Evaluation Methods for Friction Materials in a Four-Wheel Drive Drivetrain

(1)

Evaluation Methods for Friction Materials in a Four-Wheel Drive Drivetrain

Khashayar Shahrezaei Edvin Öberg

Automotive Engineering, bachelor's level 2018

Luleå University of Technology

Department of Engineering Sciences and Mathematics

(2)

Abstract

In 1992 the Haldex group bought Sigvard Johanssons patent from 1988. The original design is based on a rotational speed difference dependent pump that creates a hydraulic pressure to engage the clutch that locks up the rotational difference. From the originally rather simple solution, this has developed to gain in controllability and engagement speed to become a market leading system. The traction division of the Haldex group was re- cently bought by BorgWarner.

BorgWarner located in Landskrona are developing components for the four-wheeldrive drivetrain that are being massed produced in the automotive industry. Thanks to Borg- Warner’s unique product’s properties they have achieved a worldwide leading position as a provider of systems for advanced four-wheeldrive drivetrain (all-wheel drivetrain).

The assignment from the BorgWarner is to create a better understanding about when vibrations occur and what properties of a friction disc affect vibrations in the wet clutch.

The goals of this thesis are to:

• Map system parameters like Surface topografy.

• Measure pressure distribution between the friction discs.

• Measure the Youngs modulus, also known as stiffness.

It is not sure that a bad pressure distribution could cause vibrations in the wet clutch.

The results from the topography explain the appearance of the pressure distribution.

Varying stiffness means that the density of the material is also varying because the stiffness depends on the material composition. When forces are applied on the friction disc, it leads to varying deformation on samples. When the samples all deforms different, the result of the different deformation could be slanting surface. When a disc with non- parallel friction surface rotates it could generate vibration in the wet clutch.

(3)

Acknowledgements

This report is created in our thesis work and is the last step in our Bachelor of Science education in Automotive Engineering at the Lulea University of Technology.

We would like to thank our supervisors, Professor P¨ar Marklund and Dr Kim Berglund, for their guidance and encouragement in our work. Their knowledge and experience in the field have been very helpful in the discussion and planning of our research as well as in its execution.

We would like to thank BorgWarner in Landskrona for giving us the opportunity and their reflections on our work.

Khashayar Shahrezaei Edvin Oberg

Lulea, September 2018

ii

(4)

Abbreviations

LSC Limited Slip Coupling

HLSC Haldex Limited Slip Coupling AWD All Wheel Drive

Rp Roughness Average Rq Root mean square Rp Roughness peak Rv Roughness value Rt Rougness peak to valley

VSI Vertical Scaning Interferometry LW Low Pressure

v

(7)

Symbols

Strain

σ Stress [M P a]

E Young’s modulus [M P a]

vi

(8)

Chapter 1 Introduction

BorgWarner TorqTransfer Systems, located in Landskrona, is developing wet clutches used in AWD systems for automotive companies like Volkswagen, Volvo, Land Rover and Lamborghini. Thanks to BorgWarner’s unique products properties they have achieved a worldwide leading position as a provider of systems for advanced four-wheel drivetrain (all-wheel drivetrain). The demand for four-wheel drivetrain cars has increased since the mid-1980’s and is today an option in most production cars.

One of the components in a four-wheel drivetrain system is a wet clutch that regulates torque transformation and an electric device with appurtenant software. The friction system in the clutch is constantly developed by BorgWarner. Figure 1.1 is a picture of an all-wheel-drive unit for cars. It is called a Limited Slip Coupling (LSC) and is mounted in the rear axle of a vehicle.

Figure 1.1: All Wheel Drive unit for cars. Limited Slip Differential mounted in the rear axle of a vehicle [1].

1

(9)

2

The system is developed to avoid vibrations and noise generation in the drivetrain. The emergence of vibrations and noise in the drivetrain are a highly complex interaction between properties such as the natural frequency, stiffness, damping and also friction properties. The geometry of the friction discs as well as the composition of the friction material can also be important factors affecting the occurrence of vibrations. This project is therefore based on investigating friction discs properties.

