8/13/2019 607279

1/12

Hindawi Publishing CorporationISRN TribologyVolume 2013, Article ID 607279, 11 pageshttp://dx.doi.org/10.5402/2013/607279

Research ArticleFriction Behavior of a Wet Clutch Subjected to AcceleratedDegradation

Agusmian Partogi Ompusunggu,1

Thierry Janssens,2

and Paul Sas2

1Flanders Mechatronics Technology Centre (FMTC), Celestijnenlaan 300D, 3001 Heverlee, Belgium

2Division PMA, Department of Mechanical Engineering, Katholieke Universiteit Leuven (KU Leuven), Celestijnenlaan 300B,3001 Heverlee, Belgium

Correspondence should be addressed to Agusmian Partogi Ompusunggu; agusmian.ompusunggu@gmail.com

Received 19 November 2012; Accepted 10 December 2012

Academic Editors: D. Das, M. Dienwiebel, and J. Wang

Copyright 2013 Agusmian Partogi Ompusunggu et al. is is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

is study aims at experimentally investigating the sliding friction characteristics of a wet clutch during its lifetime. More precisely,the objective is to understand how the Stribeck and the frictional lag (i.e, sliding hysteresis) parameters evolve as the clutchdegradation progresses. For this purpose, a novel test procedure has been proposed and a set of experiments has been carriedout on a fully assembled (commercial) clutch using a modied SAE#2 test setup. Furthermore, a systematic methodology for theStribeck and the frictional lag parameters identication is developed. Regardless of the applied pressure, it appears that the rstthree identied Stribeck parameters tend to decrease with the progression of the degradation, while the last parameter tends toincrease. In regard to the frictional lag parameter, the trend shows pressure dependency.

1. Introduction

Adhesive wear and thermal degradation are the main agingsources of clutch friction materials, which are unavoidablypresent when clutches are in operation. e dominance ofthese aging sources is determined by many factors, such asthe used friction material, oil, and operational condition,

Regarding the sliding friction, the characteristics can beclassied into two categories, namely, (1) the stationaryfriction characteristic and (2) the dynamic friction charac-teristic [1, 2]. As discussed in the literature, the stationaryfriction characteristic is typied by the Stribeck curve whilethe dynamic one is typied by the frictional lag. To theauthors knowledge, how the Stribeck curve andthe frictionallag evolve with the progression of the friction materialdegradation is not fully understood yet. No articles have beenfound in the open literature that address this issue.

is study aimsat experimentally investigatingthe typicalsliding friction characteristics of a wet friction clutch duringits lifetime. More precisely, the objective of the study is

to understand how the Stribeck and the frictional lag (i.e.,sliding hysteresis) parameters evolve as the degradationprogresses. A profound understanding of the evolution of theStribeck parameters and the sliding hysteresis loop duringclutch lifetime may allow to model the evolution of the clutchfriction characteristics during the lifetime. is model canthen be integrated to a clutch model such that simulations

of the dynamic engagement behavior of the clutch with theprogression of the friction material degradation is possible.Eventually, the gained knowledge can lead to the derivationof physical features, which are useful for developing a clutchmonitoring and prognostic system.

For this purpose, a novel test procedure has been pro-posed and applied on a fully assembled (commercial) clutchtested on an SAE#2 test setup. e accelerated life test (ALT)concept is utilized to accelerate the progression of the clutchfriction material degradation, wherein two additional tests,namely, (i) stationary Stribeck test and (ii) dynamic Stribecktest, are performed between predened number of dutycycles. is way, important sliding friction characteristics

8/13/2019 607279

2/12

2 ISRN Tribology

of clutches can be systematically evaluated. e used clutchconsists of a predened number of commercial friction discsand separator discs that are lubricated with a commercialATF.

e remainder of this paper is organized as follows.e experimental aspects comprising the test setup used in

the study and the experimental procedures are presentedand discussed in Section 2. e results obtained from theexperiments are then presented and discussed in Section 3,where the parameter identication technique and the identi-cation results are also discussed in thissection. Finally, someimportant conclusions drawn from the study are summarizedin Section 4.

2. Experimental Aspects

is section begins with some details of the SAE#2 test setupused in the study. An overview concerning the used frictiondisc, the separator disc, and the ATF is also given. Finally,

the experimental procedures carried out in the study are thendiscussed in the last part of this section.

