1 Clutch systems and torsional dampers ... - Schaeffler · PDF fileSchaeﬄ er SYMPOSIUM 2010 1 Clutch systems and torsional dampers 12 Schaeﬄ er SYMPOSIUM 2010 13 Clutch systems

Schaeffl er SYMPOSIUM 2010

1 Clutch systems and torsional dampers

8 Schaeffl er SYMPOSIUM 2010 9

1Clutch systems and torsional dampers

1


9Schaeffl er SYMPOSIUM 2010Schaeffl er SYMPOSIUM 20108

LuK clutch systems and torsional dampers

Key elements for effi cient drive trains

Matt hias ZinkMarkus Hausner





1Introducti onUp until a few years ago, engine development showed a strong, consistent trend toward raising the specific engine torque in diesel engines us-ing new injection technologies combined with supercharging. Since the early 1990s, the possi-ble engine torque based on piston displacement has increased by some 300 %. The typical diesel engine at that time had four cylinders and a pis-ton displacement of approximately two liters, with which today’s engines can today generate 450 Nm or more [1]. This evolution in diesel en-gines has, to a large degree, increased the enjoy-ment and efficiency of driving. So that good re-sults as far as comfort can also be achieved without compromise, the requirements for the torsional vibration damper, which must transmit the high torques on the one hand and effectively keep the high alternating torques away from the drive train on the other, have grown by compa-rable degrees.

Up to now, when selecti ng the transmission, the customer mainly had to choose between a manual or an automati c transmission, and in comparati ve-ly rare cases, also a CVT or automated shift trans-mission.

For several years now the pressure has been grow-ing worldwide to reduce fuel consumpti on and CO2 output, while maintaining comfort and enjoyment. According to one study [2], for example, the aver-age fuel consumpti on in Germany will fall from

7.6 liters today at least to 6.1 liters by 2030 (-20 %). With a high degree of hybridizati on and accelerat-ed technology change, such as in alternati ve fuels and drives, it could even fall as low as 5.2 liters (-31 %).

Globally, this leads to a tendency towards small-er and lighter vehicles. The drive train requires massive effort in order to raise the overall effi-ciency.

For the combustion engine, this is done by means of further downsizing, with supercharg-ing now being increasingly used, particularly in gasoline engines. In concrete terms, this means a torque class that has to be covered with the smallest possible engines (Figure 2). 2.0 liter en-gines are replaced, for example, by 1.5 liter en-gines with comparable torque. Small engines with up to 1.3 liters piston displacement are showing strong growth, while engines with over 1.8 liters piston displacement are losing share.

In the drive train, automati on and electrifi cati on are used to enable the combusti on engine to be operated with opti mal loads and speeds.

Types of transmissionsDespite great strides in transmission technology, the manual transmission conti nues to have the highest producti on volume worldwide. The advan-

tages of this transmis-sion technology are relati vely low cost, high effi ciency and good driving performance. It will remain the world’s dominant transmission type for the next sev-eral years, and by 2018, with a projected in-crease of 11 million units, will show the highest growth in abso-lute numbers.

Compared to a manu-al transmission, an au-tomatic transmission represented the only

option up to now across the board for shifting gears “clutch-free.” Up to now, the high comfort level of the automatic transmission was always set against a comparatively high fuel consump-tion, since the torque converter clutch had to be open at low engine speeds for comfort reasons. But good fuel economy can actually be achieved

at low engine speeds. In this regard, damper technology also plays a key role for auto-matic transmissions, since the performance capability of the damper largely deter-mines the engine speed from which it is possible to drive with no slip and therefore optimal fuel consump-tion. LuK has already shown what is techni-cally possible today in that area. Using a double damper and centrifugal pendulum-type absorber (CPA), it is possible to increase the degree of isola-

tion by more than 60 % compared with a conven-tional damper. This creates the conditions re-quired for achieving significant advantages in fuel economy [3].

The desire to combine the advantages of the previ-ous manual transmission with those of the auto-

20002000 20102010 20202020 20302030

1010

88

66

44

22

00

6.16.1

5.25.2

7.67.6

19951995 20052005 20152015 20252025

Fu

el

cF

ue

lco

nsu

mp

on

sum

pti

on

inl/

10

0k

mti

on

inl/

10

0k

m

FFederederal Minisal Ministrtryyof Tof Trransportansport

Shell PShell Passengassenger Car Scenarioer Car Scenario

Year

Figure 1 Predict ed average passenger car fuel consumpti on in Germany

0%0%

20%20%

40%40%

60%60%

80%80%

100%100%

0%0%

20%20%

40%40%

60%60%

80%80%

100%100%

0%0%

20%20%

40%40%

60%60%

80%80%

100%100%

20072007 20152015 20072007 20152015 20072007 20152015

Specific tSpecific tororqueque

dieseldiesel

Specific tSpecific tororqueque

ggasolineasoline

PisPisttonon

displacemendisplacementt

up tup too 120 Nm/l120 Nm/lup tup too 140 Nm/l140 Nm/lup tup too 160 Nm/l160 Nm/lup tup too 180 Nm/l180 Nm/loovverer 180 Nm/l180 Nm/l

up tup too 80 Nm/l80 Nm/lup tup too 100 Nm/l100 Nm/lup tup too 120 Nm/l120 Nm/lup tup too 140 Nm/l140 Nm/lup tup too 160 Nm/l160 Nm/loovverer 160 Nm/l160 Nm/l

oovverer 2,8 l2,8 lup tup too 2,8 l2,8 lup tup too 2,3 l2,3 lup tup too 1,8 l1,8 lup tup too 1,3 l1,3 l

Figure 2 Engine producti on in Europe by piston displacement and specifi c torque (Source: CSM)

505030302020101000 4040

200920092012201220182018

?

1133

66

29293535

4040

111111

3366

77

1919

25252121

ManualManual

trtransmissionansmission

CVTCVT

HybridHybrid

AutAutomaomatteded

manualmanual


AutAutomaomatictic


Double-clutDouble-clutchch


TTrransmissions pransmissions produced, millions per yoduced, millions per yearear

Total change 2009 – 2018: (+49 %)(+49 %)

(+32 %)(+32 %)+6+6

(+133 %)(+133 %)+4+4

(± 0 %)(± 0 %)±0±0

(+? %)(+? %)+?+?

