Energy efficient oil supply for clutch actuation in planetary transmissions

(1)

Energy efficient oil supply for clutch actuation in

planetary transmissions

Martin Persson Fredrik Wallin

Division of Fluid and Mechatronic Systems

Master Thesis

Department of Management and Engineering LIU-IEI-TEK-A- -15/02237- -SE

(2)

Energy efficient oil supply for clutch actuation in

planetary transmissions

Master Thesis in Fluid Power

Department of Management and Engineering Division of Fluid and Mechatronic Systems

Linköping University by

Martin Persson Fredrik Wallin

LIU-IEI-TEK-A- -15/02237- -SE

Supervisors: Viktor Larsson

IEI, Linköping University Pierre Thorén

Volvo CE

Examiner: Liselott Ericson

IEI, Linköping University

(3)

Abstract

The improvement potential of the oil supply system used in Volvo CE’s planetary transmission PT9 is studied in this thesis. The main focus was to improve the energy efficiency of the system while at the same time satisfying all of the system’s functional demands. High energy efficiency is important for the environment, Volvo CE and Volvo’s customers.

Different concepts regarding how to design the system were developed and sim-ulated so as to evaluate their energy efficiency and potential of meeting the various demands of the system. An important aspect to consider when designing an energy efficient supply system is to select proper sizes of the components in the system. If the system is oversized the energy efficiency will not be as good as the system’s po-tential efficiency. On the other hand, if the system is undersized its performance will also suffer. The problem of finding suitable sized component was thus also studied in this thesis. Two methods were developed to guide the engineer in the work of selecting proper size of components. The methods provide a structured approach to dimensioning the system, to ensure that it meets its functional demands and giving high energy efficiency.

(4)

Acknowledgments

This Master’s Thesis was performed for Volvo Construction Equipment at the de-partment Driveline Systems in Eskilstuna, Sweden. The work was supervised by the Division of Fluid and Mechatronic Systems (Flumes) at Linköping University.

We would like to express our gratitude to Lena Brattberg Norman at Volvo CE for the opportunity to conduct our work at Volvo’s plant in Eskilstuna. Thanks to Pierre Thorén at Volvo for being an excellent supervisor and a worthy opponent in Ave Caesar. Thanks to our supervisor Viktor Larsson and our examiner Liselott Ericson at Linköping University for supporting us during this thesis work. We would also like to thank all employees at Driveline Systems for taking the time to answer questions and to provide us with valuable information and inputs during our work.

Linköping, June, 2015 Martin Persson Fredrik Wallin

(5)

4 Concept Development - Phase 1 22 4.1 Concept generation . . . 22 4.1.1 Concept C1 . . . 23 4.1.2 Concept C2 . . . 24 4.1.3 Concept C3 . . . 25 4.1.4 Concept C4 . . . 26 4.1.5 Concept C5 . . . 27 4.1.6 Concept C6 . . . 28 4.1.7 Concept C7 . . . 29 4.1.8 Concept C8 . . . 30 4.1.9 Concept C9 . . . 31 4.2 Concept evaluation . . . 32 4.3 Concept elimination . . . 34 iv

(6)

5 Concept Development - Phase 2 35 5.1 Design limitations . . . 35 5.2 Modeling . . . 35 5.3 Simulation . . . 37 5.4 Concept evaluation . . . 41 5.5 Concept elimination . . . 43

6 Concept Development - Phase 3 45 6.1 Design decisions . . . 45 6.2 Detailed modeling . . . 45 6.2.1 Pump . . . 45 6.2.2 Accumulator . . . 46 6.2.3 Controller . . . 48 6.3 Detailed simulation . . . 50 7 Sizing tool 52 8 Methods for dimensioning the pumps and accumulator 55 8.1 Selection of precharge pressure . . . 55

8.2 Sizing using Sizing tool . . . 56

8.3 Sizing using Simulation . . . 57

9 Results 58 9.1 Transmission data . . . 58

9.2 Selecting precharge pressure . . . 59

9.3 Example using Sizing Tool . . . 59

9.4 Example of sizing using simulation . . . 63

9.5 Energy efficiency . . . 66 10 Discussion 67 10.1 Concepts . . . 67 10.2 Modeling . . . 67 10.3 Simulation . . . 67 10.4 Accumulator . . . 68 10.5 Pump . . . 68 10.6 Torque converter . . . 68 10.7 Results . . . 69 11 Conclusions 70 12 Future work 71 Bibliography 72

A Accumulator sizing - Ideal gas 74

(7)

Contents 1

List of Figures

1.1 Visualization of the method. . . 5

2.1 A planetary gear including three planet gears. . . 8

2.2 Gear 1 active in bicycle transmission. . . 9

2.3 Bicycle transmission. . . 10

2.4 Overview of the clutch. . . 11

2.5 A person on roller skates moving up a slope. . . 12

2.6 Hydropneumatic accumulators with bladder, piston and diaphragm. . 14

2.7 Spring loaded mechanical accumulator. . . 14

2.8 Definition of an accumulator’s thermal time constant [12]. . . 15

3.1 Picture of Volvo A40G FS. Volvo Media Library. . . 18

3.2 Simplified hydraulic drawing of the clutch actuation system. . . 19

3.3 Overview of the pump used in PT9. . . 21

4.1 Hydraulic diagram over concept C1. . . 23

5.1 Amesim model including the supply system and one of the shifting elements. . . 36

5.2 The engine speed and gears used during the drive cycle used to opti-mize the sizes of the pumps. . . 38

5.3 Example of supplied flow and demanded flow to the LPS. . . 38

5.4 Engine speed and active gear used during the drive cycle to evaluate the energy consumption of the concepts. . . 40

5.5 Summation of the pressure drops occurring if two shifting elements are filled simultaneously at engine idling speed. . . 41

6.1 An external gear pump with unequally wide planet gears. . . 46

6.2 Pump model valid for pump displacements DP 1< DP 2. . . 47

6.3 Gas temperature and pressure graph from ASP 5.1. . . 47

6.4 Example of how the demanded flow and supplied flow to the HPS might look. . . 50

6.5 Contour graph of the energy consumption for concept C2. In the figure valid selections of pump sizes are also indicated. . . 51

(8)

Contents 2

7.2 An example of how the output of the tool can look. The upper graph shows the proportion of the total number of gear changes that are possible to perform without the need for supplementary flow. To be able to perform the rest of the gear changes it is necessary to have a supplementary flow. When an accumulator is used for this purpose, the required working volume is shown in the lower graph. . . 54 8.1 Flow chart for the two methods proposed for sizing the pumps and

the accumulator. . . 55 9.1 To the left is a map showing which shifting elements are actuated for

the respective gear. To the right is a map showing which gear changes are possible to make. If the box is gray for the concerned gear change it is allowed to be made. . . 59 9.2 Graphs generated for the example using Sizing Tool. . . 61 9.3 Results from simulation of example system. The upper graphs show

the pressure in the HPS and the lower graph shows the flow to the LPS during the sizing drive cycle. . . 61 9.4 Detailed view of the flow to the LPS during the period when the

demanded flow is higher than the supplied flow. . . 62 9.5 Resulting pressure drop in the HPS and flow to the LPS for two

sim-ulations with different sizes of pump. . . 64 9.6 Resulting pressure drop in the HPS and flow to the LPS for two

sim-ulations with different accumulator sizes. . . 65 B.1 The Simulink controller used with concept C2 in Phase 3. . . 75