2.1. SAE#2 Test Setup. According to the standard of Societyof Automotive Engineer (SAE) (i.e., SAE ) [3], anSAE#2 test setup is used to evaluate the friction character-istics of automatic transmission clutches with automotivetransmission uids (ATFs). It can also be used to conductdurability tests on wet friction clutch systems and to evaluatethe performance variation as a function of the number ofduty cycles. Normally, a typical SAE#2 test setup is equippedwith a ywheel driven by an electric motor and the kineticenergy of this wheel is dissipated in a tested clutch [4]. Asthe main objective of the current study is to perform thestationary and dynamic Stribeck tests that will be discussedin the forthcoming paragraphs, the presence of a ywheelin an SAE#2 might lead to difficulties in conducting bothtests, where the rotational speed must be controlled stepwise.Considering this reason, the SAE#2 test setup used in thisstudy is therefore modied without any ywheel.

Figure 1 shows the photograph and the schematic draw-ing of the SAE#2 test setup used in this study. e hydraulicpump() provides enough oil ow from the oil tank(7) forthe lubrication and actuation with the main pressure of 25bar. e safety valve (10) protects the proportional valve (11),which is connected to the actuation line of the clutch (5).

Before entering the clutch, the applied pressure is measuredby the pressure sensor(1), and the signal is sent to the dataacquisition system(17). e overload ow of the safety valveis used for the clutch lubrication. e outgoing ow of the oilfrom the clutch is fed back to the tank.e temperature of theoutgoingoil ow is measured by a thermocouple (15),andthesignal is sent to the data acquisition system. e clutch itself ison one side xed to the frame through a torque measurementsystem (6), while the other side is connected, via a exiblecoupling(), to a DC-motor () which is controlled by anexternal drive (1). e sha velocity of the motor is measuredby an encoder (3), and the signal is sentto the data acquisitionsystem.

2.2. Materials. As the torque capacity is limited by the designof the test setup, the clutch used in the study only containsone active friction disc, that is, two frictional contacts insteadof ten frictional contacts recommended by the design. elining material of the friction disc is paper-based type andcommercially available. Moreover, the friction disc has a

waffle groove pattern with the inner diameter of 99mmand the outer diameter of 133 mm. e commercial ATF

used for the clutch has a kinematicviscosity of 56 mm/s (cSt)

and 9.3 mm/s at 40C and 100C, respectively.

2.3. Experimental Procedure. Figure 2 shows the owchartdescribing the experiments carried out in the study. ereare four consecutive tests to be carried out that are discussedin the subsequent paragraphs. Given a fresh clutch, theexperimentis rst started with therun-intest,then continuedwith the accelerated life test (ALT), the stationary Stribecktest, and nally with the dynamic Stribeck test. ese testsequences are repeated until the clutch is considered to have

failed. In this approach, all the test parameters are designedsuch that the clutch would fail in a reasonable period of time.It should be mentioned here that the stationary and dynamicStribeck tests are carried out when the oil temperature is inthe range of 8090C.

2.3.1. Run-In Test. e run-in test designed for this study,in principle, mimics the engagement process of a clutchin automatic transmissions. While the clutch is in opencondition (i.e., disengaged phase), the sha velocity of themotor is controlled at a certain value. When this desired

velocity is attained, a certain current signal prole is sent tothe proportional valve such that a desired pressure prole,for example, see Figure 3, which is discussed in the followingparagraphs, is built up to actuate the clutch. e run-in test iscarried out for 200 engagement cycles.

e pressure prole is designed as follows. Initially,there is no electric current signal sent to the proportional

valve, such that the clutch pressure is zero (1 = 0) andconsequently no torque is transmitted. In order to actuate theclutch, the oil ow to the clutch is suddenly increased causingthe pressure to increase to the level of, thus allowing thepiston to move from its rest position. is pressure level ismaintained for a certain duration and aerwards the owis decreased. As a result, the pressure drops to the pressurelevel of3, which is intended to decelerate the piston motion

and to allow the piston to standstill. is condition should beselected such that the discs do not make contact yet with eachother. e previous sequence is referred to as thelling phase.In this particular phase, the pressure level3is chosen suchthat the elastic force of the returning spring is overcome, butthe pistonshould notmove at highvelocitywith respectto thediscs. Since the platesdo not make contact yet, the torque thatis already being transferred is only due to the viscous effect.e phase following the lling phase is referred to as theengagement phase, where the pressure is gradually increasedlinearly () until the maximum pressure is reached, so thatmore and more discs are pressed together. In this study, themaximum pressure is set to 8 bar.