(+500 %)(+500 %)+5+5

(+38 %)(+38 %)+11+11

+26+26 millionsmillions

millionsmillions

millionsmillions

millionsmillions

millionsmillions

millionsmillions

millionsmillions

Figure 3 Global producti on fi gures by transmission type





1mati c transmission led to the development of dou-ble clutch transmissions. This new transmission concept shows how enjoyment, consumpti on and comfort can be combined with impressive results. It is here that the highest relati ve growth will be realized in the next several years.

The next logical step in the ongoing process of au-tomati on and electrifi cati on is hybridizati on. Here, too, the goal is to further decrease fuel consump-ti on through energy management without com-promising comfort or enjoyment. The strength of its market penetrati on and the hybrid concepts with which it will develop cannot be fully foreseen at this ti me. What seems certain, however, is that hybridizati on will begin to establish itself across the board in the model series of all automakers in the years to come.

The contributi on of the transmission to favorable consumpti on values stems in large part from creat-ing the foundati on for operati ng the combusti on engine at a low specifi c consumpti on. To do this, the number of gears and the gear rati o spread are both increased. Automati on ensures that the cor-rect gear is always engaged. With manual transmis-sions, gear selecti on is left to the driver, but even here a visual shift recommendati on can provide positi ve support if the driver feels right in the rec-ommended gear.

Another area where eff orts are being made has to do with mechanical effi ciency. Lower fricti on and light constructi on, however, also bring about a greater sensiti vity to torsional vibrati on, since less mass is moving and vibrati ons are dampened less. Parti cularly with double clutch transmissions, the requirements regarding torsional irregularity sig-nifi cantly increase compared to a manual trans-mission, since a parti al transmission is not in the fl ow of power and is therefore suscepti ble to rat-tling noises. It must therefore be taken in consid-erati on overall that, depending on the starti ng situati on, opti mizati ons in this connecti on come with higher requirements for the torsional vibra-ti on damper.

Opti mal dampers for comfortable driving with low fuel consumpti onIndependent of the transmission concept, vehi-cles, apart from purely electric cars, will still be powered by the combustion engine for the fore-seeable future. This means that the fundamental

requirements at the interface between en-gine and transmission are the same at first for all transmission types. There must be a starter element, a means of transmitting the average torque, and the alternating torques must be dampened as well. The damper require-ments increase with higher torques at low engine speeds and as the number of cylin-ders is reduced. Or to put it another way: The damper require-ments rise with the objective of lowering fuel consumption while keeping driving enjoyment and com-fort the same.

Reducing consumpti on through opti mized dampersA torsional vibration damper in itself does not, of course, directly reduce consumption, but a simple example makes it clear how closely the choice of damper technology is linked with fuel economy. It is based on a current 2-liter diesel engine with about 350 Nm. The vibrations are effectively reduced with an optimally designed dual mass flywheel (DMF). If this vehicle is driv-en with a given gear ratio at a constant speed of 70 km/h in 5th gear, the specific fuel consump-tion is 385 g/kWh. This can be converted to a fuel consumption of 3.96 liters per 100 km for this vehicle. If the gear ratio is reduced by 20 %, for example, the engine speed drops by the same factor. This lowers the specific fuel con-sumption to 330 g/kWh or 3.39 liters per 100 km, which corresponds to a 14 % reduc-tion!

A comparable improvement is achieved when a speed of 70 km/h is driven in 6th gear instead of 5th gear.

Consumption simulations were carried out to determine the achievable advantage of driving at a lower engine speed in the new European driving cycle (NEDC). Figure 5 shows the compa-rable result for a city cycle. Together with the consumption characteristic map from Figure 4, a consumption advantage of 0.92 liters per 100 km, or 11.4 %, is calculated for the city cycle. Consumption falls from 6.08 to 5.44 liters per 100 km for the entire cycle. This is a savings of 10.5 %.

By lowering the engine speed, the engine torque rises by the same degree and both lead to a de-cline in driving comfort, since the crankshaft ir-regularity increases while the degree of isolati on of the DMF decreases. With a more eff ecti ve damper, this worsening can again be compensat-ed for. The CPA developed by LuK off ers adequate potenti al [4, 5] to realize the depicted consump-ti on advantage without impacti ng comfort.

LuK has long been working on using this princi-ple to reduce vibrations in the passenger car drive train. This technology was explained in connection with a dampened clutch disc as far

00 20002000 30003000 40004000

00

100100

200200

300300

250250

300300

350350400400500500

10001000

400400

225225

100100

400400

500500

275275

350350

300300

325325

00 15001500 20002000 25002500

00

2525

5050

7575

10001000

En

gin

et

En

gin

eto

ro

rqu

eq

ue

inN

min

Nm

MMee

ng

ng

En

gin

et

En

gin

eto

ro

rqu

eq

ue

inN

min

Nm

MMee

ng

ng

Engine speedEngine speed in 1/minin 1/minnneengngEngine speedEngine speed in 1/minin 1/minnneengng

TTrraction raction resisesisttance in 5th gance in 5th gearearwith rwith reduced driveduced drive tre train gain gear rear raatio ( 20 %)tio ( 20 %)

TTrraction raction resisesisttance in 5th gance in 5th gearear

ConsConsttanant pot powwer 6 kW (70 km/h)er 6 kW (70 km/h)Specific fuel cSpecific fuel consumponsumptiontion in gin g/kWh/kWhbbee

Figure 4 Consumpti on characteristi c map for a 2.0-liter diesel engine – potenti al consumpti on reducti on due to lowering engine speeds (downspeeding)

1

0

2

3

4

150150 200200

00

10001000

20002000

30003000

00 5050 10010000

22

44

66

00

2020

4040

6060

Ge

ar

Ge

ar

(de

(de

ffau

lt)

au

lt)

Fu

el

cF

ue

lco

nsu

mp

on

sum

pti

on

tio

n

inl/

hin

l/h

VVe

hic

lesp

ee

de

hic

lesp

ee

d

vvvv

eehh

ink

m/h

ink

m/h

BaseBase

RReduced geduced gear rear raatiotio

TimeTime tt in sin s

En

gin

esp

ee

dE

ng

ine

spe

ed

ininnn

een

gn

g1

/min

1/m

in

Figure 5 Eff ects of downspeeding in the NEDC – consumpti on simulati on (city cycle)





1

back as Symposium 1990 [6]. In Sympo-sium 2002, the CPA on the secondary side of a DMF was presented [7] and the technology has been successfully in production since 2008.