The assignment from BorgWarner is to create a better understanding when vibrations occur and which properties of a friction disc that affect vibrations in the systems. This thesis is therefore based on mapping system parameters like:

• Surface topografy.

• Pressure distribution on the friction discs.

• Young’s modulus known as stiffness.

and trying to make conclusions and see if there is any connection between these parameters and vibration.

1.0.1 Wet clutch principle

This section is a brief introduction to the subject of wet clutches and it could be a good help for a reader who is not familiar to the field to have a greater understanding for the rest of this thesis.

A clutch is a machine component used to transfer torque in different types of machinery.

When the clutch is engaged, torque is transferred between the sliding interfaces in the clutch by frictional forces [1]. In a wet clutch, the friction interface is lubricated. This means that there is a fluid between the steel plates and the friction plates. Lubrication helps the system to cool down and keep it clean. Figure 1.2 shows the Haldex Limited Slip Coupling (HLSC) and1.3displays the friction discs and separator disc used.

(10)

3

Figure 1.2: Haldex Generation V AWD coupling[2].

Figure 1.3: Friction- and sparator disc.

1.0.2 Haldex Limited Slip Coupling (HLSC)

The HLSC is the main component in AWD vehicles. It is based on a multi-plate wet clutch where every second plate is coated with a sintered bronze friction material (friction discs) and every second is a plain steel plate (separator discs). HLSC can be used in both rear wheel drive and front wheel drive vehicles. In a primary front wheel drive vehicle, the HLSC regulates torque transferred to the rear axle. In a primary rear wheel drive vehicle, the HLSC regulates torque transferred to the front axle. The clutch is activated by an axial hydraulic engagement force, generated from a centrifugal electro- hydraulic actuator. When the control system requests more amount of torque, the actuator activates. When the actuator is activated, the hydraulic piston is engaged and the plates are pushed together. Torque is transferred through the wet clutch by friction between friction and separator discs. The 3D view of the HLSC Gen V is shown in Figure 1.4.

The major advantage with the HLSC is the ability to maintain a controlled slip and transferring the requested torque to the secondary drive wheels of the vehicle at the same time [3] [4].

(11)

4

Figure 1.4: 3D view of the Haldex Limited Slip Coupling, generation V [5]

(12)

Chapter 2 Background and Theory

Friction discs used in the Haldex LSC are required to have a uniform thickness to ensure the desired contact pressure distribution. The manufacturing process may, however, introduce deviations from this desired uniform thickness. Figure 2.1 shows the cross- section of a friction disc, this is how Manufacturers desire that the final result should look like. This means that no height differences are allowed on the surface structure. Due to the manufacturing process used, friction discs always display some extent of thickness deviations, see Figure2.3. The extent to which these thickness deviations affect the wet clutch performance and risk of vibrations are however unknown, see Figure 2.3. Figure 2.2shows how the surface structure of a friction disc looks like.

Figure 2.1: De- sired cross-section

of friction discs

Figure 2.2: structure of friction disc

Figure 2.3: different height level structure of an bad

friction disc

Another mechanical property that may be important to consider is stiffness, which is a measure of a material’s resistance to localized elastic deformation. Before plastic deformation, elastic deformation occurs. The elasticity of the friction discs depends on the material composition. Variations in the manufacturing process may introduce differences in material composition as well as inhomogeneities which varies between individual friction discs. Five factors that can lead to scattering in measured material properties

5

(13)

6

are the following: test method, variations in specimen fabrication procedure, operator bias, apparatus calibration, and inhomogeneities and/or compositional variations from sample to sample [6]. The friction discs studied here are porous materials, which means that there are voids within the material. An uneven porosity along the circumferential of the friction discs could influence the elastic deformation of the friction discs. Figure2.4 shows the worst case that can be happened when forces are applied (non-parallell disc).

The opposing sides of the friction disc are elastically deforming differently. Different elastic deformation could affect the pressure distribution between friction and separator discs.

