure compressor
can increasingly contribute to the required total pressure ratio, until ultimately the high-pressure
turbocharger is deactivated by completely opening the turbine bypass on the exhaust side. At
this point the compressor bypass is opened, and the high pressure compressor is bypassed to
reduce throttling losses.

7
 1 2 3

Turbine Bypass

Compressor High Pressure

Bypass Stage

Low Pressure Wastegate

Stage

Figure 2.3.1: Diagram of two-stage regulated charging

A possible actuator control in the operating characteristic of a gasoline engine is represented

schematically in the following figure 2.3.2.

Boost-
Pressure
Control by
Control Range of
Engine Load

Wastegate
Turbinebypass and
Wastegate
Compressor
Bypass open Naturally
Aspirated
Engine
WOT Line

Wastegate open

Engine Speed

Figure 2.3.2: Control strategy of a regulated two-stage turbocharger system

in the map of a gasoline engine

8
The control strategy of the actuators aims to keep the charge-cycle losses of the engine to a
minimum; the results shown in chapter 3 were measured according to this strategy. From the
illustration it is evident that the high-pressure stage is not required from medium speeds on and
only the low-pressure stage provides the required boost pressure. In order to reduce losses, the
waste-gate is opened below the naturally aspirated full load line, so that boost pressure is
generated above this line only.

The interaction of the actuators is mainly dependent on the selection or pairing of the charging
components and will be examined more closely in the following chapter.

2.4 Design of regulated two-stage turbocharging

During the course of these investigations, the design of a two-stage regulated turbocharging
system for a direct injection gasoline engine for delivery of a specific torque of 200 Nm/l from
1500 RPM and a specific engine power of 100 kW/l was undertaken. A further demand was the
consideration of high-altitude reserves of 1000 m, whereby the torque and power objectives
equal to those at sea level were to be retained.

In the first step, the low-pressure compressor was selected for the required power, taking into
consideration that the compressor at its rated power point is sufficiently far enough away from
the choke line and that it can be operated in the good isentropic compressor efficiency (ηis,V ≈
75%) range (see Fig. 2.4.1). The map of the selected compressor shows sufficient potential to
increase the circumferential speed and accordingly the charging pressure, in order to deliver the
required high-level operation with a reduction in power.

The BMEP plateau of 25 bar requires a high overall compressor pressure ratio of πV,total > 2.7,
which can be served solely in a speed range of n = 3000 – 4500 RPM by the low-pressure
compressor, just like the required pressure ratio in the rated power range. At speeds < 3000
RPM and the associated low flow rates, the low-pressure compressor alone cannot maintain the
required pressure ratio because of the limitation by the surge line. In order to hold the BMEP
level or to enable the necessary torque curve, the high-pressure compressor must increasingly
deliver a share of the required total pressure ratio, whereby the share of the low-pressure
compressor must be considered with a certain safety margin to the statically on the gas stand
determined surge line.

Furthermore, Fig. 2.4.1 shows that the full load operating points in the compressor map are
within the area of moderate efficiencies. This is due to the fact that the compressor design is
primarily intended for one-stage high-pressure usage and not for a two-stage system. The
position of the isolines of constant compressor efficiency can be optimized by adaptation of the
9
compressor geometry (outlet angle, number of blades, etc.), as well as measures on the diffusor
and casing for the two-stage application examined here (see chapter 2.2).

The response of the two-stage turbocharging system with highly transient acceleration processes
is also mainly influenced by the choice of compressor and makes the selection of a suitable
control structure necessary. The low-pressure compressor must be able to support the dynamic
acceleration of the high-pressure compressor at the switchover point without itself having to
surge or cause a dip in the boost pressure. Good controllability requires a certain amount of
overlap of the compressor map as well as coordination of the dynamic response of the
compressors to one another. For a good transient response it is not simply sufficient to only
optimize the high-pressure turbocharger for acceleration. Configuration of the regulated two-
stage turbocharger just for stationary aspects would leave a considerable amount of potential
unused.