The potential consumption reduction is con-firmed in real driving operation for a DMF with

CPA. Together with other measures, fuel consumption could be reduced by 0.7 liters with no reduction in comfort in a produc-tion vehicle with man-ual transmission. A success that would not be possible with-out an effective damper.

This potential can be transferred without restriction to all other transmission con-cepts (Figure 7). A better damper also supports driving in the speed range fa-

vorable for consumption with double-clutch or automatic transmissions. Better decoupling of vibrations also makes it possible for the torque converter to be bypassed earlier in an automat-ic transmission. Not only can the engine effi-ciency be positively affected in this way, but also the efficiency of the transmission. Espe-

cially for automatic transmissions, effec-tive dampers are of elementary impor-tance to competitive-ness in fuel economy compared to other transmission con-cepts.

The development and optimization of tor-sional vibration dam-pers was and is driven by an effort to achieve drives that are low in noise and vibration. Better dam-pers can naturally still be used purely to improve comfort as well. The possibilities of more efficient vibration reduc-tion are qualitatively summarized in Figure 8. Depending on the starting position and objec-tive, there is a great deal of flexibility in select-ing the extent to which this potential is to be used to improve comfort or reduce consump-tion. As mentioned, these options are indepen-dent of transmission type. Compromises in comfort as a result of economizing on fuel, for example, remain without effect on the con-sumption values in the NEDC. In real driving op-eration, they will, however, have a negative ef-fect on fuel economy.

New challenges Considering the enti re system, using the 3-cylinder engine as an exampleIn larger engines, it has long been the practice to reduce the number of cylinders; today’s 6-cylin-der engines replace the 8-cylinder units of for-mer days, and 4-cylinder units have replaced 6-cylinder engines. The advantages are obvious: lower masses, fewer moving parts and lower friction lead to less dissipated energy. The ad-vantages of engines with fewer cylinders are first

set against louder engine performance, particu-larly when the engine is to run in the lower speed range with optimal fuel economy.

This is one possible reason why the super-charged and thus high-torque types with 3 or even just 2 cylinders are still comparatively rare. The requirements for noise reduction rise disproportionately. If, for example, a 4-cylinder engine with DMF is replaced by a 3-cylinder type, the torsional irregularity is practically doubled at the transmission input with the same damper technology. The reason for this is the lower ignition frequency, which causes a higher irregularity of the crankshaft and a shift in the drive train resonance closer in the speed range relevant to driving.

In addition to the more critical situation on the transmission side, the requirement for the pow-er take-off is also significantly sharpened. Here, too, the irregularity rises to an extreme degree for the reduced number of cylinders, despite the same engine torque. A single-mass flywheel (SMF) would be advantageous for the power take-off.

On the drive train side, the resonance likewise shifts to higher speeds. This is positive, however, in contrast to a DMF, since the resonance lies within the drivable speed range as it is. Due to the shift to higher speeds, the resonance in an SMF is not as strongly excited, nevertheless the speed fluctuations lie clearly above those of a 4-cylinder engine (Figure 9).

4040

2020

1010

00

10001000 15001500 20002000 25002500

3030

4040

2020

1010

00

10001000 15001500 20002000 25002500

3030

4040

2020

1010

00

10001000 15001500 20002000 25002500

3030

Double-clutch transmission

Automatic transmission

Engine

Manual transmission

Standard damper

Damper with centrifugal pendulum absorber

Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nti

on

in1

/min

in1

/min

Speed in 1/minSpeed in 1/min

Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nti

on

in1

/min

in1

/min


Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nti

on

in1

/min

in1

/min


Figure 7 LuK CPA – the standard for torsional vibrati on isolati on in all transmission typesF

ue

lc

Fu

el

con

sum

po

nsu

mp

tio

n/

Co

tio

n/

Co

rred

uct

ion

ed

uct

ion

22

TTororsional vibrsional vibraation rtion reductioneduction

ConConvvenentional damper (base)tional damper (base)

Dual-mass flyDual-mass flywheelwheel

Dual-mass flyDual-mass flywheelwheelwith cenwith centrifugtrifugal pendulum abal pendulum absorbersorber

Figure 8 Flexible opti ons in connecti on with improved damper technology, using the example of the manual transmission

Of course, the heart of Operation

Spritspar was on board: a centrifugal

pendulum absorber integrated in the

dual mass flywheel, which suppresses

the unpleasant surging in the drive train

at low engine speeds.

AAuuttooSStrtraasssseennvveerrkkeehhrr 1199/2/2000099

And so that the low-speed diesel doesn’t bump around

like a sack of nuts, a vibration-damping centrifugal pendulum-flywheel

is very successful at achieving smooth running.AutAutoZoZeitung 19/2009eitung 19/2009

AAuuttooMoMottoorSprSpoort 19/2009rt 19/2009

The centrifugal pendulum absorber

in the dual-mass flywheel really

impressed us. ... We like that:

tooling around with no jerking ...

Even at 1000 rpms, the 320d doesn’t

grumble – good job.

AAuuttooBBIILLDD 3355//22000099

This is made possible by a new type of centrifugal pendulumabsorber integrated in the familiar dual mass flywheel.This technologylowers the shifting speed by up to 400 rpms because“the driver feels better in the low-speed range with no surging," ...

Figure 6 LuK CPA – successful implementati on in producti on





1

In order to achieve better results with conven-tional damper technology, a shift in the drive train resonance would be helpful. The resonance frequency would have to be reduced for a DMF, and raised for an SMF. In concrete terms, this would mean modifying the stiffness of the drive train in order to expand the application area of the damper (Figure 10).

The theoretical potential is considerable. To the extent that the boundary conditions and other requirements permit, the possibility of a drive train modification should be considered as ear-ly as possible in the concept phase. With a cor-responding reduction in drive train stiffness, a

DMF could remain the first choice for super-charged 3-cylinder engines as far as torsional vibrations on the transmission. On the other hand, the potential of an SMF can be signifi-cantly increased by raising the drive train spring rate.