Figure 2.4: Cross-section of an friction disc that shows an schematic extreme case of different elastic deformation

(14)

7

2.1 Objectives

Previous work regarding LSC’s has been focused on the clutch and drivetrain, not the influence of friction disc properties like stiffness that affects deformation of the disc and topography of surface that can affect pressure distribution. Pressure distribution is not a property for the friction disc, but a parameter that depends on topography, the global geometry and also the applied force. An important aspect is to investigate the stiffness and surface topography to see how it can affect the pressure distribution on a friction disc in a multi-disc wet clutch. These parameters and properties have been investigated to see how they can affect vibrations in the drivetrain.

2.2 Limitation

The emergence of vibration in a wet-clutch is very complex. There are a lot of machine properties that can affect the emergence of vibration. Due to time constraints, it is important to follow the project boundaries. Figure 2.5 shows that the occurrence of vibrations can depend on an interaction between different parts in a multi-disc wet clutch. Main focus is friction disc and its appurtenant measurable parameters.

Therefore this thesis focus on properties such as surface topography, pressure distribution and stiffness of the material. See 2.2.1,2.2.2and2.2.3.

Figure 2.5: Component flowchart of the multi-discs wet clutch and its measurable parameters

(15)

8

2.2.1 Surface topography

Surface topography is the local divergence of a surface from a perfectly flat plane. Surface topography can be measured in some parameters such as [7]:

• Roughness average (Ra) is the measure of the surface profile arithmetic average deviation from the center line.

• Root mean square roughness avarage (Rq) is the root mean square of the distance of the filtered or unfiltered roughness profile from its mean line.

• Roughness peak (Rp) is the highest peak distance from the mean line.

• Roughness valley (Rv) is the lowest valley distance from the mean line.

• Roughness peak to valley (Rt) is the distance from the lowest valley to the highest peak.

Figure 2.6shows a graphic description of Rv, Rt, Rp.

Figure 2.6: Graphic description of Rv, Rt and Rp [8].

2.2.2 Pressure Distribution

Another important factor that can affect vibrations in the wet clutch is the pressure distribution. When the hydraulic piston press on the clutch discs, it is important that pressure is distributed evenly among the discs. This factor might be interrelated with surface topography and stiffness, also the global disc deformation should be considered.

(16)

9

2.2.3 Stiffness

Deformation, where stress and strain are proportional, is called elastic deformation. A plot of stress against strain results in a linear relationship through equation2.1, as shown in figure2.7. The slope of this linear segment corresponds to the module of elasticity E.

σ = E (2.1)

Figure 2.7: Schematic stress-strain diagram showing linear elastic deformation for loading and unloading cycles, [6].

This is known as hooke’s law and the constant of proportionality E (GP a or psi) is the modulus of elasticity, or Young’s modulus. This modulus may be thought of as stiffness, or the resistance of a material to elastic deformation. The greater the modulus the stiffer the material [6]. Elastic deformation is non-permanent, which means that when the applied load is released, the piece returns to its original shape as shown in the stress-strain plot (Figure 2.7). Stiffness might effect pressure distribution in order of inhomogeneously material composition.

(17)

Chapter 3 Experimental Setup

There are a total of six friction discs involved in this thesis for the investigation. Three of those friction discs were classified as ”non-working” by the research team at BorgWarner, and three of the investigated discs were classified as ”working”, remain that all friction discs were manufactured by BW sub-supplier. Each friction disc was named after an alphabetic letter A to F, and each side of the friction disc was named green and red.

One cog of each side got either a green or a red mark to avoid mixing and confusion.

That cog with a marking was used as a starting position. One example is AG, it means disc A, green side. Figure 3.1 demonstrates which friction disc are ”non-working” and

”working”.

Figure 3.1: 6 Frictions discs,A-C classified as non-working and D-F classified as working .

10

(18)

11

3.1 Surface characterization

Samples on each friction disc were chosen to be investigated with an objective of 5X magnification and a field zoom lens of 0.5X. Chosen samples are based on distinctive features of the friction discs. The main reason for the chosen areas is that they look different from the rest of the disc. The colour of the surface is darker, black dots have been observed, also stripes can be detected in different light angle. Figure 3.2 shows three different sample areas that have been chosen because of their different appearance.