The selection of the turbine size with one-stage turbocharger systems is characterized by the
conflict of goals between acceleration or low-end torque response as well as acceptable
pressures from the turbine and the associated consequences for gasoline engine charge cycles,
see chapter 2.1. This conflict is practically solved using two-stage regulated turbocharging. On
these systems, the focus of the turbine side considerations is concentrated mainly on the
selection of suitable turbine pairs. On the one hand, the respective operating range should be
optimally covered by the corresponding stage, and on the other hand, a favorable switchover
point should be guaranteed in order to enable both good stationary and transient operating
conditions. A small low-pressure turbine shifts the switchover point in the direction of low engine
speeds, allows less scope for corresponding high-altitude operation or leads to bad fuel
consumption values in the rated power range. Inversely, by using large low-pressure turbines
that allow the high-pressure turbine to remain operational for longer at full load, so that
unfavorable pressure ratios before the high-pressure turbine can result in higher levels of
residual gas content as well as poorer fuel consumption values in the mid engine speed range.
Considering these aspects, for this design a switchover point in the range from n = 2600 – 2800
RPM has been selected.

10
 4.2

uredC = 560 m/s

3.8
Compressor Pressure Ratio t-t [-]

520
3.4

490
3.0
uredC = 490 m/s 460

0.72
2.6 460 ηisC = 0.76

0.68
0.74 0.75
420
2.2
0.64 0.73
ηisC = 0.75
370
0.70
1.8
420
310 0.68

370
1.4 230
0.60 310

150 230
150
1.0
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22
3
Reduced Compressor Flow Rate [m /s]

Figure 2.4.1: Matching of the regulated two-stage turbocharging system

in the maps of the high- and low-pressure compressor

2.5 The regulated two-stage turbocharging in high-altitude and hot

conditions
The cylinder charge and therefore the power of normally aspirated engines decline because of
the lower air density at high altitudes. The same applies for the operation in warm ambient
temperatures. One big advantage of turbocharged engines is that this reduction in power can be
compensated for to a large degree, as the design reserves of the compressor are in a position to
provide the necessary higher pressure ratio required to deliver nominal charging pressure. The
turbocharger will operate at increased speed, and the operation point will shift in the compressor
map in the direction of larger pressure ratios and volume flows. Up to which height or
temperature the power of the engine can be kept constant before the turbocharger reaches its
speed limit, is a parameter that is part of the matching strategy of the turbocharger to the engine.
Different specifications exist depending on manufacturer and vehicle type.

11
A matching strategy frequently used is to keep the power constant up to a height of 1000 m and
an ambient temperature of 40 °C. The full load curve for such extreme operation in comparison
to the standard conditions for the engine under investigation can be seen in figure 2.5.1 for a
specific power of 85 kW/l.

3,4

uredV = 520 m/s

3,0 490
Compressor Pressure Ratio t-t [-]

Full Load Curve

460 under Extreme 0.68

2,6 Conditions

420

2,2

370
0.70 0.74
0.65
0.68 ηisV = 0.76 0.7 Full Load Curve
1,8 0.55
(Standard
310 0.65
Conditions)
0.6

1,4 230
0.6

150

1,0
0,00 0,02 0,04 0,06 0,08 0,10 0,12 0,14 0,16 0,18
Reduced Compressor Flow Rate [m³/s]

Figure 2.5.1: Full load curve of a single-stage turbocharger in the compressor map,
standard conditions and extreme conditions

It is clearly evident that the turbocharger reaches its operating limits under extreme conditions.
On the one hand the surge limit is reached, on the other hand the rated speed for the flow rate
limit is achieved. The last operating point is also on the speed limit, which presents an operating
risk. The calculated application or safety reserves in the basis design are no longer sufficient
under these operating conditions. A design of this type would not be practical for series
applications without a clearly power limitation in extreme conditions. A larger charger would
remedy the situation and would include all the known disadvantages of a comparable delayed
response and operation within the area of moderate efficiencies as well suffering to provide
enough boost pressure at low engine speeds. A further result of the operation under extreme
conditions can be seen in the following figure 2.5.2.
12
 1,4
Turbine Flow Parameter [kg/s K^0.5/ bar]

1,2

Full Load Curve uredT = 430 m/s

390
(Standard Conditions)
340
1,0

280

Full Load Curve under

0,8 Extreme Conditions
210
(From 1000-1600 rpm 14%
of Target Torque is Missing)
0,6 140

0,4
1,0 1,2 1,4 1,6 1,8 2,0 2,2 2,4 2,6 2,8 3,0 3,2 3,4
Turbine Pressure Ratio t-s [-]

Figure 2.5.2: Full load curve of a single-stage turbocharger in the turbine map,
standard conditions and extreme conditions

Here both the engine characteristics for standard and extreme design conditions in the turbine
map are shown. It is evident that the flow rate curve of the selected turbine under extreme
conditions in the speed range from 1000 – 1600 RPM cannot be achieved, i.e. the turbine
appears to be “too large” and the required turbine power cannot be achieved. This leads to a
reduction of the attainable engine torque of about 14 % in the described speed range, where the
additional problem of rated power limitation is associated with a tangibly lower startup torque.