In case an adaptati on of the drive train is not pos-sible and no sati sfactory result can be achieved with conventi onal damper technology, alternati ve concepts have to be considered early on. The po-tenti al of the LuK-developed CPA technology can be used here as well. Considering the overall situa-ti on, a pendulum in combinati on with either a DMF or SMF can present interesti ng soluti ons.

In the case of an SMF it has been shown that the doubled primary excitation force is not negligi-ble. The most favorable results are achieved when both the single and double primary excita-tion forces are canceled out with the CPA.

From a drive train perspective, a DMF with CPA is the best solution. Outstanding isolation val-ues are achieved from an engine speed of 1000 1/min. The advantage of a solution with SMF especially benefits the power take-off,

00

100100

150150

200200

20002000 30003000 400040001000100000

250250

5050

00

100100

150150

200200

250250

5050

20002000 30003000 400040001000100000

Dual-mass flyDual-mass flywheelwheel Single-mass flySingle-mass flywheelwheel

Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min

Engine speedEngine speed in 1/minin 1/minnneengng

Engine 3-cylinderEngine 3-cylinder TTrransmission 3-cylinderansmission 3-cylinderTTrransmission 4-cylinderansmission 4-cylinder

Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min


Engine 4-cylinderEngine 4-cylinder

Figure 9 Increasing requirements when reducing the number of cylinders – Example: 3 cylinders instead of 4, with the same maximum engine torque

EngineEngine

DiffDiffererenentialtial

Axle shaAxle shaftsfts

EngineEngine

TTrransmissionansmission

VVehicleehicleDiffDiffererenentialtial

Axle shaAxle shaftsfts

TTrransmissionansmission

Figure 10 Taking the enti re drive train into considerati on to reduce torsional vibrati ons

00

100100

150150

200200

20002000 30003000 400040001000100000

250250

5050

00

100100

150150

200200

250250

5050

20002000 30003000 400040001000100000

Dual-mass flyDual-mass flywheelwheel Single-mass flySingle-mass flywheelwheel

Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min


Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min


Engine, DrivEngine, Drive tre train with loain with low sw stiffnesstiffness TTrransmission, Drivansmission, Drive tre train with loain with low sw stiffnesstiffnessTTrransmission, Drivansmission, Drive tre train with high sain with high stiffnesstiffnessEngine, DrivEngine, Drive tre train with high sain with high stiffnesstiffness

Figure 11 Performance limits of torsional vibrati on dampers in the enti re drive train

00

100100

150150

200200

20002000 30003000 400040001000100000

250250

5050

00

100100

150150

200200

250250

5050

20002000 30003000 400040001000100000

Engine irrEngine irregularityegularity TTrransmission irransmission irregularityegularity

Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min


Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min


SMF without cenSMF without centrifugtrifugal pendulum abal pendulum absorbersorber DMF without cenDMF without centrifugtrifugal pendulum abal pendulum absorbersorberDMF with cenDMF with centrifugtrifugal pendulum abal pendulum absorbersorberSMF with cenSMF with centrifugtrifugal pendulum abal pendulum absorbersorber

Figure 12 Isolati on potenti al of alternati ve damper types for 3-cylinder engines





1since the CPA directly reduces the irregularity of the engine.

Which type is the best for a concrete applicati on must be evaluated. For the drive train, an opti mal result can be achieved with 3-cylinder engines with DMF. LuK has likewise developed a soluti on for possibly criti cal vibrati ons on the power take-off side, which will be explained in the next sec-ti on.

Soluti on for the power take-off Among the options shown above for reducing vibration in supercharged 3-cylinder engines, variants with DMF also offer the highest poten-tial. The best isolation results can theoretically be achieved even from low engine speeds, par-ticularly in combination with a CPA. Greater vi-brations on the power take-off side are the dis-advantage compared to types with SMF. LuK’s solution for this is to bring the vibrations to a lower level. The model is the proven principle of the DMF. This technology, made appropri-ately smaller, can also be used to enable over-

critical operation for the power take-off. Fig-ure 13 shows a sample design for a LuK belt pulley decoupler (BPD) [8]. The BPD is arranged directly on the free end of the crankshaft. The entire belt drive thus benefits from the isola-tion potential. The BPD is naturally not reserved only for 3-cylinder engines, but can be used in all applications where alternating torques need to be isolated as efficiently as possible from the power take-off.

Figure 14 shows comparative measurements with a conventional generator freewheel on a 4-cylinder diesel engine with 240 Nm. The vibra-tion level on the generator is significantly lower with the BPD. Another advantage is that the dy-namic torques between the crankshaft and belt pulley are isolated immediately. The belt must therefore only transfer low alternating torques and the belt pretensioning can be correspond-ingly low. A lower belt pretensioning in turn means lower friction, which, in the end, can re-duce the losses in the belt drive [8]. Thus, in ad-dition to quieting the power take-off, the BPD also allows a further contribution to reducing consumption and CO2 output.

Stop-start systems, leading the way to hybridizati onFuel economy benefi ts of stop-start systemsMany cars already have it on board today: a stop-start system as a lead-in to hybridization.

One forecast [9] concludes that as early as 2012 every second new vehicle will be equipped with such a system. The consumption advantages are considerable: in the NEDC, the stop time to-tals 240 s, which is 20 % of the entire cycle. This amounts to up to 5 %, or in the pure city cycle, even up to 8 % of the total fuel consumed. Con-cretely, 4.5 % can be saved in the entire cycle for the vehicle considered above with an idle range of about 0.6 L/h (Figure 5).

33 …… 4 Nm/°4 Nm/°

100 Nm100 Nm

Torsion angle α in °

Torq

ue

inN

mM

approx.