Figure 3.2: Three different chosen sample areas.

Vertical scanning interferometry (VSI) is a useful and quick method for evaluating the height of different surface features by mapping out interference patterns that are com- bined in the software. This was used to see the depth and shape of the chosen sample areas. Samples are analyzed in an optical microscope Zygo NewView 7300. Optical profilers from Zygo are white light interferometer systems, non-contact, high precision 3D metrology of surface features.

Surface characterization of the friction material is extremely difficult to make a measurement because of the surface of the friction disc, and also massively time-consuming especially trying to make stitch measurement for bigger areas. Stitch mode is an option in the optical microscope for bigger areas to measure. The optical microscope could not handle stitch measurement because of the surface roughness (no valid data found by stitching algorithm). So this method got eliminated after a day of measurements.

Instead of stitching measurements, the surface characterization was applied before and after compressive tests on the chosen samples. Those samples were easier to measure because the stitching was not needed. Size of the chosen samples was perfectly suitable for the lowest magnification of the optical microscope.

(19)

12

3.2 Pressure distribution measurement

Prescale film can precisely measure the pressure distribution. Red patches will appear on the film when pressure is applied and the colour density changes according to the various pressure levels. There are seven types of Prescale available to fulfil varying pressure range (0.05−300 MPa). There are two types of Prescale film types, the mono- sheet type is composed of a polyester base on which the colour-developing material is coated, with the micro-encapsulated colour-forming material layered on top. Two-sheet type is composed of two polyester bases. One is coated with a layer of micro-encapsulated colour forming material and the other with a layer of the colour-developing material [9].

Figure 3.3 shows two films facing the coated sides of each other. When pressure is applied, the microcapsules are broken and the colour-forming material reacts with the colour-developing material and red patches appear on the film, see Figure 3.3.

The nominal contact pressure in the Haldex LSC can be up to 4 MPa. Therefore, the

Figure 3.3: A-film and C-film facing the coated side of each other during measurements. [9]

low-pressure mono-type of prescale film with a pressure range of 2.5-10 MPa is used.

Figure 3.4shows steps by step how to use the Prescale film.

Figure 3.4: Step by step demonstration. [9]

The pressure distribution test rig was made by two steel discs with 15 mm thickness and two glass made discs with 16 mm thickness. The pressure machine applied a controlled force and was operated manually. Glass discs were chosen to make sure that the applied

(20)

13

pressure was evenly distributed. The steel discs were used to avoid damaging the glass discs, see Figure 3.5. Glass discs and prescale film was marked by green, red, blue and black dots, for a better understanding of the localization. Each side of the friction disc was marked by red and green and those marked cogs where always placed at the green dot of the glass made discs. This ensured that every test had the same condition.

Figure 3.5: Pressure distribution test rig.

When the prescale measurements were done, each film was scanned for validation and analysis in MATLAB. The applied loads during the experiment were 1200 Kg and 2000 Kg. This will give pressure about 2.5 MPa and 4 MPa nominally. An example of how the real pressure distribution may look like can be seen in Figure 3.6 where a load of 2000 kg was used. With help of MATLAB, the density of the red patterns on the prescale film was translated to pressure scale.

(21)

14

Figure 3.6: One example from pressure distribution test. Disc BG.

After that all pressure distribution tests were done, it was time to discuss which disc had the best load distribution. In this phase of the project, it is important to do an objective analysis. Therefore 10 students reviewed a collection of all pressure distribution appearance, all pressure films together at the same time as a map. The task of the review was to rank the best disc who had the best pressure distribution appearance.

Those students did not know anything about the discs background.

3.3 Youngs modulus of the friction material

In this work, an ElectroPuls E3000 test instrument was used to perform compressive tests on different parts of the friction discs. The precision of the compressive test machine used is ± 0.5 % of the indicated load. One calibration measurement was performed at the beginning of the test and one at the end of the test, to ensure that the compressive test machine has the same behaviour along all tests with the specimen. The test specimen was loaded with a compressive load up to 70 newtons in the test machine, see Figure3.7.