In comparison, the regulated two-stage R2S system performs decisively better, as can be seen
in figure 2.5.3. As the operating range of the high-pressure stage (characteristic on the left of the
graphic) is not fully exploited in standard operation, high-altitude operation or an increased
ambient temperature only has a minor effect on the operation of the turbocharger. This means
that the steady state response of the engine does not measurably worsen.

Also in the characteristic of the low-pressure stage, the extreme operation conditions are shown
to be uncritical. The operating points at higher engine speed are shifted to higher turbocharger

13
speeds analog to a single-stage design. The comparably larger turbocharger, however, allows
considerably higher flow rates, so that even when operating under so-called extreme operating
conditions, there are sufficient application and safety reserves available, and the operating points
are still within the range of the operating limits.

4,2
uredC = 560 m/s
High Pressure Stage, Standard Conditions, 100kW/l
3,8 High Pressure Stage, Extreme Conditions, 100kW/l
Low Pressure Stage, Standard Conditions, 100kW/l
Low Pressure Stage, Extreme Conditions, 100kW/l
Compressor Pressure Ratio t-t [-]

520
3,4 Low Pressure Stage 110 kW/l

490
3,0
uredC = 490 m/s 460

0.72
2,6 460 ηisC = 0.76

0.68
0.74 0.75
420
2,2
0.64 0.73
ηisC = 0.75
370
0.70
1,8
420
310 0.68

370
1,4 230
0.60 310

150 230
150
1,0
0,00 0,02 0,04 0,06 0,08 0,10 0,12 0,14 0,16 0,18 0,20 0,22
3
Reduced Compressor Flow Rate [m /s]

Figure 2.5.3: Full load curve for a R2S system in the compressor map,
standard conditions, extreme conditions and higher rated power

The design shown in figure 2.5.3 is well suited for use in practice and the rated power of the
engine would be available at an ambient temperature of 40 °C up to a height of 1000 m. With
regard to fuel consumption, a moderate rise can occur, as the extension of the regulated two-
stage operation leads to higher engine speeds and corresponding higher afflux of the smaller
sized high-pressure turbine.

Should the design reserves not be used for high-altitude and hot conditions, they can be used
alternatively to deliver a higher engine power. The additional characteristic curve (dotted line) in
figure 2.5.3 indicates the potential for a matching to 110 kW/l for a high power variant.

14
2.6 Design and function
The 2-stage turbocharging system can be designed under different priorities. In this case the
emphasis was put on a compact design that can be easily integrated in the typical engine
compartment of a B-Class vehicle. This compact design incorporates the arrangement of the
high-pressure turbocharger directly on the connection flange of the exhaust manifold with the
downstream low-pressure turbocharger. In figure 2.6.1, an overall view of the turbocharger with
the schematic representation of the gas ducting as well as the control elements is shown.

Wastegate
Low Pressure Stage
Actuator
Fresh Air Supply
Exhaust
System

Compressor High
Bypass
Pressure

Regulated Two Stage

Turbine
Stage Mode
Bypass
Exhaust Gas
from Engine

Single Stage Mode

To Charge Air Cooler

Figure 2.6.1: Overall view of the regulated two-stage turbocharger system

Using this figure, the gas ducting with regulated two-stage turbocharging is indicated, i.e. the
turbine and compressor bypass as well as the waste-gate are closed, and the operating medium
15
is compressed or expanded in two stages. Because of the dynamically active gas forces, the
actuator of the turbine bypass is dimensioned to be larger than the other two actuators. The
optimised design and dimensioning of the actuator guarantee good controllability of the turbine
power in order to also safely enable transient acceleration conditions. With respect to the design
and the kinematics of the turbine bypass, the knowledge gained in the Diesel area was used to
assure the necessary functions. If the high-pressure stage is not required, the turbine and
compressor bypass are opened, and the low-pressure stage takes over the charging pressure
generation.