(9)

Contents 3

List of Tables

4.1 Pugh matrix from the concept evaluation in Phase 1. . . 33 5.1 Optimal pump displacement in cm3/rev. Pump P1 and P2 are the

sizes of the respective pumps if they are allowed to be of unequal size. The column Equal is the size of the pumps when they are constrained to be of equal size. Note that these values are valid for an ideal pump. 39 5.2 Energy consumption expressed as a percentage of the consumption of

the reference system with unequal pump sizes. The reference system has the same layout as the system used in PT9 but with resized pumps. A low value indicates high efficiency. . . 39 5.3 Cost evaluation for the concepts. The score for the concept with equal

pump sizes is shown in the parentheses. If the score applies for both equal and unequal pump sizes only one value is shown. . . 42 5.4 Pugh matrix from the concept evaluation in Phase 2. The score for

the concept with equal pump sizes is shown in parentheses. If the score applies for both equal and unequal pump sizes only one value is shown. . . 43 6.1 Parameters for the accumulator used in Phase 3. . . 48 9.1 The volume and targeted duration of the filling phase for the respective

shifting elements. . . 58 9.2 Parameters used in example calculation of the minimum pump size. . 60 9.3 Sizes of the pumps and accumulator for the simulations conducted on

the example transmission. . . 63 9.4 Mean power consumption for the two simulated systems. . . 66

(10)

Chapter 1

Introduction

1.1 Background

Gear changes in Volvo’s articulated haulers are performed with a gear change method called Powershift. In order to perform a controlled gear change, the transmission is dependent on a reliable oil supply. An insufficient supply of oil can affect the performance of the gear changes. The driving experience would hence be negatively affected, which would harm Volvo’s reputation. A poorly executed gear change can also increase wear of the driveline. During the development of an oil supply system for a transmission the energy efficiency is an important element. Increased energy efficiency will reduce the fuel consumption, thus reducing the emission of greenhouse gases to the environment.

An articulated hauler is used to transport materials in areas with rough terrain where trucks are not a suitable option. Volvo’s haulers have a planetary transmission to transfer power from the engine to the wheels at various gear ratios. The transmis-sion consists of multiple planetary gears connected to each other. The ratio between the input shaft and the output shaft of the transmission is controlled by locking axles and gears between the different planetary gears. The locking is performed by wet clutches that are actuated by hydraulic pistons.

The hauler uses powershift in order to always have power at the wheels, even when changing gear. The powershift is obtained by releasing one clutch as another clutch is engaged. A quick but controlled gear change is desirable to achieve high performance in terms of comfort for the driver, hauler drivability and transmission durability. This in turns sets high demands on the transmission’s clutch actuation system including the pump, control valves and pistons. In this study the focus is on ensuring a reliable hydraulic flow to the pistons used for clutch actuation.

The transmission considered in this study is a Volvo PT9, a nine speed planetary transmission. The clutch actuation system has a variable pressure level and is fed by a fixed displacement external gear pump, whose speed is equal to the engine speed. This system has a poor energy efficiency when the pump speed is high, resulting in a high flow, when the demanded flow is low. The excess flow from the pump leads to energy losses. The studied transmission has a high excess flow when no gear change is performed and the engine speed is high. It is thus desirable to use a small pump in order to reduce the energy losses, but the pump still has to fulfill the flow demand during the gear change. These two contradictory requirements make it difficult to

(11)

1.2 Purpose 5

dimension the components of the supply system.

Continuously improvement of the control strategy used for the gear changes has shown the importance of a stable main pressure, setting higher demands on a reliable oil supply. This may make it necessary to upgrade the clutch actuation system. In Volvo’s transmission it has been found that the flow from the pump is the limiting factor for further improvement of the gear change performance. Insufficient flow from the pump results in a pressure drop in the actuation system that may cause a slower gear change and unwanted slip in the clutches.

The pump is located inside the transmission due to space limitations. This made it necessary for Volvo to design an application specific pump and it is thus not an option to replace the current external gear pump with another pump type.

1.2 Purpose

The purpose of this Master’s thesis is to develop an energy efficient oil supply system for the clutch actuation in a transmission containing wet clutches. In addition, a dimensioning method for the new system is to be proposed. The method will dimension the components of the system to ensure good performance in all of its working conditions, and at the same time be energy efficient. Volvo CE has expressed a desire for a system including a hydraulic accumulator to be investigated. The developed system should also be cost effective and fulfill all of its functional demands.

1.3 Method

The method used in this thesis consists of three phases. Figure 1.1 shows the three phases and their activities.

Phase 1

Phase 2

Phase 3

• Modeling • Simulation • Concept evaluation • Concept elimination • Detailed modeling • Detailed simulation • Concept generation • Concept evaluation • Concept elimination

(12)

1.4 Delimitations 6

The benefit of dividing the process into the three phases is the time required. By eliminating concepts without potential to be the final solution in an early stage two benefits are obtained: A high number of solutions can be investigated, without spending too much time developing them. The reduced number of thoroughly inves-tigated concepts will make it possible to develop them closer to a finalized solution. A detailed description of the phases is presented below.

Phase 1: The concept generation in the first phase results in simple diagrams describing the concepts. The generated concepts were evaluated with respect to criteria concerning the following system aspects:

• energy efficiency

• ability to maintain a desired pressure • estimated cost

• packaging

• robustness and complexity.

The evaluation results in a score for each of the concepts that are then used to eliminate the least promising concepts. The elimination is not based only on this score, but also on the concept’s improvement potential. The knowledge of Volvo’s employees is of great value during the concept evaluation and elimination in Phase 1.

Phase 2: The concepts that proceed to the second phase are modeled in LMS Amesim to allow a more objective comparison. All concepts are simulated with a flow and pressure demand as per those found in a dimensioning drive cycle. The drive cycle consists of a sequence of rapid gear changes that test the performance of the concepts. The energy efficiency of the concepts are evaluated using another drive cycle simulating how an articulated hauler is used in the field. A new evaluation is then performed using results from the simulations. Only one concept proceeds to the last phase.

Phase 3: In the final phase the model of the remaining concept is further devel-oped. This makes it possible to propose a design specification containing the sizes of the components. This also makes it possible to compare the proposed system with the system used in PT9.

Methods are developed that can be used when selecting component sizes for the proposed system. The methods should generate component sizes that give an energy efficient system that meets the functional demands of the system.

1.4 Delimitations

To complete the project in the set time delimitations were made:

• Only the supply system for the clutch actuation is modified. No changes are made to the clutch actuation hydraulics, including control valves and cylinders used for actuating the clutches.