Figure 13 LuK belt pulley decoupler (BPD) – the DMF for the power take-off

150150

0000

5050

100100

20002000 3000300010001000

00 0,0750,075 0,1000,1000,0250,025 0,0500,05000 0,0750,075 0,1000,100800800

900900

10001000

11001100

12001200

0,0250,025 0,0500,050800800

900900

10001000

11001100

12001200

150150

0000

5050

100100

20002000 3000300010001000

TimeTime in sin stt

Sp

ee

dS

pe

ed

in1

/min

in1

/min

nn

TimeTime in sin stt


Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min

Sp

ee

dfl

uct

ua

Sp

ee

dfl

uct

ua

tio

nin

1/m

inti

on

in1

/min

Sp

ee

dS

pe

ed

in1

/min

in1

/min

nn


FlyFlywheelwheel PullePulleyy GenerGeneraattoror

Figure 14 LuK belt pulley decoupler (right) – compared with a conventi onal generator freewheel (left )

00

6060

4040

2020

200200 25025000

150150

100100

5050

00 12001200200200 400400 600600 800800 10001000 400400300300 350350

VVe

hic

lesp

ee

de

hic

lesp

ee

din

km

/hin

km

/hvv

vvee

hh

TimeTime in sin stt TimeTime in sin stt

VVe

hic

lesp

ee

de

hic

lesp

ee

din

km

/hin

km

/hvv

vvee

hh

NEDC tNEDC tototal cyal cyclecle(0 km/h r(0 km/h raatte = 20 %)e = 20 %) (0 km/h r(0 km/h raatte = 30 %)e = 30 %)

NEDC city cyNEDC city cyclecle

Figure 15 Consumpti on reducti on potenti al of a stop-start system in NEDC





1

Necessity and opti on in connecti on with stop-start systemsThe fundamental requirements of a clutch and damper system for applications with stop-start function change only slightly. In the case of a DMF, more stop and start operations must be able to be tolerated and the starter sprocket may have to be made somewhat more wear re-sistant. As far as comfort, stop-start behavior must certainly be viewed more critically. Since the driver does not, as usual, actively turn the engine on and off, there are no higher require-ments to deal with from this standpoint.

The consequences are clearer for release system components, since stop-start systems require in-

formati on on the clutch positi on. Positi on record-ing in itself is not new in clutch systems; there are many systems that require this informati on:

Stop-start systems

Cruise control

Start lock

Clutches for hybrid applicati ons

Electric parking brake

Hill-holder system

Input signal for automated manual and double clutch transmissions

Due to the trend toward increased electrifi cati on, strongly driven by stop-start applicati ons in man-ual transmissions, the number of projects in re-

lease cylinders with travel measurement technol-ogy is growing to a great degree. At LuK, intensive development is underway to fi nd the best mea-surement principle from the standpoint of space, performance and cost [10].

One scenario being discussed in connection with stop-start systems is the system behavior in case of a “change of mind.” In this case, the system has decided, based on the driving situation, to stop the engine, but the driver wants to drive on before the stopping process is complete. A re-start of the engine is not possible during the run-out phase. The engine must first be allowed to come to a complete standstill before the starter can be engaged. In this situation there can be a delay of a few tenths of a second, which is sub-jectively perceived as negative.

In order to eliminate a possible delay with a change-of-mind, there are developments in which the starter pinion continuously meshes with the starter sprocket. The starter thus al-ways remains engaged. In order to make this possible, an additional freewheel is required between the starter sprocket and crankshaft, which couples with the starter at low speeds and decouples when a certain engine speed is exceeded (n

A in Figure 17). This enables the en-gine to be started immediately and without de-lay even in the run-out phase. Tests have al-ready shown that such a strategy brings considerable benefits in the subjective evalua-

tion. A restart when the engine is at a complete standstill is shown in Figure 17 on the right. In this concrete example, the time difference is 0.2 seconds better than solutions without a permanently engaged starter.

LuK soluti on for a permanently engaged starter (PES)The challenge for a stop-start system with PES lies in the design of the freewheel. In a design with a conventional freewheel, the starter sprocket is mounted to the output end of the crankshaft with a rolling bearing; at low speeds (starter speed) the starter sprocket is connect-

Slave cylinder Master cylinder

Sensor

Figure 16 Travel sensors in the release system – not only for stop-start systems

nnAA nnAA

CrCrankankshashaftftStarter

TimeTime

Sp

ee

dS

pe

ed

TimeTime

Sp

ee

dS

pe

ed

Time advTime advananttagagee

Normal sNormal sttart prart processocess changchangee of mindof mind

Figure 17 Normal start and restart of a sti ll-running engine (change-of-mind)

Starter ring gear

Crank house

Freewheel

Rolling bearing

Crankshaft

Figure 18 Conventi onal freewheel arrangement for a stop-start system with PES

nn >> nnAA

nn << nnAA

RRolling bearingolling bearing

StStartarter ring ger ring gearear

DrivDriven plaen plattee

FrFreeeewheelwheel

Figure 19 LuK freewheel concept for PES (permanently engaged starter)





1ed with the crankshaft by means of the locked freewheel, while at higher speeds, the crank-shaft outruns the starter, thus uncoupling it. With such an arrangement, the rolling bearing is under stress during the entire operation, spe-cifically with the engine speed. The demands on the bearing are thus extremely high, as it must be designed for the maximum engine speed and the full engine operating hours. The freewheel is likewise constantly turning with the engine speed. Both components, bearing and freewheel, thus generate a drag torque at all times during engine operation and thus ad-ditional dissipated energy.

Simple rolling bearing technology can be used for the LuK solution, since the load turns out to be many times lower than in a conventional free-wheel arrangement. The starter sprocket is no longer mounted on the rotating output end of the crankshaft, but on the stationary crankcase (Figure 19). The freewheel is designed as a cen-trifugal force clutch and bolted to the crank-shaft. When the starter speed range is exceeded (nA in Figure 17), the centrifugal force clutch opens, thus uncoupling the starter sprocket from the crankshaft. In this arrangement, the rolling bearing only has to be designed for the starter speed and the duration of the start pro-cess. After the engine start, this arrangement creates no additional drag torque and thus no dissipated energy. The LuK freewheel for a PES is thus highly robust and has no disadvantages for the overall efficiency.

Higher requirements for the clutch systemOptimizations regarding fuel economy and tor-sional vibration in modern drive trains with com-bustion engines have almost exclusively to do with driving with a closed clutch. The interface between engine and transmission is definitively characterized by the torsional vibration damper. The circumstances change completely for oper-ating conditions with a slipping or actuated clutch. With ideal friction coefficient perfor-mance by the clutch, there would theoretically no longer be any alternating torque transmitted directly from the engine to the transmission at a sufficiently high slip speed. An effect of slip-con-trolled systems for torsional vibration isolation is used.