(22)

15

Figure 3.7: Compressive test rig, under measurements.

The measured curves are also influenced by the stiffness of the machine and machine holder. Thus, the estimated values for Youngs modulus depends both on the properties of the measurement system as well as the actual material properties, and should only be used to compare differences between friction discs and measurement positions. Figure 3.8shows a plot of the measurement point from the compressive test.

(23)

16

Figure 3.8: Plot of the measurement point from the compressive tests (two different samples).

It is a total of 96 samples that have been investigated, 16 samples on each friction disc and 8 samples on each side of the friction discs. Those samples were named the same way as friction discs with only a number was added. One example could be BG4 which means disc B, the green side and sample number 4. The unit of the presented values are in [MPa].

The placement of the measurement positions was based on the results from the pressure distribution measurements. Two measurement positions were placed in each of four different friction disc areas (red, black, blue, green), see Figure3.9. In each area, one of the measurement positions were placed in a high-pressure zone and the other measurement position in a low-pressure zone.

(24)

17

Figure 3.9: One example of how sample for stiffness measurements was made

Samples were also analyzed in the optical microscope (Zygo New View 7300) before and after the compressive test to see how friction materials behaved. Since there was a lot of samples during the compressive test, there was no time to do topography measurements for each sample, but five samples from each disc were measured (one before the compressive test, one after). See AppendixBfor results of topography measurements.

It is appropriate to do statistical analysis from the result of the compressive test. It helps to create a better understanding of how sample values varying on the friction disc, and also variation between discs.

Box plot, is a graphic representation method used in data analysis. Standard deviation is used to quantified the amount of variation or dispersion of a set of data values. A low standard diversion indicates that the data points tend to be close to the mean of the set, while a high standard deviation indicates that a data points are spread out over a wider range of values.

(25)

Chapter 4 Results and Discussion

In this chapter, results from experimental methods are presented and discussed shortly.

4.1 Surface characterization.

The results from the surface characterization were inconclusive due to large variation in the surface profile from sample to sample. Every sample had a unique surface profile.

See appendix B for the result of the surface characterization. It would be very interesting to see how the surface profile looks like after doing rig-test. At the moment it is hard to make a conclusion about generated vibration only by the result of surface characterization. By doing surface characterization after doing rig-test, it will help to understand the behaviour of friction material.

18

(26)

19

4.2 Pressure distribution measurement

The pressure distribution tests show that the edges on the outside of all friction discs carry less load than the inner side of the friction discs. This might depend on how discs are manufactured. The transport band during manufacturing makes a lot of vibrations that, when the metal powder falls on the disc the powder will fall from the edges before they make it to the oven. The result of the pressure distribution of the disc D is a good example that shows edges on the outside of the disc carries less load than the inner side of the friction disc. Especially near around black mark, see Figure4.1. See appendix A for the result of the pressure distribution measurement.

Figure 4.1: Pressure distribution DG, calculation model 2000 Kg load.

(27)

20

The result of the review by the 10 students was different from the expected. Discs classified as ”non-working” discs was best ranked. See Table4.1.

Table 4.1: Result from review, ranking 1 is the best.

Disc Rank

C 1

B 2

A 3

D 4

F 5

E 6

The results from the topography explain the appearance of the pressure distribution.

Sample BG1 proves assertion about how topography and pressure distribution correlates.

Table 4.2 shows the correlation between pressure distribution appearance and surface topography. See Table4.2.

Sample Topography Pressure Distribution

BG1

Table 4.2: Shows how pressure distribution and surface topography connects.

There are several other examples of samples that prove there is a correlation between topography and pressure distribution appearance. See Table4.3.

(28)

21

Sample Topography Pressure Distribution

EG3

CG3

CG8

DG3

Table 4.3: Shows how pressure distribution and surface topography connects.

It is not sure that a bad pressure distribution could cause vibrations in the wet clutch.