8/13/2019 607279

3/12

ISRN Tribology 3

6 5

3

2

(a)

1

2

3

45

6

7

8

9

10

11

12

13

14

15

16

17

PC

V/A

D/AA/D

(b)

F 1: odied SAE#2 test setup without a ywheel.

Start

Run-in test

ALT

integer?

StationaryS

tribeck test

DynamicStribeck test

observed?

Stop

No

Yes

No

Yes

Failure

F 2: A owchart describing the test sequences.

e control mode of the motor is a velocity control inwhich the maximum torque of the motor is limited to acertain value. In this study, the velocity is set to 1200 rpmand the maximum torque is set to 20 Nm during the test. Asthe applied pressure increases, the resulting friction torqueincreases and the motor delivers more torque to maintain thedesired velocity. However, since the maximum torque of themotor is limited, the motor will decelerate when the resultingfriction torque exceeds the limit torque and eventually willcome to a standstill. e procedure proposed in this paperdiffers from the standard procedure in an SAE#2 test setupas discussed in [4], where the electric motor is switched offduring the engagement. Here, the standard procedure is not

possible to implement because of the absence of a ywheelon the test setup. As the inertia is very low on the test setup,the input sha will directly stop when the motor is switchedoff. An additional advantage of the motor torque being keptconstant during the test is that there is more energy dissipatedin the clutch, so that the lifetime of the clutch could bereached in a shorter time.

2.3.2. Accelerated Life Test (ALT). In principle, the acceler-ated life test (ALT) is carried out in the same manner as therun-in test discussed previously. However, the energy appliedduring the ALT is higher than that in the run-in test. For thispurpose, the sha velocity of the motor is set to 2000 rpm.

8/13/2019 607279

4/12

4 ISRN Tribology

F 3: A typical pressure prole applied in the run-in test andALT.

2.3.3. Stationary Stribeck Test. e main objective of thestationaryStribeck test is to determine the stationary Stribeck

curve as a function of the clutch lifetime. It is important tonote here that the stationary Stribeck curve is derived from aset of friction torques measured at constant relative velocityand constant applied pressure. In this study, the stationaryStribeck curve is determined for the applied pressure rangingfrom 4 to 7 bar and for the relative velocity ranging from 20to 1970 rpm. e pressure step is 0.5 bar and the velocity stepis 50 rpm.

For a given pressure level, the velocity is increased step-wise from 20 to 1970 rpm, where the duration for a given

velocity level is set to 5 s. Aer the friction torques for theentire velocity range have been measured, the pressure isincreased by 0.5 bar, andin the new pressure level the velocity

is restarted from 20 to 1970 rpm. is procedure is repeateduntil the maximum pressure level, that is, 7 bar, is attained.

2.3.4. Dynamic Stribeck Test. e dynamic Stribeck test aimsat determining the dynamic frictional behavior of the clutchin nonstationary conditions as a function of the clutchlifetime. As mentioned previously, the dynamic frictionalbehavior is typied by the frictional lag phenomenon. henthe relative motion is accelerated the resulting friction torqueis higher than the friction torque in stationary condition. Incontrast, when the relative motion is decelerated, the result-ingfriction torqueis lower than the stationaryfriction torque.For a sinusoidal relative motion where acceleration and

deceleration are evident, a loop is observable around the sta-tionary Stribeck curve. e loop area is strongly determinedby the frequency of the motion. In the GMSfriction modelingframework [5, 6], the frictional lag phenomenon is deter-mined by the attractor parameter. e higher the attractorparameter is, the faster the resulting friction torque willfollow the stationaryStribeck curve, that is, smaller loop area.

e dynamic Stribeck test is carried out at differentconstant pressures ranging from 4 to 7 bar. For a givenpressure level, a sinusoidal velocity prole having frequencyof 0.1 Hz, amplitude, and offset of 1000 rpm, respectively, isimposed. e total duration of the dynamic Stribeck test fora given velocity prole is 15s.

3. Results and Discussion

is section discusses the results of the stationary anddynamic Stribeck tests. From the tests, the friction torque as afunction of three variables, namely, (i) velocity, (ii) pressure,and (iii) degradation level (i.e., number of duty cycles) areobtained. e experimental results of the stationary Stribecktests are rst discussed. Aerwards, the Stribeck parametersidentied at different degradation levels and their evolutionsduring the clutch lifetime are discussed. Aer addressing thestationary Stribeck characteristics and the evolution of theStribeck parameters, the dynamic Stribeck behavior, which ischaracterized by the attractor parameter, is then discussed.Furthermore, the characteristics of the attractor parameteridentied at different pressures, frequency excitations, anddegradation levels are addressed.