Nevertheless, there are known to be a number of comfort and NVH problems for which the clutch system or the entire vehicle has to be de-signed. In past years, this was also strongly driv-en by the evolution of diesel engines. The sig-nificantly increased engine torques from turbocharging led to higher operating loads in manual transmissions [1], and the comparatively low available engine torques in the lower speed range in some cases led to weak start-up. The latter is mainly a comfort issue, but also makes

the start-up situation more critical. This can lead to a reduced service life, but also, in particular, to high clutch temperatures when starting up on a hill with a trailer. Here too there are increas-ingly deficits to deal with in supercharged, small-volume gasoline engines. On the engine side, effective counter-measures can be taken. Special engine characteristic maps, with declining par-tial load curves and the highest possible dynam-ic during the start-up phase, stabilize the start-up process. This is equally positive for the start-up energy as for the start-up comfort.

An example is shown in Figure 20. In cooperati on with an automobile manufacturer, the engine-side opti mizati on potenti al was studied with re-gard to start-up quality for a small-volume, super-charged diesel engine. The modifi cati on primarily had to do with the course of the stati c parti al load curves (engine torque plott ed against engine speed for constant thrott le positi on). Compared to the base (producti on design, gray in Figure 20), the start-up energy level has been decreased sig-nifi cantly. Criti cal start-ups with high energy in-puts in parti cular can be largely avoided by this measure. The diff erence was also considerable in the subjecti ve evaluati on, with rati ngs improving by 1.5 points on average.

Overall, tightened engine and transmission boundary conditions lead to higher require-ments on the clutch system (Figure 21). The engine-side excitement and the sensitivity of the transmission or the entire drive train in-crease. In particular, a trend can be seen to-wards greater variety of vibration and noise phenomena in the slip phase. A number of problems can be caused by axial vibrations of the crankshaft. Some important examples of this include:

Pedal vibrati ons

Operati on noises

Chatt er (modulati on excitati on)

Drive train noises (ratt ling, clatt er, whoop)

These forced or impulse-excited phenomena can be most efficiently solved with a cover-mounted release system (CMR) [10], since the excitation mechanism is compensated for. The CMR has been in production for a hybrid application since 2009 and is in the concept phase for manual transmissions at a few automotive manufactur-ers to solve various problems.

00

1010

2020

4040

5050

3030

11

77

1010

7.27.2

6.16.1

6.86.8

8.98.9

7.37.3

8.08.0

44

50

-59

50

-59

60

-69

60

-69

70

-79

70

-79

80

-89

80

-89

90

-99

90

-99

10

0-1

09

10

0-1

09

11

0-1

19

11

0-1

19

12

0-1

29

12

0-1

29

13

0-1

39

13

0-1

39

14

0-1

49

14

0-1

49

Engine speedEngine speed

En

gin

et

En

gin

eto

ro

rqu

eq

ue

HorizHorizononttal partial load cural partial load curvvee

Descending partial load curDescending partial load curvvee

RRe

lae

lati

vti

ve

fre

fre

qu

en

cye

qu

en

cyin

%in

%

LeLevv

el

el

no

rma

ln

orm

al

LeLevv

el

el

spo

rty

spo

rty

12

%u

ph

ill

12

%u

ph

ill

gr

gr a

de

ad

ew

ith

trw

ith

tra

ile

ra

ile

r

Su

bje

ctiv

Su

bje

ctiv

er

eraa

tin

gti

ng

DisDistribution of stribution of sttart-up enerart-up energygy

(12 % uphill gr(12 % uphill grade with trade with trailer)ailer)SubjectivSubjective ee evvaluaaluationtion

StStart-up enerart-up energygy in kJin kJ Driving situaDriving situationtion

Engine charEngine charactacteriseristic maptic map

Figure 20 Eff ect of engine characteristi c map on start-up stability – driving test with 10 testers

Sp

ee

dS

pe

ed

TimeTime

Sp

ee

dS

pe

ed

TimeTime

WWe

ar

re

ar

raattee

TTemperemperaaturturee

SpecificSpecificengine tengine tororqueque

Full loadFull loadcharcharactacteriseristictic

DoDownspeedingwnspeeding

FFeewwerercylindercylinderss

GrGreaeatteesensitivitysensitivity LoLowweringering

idle speedidle speedCrCrankankshashaftft

aaxial vibrxial vibraationstions

IncrIncreasedeasedadditional cadditional consumeronsumerss

EEffffectivectiveevvehicle wehicle weigheightt

En

gin

eto

rqu

e

En

gin

eto

rqu

e

En

gin

eto

rqu

e

EEffff

.v

.v

eh

.w

eh

.w

eig

he

igh

tt

IsolaIsolationtion ImpactsImpactsMechan.Mechan.loadload

StStart-upart-upqualityqualityNoiseNoise

StStartartbehabehaviorvior

ThermalThermalsstrtressess

PPedaledalvibrvibraationstions

PPedal loadedal loadmodulamodulationtion

JudderJudderFFacingacingwwearear

RRequirequiremenementsts

BoundarBoundary cy conditionsonditions

TTor

orq

ue

fq

ue

fact

act

or

or

TimeTime

Sp

ee

dS

pe

ed

TTimeime

Loa

d/

TLo

ad

/To

ro

rqu

eq

ue

PPedal tredal traavvelel

MP

MP

EP

EP

DP

DP

TTem

pe

re

mp

eraa

tur

turee

TimeTime

Sp

ee

dS

pe

ed

TimeTime

Acc

ele

rA

cce

leraa

tio

nti

on

TimeTime

Fr

Fre

qu

en

cye

qu

en

cy

TimeTime

YYearear



YYearear

Figure 21 Tightened boundary conditi ons and requirements for a clutch system





1

Figure 22 shows an example of a noise problem (ratt ling) in the slip phase, which is characterized by structure-borne noise peaks on the transmis-sion occurring every 0.5th engine order. The start-ing situati on with a conventi onal release system is scored with a subjecti ve rati ng of 4. With a CMR, it

achieves a rati ng of 10; the peaks in the structure-borne noise signal during the engagement phase are completely eliminated.

Figure 23 shows another example of the eff ecti ve-ness of the CMR. With a CMR, a rati ng of 10 can

also be achieved in the slip evaluati on in this vehicle. The excitati on is demonstrably caused by crankshaft vibra-ti ons.