(13)

1.4 Delimitations 7

• The pump type is not changed. An external gear pump is used in all concepts. For all concepts the location of the pump is at the same position as in PT9. • The performance of the concepts is evaluated using simulations. No physical

(14)

Chapter 2

Theory

In this chapter, a theoretical basis is given to help reader comprehension of this the-sis. General information regarding a planetary transmission and a clutch actuation system is presented. In the last section, the hydraulic accumulator will be addressed.

2.1 Planetary transmission

The planetary transmission is built up of a number of planetary gears. A planetary gear consists of a sun gear, one or more planet gears and a ring gear [4]. Figure 2.1 shows a planetary gear with three planet gears. The planet gears are coupled via a planet carrier. By varying where the input and output shafts are connected different gear ratios are achieved.

z₂ z₁ z₃ Ring gear Planet gear Planet carrier Sun gear

Figure 2.1: A planetary gear including three planet gears.

The gear ratio of a planetary gear is described by eq. (2.1) and eq. (2.2) [6].

R = −z2

z1

(2.1)

(15)

2.1 Planetary transmission 9

R = nsun− ncarr

nring− ncarr

(2.2) The ratio between the input axis and the output axis is described by eq. (2.3) [6].

u = nin

nout

(2.3) An example where a planetary gear is used as a transmission is in an old three-speed bicycle. In this example the sun gear is locked to the bicycle frame via a shaft and has no rotational velocity (nsun = 0). The number of teeth on the sun gear and the ring gear are for this example selected to give a ratio R = −3 (see eq.2.1). Eq. (2.2) gives the ratio between the ring gear and planet carrier:

nring =

(R − 1)

R ncarr=

4

3ncarr (2.4)

The first gear is active when the sprocket is connected to the ring gear while the wheel is connected to the planet carrier (see figure 2.2). The ratio between the sprocket and the wheel is given by eq. (2.5), which is derived from eq. (2.3) and eq. (2.4). For the first gear the sprocket rotates four turns for every three turns of the wheel. u = nring ncarr = 4 3 (2.5) Whe e l Sprocke t Sun ge a r Pla ne t ge a r Ring ge a r Pla ne t ca rrie r Ce ntrum a xis

Figure 2.2: Gear 1 active in bicycle transmission.

For the second gear both the sprocket and the wheel are connected to the ring gear. The input power is transferred directly to the wheel without passing the planet gears (see figure 2.3a). The ratio calculated from eq. (2.3) gives:

u = 1

(16)

2.2 Clutch actuation system 10

For the third gear the sprocket is connected to the planet carrier and the wheel connected to the ring gear (see figure 2.3b). The wheel will now rotate four turns for every three turns of the sprocket (see eq. (2.7)).

u = ncarr nring

= 3

4 (2.7)

(a) Gear 2 active. (b) Gear 3 active.

Figure 2.3: Bicycle transmission.

As seen in the example, various gear ratios can be achieved by changing where the input and output are connected. In a planetary transmission, where several planetary gears are combined in series an even larger number of different gear ratios can be achieved.

In some applications torque converters are used to connect the transmission to the engine. A torque converter is a hydrodynamic transmission that allows the engine to run even though the transmission is in gear and the vehicle is standing still [11]. To decrease the energy losses in the torque converter, the transmission can be equipped with a lock up. A lock up locks the engine and transmission with a clutch. A hydraulic piston actuates the clutch (see section 2.2 for a similar arrangement).

2.2 Clutch actuation system

In order to achieve the different gear ratios of the transmission, the gears are con-nected and locked in different combinations. They can be concon-nected to each other, to a shaft or to the transmission chassis. A brake is used to lock the gear to the chassis. Otherwise a clutch is used. Both clutches and brakes are built up of friction discs and steel discs stacked alternately (see figure 2.4). The difference between a clutch and a brake is that in a brake the steel discs are connected to the chassis and are therefore stationary. In a clutch all discs are connected to parts that can rotate. The function is however the same, so only the clutches will be described in this section. The clutches used are wet clutches, meaning they are immersed in oil.

The clutch is engaged by the piston, which pushes the clutch discs together (see figure 2.4). This is done by filling the cylinder with oil. The pressure demand can be divided into two stages. The first stage is to overcome the spring holding the piston in position while it is disengaged. This will move the piston towards the discs.

(17)

2.3 Powershift 11

Gear Clutch disc Steel disc

Oil

Piston

Spring

Shaft Bearing

Figure 2.4: Overview of the clutch.

When the piston reaches the discs, the pressure will be increased in order to lock the clutch. The hydraulic pressure needed depends on the clutch design and the amount of torque transferred over it [6].

When the clutch is disengaged, the clutch discs must be separated to create a spacing between them to lower the drag losses [9]. This means that the piston has to be pushed back into the cylinder more than just to release the pressure on the discs. This is done by the spring. The total distance that the piston has to move in order to engage and disengage the clutch is dependent on the number of discs. More discs require more total spacing.

2.3 Powershift

A powershift transmission is designed to allow a gear change to be performed with-out interruption of transmitted power. Hydraulically actuated clutches are used to change active gear ratio.[14]

How the actual disengaging and engaging of a gear is done is described in sec-tion 2.2. In order to change gears without an interrupsec-tion in the transmitted power, the next gear’s clutch is engaged at the same time as the previous gear’s clutch is released. The procedure can be divided into the following phases:

• Filling phase: During this phase the cylinder is filled with oil. The pressure is then increased to the point where the piston just reaches the clutch discs (kissing point). No torque is transferred through the clutch during this phase. • Torque phase: By increasing the pressure in the engaging clutch, the torque

gradually transfers from the disengaging clutch to the engaging.

• Inertia phase: At the start of this phase there will be no slippage at the dis-engaging clutch even though no torque is transferred through this clutch. The

(18)

2.3 Powershift 12

slippage is at the engaging clutch. Pressure in the engaging clutch is increased even more, locking the engaging clutch. Slippage will now occur at the disen-gaging clutch.

The torque and inertia phases can easier be explained in the following example illustrating the phases translational instead of rotational. This example is inspired by [7]. A person on roller skates is holding each hand on long rods which are moving up a slope (see figure 2.5). The person is holding onto the right hand rod and is thus moving at the same speed as the rod. The left rod is moving faster. The person wants to increase its speed by gripping the faster moving rod.

2 m/s

3 m/s

Figure 2.5: A person on roller skates moving up a slope.

In what order should the person release and grip the rods? The procedure can be divided into the following phases:

• Filling phase: This phase is not described by this example.

• Torque phase: Immediately as the person starts to grip the faster moving rod with the left hand the friction force in this hand will increase. As this happens the force pulling the person upwards will gradually move from the right hand to the left hand. At a certain pressure, all the force will be transferred through the left hand which means the force on the right hand will be equal to zero, even if this hand holds with full pressure. Since the right hand is no longer transferring any force, it is time to release it. Now the person will have the same speed as before but the force pulling the person upwards comes from the persons left hand slipping on the faster moving rod. This will produce a lot of heat.

• Inertia phase: In the inertia phase the person is accelerated to the speed of the faster moving rod. This is done by increasing the pressure of the left hand until there is no slippage.