For comparison purposes, photos, and surface prolesof the active friction disc are taken from the same waeblock (see Figure 4) prior and aer the test. e photos arecaptured using an optical microscope and the surface prolesare measured along the sliding direction using a talysurfprolometer. At the same wae block, the surface proles aremeasured on eight different radial locations before and aerthe tests. e representative surface proles of the wae blockbefore and aer the tests are depicted in Figure 5. e averagesurface roughnessfrom the surface prole measurementsbefore the tests is 9.52m and aer the tests itis 1.94m.eaverage skewnessskbefore the test is0.87and aer the testit is3.4.

Figures 4 and 5 clearly show that the friction material hasbecome smooth and at aer the tests, which indicates thatthe friction material has degraded during the tests. One can

also notice in Figure 4 that the surface of the friction materialbefore the tests is more porous compared to aer the tests.e reduction of the surface porosity is believed to be causedby the pores blocking [7, 8], resulting from the deposition ofdebris particles of the friction material and/or degradationproducts of the ATF.

3.1. Stationary Stribeck

3.1.1. Experimental Results. Figure 6 shows the stationaryStribeck curve measured aer the run-in test (i.e., initialcondition), which is plotted both in 2D and 3D. From the2D Stribeck curve shown in Figure 6(a), the viscous effect

dominates the friction torque at lower pressure (i.e., 4 bar),where the friction torque slightly increases with the velocity.At higher pressure, a different behavior can be observedin the gure, where the friction torque increases at lower

velocity (positive slope) and decreases for higher velocity(i.e., Stribeck effect). e positive slope exhibited at lower

velocity indicates that the clutch system has an anti-shudderproperty, which is possibly activated by the presence offriction modier additive in the used ATF. In addition, itis also clear from the gure that the slope of the curvesat lower velocity becomes more positive for higher appliedpressure implying that the antishudder property becomesmore pronounced at higher pressure.

8/13/2019 607279

5/12

ISRN Tribology 5

(a) (b)

F 4: Photographs of the friction disc (a) before and (b) aer the tests.

0 1 2 0 0.05 0.1

20

0

20

40

(a)

20

0

20

40

0 1 2 0 0.2 0.4

(b)

F 5: Surface proles and the distributionsof the friction disc (a) before and (b) aer the tests.

e evolution of the Stribeck curve with the progressionof degradation is shown in Figure 7. Figure 7(a) shows theevolution of the Stribeck curve for all pressures, and Figure7(b) shows the evolution of the Stribeck curve at 5 bar. eStribeck curves represent the measurements obtained fromthe initial state, aer 1000, 7000, and 15000 duty cycles ofALT. e condition aer 15000 duty cycles is considered tobe fully degraded state, where the mean coefficient of friction(COF) has been reduced to 50%from the initial mean COF[9].

It is evident in Figure 7 that the overall friction torque

decreases with the progression of the clutch degradation.A drastic decrease of the friction torque is observed in thebeginning of the clutch lifetime and the friction torque tendsto saturate when the clutch degradation proceeds further.is implies that the overall (i.e., averaged) clutch coefficientof friction (COF) decreases with the progression of theclutch degradation, which is consistent with the experimentalobservations reported in the literature [4, 10]. In many worksof literature, the decrease of the COF is widely accepted tobe caused by the loss of the surface porosity of the frictionmaterial (see Figure 4). Low surface porosity implies that theATF is not easily evacuated from the contacting surfaces,so that the contact between the friction and separator discs

is predominantly controlled by the ATF. Nevertheless, thethermal degradation occurring in the friction material canalso contribute to the reduction of the COF [11].

One can also notice the effect of the clutch degradationon the slope of the Stribeck curve in Figure 7(b). e positiveslope at lower velocity, that is,the antishudder property, is lostdue to the clutch degradation and the slope becomes morenegative at lower velocity with the progression of the clutchdegradation.