On the compressor end, an actively switched compressor bypass was used for the investigation
presented here to examine further degrees of freedom. Alternatively, a self-actuating compressor
bypass used as standard in the Diesel area can be used. This enables a simplification in the
design of the charging air circuit with practically identical functional properties and leads to easier
package conditions. The following figure 2.6.2 shows a different view of the turbocharger.

Turbine
Bypass

Wastegate

Figure 2.6.2: Regulated two-stage turbocharger system, view of the low pressure stage

In this figure, the compact construction of the turbocharger assembly can be seen. Due to the
arrangement of both turbochargers above one another, only a slight additional amount of fitting
depth compared to conventional turbochargers is required, and it is thus suitable for longitudinal
and transverse engine installation.

16
2.7 System optimization for regulated two-stage operation
In chapter 2.4 the influencing parameters for the adaptation of an R2S system to an engine have
already been discussed, and the position of the full load curve of the engine in the compressor
performance characteristics has been shown. It becomes clear that a compressor designed and
optimized for one-stage operation cannot optimally meet the demands when used in a R2S
system. As the high-pressure turbocharger only contributes at medium to high loads and up to
moderate engine speeds to the charging pressure generation, it is only used in the lower left
area of the compressor map under steady state conditions, that is to say with low pressure ratios
and flow rates. It is obvious that to optimize the high-pressure compressor, the area of maximum
efficiency is to be shifted to lower pressure ratios and the efficiency in the lower characteristic
map region is to be increased in total, as can be seen in the left characteristic in figure 2.7.2.

Characteristic for the operation of the low-pressure stage in the R2S system is that a wide
variation in the pressure ratio is required. As the low-pressure stage has to provide the charging
pressure for the target torque up until rated output without the support of the high-pressure stage
from moderate speeds, the operating points are shifted to higher pressure ratios in comparison
to the one-stage design. In consequence, the characteristic for a R2S application should enable
high pressure ratios and exhibit high degrees of efficiency in the upper characteristic range. This
is illustrated in the right section of figure 2.7.1.

Typical Operation Range

of LP-Stage R2S™
uredC = 560 m/s u = 560 m/s Πmax
Total Pressure Ratio t-t [-]

0.68 redC
Applicaion at Rated Power 0.68

520 520

490 0.7 490 0.7

0.72 0.72
460 460

0.65
η isC = 0.73
0.65
η isC = 0.73
ηmax
420 420

370
ηmax 370
0.6 0.6

η↑
0.55 0.55
310 310

230
Typical Operation range 230

150 0.5 of HP-stage R2S™ application 150 0.5

Volumetric Flow Rate [m^3/s] Volumetric Flow Rate [m^3/s]

Figure 2.7.1: Compressor optimisation for the R2S system

17
The characteristic of a HP-compressor optimized to these demands in comparison to a
compressor for one-stage applications is shown in figure 2.7.2.

Standard Compressor Wheel

Total Pressure Ratio t-t [-]

R2S™ Optimized High Pressure Compressor

Comp. Efficiency t-t [-]

Δη = +6 %- points

Standard Compressor Wheel

R2S™ Optimized High Pressure Compressor

Volumetric Flow Rate [m3/s] Volumetric Flow Rate [m3/s]

Figure 2.7.2: Map of a compressor optimized for operation as a high-pressure stage in the R2S
system

It is obvious the surge limit is shifted moderately towards higher flow rates and the map is a little
narrower, what is not considered as a drawback for this application. However, the efficiency is
increased and retained at a high level in a wide range, especially in the area of small pressure
ratios which is most important for the high pressure turbo. This will be immediately recognized in
a better transient behaviour at low engine speeds, is beneficial for low pumping losses and
therefore leads to reduced fuel consumption.

3 Test engine and measurement results

3.1 Test engine

The tests were performed on a modified gasoline engine with direct injection and camshaft
phase adjusters on the inlet and outlet sides. Table 1 gives an overview of the most important
engine characteristics.