The strength of the ef-fect of an excitation on the subjective rat-ing depends defini-tively on the vibration ability of the system that is excited. The same excitation can thus lead to a good rating in one vehicle and a poor rating in another. For example, different vehicles have significant differences when it comes to slip sen-sitivity, as the comparison of some 40 vehicles in Figure 24 shows.

The vehicle factor on the x axis represents the effective vehicle weight or the vehicle size. The slip sensitivity is a vehicle constant. It is the fac-tor between cause (excitation) and effect (vehi-cle vibration) and applies to a harmonic excita-tion in the slip frequency, smaller values being desirable.

Sedans A and B are in the same vehicle class and have similar engines. The slip sensiti vity of se-dan B is more than double that of sedan A. In con-crete terms, this means that more than double the vehicle vibrati ons can be expected with the same slip excitati on. The same applies to the ex-ample of the two small cars. In general, smaller vehicles are more sensiti ve than larger ones, since less mass must be accelerated. This is parti ally compensated for, however, since the engines are smaller and the torque capaciti es of the clutches are less.

The diff erences in sensiti vity result largely from the interacti ons of the drive train with the eigenmodes of the engine block (rotati on, translati on) and the wheel suspension (translati on).

Vehicles with high slip sensitivity can be mas-tered only with difficulty with the current state of the art. Special measures can be necessary to avoid complaints. In addition to component op-timization, a cover-mounted release system (Fig-ure 23) or an impact compensation bearing (Fig-

ure 26) can help, depending on the type of excitation. Another solution is offered by a slip damper [11], which has the advantage of effi-ciently reducing or preventing the slip type of excitation.

The possible problems are multi-layered, and are often first identified in vehicle trials, and thus late in the development process. At LuK, in-tensive measurement and simulation methods are developed to duplicate the various phenom-ena at the system level as well as the component level. The objectives are the earliest possible de-tection of potential problems and the develop-ment of robust designs and remedies. For exam-ple, the “eek noise,” as it is called, an instability of eigenmodes of the transmission input shaft, has been simulated virtually. That was the only way stabilizing measures could be determined. In concrete terms, clutch discs were developed with tilt and radial flexibility which significantly reduce the probability of unstable system be-havior [11].

In connection with pedal vibrations, the entire system can be analyzed and optimized from fly-wheel to pedal using well-grounded simulation models [12]. This approach is already being used intensively in the development process to pre-cisely tune hydraulic lines and dampers [10]. Supposedly small details are also being closely modeled and tested. An example of this is the interface between the diaphragm spring finger and the release bearing. Depending on the con-figuration of the finger design and the friction coefficient in the contact zone, appreciable im-

00 22 3311

00 22 3311

20002000

40004000

30003000

-600-600

600600

00

20002000

40004000

30003000

00 22 3311

00 22 3311-600-600

600600

00

CoCovver-mouner-mountted red release selease syyssttememConConvvenentional rtional release selease syyssttemem

SubjectivSubjective re raating 10ting 10SubjectivSubjective re raating 4ting 4


TimeTime tt in sin sS

pe

ed

Sp

ee

dnn

in1

/min

in1

/min

Sp

ee

dS

pe

ed

nnin

1/m

inin

1/m

in

Str

uct

ur

Str

uct

ure

-bo

rne

no

ise

e-b

orn

en

ois

e

inP

inP

aa



Str

uct

ur

Str

uct

ure

-bo

rne

no

ise

e-b

orn

en

ois

e

inP

inP

aa

TransmissionEngine

Figure 22 Ratt ling noise in the slip phase – eff ecti veness of the cover-mounted release system (CMR)

22

00

11

00 22 3311

500500

15001500

10001000

00 88 10102200

6644

500500

15001500

10001000

00 88 10102200

6644

22

00

11

00 22 3311

CoCovver-mouner-mountted red release selease syyssttememConConvvenentional rtional release selease syyssttemem

SubjectivSubjective re raating 10ting 10SubjectivSubjective re raating 6.5ting 6.5





Sp

ee

dS

pe

ed

nnin

1/m

inin

1/m

inA

cce

ler

Acc

ele

raati

on

tio

n

aavv

eehh

inm

/in

m/ss

22

Sp

ee

dS

pe

ed

nnin

1/m

inin

1/m

inA

cce

ler

Acc

ele

raati

on

tio

n

aavv

eehh

inm

/in

m/ss

22

TransmissionEngine

Figure 23 Crankshaft -excited slip – eff ecti veness of a cover-mounted release system (CMR)

55

0,00,0

0,50,5

1,01,0

1,51,5

0,000,00 0,020,02 0,040,04 0,060,06

2020

4040

1010

kk ==vveehh mveh

iaxleigear

rdyn

(1,44 kgm(1,44 kgm ))--11

(( kgmkgm ))--11

0,700,70(0,60 kgm(0,60 kgm ))

--11

(0,26 kgm(0,26 kgm ))--11

VVehicle fehicle factactoror kkvveehh in 1/kgmin 1/kgm

Jud

de

rse

nsi

tiv

ity

Jud

de

rse

nsi

tiv

ity

in1

/kg

min

1/k

gm

AmplificAmplificaationtion

SedanSedan AA

SedanSedan BBSmall cSmall carar AA

Small cSmall carar BB

Figure 24 Benchmark slip sensiti vity of automobiles in 1st gear





1provements may also be made at this point, depending on the starting situation (Fig-ure 25). The lowest possible friction coef-ficient that can be reliably obtained with an additional contact disc on the release bearing is desirable [10].

With tightened boun-dary conditions, it has become even more important for auto-maker and supplier to work closely together in order to identify possible problems early on. This is the only way to find the need for action as early as possible so that the best solu-tion, from both a technical and a commercial standpoint, can be developed jointly. Figure 26 shows some known as well as some new exam-ples of problem- and comfort-oriented system solutions.

SummaryIn competi ti on with hybridizati on and ulti mately electrifi cati on of the drive train, the combusti on engine will conti nue to claim a leading role for some ti me to come.

Opti mizati ons on the engine side as well as the se-lecti on of a suitable transmission concept to make full use of any potenti al for reducing fuel consump-ti on will turn the confl ict of interest between plea-surable driving or effi ciency into the goal of plea-surable driving and effi ciency.