In order to characterize the

evolution of the stationary Stribeck curve, the parametersgoverning the curve, that is, the Stribeck parameters, areidentied at different degradation levels. In the followinganalysis, the rotational velocity in rpm, which is obtainedfrom the tests, is converted to the sliding velocity by thefollowing equation:

2

60 , (1)

with +/2 denoting the mean radius of the frictiondisc. At a given state (i.e. degradation level), the Stribeck

8/13/2019 607279

6/12

6 ISRN Tribology

0 500 1000 1500 2000

0

25

50

(a)

4

5.5

7

0 500

1000 1500

2000

0

25

50

(b)

F 6: e stationary Stribeck curve measured at initial condi-tion. (a) e 2D Stribeck curve and (b) the 3D Stribeck curve.

curve as a function of pressure and sliding velocity canbe modeled according to the following equation [12]:

=sign () 1 + 1 |()|

+ ||

(2)

where()is the static friction torque,()is the ratiobetween the Coulomb friction torque and the static frictiontorque, () is the Stribeck velocity, and() is the viscousconstant at pressure.

In [12], it has been shown that the typical value ofthe parameter is 0.3. By keeping this parameter constant,the complexity of the model in (2) is reduced from Stribeck parameters to four Stribeck parameters. Hence,the shape of the Stribeck curve is now determined by the

four parameters,(),(),(), and(), whichare pressure dependent. ese pressure dependencies aredetermined from the experimental data, as will be shownlater.

Since the function expressed in (2) is highly nonlinearand in order to guarantee that the parameters to be identiedhave physical meaning, that is, the Stribeck parameters mustbe positive, the parameter identication thus constitutesa constrained nonlinear optimization problem. Here, theidentication is performed separately for different pressures

4

5.5

701000

2000

0

25

50

(a)

0 1000 2000

10

20

30

0

1000

7000

15000

(b)

F 7: e stationary Stribeck curves measured for differentdegradation levels. (a) the evolution of the 3D Stribeck curve and(b) e evolution of the 2D Stribeck curve at 5 bar. Note that thearrow indicates the progression of degradation.

and states (i.e., degradation level). At a certain pressure andstate, the Stribeck curve is therefore expressible as

()=sign () 1 + 1 ||+ ||

0.3 .(3)

us, the optimization problem can now be formulatedas

minimize

0()

subject to 0

(4)

where0() represents the objective function, that is, thepercentage mean-square-error (S), which is dened by

0()= 1

2exp

=1

() exp2 100 [%] (5)

with = [ ] denoting a vector containingthe Stribeck parameters at a certain state and pressure,expdenoting the measured torque at sliding velocity,2expdenoting the variance ofexp, anddenoting the number

8/13/2019 607279

7/12

ISRN Tribology 7

0 3

6 9

12

4

5.5

7

0

25

50

F 8: Comparison between the Stribeck curve measured atinitial condition and the model.

of the velocity steps. e MSE can be viewed as the goodnessof t between the experimental data and the model, where, asa rule of thumb, the tness is good when the MSE is smallerthan 5%.

For the Stribeck parameters identication, the interior-point algorithm [13, 14] is used to solve the optimizationproblem formulated in (5). Figure 8 shows the comparisonbetween the 3D Stribeck curve (the circlemarkers), measuredat initial condition, andthe modeled curve (surface) obtainedbased on the aforementioned identication procedure. eaverage MSE obtained from the identication of the Stribeckcurve measured at initial condition for different pressures is1.74%and the maximum MSE is 3.61%.

Figure 9 shows the pressure dependence of the Stribeckparameters. e circle markers denote the identied valuesfor different pressures and the solid lines denote the trends.It can be seen that the three Stribeck parameters (,,and) increase linearly with the applied pressure, whilethe parameterdecreases asymptotically with the appliedpressure. e pressure-dependent relationships of theStribeck parameters are based on the formulations discussedin [12].

Figure 10 shows the evolution of the Stribeck parametersidentied at a pressure of 5 bar (circle markers) in functionof the ALT duty cycles and the general trend (the solid lines)estimated using the degradation model expressed as [15]

0+ 1 11 + , (6)

with

denoting a Stribeck parameter at an arbitrary state,

0denoting the Stribeck parameter at an initial state,,, andbeing the degradation model coefficients that, respectivelydenote the maximum deviation from0, the scaling factor ofthe abscissa, and the characteristic exponent, anddenotinga variable that represents the progress of degradation, forexample, number of cycles or sliding distance.

For this particular pressure, one can see that the threeparameters, namely,, and, tend to decrease withthe progression of clutch degradation. In contrast, the viscousparameter tends to increase with the progression ofclutch degradation. e increasing trend of implies thatthe viscous effect becomes more dominant as the clutchdegradation progresses. is tendency is consistent with the

visual observation on the surface of the friction disc aerthe tests (see Figure 4), where the contact surface has lost itsporosity, so that, as mentioned before, the resulting frictiontorque is predominantly controlled by the ATF property.