18
 Engine Type - Inline-4-cylinder
Firing Order - 1-3-4-2
Combustion System - Gasoline Direct Injection
DOHC, Cam Phaser on Intake- and Outlet
Valvetrain -
Camshaft
-1
Spec. Power kW/l @ min 100 @ 5500
max. Mean Effective -1
bar @ min 25 @ 1500-4500
Pressure BMEP
Low-Pressure Stage: K04
Turbocharging System -
High-Pressure Stage: KP35

Table 1: Technical data of the test engine

The adaptation of the engine-specific setting parameters on the regulated two-stage

turbocharging occurs via access to the required setting variables of the control device. The
overall system of engine and turbocharger is operated under defined adjusting conditions in
order to guarantee reproducible results: The following test boundary conditions were defined:

• Break mean effective pressure (BMEP): Target is 25 bar, to be achieved as early as

possible

• Air ratio: λ = 1 (as far as possible)

• Maximum exhaust gas temperature: T3 = 950 °C (enrichment if required)

• Spark timing: For efficiency-optimum for maximum BMEP or knock limited

• Minimum air ratio: λ = 0.75 (power reduction if necessary)

• In the case valve timing variations: control of the air ratio by lambda sensor upstream of
catalytic converter

• Latest 50% BMF (burnt mass fraction) point: 35°KW n.OT

3.2 The regulated two-stage turbocharging used at full load

In a first step, a comparison is made between a single-stage and regulated two-stage
turbocharger set-up. Essential components of the engine (compression ratio, valve timing and
duration, exhaust manifold, exhaust system) are identical in order to facilitate a direct
comparison of the turbocharging system. The cam phasers have been set for a minimum valve
overlap. A comparison with the same power target values was not performed for the reasons
specified in chapters 2 and 3.3. Selected results are represented in figure 3.2.1.

19
 28
25
22
pme [bar]

19
16 1-stage, Mono Scroll Turbine
Regulated Two-Stage (R2S)
13 Potential with opt. Combustion
10
120
Spec. Power [kW/l]

100
80
60
40
20
0

360
340
BSFC [g/kWh]

320
300
280
260
240
Abs. Manifold Pressure [mbar]

3000

2600

2200

1800

1400

1000
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500

Engine Speed [min-1]

Figure 3.2.1: Comparison between one-stage and regulated two-stage turbocharging

at full load

20
It can be seen that with the R2S system a significantly higher BMEP is achieved. The dip in the
progression between 1400 RPM and 1800 RPM results from an NVH related retard of the
combustion. Without this NVH limitation, it can be assumed from the torque gradient between
1000 RPM and 1200 RPM that the target BMEP would be achieved at a speed of about 1400
RPM. From a speed of 2000 RPM, a BMEP level of 25 bar has been achieved, whereby a further
increase is not limited by the turbocharging, but rather in the current case it is limited by the base
engine. Together with the desired increase of BMEP, there is an associated simultaneous
increase of the design power from 85 kW/l to 100 kW/l. This becomes apparent in a moderate
increase of the specific fuel consumption that results from the necessary ignition timing retard
with a higher BMEP, and which follows the known sequence in the engine characteristic. It can
be assumed that with an extensive adaptation of the engine (charge cycle components,
combustion process) to the increased charging rate, a reduction of the specific fuel consumption
can be achieved. The observation of the absolute intake manifold pressure clearly shows the
superiority of the R2S system over one-stage basis charging. From a speed of 1000 RPM, an
absolute intake manifold pressure of about 1600 mbar can be delivered increasing up to 1400
RPM to a peak value of about 2750 mbar and thus offering the best preconditions for a high low-
end torque. Together with the progression of BMEP it is easily recognizable that for effective
implementation of higher charging pressures in high BMEP, an optimization of the design of the
overall engine system with turbocharging is necessary. Between a speed of 1400 to 2500 RPM,
a practically constant absolute intake manifold pressure of about 2800 mbar is delivered. The
further progression becomes clear in the observation of the stage pressure ratios from high and
low-pressure compressors in figure 3.2.2.

21
 3.0

2.8
Compressor Pressure Ratios πC [-]

2.6

2.4

2.2

2.0 πC - LP-stage
πC - HP-stage
1.8 πC - entire stage

1.6

1.4

1.2

1.0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500
Engine Speed [min-1]

Figure 3.2.2: Compressor stage and total pressure ratio of the regulated two-stage turbocharging
system at full load operation

From a speed of 3000 RPM on, the low-pressure compressor alone can provide the necessary
charging pressure, and the downstream high-pressure compressor can be bypassed. Below this
speed, the high-pressure compressor must contribute a maximum pressure ratio of 1.8 in this
configuration. At speeds below 1600 RPM, the share provided by the high-pressure compressor
prevails significantly, whereby the support particularly at low speeds is clearly evident. This
progression shows that with a conventional compressor used in a high-pressure stage only a low
share of the entire compressor characteristic available is utilized (see chapter 2.7). Furthermore,
it becomes evident that the high-pressure compressor can provide a visibly higher contribution,
should the design require higher target values.