This example describes a gear change to a higher gear while driving. While driv-ing the input torque to the transmission is positive. There are three other scenarios:

(19)

2.4 Hydraulic accumulator 13

changing to a lower gear while driving, changing to a higher gear while braking (neg-ative input torque) and changing to a lower gear while braking. These three scenarios are executed in a similar way to the one described in the previous example with the exception that the phases are performed in different orders.

The quality of the gear change in terms of smoothness, transmission wear and ability to transfer torque is highly dependent on a reliable oil supply. For example, if the hydraulic pressure drops below the required pressure in a clutch, the clutch will start slipping and torque transfer will be reduced. This will also increase wear on the clutch discs. [7]

2.4 Hydraulic accumulator

Hydraulic accumulators are common components in fluid power systems. They can be used for multiple purposes, including

• as a supplement to a pump • energy storage

• shock absorption • reduction of pulsations.

In systems with long periods of low flow demand and shorter periods of higher flow demand it may be suitable to use an accumulator. The accumulator will supplement the pump during the periods with high flow demand meaning the pump can thus be downsized. [3]

2.4.1 Types of accumulator

Accumulators can be divided into two groups depending on their design, namely mechanical accumulators and hydropneumatic accumulators. Hydropneumatic ac-cumulators are the most common. They all utilize the compressibility of a gas, often nitrogen (N2), to store energy. The accumulator is charged by letting a pressurized

liquid enter the accumulator, causing the gas to be compressed. The liquid flows into the accumulator until the pressure of the gas and the liquid are equal. Depending on the volume and the initial (precharge) pressure of the gas, different amounts of energy can be stored. It is desirable to keep the gas and the fluid inside the accumulator separated. This is to prevent the gas from entering the liquid. The most frequently used separating elements are bladders, pistons and diaphragms (see figure 2.6). [2]

Mechanical accumulators typically consist of a spring-loaded piston (see fig-ure 2.7). There also exist mechanical accumulators that use the weight of a mass as a load. The gravitational force acting on the mass will work as the spring in the pre-viously described accumulator. In contrast to the prepre-viously described accumulator, the pressure in this type of accumulator is not dependent on the stored amount of liquid, due to the constant gravitational force. [2]

Depending on whether a bladder or a piston is used to separate the gas and liquid, the properties of the accumulator will differ. Some of the advantages of using a piston instead of a bladder are that it allows a higher flow rate, tolerates a wider

(20)

2.4 Hydraulic accumulator 14 Gas Oil Diaphragm Gas Gas valve Piston Liquid Hydraulic port Bladder Gas Liquid Hydraulic port Gas valve

Figure 2.6: Hydropneumatic accumulators with bladder, piston and diaphragm.

Hydraulic port Liquid Piston Spring

Figure 2.7: Spring loaded mechanical accumulator.

range of temperatures and withstands a higher compression ratio. However, by using a bladder the accumulator will be able to quickly respond to pressure changes and be tolerant against contamination in the liquid. [3]

According to [12], hydropneumatic accumulators are preferred over mechanical accumulators in most applications due to their lighter weight, lower cost and com-pactness.

2.4.2 Gas model

The main part of hydropneumatic accumulator’s energy losses results from the ther-modynamical properties of the gas. As the accumulator is charged the gas is com-pressed. As a consequence the gas temperature is also increased. Heat is then transferred from the gas to the surroundings and the temperature will hence de-crease. This will decrease the pressure of the fluid in the accumulator and the stored potential energy will decrease. Multiple ways of modeling this process exists and a commonly used method is to assume that the gas behaves as an ideal gas. It is then modeled as in eq. (2.8), where p and V are, respectively, the pressure and the specific volume of the gas and n is the polytropic index [11].

pVn= constant (2.8)

The value of the polytropic index is dependent on how the accumulator is used and if the process is considered to be adiabatic or isothermal. If the charge and discharge of the accumulator is fast (less than one minute) the process is considered to be adiabatic due to the low amount of heat transferred from the gas. The index is often

(21)

selected to 1.4 − 1.5 for N₂. If the process is slower (takes more than three minutes) it is considered to be isothermal and all heat generated when the gas is compressed is transferred to the surroundings. The index is thus selected to 1.0 for N2. These

values are approximations made for simplicity. The index value is dependent on the gas type and at which pressures and temperatures the accumulator is working. The index should be adjusted with respect to these factors to give a good agreement with reality. [11]

A more accurate model of the gas in the accumulator can be achieved by assuming that it is a real gas instead of an ideal gas. In [13] a real gas model based on the Benedict-Webb-Rubin equation is used. The real gas model is compared to an ideal gas model and the conclusion is that for low pressures (below 100 bar) the thermal properties of the gas can be determined using the simpler ideal gas model with good accuracy. Due to the small deviation between the ideal gas model and the real gas model it is sufficient to use methods assuming an ideal gas when sizing accumulators for a system working at low pressure levels. [13]

T

1

t

T

τ

T

2

t

1

t

2

0.632(T

1

-T

2

)

T

t2

Figure 2.8: Definition of an accumulator’s thermal time constant [12]. The thermal losses are ignored when assuming that the charge and discharge of the accumulator can be described as an isothermal or an adiabatic process. When the accumulator is charged the gas inside it is compressed which will result in in-creased gas temperature. Heat energy is then transferred away from the gas leading to decreased gas pressure and thus the amount of stored energy will be lowered, even though there is no fluid leaving the accumulator. If the efficiency of the accumulator is of interest it is necessary to use a model which includes heat transfer. One model which includes these losses is presented in [12]. It is based on that every accumulator has a thermal time constant that describes how heat is transferred between the gas and its surroundings. It is possible to calculate how temperature varies with time if this constant is known. Thus the thermal losses can be determined. The thermal time constant can either be measured for a specific accumulator, or estimated based on an accumulator’s design and size. One definition of the thermal time constant is the time it takes for the temperature to drop after rapid compression of the gas. If the gas temperature before the compression is T₂, and the time and temperature

(22)

im-2.4 Hydraulic accumulator 16

mediately after the compression are t₁ and T₁respectively, the thermal time constant is calculated as:

τ = t2− t1 (2.9)

where t₂ is the time when the gas temperature equals T_t2 (see figure 2.8).

Tt2 = T1− 0.632(T1− T2) (2.10)

In reality the time constant is in fact not a constant but will vary with how the accumulator is used. According to Otis et al. this approximation is however still quite a good representation of the actual heat transfer and agrees with experimental data. [12]

2.4.3 Sizing

A sizing method for a hydropneumatic accumulator using the ideal gas model can be derived from eq. (2.8). Five parameters need to be specified to determine the accumulator size. The parameters are:

• minimum system pressure p1,

• maximum system pressure p₂, • precharge pressure p0,

• working volume ∆V and • polytropic index n.

The minimum system pressure is the lowest acceptable pressure for which the system still works as intended. Maximum system pressure is the highest pressure in the system. The precharge pressure is typically set below the selected minimum pressure (0.9·p1). The working volume is the useful volume of the accumulator. The polytropic

index is dependent on how the accumulator is used (described above). The total volume of the accumulator, V0, can then be calculated using eq. (2.11).