The damper and clutch systems will be key elements for such drive trains – they will be given the Hercu-lean task of off ering opti mal degrees of isolati on in the face of increasing irregularity, rising comfort de-mands and fricti on-opti mized drive trains.

Shift recommendati ons, stop-start systems, and, depending on the vehicle class, transmission auto-mati on will soon be a matt er of course and will no longer have to court the car buyer’s acceptance. The vast majority of these technical soluti ons can be found in the intelligent refi nement of existi ng components, such as the DMF, clutch and release system.

Close cooperati on between automaker and suppli-er is required now more than ever, both in concept selecti on, or even drive train defi niti on, as well as in producti on development.

Literature[1] Zink, M.; Hausner, M.; Welter, R.; Shead, R.:

Clutch and release system – Enjoyable clutch actuati on! 8th LuK Symposium, 2006

[2] Shell Passenger Car Scenarios up to 2030 – Facts, Trends and Opti ons for Sustainable Auto Mobility. htt p://www.shell.de/pkw-szenarien

[3] Müller, B.: LuK torque converter – strategic converter for new automati c transmis-sions, VDI Report No. 2029, Düsseldorf 2008

[4] Zink, M.; Hausner, M.: The centrifugal pen-dulum-type absorber – applicati on, perfor-mance and limits of speed-adapti ve damp-ers. ATZ, Issue 07/08 2009

[5] Kroll, J.; Kooy, A.; Seebacher, R.: Land ahoy? - Torsional dampers for engines of the fu-ture. 9th LuK Symposium, 2010

[6] Reik, W.: Torsional Vibrati on Isolati on in the Drive Train – An Evaluati ve Study, 4th LuK Symposium, 1990

[7] Kooy, A.; Gillmann, A.; Jäckle, J.; Bosse, M.: DMF – Nothing New?, 7th LuK Symposium, 2002

[8] Sti ef, H.; Pfl ug, R.; Schmidt, T.; Fechler, C.: Belt Drive Systems – Potenti al for CO2 re-ducti ons and how to achieve them, 9th LuK Symposium, 2010

[9] Bosch: Press Release PI6753, August 2009

[10] Welter, R.; Lang, V.; Wolf, B.: Clutch Opera-ti on, 9th LuK Symposium, 2010

[11] Freitag, J.; Gerhardt, F.; Hausner, M.; Witt -mann, C.: The Clutch System of the Future – More than Separati ng and Connecti ng, 9th LuK Symposium, 2010

[12] Fidlin, A.; Ineichen, L.; Kremer, E.; Klünder, D.; Tikhomolov, A.: Vibrati ons in the Clutch System: From the Crankshaft to the Pedal, VDI Report No. 2077, Düsseldorf 2009

CoCovver-mouner-mountted red release selease syyssttemem

EliminaEliminattes NVH pres NVH problems due toblems due tooaaxial crxial crankankshashaft vibrft vibraations [10]tions [10]

Run-out cRun-out compensaompensation bearingtion bearingRReduces geduces geomeeometrictricjudder ejudder exxcitcitaation [10]tion [10]

AnAnti-vibrti-vibraation unittion unitRReduces pedal vibreduces pedal vibraationstions(particularly lo(particularly low-frw-frequency pulses) [10]equency pulses) [10]

RRelease bearing with plaselease bearing with plastic ringtic ringLoLow friction in the actuaw friction in the actuation stion syyssttem,em,rreduces preduces pressuressure vibre vibraations [10]tions [10]

HyHydrdraulic damperaulic damperRReduces preduces pressuressure vibre vibraationstionsin the rin the release selease syyssttemem(fr(frequency requency rangange ~ 100e ~ 100 –– 500 Hz)[10]500 Hz)[10]

ClutClutch pedal with spring accumulach pedal with spring accumulattororRReduces opereduces operaating loads, opting loads, optimiztimizeses

operoperaating load curting load curvve [10]e [10] VVolume floolume flow rw regulaegulatted Ped PTLTL

PrPrototects DMF and drivects DMF and drive tre train frain fromompeak tpeak tororques, loques, low thrw throtottle lossestle losses

with normal operwith normal operaation speedtion speedSerServvo supporto supportLoLow operw operaating loads andting loads andoperoperaating hting hyyssttereresis,esis,opoptimal load curtimal load curvve [1]e [1] Judder damperJudder damper

RReduces all feduces all fororced chaced chatttter vibrer vibraations,tions,prpreevvenents self-ets self-exxcitcitaation (ftion (facing chaacing chatttter) [11]er) [11]

DecDecoupling drivoupling driven plaen platteeRReduces noise in the slip phase dueeduces noise in the slip phase duetto unso unsttable trable transmission input shaansmission input shaftft

modes [11]modes [11]

Figure 26 Examples of effi cient products to opti mize the system behavior in the slip phase

-0,5-0,5

0,00,0

0,50,5

-0,5-0,5

0,00,0

0,50,5

-0,5-0,5

0,00,0

0,50,5

-0,5-0,5

0,00,0

0,50,5

Axial eAxial exxcitcitaationtion

Axial diaphrAxial diaphragm spring fingagm spring finger vibrer vibraationtion

Radial diaphrRadial diaphragm spring fingagm spring finger vibrer vibraationtion

LoLow fingw finger heigher heightt

LoLow friction cw friction coeoefficienfficientt

PPe

da

lacc

ele

re

da

l acc

ele

raati

on

tio

n

(e(effff

.v

.va

lue

)in

m/

alu

e)

inm

/s²s²

High fingHigh finger heigher heighttHigh friction cHigh friction coeoefficienfficientt

PPe

da

lacc

ele

re

da

l acc

ele

raati

on

tio

n

(e(effff

.v

.va

lue

)in

m/

alu

e)

inm

/s²s²

MeasurMeasuremenementtSimulaSimulationtionVVarianariantt

a)a)

b)b)

Figure 25 Pedal vibrati ons – detail opti mizati on at the contact point between diaphragm spring fi nger and release bearing

Documents

1 Clutch systems and torsional dampers ... - Schaeffler · PDF fileSchaeﬄ er SYMPOSIUM 2010 1 Clutch systems and torsional dampers 12 Schaeﬄ er SYMPOSIUM 2010 13 Clutch systems