Furthermore, Figure 11 shows the overall evolutions ofthe identied parameters (circle markers) and the trends

(surfaces) at different pressures. One can clearly see in thegure that the Stribeck parameters have similar trends infunction of the number of duty cycles when observed atdifferent pressures.

3.2. Dynamic Stribeck Tests. As mentioned previously, thedynamic Stribeck test in this study is carried out by applyingan imposed sinusoidal velocity prole with a frequencyof 0.1 Hz, where the pressure is kept constant during the

velocity excitation. In this way, the effects of accelerationand deceleration on the resulting friction torque can beobserved. Due to acceleration and deceleration, the frictiondeviates with respect to the steady-state behavior, that is,

the stationary Stribeck curve, thus resulting in the formationof a hysteresis loop which is located around the stationaryStribeck curve, that is referred to as the frictional lag effect.

Figure 12 shows the dynamic Stribeck curve measured at5 bar, where the effects of acceleration and deceleration onthe resulting friction are visible. It can be seen in the gurethat the resulting friction is higher during the accelerationthan that during the deceleration. e physical explanationfor this phenomenon is widely associated with the squeezelm effect [16]. As discussed in [16], the squeeze lm forcesqon a lubricated line contact is given by

sq

2

3 ,

(7)

with denoting the effective dynamic viscosity,denotingthe length of line contact, denoting the half width ofHertzian contact, anddenoting the average lm thickness.Duringthe acceleration, thelm thickness increases()implying that the squeeze force is negative. In other words,the squeeze force acts as an external load during the accel-eration. As a result, the lm thickness during acceleration isless than that during stationary condition. is suggests thatthe number of asperity-to-asperity contacts becomes higherin the acceleration with respect to the stationary condition,thus resulting higher friction. In contrast, the lm thickness

decreases during the deceleration phase, implying that thesqueeze pressure becomes positive which contributes to theload-carrying capacity. is results in a higher lm thicknesscompared with the corresponding value in the stationarylubrication, which translates to a lower friction. A measurethat represents the frictional lag effect can be simply denedas the loop arealagof the friction curve, which is expressibleas

lag , (8)with and, respectively, denoting the torque and the

velocity.

8/13/2019 607279

8/12

8 ISRN Tribology

4 5.5 7

10

30

50

(a)

4 5.5 7

0

0.4

0.8

(b)

4 5.5 7

0

2.5

5

(c)

4 5.5 7

0

3.5

7

(d)

F 9: e ideied Sribec araeers a iiial codiio i cio o alied ressre.

1 5000 10000 15000

22

25

28

(a)

1 5000 10000 15000

0.1

0.3

0.5

(b)

1 5000 10000 15000

0.5

1.5

2.5

(c)

1 5000 10000 15000

0

1.5

3

(d)

F 10: e eolio o e ideied araeers o e Sribec cre easred a alied ressre o bar i cio o e dycycles.

8/13/2019 607279

9/12

ISRN Tribology 9

4

5.5

7

0

5000

10000

15000

0

25

50

(a)

4

5.5

7

0

5000

10000

15000

0

0.4

0.8

(b)

4

5.5

7

0

5000

10000

15000

0

3

6

(c)

4

5.5

7

0

5000

10000

15000

0

2

4

(d)

F 11: e overall evolutions of the identied parameters of the Stribeck curves measured at different pressures in function of the dutycycles.

0 500 1000 1500 2000

18

24

30

Deceleration

Acceleration

F 12: Dynamic Stribeck curve of the fresh clutch measured at5 bar.

e effect of the applied pressure on the frictional lagphenomenon at initial condition (fresh clutch) is shownin Figure 13. Visually, one may notice that the loop areaincreases with an increase of applied pressure. is obser-

vation is qualitatively identical to the simulation resultspresented in [16. As the pressure increases, the average lmthickness decreases. As revealed by (7), the squeeze lm forceis inversely proportional to the average lm thickness. issuggests that the squeeze lm effect is more pronounced athigher pressure. Hence, the loop area lag becomes largerat higher pressure, which can be clearly seen in Figure 14.oreover, the gure also depicts that the loop area increasesquite proportionally to the applied pressure.