Since the surface topography explicates pressure distribution appearance, therefore pressure distributions appearance can indicate the difference in the manufacturing process.

It is interesting to see how pressure distribution would look like after they have been run in the wet clutch.

(29)

22

4.3 Young’s modulus of the friction material.

See Table4.4for compressive test results.

Table 4.4: Young’s modulus of the investigated samples, as a comparison values. Unit MPa

Sample/Disc A B C D E F

G1 30.5 22.2 34 23.6 31.7 31.3

G2 18 35.8 35 36.9 31.6 34.8

G3 32.3 36 27.8 33.3 28.3 33.2

G4 34.3 31 35.6 40.6 36 36

G5 25 36.2 38 37.5 27.2 26

G6 18.5 35.3 31 33.6 26.7 27.6

G7 26.9 27.1 28.7 36.4 36.8 35

G8 39.8 36 36.6 24.7 34.6 26.7

R1 27 36.2 33.2 34.7 23 25.7

R2 28.5 41.8 36.5 33.3 26 29.5

R3 31.5 31.7 29.5 36.8 30.4 26.9

R4 41.1 28.1 40.5 41 28.7 34.1

R5 28.4 32.5 27.7 28.5 35.1 36.4

R6 40 27.3 24.3 34.2 29.8 31.3

R7 29.8 29.7 31.5 40.7 34.3 34 R8 38.7 28.6 34.4 34.9 35.7 31.8

Figure 4.2 shows box plot for all six discs (Young’s modulus of friction material). The range between lower bracket and the highest bracket is wider on data values for disc A, B and C compared to D, E and F discs. The standard deviation is also higher on disc A, B and C. See Figur4.2.

(30)

23

Figure 4.2: Box plot grafic representation of data values from compressive test.

One reason for the larger variation on stiffness of discs A, B and might be the divergent appearance of disc A, B and C. Figure 4.3shows the surface of one ”non-working” disc and Figure4.4shows the surface of one ”working” disc. As it seems the ”non-working”

discs have a divergent surface (black dot and darker areas). The squares that have dots involved in the material composition has also higher stiffness. Figure4.3and Figure4.4 demonstrates the typical appearance of the surface.

Figure 4.3: Typical ”non-working”

surface appearance of friction disc

Figure 4.4: Typical ”working” surface appearance of friction disc

Varying stiffness means the density of the material are also varying since stiffness depends on material composition. When the force is applied on the friction disc the result will be varying deformation on samples. When the samples all deforms different, the result of the different deformation could be slanting surface. When the slantwise discs rotates it could generate vibration in the wet clutch.

(31)

Chapter 5 Conclusion

The conclusions of this thesis are:

• Surface topography and pressure distribution:

The surface topography and pressure distribution cannot be easily linked to the friction disc quality. More experiments are needed to approve the link.

• Pressure distribution measurements with Fuji prescale film is a very quick method to indicate differences in the manufacturing quality of the friction discs and it gives a quick picture of the surface without involving any optical microscope. This method could potentially be used as a quality manufacture control.

• Young’s modulus of the friction material:

Since the stiffness had much-varying values on ”non-working” discs this could be the parameter causing vibration. It is important to create a bigger database with measured stiffness (more experiment) to approve this statement.

24

(32)

Chapter 6 Future work

In this work, the main focus was to come up with good evaluation methods for mapping the friction discs parameters and their mechanical properties. Further work should be focused on:

• Comparing results from unused and used friction discs in the wet clutch.

• Doing a compressive test on a whole disc package, before using them in the wet clutch to see vibrations behaviour.

• Rig tests for the investigated friction discs to see at what point vibrations appear;

and how many ”non-working” disc are needed to cause vibrations?

• Deeper study on material composition, create greater understanding about the reason behind varying stiffness.

25

(33)

Appendix A

Pressure Distribution

26

(34)

27

A.1 Disc A

Figure A.1: Pressure distribution AG, calculation model 1200 Kg load.

Figure A.2: Pressure distribution AR, calculation model 1200 Kg load.

(35)

28

A.2 Disc B

Figure A.3: Pressure distribution BG, calculation model 1200 Kg load.