V0= ∆V p1 p0 1 − (p1 p2) 1 n (2.11)

This equation assumes that both the charge and the discharge are adiabatic or isothermal. If the charging of the accumulator is isothermal (slow) and the dis-charge is adiabatic (fast) another way of calculating the accumulator volume, V₀, is used. This equation, and the derivation of eq. (2.11), are given in appendix A. [11]

2.4.4 Gas permeability

The amount of gas in hydropneumatic accumulators decreases with time as the gas permeates through the diaphragm or bladder and enters the liquid. This effect is also present in piston accumulators, but the process is slower since the surface area for gas permeability is only that of piston seal. Below are some factors that determine the rate of gas permeability:

(23)

• The permeability coefficient of the material used to separate the gas and the liquid.

• How large area is in contact with the gas.

• The thickness of the material used to separate the gas and the liquid. • The pressure of the gas.

• The temperature of the gas. [10][5]

(24)

Chapter 3

System description

The system used as a reference in this thesis is the planetary transmission PT9, developed by Volvo CE in Eskilstuna. This transmission is used in Volvo’s articulated haulers. Figure 3.1 shows one of Volvo’s articulated haulers.

Figure 3.1: Picture of Volvo A40G FS. Volvo Media Library.

PT9 uses five planetary gears connected in series to achieve its nine speeds forward and three speeds backwards. Figure 3.2 shows a simplified drawing of the hydraulic circuit that is used in the transmission. The hydraulic circuit can be divided into two subsystems: a high-pressure system (HPS) and a low-pressure system (LPS). The purpose of the HPS is to control the pistons that are used to control the clutches (C1-C3), the brakes (B1-B5) and the lock-up function (LU). For simplicity the clutches, brakes and lock-up are referred to as shifting elements. The LPS provides oil to cool the torque converter and the clutch discs plus lubrication to the moving parts in the transmission. Pump P2, 2 in figure 3.2, is the main oil supplier to the LPS system.

(25)

3.1 Oil supply system 19

P1 P1 M

P2

Cooling and lubrication of clutch discs and bearings

10 9 1 2 3 4 5 6 7 8 11 12 Pump for HPS Pump for LPS Piston for lock-up Pressure relief valve, LPS Filter, HPS

Torque converter Cooler, LPS Filter, LPS

Pilot-operated pressure relief valve Pilot valve for pressure relief valve Control valve, PWM input signal Pistons for clutch actuation 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

HPS

LPS

Figure 3.2: Simplified hydraulic drawing of the clutch actuation system.

3.1 Oil supply system

The pilot-operated pressure relief valve (PRV) 9 is used to control the pressure in the HPS and is thus normally open to allow excess flow from the HPS to enter the LPS. The opening pressure is controlled by the pilot valve 10, which in turn is controlled by a PWM-signal. While driving the pressure in the HPS is set by the shifting element that currently needs the highest pressure to prevent it from slipping. The pressure will vary with how much torque is being transferred through the transmission, but it is typically around 11-16 bar. While driving the flow demand to the HPS is only the leakage flow. The main part of the leakage originates from rotating seals located between the control valves and the pistons (between 11 and 12 in drawing, not included) and leakage in the control valves 11. The leakage flow varies with the shifting elements being actuated and the pressure in the HPS. The clutches (C1-C3) and the lock-up (LU) are drained. The drainage flow will be referred to as a leakage flow in the rest of this thesis. The total leakage flow is smaller than the flow supplied by pump P1 1 , and the main part of its flow is sent to the LPS.

When a gear change is performed the control valve 11 is opened to initiate the filling phase of the shifting element 12 that needs to be engaged for a new gear. The volume required to fill the different shifting elements varies. After the filling phase the pressure is decreased in a disengaging element at the same time as the pressure is increased in an engaging element. This is done in order to perform a powershift (see section 2.3). Depending on which gears the gear change is done between, a varying number of element pairs are switched. Only one pair of shifting elements needs to

(26)

3.2 Pump 20

be switched for the majority of gear changes, but for some gear changes up to three pairs are switched. During a gear change the pressure in the HPS is increased to around 20 bar. The flow demand is also increased during the filling phase of the gear change, since one or more elements need to be supplied with oil. The reason for increasing the pressure during a gear change is mainly to make it easier to calibrate the control valves 11. With a constant pressure level it is easier to predict the flow through the valve and thus the duration of the filling phase. In order to perform a good gear change, the duration of this phase needs to be known.

A pressure transducer is located at the filter in the HPS 5 that is used to monitor the contamination level of the filter. If the pressure drop over the filter reaches a certain level the PRV located at the filter is opened. Another PRV is located in the LPS 4 which prevents the pressure in the LPS from getting too high preventing possible damage the torque converter 6 . The torque converter will increase the oil temperature when active. Therefore is the oil passed through the cooler 7 before it is used to cool the clutch discs. The torque converter should be used as little as possible, improving the overall efficiency of the transmission. The lock-up function 3 is used to mechanically connect the input and output shafts of the converter to bypass its function and thus reduce the losses associated with the converter.

3.2 Pump

The pump used in the transmission is an external gear pump. The benefits of an external gear pump are that it is both compact and cheap compared to piston pumps. Unlike a common external gear pump, with two gears, this pump is designed to supply two separate circuits with different pressure levels. The pump consists of a sun gear and two planet gears (see figure 3.3). The sun gear is connected to the engine’s output shaft and they are therefore rotating with the same speed. The oil is pumped by the sun gear and the planet gears towards the pump outlets 4 and 2 . In this thesis the pump is referred to as two pumps: pump P1 and pump P2, as in figure 3.2. The pump displacements for pump P1 and pump P2 depend on the delivery area and the gear width. The flow from the pumps is described by the following equations:

q2 = q12+ q32 (3.1)

(27)

3.2 Pump 21 1 2 3 4 q₁₂ q32 q₃₄ q₁₄ q₂ q₄

(28)

Chapter 4

Concept Development - Phase 1

Concepts are generated in the first concept development phase. The concepts are not intended to be very detailed, but are more as simple ideas to solutions to allow comparison between the concepts.

4.1 Concept generation

The concepts were created with the goal to fulfill functional requirements. The main requirements were to maintain a steady pressure in the HPS during gear changes and to supply the LPS with a reliable flow. If the oil supply to the HPS is too low while changing gears a pressure drop will occur and therefore the gear change performance will decrease. If the flow to the LPS is insufficient this can have fatal consequences for the transmission. For example the clutch discs can overheat if they are not properly cooled.

To fulfill the functional requirements, the pump sizes could be increased. Larger pump sizes would result in increased energy consumption but since this thesis is about creating an energy efficient oil supply system this was not considered as being a suitable solution. When the concepts were generated the main focus was to keep the flow over the pilot-operated PRV low (9 in figure 3.2). The excess flow will be throttled over the PRV, resulting in energy losses. Cost has also been a factor when generating the concepts. Some of the concepts may be more expensive than others but if they perform better they might be worth the higher cost. This will be evaluated in section 4.2.