0 500 1000 1500 2000

0

25

50

F 13: Dynamic Stribeck of the fresh clutch at different appliedpressures.

Figure 15 shows the dynamic Stribeck measurementsduring the clutch lifetime at the applied pressure of 5 bar. Ascan be seen in the gure, the friction torque decreases withthe progression of the clutch degradation, which is consistentwith the stationary Stribeck measurements. Interestingly, onecan see that the loop area increases during the clutch lifetimeat this particular applied pressure, which is more visible inFigure 16. e possible physical explanation is as follows. Ithas been mentioned in the previous section that the frictionmaterial surface becomes smoother with the progression ofclutch degradation (see Figure 5). For a rough surface, theasperities have a greater contribution in carrying the loadwhich suggests that the squeeze effect does not play any

8/13/2019 607279

10/12

10 ISRN Tribology

4 5.5 7

1400

1800

2200

F 14:e loop area resultingfromthe frictionallag in functionof applied pressure calculated based on the friction curves in Figure13.

signicant role in a rough surface if compared to a smoothsurface [16].

It is now interesting to observe how the loop arealag evolves during the clutch lifetime for different appliedpressures. Surprisingly, the trends of lag are exceptional athigher pressures. As can be seen in Figure 17,lagincreaseswith the progression of the clutch degradation at relativelylower pressure (46 bar). On the other hand, the evolutionof lag exhibits an opposite behavior at relatively higherpressures, where it tends to decrease with the progressionof the clutch degradation. e possible physical explanationfor this evidence is as follows. It is assumed that the surfaceporosity plays a signicant role at higher pressure on thesqueeze lm due to the dynamic conditions. As the clutchdegradation progresses, the friction material looses its surface

porosity, namely, more surface pores are blocked (see Figure4), thus resulting in the deterioration of the clutch capabilityto evacuate the oil from the contacting surfaces. is suggeststhat the average lm thickness does not necessarilydecrease for degraded friction material when the appliedpressure is increased. In addition, the loss of surface porositymay also suggest that the rate of change of the lm thickness/ at higher pressure becomes less pronounced in thepresence of acceleration or deceleration.

4. Conclusion

e stationary Stribeck tests reveal that the stationaryfriction

curve drops globally as the friction material degradationprogresses. Four parameters, namely, (i) the static frictiontorque , (ii) the ratio between the Coulomb friction

torque and the static friction torque , (iii) the Stribeckvelocity, and (iv) the viscous effect, are referred to asthe Stribeck parameters that govern the stationary frictioncharacteristics, that is, Stribeck curve, at a certain state.In order to characterize the change of the sliding frictioncharacteristics, the four parameters need to be identiedat different states. e identication results show that theStribeck parameters appear to evolve deterministically during

the clutch lifetime. e rst three parameters (,, and) tend to decrease with the progression of the degradation,

0 500 1000 1500 2000

10

20

30

0

1000

7000

15000

F 15: e evolution of the dynamic Stribeck curve at appliedpressure of 5 bar.

0 5000 10000 15000

1500

2000

2500

F 16: e evolution the frictional lag loop during the clutchservice-life at applied pressure of 5 bar.

while the last parameter () tends to increase. Increasingimplies that the viscous effect becomes more pronouncedas the clutch degradation progresses. is implication isconrmed by the visual observation on the surface of the fullydegraded friction material. Based on the observation, it isevident that the degraded friction surface has lost its porosity.us, the ability of the degraded friction surface to evacuatethe lubricant from the interface deteriorates, suggesting thatthe friction torque occurring in the clutch is predominantlycontrolled by the lubricant.

Under acceleration or deceleration, the resulting friction

will deviate from the stationary friction (stationary Stribeckcurve). To investigate this effect, the dynamic Stribeck testshave been carried out, wherein a sinusoidal velocity exci-tation is applied on the clutch. In this way, both accelera-tion and deceleration effects are present simultaneously inthe measured friction, eventually creating a hysteresis looparound the stationary curve that is called the frictional lageffect. e loop area, which is simple to compute, can be con-sidered as a measure of the frictional lag effect. Furthermore,it has been shown in this study that the applied pressure hasalso an effect on the loop area of the sliding hysteresis. In thebeginning of clutch lifetime, the loop area increases with theapplied pressure. However, this characteristic changes with

8/13/2019 607279

11/12

ISRN Tribology 11

4

5.5

7

0

5000

10000

15000

1500

2000

2500

F 17: e evolution the frictional lag loop during the clutchservice-life at different applied pressures.

the degradation progression, where the hysteresis loop areatends to decrease with the applied pressure at the end of theclutch lifetime.