Figure A.4: Pressure distribution BR, calculation model 1200 Kg load.

(36)

29

A.3 Disc C

Figure A.5: Pressure distribution CG, calculation model 1200 Kg load.

Figure A.6: Pressure distribution CR, calculation model 1200 Kg load.

(37)

30

A.4 Disc D

Figure A.7: Pressure distribution DG, calculation model 2000 Kg load.

Figure A.8: Pressure distribution DR, calculation model 2000 Kg load.

(38)

31

A.5 Disc E

Figure A.9: Pressure distribution EG, calculation model 2000 Kg load.

Figure A.10: Pressure distribution ER, calculation model 2000 Kg load.

(39)

32

A.6 Disc F

Figure A.11: Pressure distribution FG, calculation model 2000 Kg load.

Figure A.12: Pressure distribution FR, calculation model 2000 Kg load.

(40)

Appendix B

Topography

33

(41)

34

B.1 Disc A

Sample Before compressive test After compressive test

AG1

AG3

AG4

AR4

AR8

Table B.1: Disc A, Topography measurement before and after compressive test.

(42)

35

B.2 Disc B

BG1

BG2

BR3

BR4

BR5

Table B.2: Disc B, Topography measurement before and after compressive test.

(43)

36

B.3 Disc C

CG1

CG8

CR2

CR6

CR7

Table B.3: Disc C, Topography measurement before and after compressive test.

(44)

37

B.4 Disc D

DG1

DG2

DG3

DR1

DR8

Table B.4: Disc D, Topography measurement before and after compressive test.

(45)

38

B.5 Disc E

EG1

EG8

ER1

ER5

ER6

Table B.5: Disc E, Topography measurement before and after compressive test.

(46)

39

B.6 Disc F

FG1

FG8

FR2

FR5

FR6

Table B.6: Disc F, Topography measurement before and after compressive test.

(47)

Bibliography

[1] PR MARKLUND. Wet clutch tribological performance optimization methods. Re- view of Scientific Instruments, 2008. URL http://ltu.diva-portal.org/smash/

search.jsf?dswid=6938.

[2] BorgWarner. Haldex 5. generation - how it works. 2013. URL https://youtu.be/

CDRVTjMPK9Q.

[3] Michelle Collins. Gripping performance borgwarners awd coupling of- fers enhanced traction for volvo xc40. January 2018. URL https:

//cdn.borgwarner.com/docs/default-source/press-release-downloads/

bw-00474_en_borgwarner_awd_coupling_volvo.pdf?sfvrsn=4f5b03c_2.

[4] Henrik Lundh. Simulation of self induced oscillations in four wheel drive train. pages 9–10, May 2011. URL urn:nbn:se:ltu:diva-55576.

[5] Martin Sderlund Fredrik hman. Simulering av hydrauliken i haldex-kopplingen.

page 5, Februari 2005. URL http://www.diva-portal.org/smash/get/diva2:

20157/FULLTEXT01.pdf.

[6] JR DAVID G. RETHWISCH WILLIAM D, CALLISTER. Materials science and engineering. Department of Metallurgical Engineering, (9th Edition):204–205, October 2013.

[7] Andreas Luttge Mark R. Wiesner Ismail Koyuncu, Jonathan Brant. A comparison of vertical scanning interferometry (vsi) and atomic force microscopy (afm) for char- acterizing membrane surface topography. Journal of Membrane Science 278 (2006) 410417, pages 411–412, November 2005. URL www.sciencedirect.com/science/

article/pii/S0376738805008355.

40

(48)

Bibliography 41

[8] Process metrology Veeco. Vertical scanning interferometry. pages 1–

5. URL www3.kau.se/kurstorg/files/i/C10B9B671695d1B009NvFF7EA095/

interferometri.ppt.

[9] FUJIFILM Corporation. Structure. 2018. URL http://www.fujifilm.com/

products/prescale/prescalefilm/#features.

Evaluation Methods for Friction Materials in a Four-Wheel Drive Drivetrain