The nine concepts generated are presented below.

(29)

4.1 Concept generation 23

4.1.1 Concept C1

- Pump P2 works as a supplement to pump P1.

A 9

B

Shifting Elements

Cooling & Lubrication

Figure 4.1: Hydraulic diagram over concept C1.

In this concept, pump P2 is switched over to the HPS to support pump P1 when changing gears. This is done by closing the on/off-valve, A in figure 4.1, located after pump P2. When this valve is closed the flow from pump P2 will be directed to the HPS, the flow from both pumps being used to fill the shifting elements. By increasing the flow supply during the filling phase the pressure drop will be smaller compared to the pressure drop occurring in the reference system. Excess flow in the HPS is directed back to the LPS through the pilot-operated PRV 9 . The restriction B is used to ensure a minimum flow to the LPS. It also dampens the pulsations that may occur when the on/off-valve is closed. The valve is normally closed. If an error occurs and the valve can not be controlled the valve returns to closed position. The benefit is the transmission still will be able to perform gear changes with maintained performance. The negative effect is the increased energy consumption since more oil flow over the pilot-operated PRV 9 .

(30)

4.1.2 Concept C2

- Actively controlled accumulator in the HPS.

A B 9 Shifting Elements Cooling & Lubrication

Concept C2 uses an accumulator that is recharged when no gear change is be-ing performed and there is excess flow supplied to the HPS. The oil stored in the accumulator can then be used to supplement pump P1 during a gear change. An on/off-valve, A in figure 4.2, located at the entrance to the accumulator allows the pressure in the HPS to be lowered without lowering the pressure in the accumulator. To monitor the pressure level in the accumulator a pressure transducer B is used. The transducer is needed since the pressure changes with time, even if the on/off-valve is closed. A declining gas pressure may cause problems if not handled properly. If the usable oil volume in the accumulator is smaller than expected due to decreased pressure, it may lead to impaired gear change performance. This problem is present in hydropneumatic accumulators.

(31)

4.1.3 Concept C3

- Pump P2 works as a supplement to pump P1. The HPS has an actively controlled accumulator.

Shifting Elements

Concept C3 is a combination of C1 and C2. The possibility to direct pump P2’s flow to the HPS can be seen as a flow backup to pump P1 and the accumulator. If pump P1’s size is selected as small and the accumulator has been drained due to continuous gear changes, the supplied flow to the HPS may be insufficient. When this scenario occurs pump P2’s flow can be redirected to support pump P1. The accumulator will have a dampening effect when pump P2 is redirected.

Compared to concept C2 a wider range of pumps and accumulator sizes can be chosen. If pump P1 is dimensioned to only cover the leakage flow in the HPS at engine idle speed, the accumulator size will depend on how often pump P2 must be redirected. A larger accumulator might reduce the number of times pump P2 has to be switched over. If there is a space limitation, a smaller accumulator might be necessary, leading to a more frequent use of pump P2 supplying the HPS.

(32)

4.1.4 Concept C4

- Pump P2 works as a supplement to pump P1. The HPS has a constant pressure level and a passive accumulator.

Cylinders

A

In this concept the pressure level in the HPS is constant and it is thus possible to replace the pilot-operated PRV seen in the previous concepts with a simpler PRV A in figure 4.4. There is also no need to actively control the accumulator. This will make the system easier to control and decrease the component cost.

A possible downside with this concept could be its energy efficiency. Due to the constant pressure in the HPS the pressure drop over the PRV 9 will always be large. In the reference system energy is saved by lowering the pressure when no gear change is performed. The downside of not being able to lower the system pressure will be emphasized if the accumulator is fully charged and the engine speed is high. The power losses can however be lowered by selecting a small pump P1. If the accumulator is large some of the excess flow from pump P1 can be stored in the accumulator. If the combined flow from pump P1 and the accumulator is insufficient pump P2 can be redirected to supply the HPS.

(33)

4.1.5 Concept C5

- Pump P2 works as a supplement to pump P1. An accumulator is located between the two systems.

A

B

Shifting Elements

In this concept an accumulator is located between the LPS and the HPS. An advantage of this location is the possibility to charge the accumulator without the need to change the pressure in the HPS. For example this can be useful after a filling phase where the accumulator was used and needs to be recharged. If the accumulator was directly connected to the HPS, it would be necessary to recharge the accumulator in order to increase the main pressure. This would result in a pressure below the desired value during the subsequent shifting phases. The time needed to increase the pressure can be shortened by closing the on/off-valve, A in figure 4.5, immediately after the filling phase is done. The pressure will then quickly be rebuilt by pump P1. The accumulator is then recharged by pump P2.

When filling shifting elements requiring only small volumes, the accumulator will not be depleted after the filling phase. It may then be enough to only use pump P1 to recharge the accumulator. This can be done by leaving both valve A and B open after the filling phase. The pressure in the accumulator will probably not be lower than what is acceptable during the torque phase due to the small volume drained from the accumulator. Thus it may not be necessary to quickly restore the pressure, as described in the paragraph above.

One possible downside with using pump P2 to charge the accumulator is that during charging the flow to the LPS will be reduced.

(34)

4.1.6 Concept C6

- Pump P2 works as a supplement to pump P1. A high-pressure accumulator is located between the two systems.

A

9 Shifting Elements

In this concept the pressure in the accumulator is increased. By increasing the maximum pressure, it is possible to have a larger pressure difference between when the accumulator is fully charged, p2, and when depleted, p1, in eq. (2.11). This will

make it possible to install a smaller accumulator with a preserved working volume. A smaller accumulator is easier to install in the transmission.

Control valve A is used for regulating the flow from the accumulator to the HPS. The valve works as a pressure reducer and strives to maintain the pressure downstream of the valve at a desired level. The pressure drop, occurring when changing gear in the previous concepts including an accumulator, may be smaller in this concept. This is because the lowest pressure in the accumulator will still be higher than the desired pressure in the HPS. The desired pressure in the HPS can thus be maintained with reduced pressure drop.

The increased pressure in the accumulator is not favorable when it comes to energy efficiency. The pressure difference over the pump supplying the accumulator will increase, and thus the volumetric efficiency of the pump will be lowered leading to energy losses. In addition, the accumulator is charged to a higher pressure than used in the actuation system. Thus it will be necessary to throttle the oil pressure before it enters the HPS.

(35)

4.1.7 Concept C7

- The HPS uses a high-pressure accumulator.

In concept C7 the maximum accumulator pressure is increased to obtain the same benefits as in concept C6 . In this concept the accumulator is located in the HPS and charged by pump P1. As in concept C6, a control valve, A in figure 4.7, is used to regulate the flow from the accumulator. It is necessary to increase the pressure in the HPS to allow the accumulator to be charged to the higher pressure. The pressure is increased by setting a higher opening pressure of the pilot-operated PRV 9 . The check valve B allows oil to enter the accumulator if the pressure in the HPS is higher than in the accumulator. The increased pressure during charging will affect the whole HPS, including all of the control valves used for the clutch actuation (11 in figure 3.2). This sets higher demands on the maximum pressure rating of these valves. The increased pressure might also contribute to increased leakage in the HPS.