The progression of the turbine stages and the overall expansion ratio is shown in figure 3.2.3.

22
 3.0

2.8

2.6
Turbine Pressure Ratios πT [-]

2.4

2.2

2.0 πT - LP-stage
πT - HP-stage
1.8 πT - entire stage

1.6

1.4

1.2

1.0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500
Engine Speed [min-1]

Figure 3.2.3: Turbine stage and total pressure ratio of the regulated two-stage turbocharging
system at full load operation

The progression of the turbine expansion ratio shows a similar response to the compressor side.
With regulated two-stage charging operation, the high-pressure turbine contributes a higher
share of the necessary expansion ratio up to about 1800 RPM, until the flow rate matches the
characteristic of the larger low-pressure turbine. On the turbine end, it must be considered that
by regulating the power of the high-pressure turbine by opening the turbine bypass in stages with
rising speeds, a respective mass flow bypasses the high-pressure turbine and can provide direct
expansion assistance on the low-pressure turbine. From a speed of 3000 RPM, the expansion
occurs using a single-stage via the low-pressure turbine. Because of the associated reduction on
the turbine end pressure build-up behaviour, the exhaust gas pressure is reduced significantly in
comparison to the regulated two-stage charging operation, which becomes visible from the
progression of the overall expansion ratio. The lower share of the high-pressure turbine on the
overall pressure ratio at medium to high engine speeds results from the structural-dependent
arrangement of the high-pressure turbine casing and turbine bypass, where gas flows through
with an opened turbine bypass, before the exhaust reaches the low-pressure turbine (also refer
to chapter 2.6). Because of this arrangement, the rotating parts of the high-pressure charger
23
achieve a basic speed level even though the bypass is opened, which on the one hand can be
beneficial with quick changes in load; however, on the other hand it supports the tribological
function of the bearing system. Here too it becomes evident that with a maximum expansion
figure of 1.8, the characteristic map width of the high-pressure turbine is only slightly affected. In
the context of the available series and contour stages, this can be considered for an optimization
with the selection of the turbine/compressor pairing.

3.3 Regulated two-stage turbocharging at full load with timing adjustment

In the next step, the interaction of regulated two-stage turbocharging with variable timing using
camshaft phase adjusters is examined. For better understanding, in the following figure 3.3.1 the
valve events with the resulting overlap areas are shown.

Shift Range Ex.Camshaft Shift Range In. Camshaft

Valve Lift [mm]

90 180 270 360 450 540 630

max. Valve Overlap Crank Angle [°]

Figure 3.3.1: Valve events with minimum and maximum valve overlap

It can be seen that with this timing configuration the resulting valve overlap areas can be varied
from high overlap to minimum overlap (dashed line). The associated effects on BMEP and fuel
consumption are shown in the following figure 3.3.2.

24
The use of camshaft phase adjusters for generating maximum valve overlaps has different
effects: At low speeds and maximum overlap, a significant reduction of BMEP is evident; from a
speed of about 1100 RPM, BMEP can be increased by up to 2 bar.

This progression is surprising insofar as with one-stage turbocharging, significantly higher BMEP
increases are known with the use of phase adjusters [1]. A more detailed analysis not displayed
here has shown that starting from operation with minimal valve overlap with ever increasing
overlap, a “local minimum” of BMEP is achieved first, before it again rises with an increasing
overlap. This leads to the conclusion that an overall system optimization can be reached by
adjustment of valve timing and duration, see also [1], [5].
BMEP [bar]

min. Valve Overlap

Δ = 4 bar max. Valve Overlap

Δ = 20 g/kWh
BSFC [g/kWh]

800 1000 1200 1400 1600 1800 2000 2200

Engine Speed [min-1]

Figure 3.3.2: BMEP and specific fuel consumption with minimum and maximum valve overlap of
the regulated two-stage turbocharger system at full load

A similar progression is evident with the specific fuel consumption, with the exception of a
reduction at 1000 RPM, that is otherwise increased by up to 30 g/kWh. An analysis of the cause
for this behaviour is made possible by observing figure 3.3.3.