Acknowledgments

All the authors would like to thank Tycho Van Peteghem andWout Vandelaer for performing the experiments. Valuablecomments of Professor Hendrik Van Brussel and ProfessorFarid Al-Bender on this study are appreciated. e experi-mental support from Dr. Mark Versteyhe of Dana-Spicer OffHighway Belgium is also acknowledged.

References

[1] F. Al-Bender, V. Lampaert, and J. Swevers, A novel generic

model at asperity level for dry friction force dynamics, Tribol-ogy Letters, vol. 16, no. 1-2, pp. 8193, 2004.

[2] T. Tjahjowidodo, Characterization, modelling and control ofmechanical systems comprising material and geometrical nonlin-earities [Ph.D. thesis], Katholieke Unversiteit Leuven, Depart-ment of Mechanical Engineering, Division PMA, Leuven,Belgium, 2006.

[3] SAE-International, SAE No. 2 Friction Test Machine Durabil-ity Test, Vol. SAE J2489, 2012.

[4] W. Ost, P. De Baets,and J.Degrieck,e tribologicalbehaviourof paper friction plates for wet clutch application investigatedon SAE#II and pin-on-disk test rigs, Wear, vol. 249, no. 5-6,pp. 361371, 2001.

[5] V. Lampaert, F. Al-Bender, and J. Swevers, A generalizedmaxwell-slip friction model appropriate for control purposes,inProceedings of the IEEE International Conference on Physicsand Control, 2003.

[6] F. Al-Bender, V. Lampaert, and J. Swevers, e generalizedMaxwell-slip model: a novel model for friction simulation andcompensation,IEEE Transactions on Automatic Control, vol.50, no. 11, pp. 18831887, 2005.

[7] S. Li,M. Devlin, S. H. Tersigni, T. C. Jao,K. Yatsunami, andT. M.Cameron, Fundamentals of Anti-Shudder Durability: Part I-Clutch Plate Study, SAE Technical Paper 2003-01-1983, 5162,2003.

[8] T. Newcomb, M. Sparrow, and B. Ciupak, Glaze Analysis ofFriction Plates, SAE Technical Paper 2006-01-3244.

[9] T. VanPeteghem and W. Vandelaer, Levensduurvoorspellingvan Transmissiesystemen voor Off-road Voertuigen [M.S. thesis],Katholieke Unversiteit Leuven, Department of MechanicalEngineering, Division PMA, Leuven, Belgium, 2011.

[10] J. Fei, H.-J. Li, L.-H. i, Y.-W. Fu, and .-T. Li, Carbon-berreinforced paper-based friction material: study on friction sta-bility as a function of operating variables,Journal of Tribology,vol. 130, no. 4, Article ID 041605, 2008.

[11] A. P. Ompusunggu, P. Sas, and H. VanBrussel, Inuenceof adhesive wear and thermal degradation on the frictionalcharacteristics of paperbased friction materials: a comparativestudy,ISRN Tribology, vol. 2013, Article ID 739202, 2013.

[12] T. Janssens,Dynamic characterisation and modelling of dry andboundary lubricated friction for stabilisation and control pur-poses [Ph.D. thesis], Katholieke Unversiteit Leuven, Departmentof Mechanical Engineering, Division PMA, Leuven, Belgium,2010.

[13] R. H. Byrd, J. C. Gilbert, and J. Nocedal, A trust region methodbasedon interior point techniques fornonlinearprogramming,Mathematical Programming B, vol. 89, no. 1, pp. 149185, 2000.

[14] R. H. Byrd, M. E. Hribar, and J. Nocedal, An interior pointalgorithm for large-scale nonlinear programming,SIAM Jour-nal on Optimization, vol. 9, no. 4, pp. 877900, 1999.

[15] A. P. Ompusunggu, Intelligent monitoring and prognostics ofautomotive clutches [Ph.D. thesis], Katholieke Unversiteit Leu-ven, Department of Mechanical Engineering, Division PMA,Leuven, Belgium, 2012.

[16] H. Sojoudi and M. M. Khonsari, On the modeling of quasi-steady and unsteady dynamic friction in sliding lubricated linecontact,Journal of Tribology, vol. 132, no. 1, Article ID 012101,9 pages, 2010.

8/13/2019 607279

12/12

Submit your manuscripts at

http://www.hindawi.com