(36)

4.1.8 Concept C8

- The HPS uses a high-pressure accumulator.

Valve A in figure 4.8 is used to control when the accumulator should be used, as in concepts C6 and C7. When no gear change is performed, and the main pressure is lowered, valve B opens to allow the flow from pump P1 to enter the HPS. To prevent the accumulator from discharging, control valve A is closed.

When changing gear the HPS pressure is increased by valve 9 . Valve A is also set to maintain the higher HPS pressure. As the filling phase begins, and the flow to the shifting element starts, valve A will gradually open to supply flow from the accumulator. After the gear change, valve B is closed to allow the accumulator to recharge. Valve 9 is set to maintain a lower main pressure. Valve B has a pressure relief function that prevents the accumulator from being overcharged.

In this concept fewer components will be subjected to the increased HPS pressure compared to concept C7. This comes with a cost of a larger number of components and more complex control.

(37)

4.1.9 Concept C9

- Three pumps combined in different ways to supply both systems.

Cylinders

This concept contains three pumps controlled by on/off-valves. The pumps are intended to be arranged as the pump used in PT9 with the difference that it has three planet gears instead of two. The pump will however remain compact.

By closing the on/off-valves in different combinations, the flow from the different pumps will be directed to the HPS, the LPS or to the suction side of the pumps. Check valves ensure that the oil flows in the intended direction. If a pump is not needed for either the HPS or the LPS, its on/off-valves will be open, leading the oil back to the suction side. This reduces energy consumption since there is no pressure increase over the pump. By directing the oil back to the suction side instead of to the tank the energy consumption is reduced due to reduced flow losses. The idea is to combine the pumps to supply both systems without excess flow. Low flow over the pilot-operated PRV results in higher energy efficiency.

The sizes of the pumps are chosen to give high efficiency at selected working points. There are a lot of different ways to select the size of the pumps and when they are to be used. The following paragraph explains one way.

The three pumps are of the sizes small, medium and large. At engine idle speed the medium pump is used to cover the leakage in the HPS. The large pump covers the flow demand to the LPS. The small pump is directed to the suction side to save energy. While driving, when the engine speed is around twice as high as the engine idle speed, the medium pump will deliver too much flow to just cover the leakage flow in the HPS. At this point the small pump cab be directed to the HPS while the medium pump replaces the large pump to deliver flow to the LPS. The large pump is directed to the suction side and therefore consumes less energy.

The HPS requires a higher flow when changing gear. To supply the HPS with sufficient flow the pumps may be redirected. The large pump may be directed over

(38)

4.2 Concept evaluation 32

to the HPS if the gear change is performed at lower engine speeds. Otherwise the combined flow from the small and medium pump may be sufficient. If the pump that supplies the LPS is directed over to the HPS another pump must be directed to the LPS to ensure sufficient flow.

4.2 Concept evaluation

A concept evaluation was performed to allow elimination of the least promising con-cepts. In this first phase the concepts were evaluated in consultation with the staff at Volvo CE. Due to the large number of concepts it was not possible to model and simulate the concepts to obtain numerical values such as energy efficiency of the con-cepts. The concepts were instead rated during a group discussion where they were rated by how well they were estimated to perform with respect to a number of cri-teria. The evaluation criteria were generated and weighted by importance together with the staff at Volvo. The evaluation criteria were:

• Ability to maintain pressure • Energy efficiency

• Component cost • Continuous shifting • Control complexity • Packaging

All concepts must be able to supply the LPS with sufficient flow. A criterion for this has thus not been included in the evaluation since it was considered a require-ment.

The criterion Ability to maintain pressure describes how well the system keeps desired pressure in the HPS during a gear change. It was assumed that concepts including an accumulator were better at maintaining the pressure compared to con-cepts without one. The accumulator can deliver a high maximum flow if the demand is high, whereas a pump will deliver the same amount of flow, regardless of the flow needed. This criterion was considered to be important since it has a large impact on the quality of a gear change.

The system’s Energy efficiency is also considered to be an important criterion. With increased energy efficiency, the fuel consumption of the vehicle can be de-creased. At Volvo CE and other companies, low fuel consumption is an important sales argument. Increased efficiency may also result in more power at the wheels, which will increase the performance of the vehicle. When evaluating the energy ef-ficiency, concepts with low amounts of excess flow in the HPS were rewarded with a higher score.

The Component cost is, as always, an important aspect when developing products. Concepts with a large number of complex components will be a more expensive solution.

(39)

4.2 Concept evaluation 33

It is important that the concepts work well for a series of rapid gear changes. The concepts’ capacity to perform sequential gear changes is evaluated by the crite-rion Continuous shifting. For example concepts including an accumulator may have problems if the time between two subsequent gear changes is too short to allow the accumulator to be recharged. The oil volume stored in the accumulator may then be insufficient, resulting in reduced performance of the gear change.

Concepts including valves that are difficult to control are penalized by the crite-rion Control complexity.

Packaging penalizes concepts with a large space requirement, e.g. concepts with

a large accumulator or many valves will get a low score for this criterion.

The Pugh method was used to perform the concept evaluation. This method was selected since it is a systematic way to execute the evaluation and was familiar to the staff at Volvo CE. The completed Pugh matrix from the evaluation is presented in table 4.1. The concepts were compared against the reference system, used in PT9. Depending on how well the concepts performed on each of the criteria they were given a value between -5 and 5. The given value is then multiplied by that criterion’s Weight and the Total Score of each of the concepts is calculated as the sum these products.

Table 4.1: Pugh matrix from the concept evaluation in Phase 1.

Criteria Weight PT9 C1 C2 C3 C4 C5 C6 C7 C8 C9

Ability to maintain pressure 10 0 2 4 4 4 4 5 5 5 3

Energy efficiency 9 0 2 1 3 -1 3 3 0 0 4 Component cost 5 0 -1 -2 -3 -2 -4 -4 -2 -2 -5 Continuous shifting 6 0 0 0 0 0 0 0 0 0 0 Control complexity 5 0 -1 -2 -3 0 -3 -3 -3 -3 -3 Packaging 3 0 -1 -3 -3 -4 -4 -3 -3 -3 -2 Total Score 0 25 20 28 9 20 33 16 16 20

It can be seen in the weight column in table 4.1 that the Ability to maintain

pressure and the Energy efficiency were considered to be the most important criteria.

Their weights were selected to 10 and 9, respectively. The criterion Continuous

shifting was also regarded as important, but during the evaluation it was realized

that it was difficult to estimate the concepts’ performance regarding this criterion. Thus, values were set to zero for this criterion for all the concepts.

The concept that got the highest total score was C6. This concept was considered to be very good in the criterion Ability to maintain pressure due to the increased accumulator pressure. This concept also scored high on the Energy efficiency crite-rion. The Component cost of this concept was, on the other hand, considered to be high and was the concept’s weak point.

Concept C3 got the second highest score. The difference between concept C3 and C6 was that C6 scored better for the criterion Ability to maintain pressure, but has a higher Component cost.