25
The progression of the BMEP of the engine charge cycle clearly shows disadvantages by
operation with maximum valve overlap of up to 0.8 bar. The increased charge cycle leads to a
shift of the load point with corresponding effects on BMEP and fuel consumption. This behaviour
is enhanced by the progression of the pressure drop across the engine (pressure difference from
intake manifold up to the turbine inlet).

min. Valve Overlap

max. Valve Overlap
PMEP [bar]

Δ = 0,4 bar
Δp over Engine [mbar]

Δ = 100 mbar
Ignition Advance [°CA]

Δ = 4 °CA

800 1000 1200 1400 1600 1800 2000 2200

Engine Speed [min-1]

Figure 3.3.3: Pumping losses (PMEP), pressure gradient across the engine and spark advance
with minimum and maximum valve overlap of the regulated two-stage
turbocharging system at full load

At maximum valve overlap, a constant decrease in the direction of negative values is also
recognizable here. The ignition advance shown indicates a large ignition advance angle with the
26
exception of 1200 RPM, indicating more favourable thermodynamic conditions at the ignition
point. Clearly – irrespective of the reduced pressure difference – the opening cross-sections at
maximum valve overlap enable scavenging of the residual gas and thus a reduction of the
charge temperature.

In conclusion, by using cam phasers a relatively moderate increase of BMEP can be achieved at
lower engine speeds. Obviously, the regulated two-stage turbocharging reacts less to these
valve timing variabilities than a one-stage system.

4 Further potential of regulated two-stage turbocharging

In chapter 2.6 it was already shown that the design reserves can be used for increasing the
engine power further instead of using them to maintain power and low-end torque under extreme
conditions. In figure 4.1 possible power variations with the engine displacement as a parameter
are shown.

Figure 4.1: Extension of the power range as a function of engine displacement

and specific power

27
It becomes evident that an impressive power variation of gasoline engines is possible by
inventing regulated two-stage turbocharging. On the basis of a specific power of 85 kW/l the
power can be extended to 110 kW/l and even a further increase up to 120 kW/l is not limited by
the turbocharger system.

Two-stage turbocharger systems are the basis for a high degree of downsizing. The system
consists of standard components so that a wide range of charger sizes is available. With a
single-stage turbocharger representing the basis an attractive power range within an engine
family can be realized thus avoiding higher cylinder numbers and, to a certain extend, larger
displacements.

Because of the beneficial properties of regulated two-stage turbochargers applications can be

identified that offer further customer benefits:

• Combination with high and/or low-pressure EGR to reduce emissions and exhaust
temperatures and a reduction in material requirements
• Combination with Miller/Atkinson process (reduction of effective compression ratio)
• Combination with alternative knock-resistant fuels (example: Compressed Natural Gas)
for realization of highly-efficient engines [8]

5 Summary
The investigations undertaken show that in the trade off between low-end torque and achievable
rated power to realize higher degrees of downsizing in tubocharged gasoline engines, one-stage
charging systems reach their limits. An extension of the performance beyond a certain degree
can only be achieved by the use of regulated two-stage turbocharging.

The results achieved with an R2S turbocharging system of this type can be summarized as
follows:

• Even with a rated power of 100 kW/l high intake pressures and BMEP can be reached
even at lowest engine speeds (absolute intake pressure of 2,8 bar more than realistic at
1400 rpm)
• Power loss and disadvantages in low-end torque can be avoided under extreme
conditions (operation in height and hot ambient conditions)
• With reduced reserves for extreme conditions there is potential to increase the specific
power to 110 kW/l and more
• With an overall system optimization (adaptation of the basie engine, valve timing, cam
phasers with adequate operating range, turbocharger system) a further increase in the
target values, especially fuel consumption, is possible
28
Furthermore, the regulated two-stage turbocharger features the following properties when used
with gasoline engines:

• Significantly higher pressure ratios possible compared to supercharging devices

• Good NVH behaviour
• Suitability for low pressure EGR (resistance to contamination has been proven with
Diesel engines)
• Use of optimized but standard components possible to reduce expenses associated with
the development and to improve reliability
• Vibration and wear free engagement of the high-pressure stage due to gas side coupling

To get the best performance out of regulated-two-stage charging systems a careful optimization
of the individual components is required. The demands on the tubochargers in such a charging
group clearly exceed the demands on the standard component. By participating from the
continuous generic development on regulated-two-stage charging systems for Diesel engines the
thermodynamic and mechanical properties could be enhanced for gasoline applications as well.

Regulated two-stage turbocharging is a further contribution from BorgWarner for the realization
of future high-performance, low-emission and especially low fuel consumption gasoline engines.

29