(40)

4.3 Concept elimination 34

4.3 Concept elimination

The concepts that scored the fewest points in the Pugh matrix were concepts C4, C7 and C8. These concepts were eliminated as none of them were considered to have the potential to become the most energy efficient system.

(41)

Chapter 5

Concept Development - Phase 2

The concepts will be modeled and simulated in the second concept development phase. All concepts that made it to the second phase, except C9, have two pumps and share the same basic idea on how they are to be controlled. Due to the complexity of the controller that would have been needed to control concept C9, this concept will not be investigated further due to time limitations. Nevertheless, the concept is still believed to have potential as an energy efficient solution.

5.1 Design limitations

During Phase 2 it was found that installing an on/off-valve after pump P2 is more difficult than expected, due to where the pump is located in the transmission. The concepts that are dependent on this valve were penalized in the following concept evaluation. As mentioned in Phase 1, the concepts performance is dependent on having unequal pump sizes. Due to the design of the pumps it would probably be more expensive to build pumps with unequal pump displacements. Due to the additional cost every concept was evaluated twice: once with unequal sizes of pump and once with equal sizes.

In the previous concept evaluation the system used in PT9 was used as reference system. Since this system is not able to maintain the pressure in the HPS, the pump sizes were increased in order to make a relevant comparison possible. The pump sizes were decided through optimization. This system was used as the reference system for concept evaluation in this phase.

5.2 Modeling

The remaining concepts (C1, C2, C3, C5 and C6) and the reference system used in PT9 were modeled in LMS Amesim. The main purpose of these models was to make it possible to determine the energy efficiency of the different systems, investigate how well they maintained the pressure in the HPS during gear changes and ensure that they supply the LPS with sufficient flow.

In figure 5.1, part of the Amesim model of PT9 is shown (annotation numbering is the same as in figure 3.2). The diagram only includes one of the shifting elements. The actual model includes all of the shifting elements, which are located further

(42)

5.2 Modeling 36

12

11

2

4 C

D

E

A

B

9

Figure 5.1: Amesim model including the supply system and one of the shifting ele-ments.

along line A. Hydraulic cylinders 12 were used to model the shifting elements. The volumes of the cylinders were set to those existing in PT9. The control valve 11 was modeled as an orifice with variable diameter. The dynamics of the valves were modeled using second order filters and time delays B. No flow measurements were available that could be used to validate the flow through the valves. The approximate time delay of the valves were however known and the mean flow through them was estimated using the durations of the filling phases and the volumes of the shifting elements (see eq. (5.1)). ¯qvalveis the mean flow, Velementis the volume of the elements that need to be filled, t_{f p} is the known duration of the filling phase and t_delay is the delay of the valve.

¯

qvalve=

Velement tf p− tdelay

(5.1) It was important to make the flow through the valves behave as realistically as possible since the flow was believed to determine whether a pressure drop would occur during a gear change and, if so, the magnitude of it. A variable orifice 2 was used to empty the cylinder when it is disengaged. The pressure reducing valve and the orifice C were used to model the leakage that is present in the clutches and the

(43)

5.3 Simulation 37

lock-up. Another orifice D was used to model the leakage in the rest of the system. Recorded or generated drive cycles were used that contain a variable engine speed, requested HPS pressure and data of when the different shifting elements were to be activated. The data was transmitted to the relevant receiver E. The variable engine speed and the HPS pressure were important to incorporate into the model to get realistic energy consumption for the concepts. All of the concepts needed a controller to function properly. The controllers were built in Matlab Simulink and then co-simulated with the Amesim model.

The size of the accumulator was selected as 1 dm3 for all models, except concept C6 whose accumulator size was selected as 0.7 dm3 . The reason for the smaller accumulator in concept C6 was the increased span of pressures over which the accu-mulator is working, (p1−p2). This enables the accumulator to have the same working

volume but with a reduced total volume, according to eq. (2.11). The size of 1 dm3 was selected since an accumulator with this size was believed to fit in the design of PT9. Preliminary simulations using this size showed good results. The precharge pressure of the accumulator was, in accordance with section 2.4.3, selected as:

0.9 · p1 = 0.9 · 18 = 16.2 bar (5.2)

5.3 Simulation

The sizes of the pumps in the concepts were determined by optimization. The con-cepts were supposed to complete a drive cycle while consuming as low an amount of energy as possible. During the cycle there should not be any large pressure drops in the HPS or too low flow to the LPS. To dimension the pumps a recorded drive cycle was used. The cycle is a recording of a hauler entering a steep uphill at ninth gear. The speed of the hauler drops as it enters the slope. To keep the engine speed up it is necessary to continuously change to a lower gear, all the way to the first gear. This was considered to be the most demanding scenario for a couple of reasons, namely: the gear changes are occurring closely together, as soon as one is performed the next is initiated; the gear changes are also performed at low engine speed resulting in low flow supply from the pumps, at the same time as the flow demand is high to both the HPS and the LPS. The demand to the LPS is high during parts of the cycle due to use of the torque converter. The additional heat generated by the converter needs to be removed. The flow demand to the LPS is thus increased from 35 lpm to 70 lpm. Figure 5.2 shows the engine speed and the gears used during the drive cycle. As seen in the figure, the engine speed is almost as low as 700 rpm (idle speed) during one of the gear changes, making it a challenge for the actuation system. Especially the time between the eighth and the first gear is very short. The strategy controlling the gear changes decides to skip the sixth, fourth and second gear in order to save time. The objective function of the optimization was to minimize the energy consump-tion of the pumps. The pressure drop in the HPS and the amount of flow to the LPS were used as constraints. A pressure drop of 2 bar was allowed as it was considered small enough to not have a negative effect on the gear changing performance. A constant flow demand of 35 lpm was assumed to be sufficient to cover the lubrication and cooling needs of the transmission. When the converter is used, an additional flow of 35 lpm was assumed to be needed. It was also assumed that shorter periods with

(44)

5.3 Simulation 38 0 2 4 6 8 10 12 14 N1 2 3 4 5 6 7 8 9 Active Gear Time [s] Gear [−] 0 2 4 6 8 10 12 14 500 1000 1500 2000 Engine Speed Time [s] Engine Speed [rpm]

Figure 5.2: The engine speed and gears used during the drive cycle used to optimize the sizes of the pumps.

a too low flow were allowed. In figure 5.3 an example is given to demonstrate the method used to determine if a lack of flow to the LPS was acceptable. During time a1 and a2 the flow to the LPS is insufficient, while during b1 the flow is sufficient. The flow demand was considered to be a violation if the duration of any period with a lack of flow (a1 or a2) was longer than 400 ms. It was also a flow violation if b1 was shorter than 400 ms, and the sum of a1, b1 and a2 was longer than 400 ms. Thus the flow to the LPS could be considered as a violation even though the durations of both a1 and a2 were shorter than 400 ms. If two periods with a lack of flow occurred closely together it was considered to be potentially harmful for the components in the transmission, and therefore was not allowed.

t

q

a1 b1 a2

supplied flow demanded flow

Figure 5.3: Example of supplied flow and demanded flow to the LPS.