Prof. Dr.-Ing. Peter Steinberg, BTU Cottbus

M.Sc. Oleg Krecker, PhD candidate, BMW

SIMULATION AND EVALUATION OF ENGINE FRICTIONEUROPEAN GT CONFERENCE, FRANKFURT/MAIN, OCTOBER 9TH, 2017

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 2

AGENDA

Motivation and objective1

Modelling of friction related components 2

Validation3

Conclusion and further developments4

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 3

01MOTIVATION AND OBJECTIVE

MOTIVATION

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 4

Thermal engine

model1D Simulation warm up

& fuel consumption

within driving cycles

Test

Warm up

0 200 400 600 800 1000 12000

50

100

150

Zeit [s]

TH

_Z

ST

_1

_2

_0

5 [°C

]

m.TH_ZST_1_2_05

eSCV zu Mean

eSCV auf Mean

0 200 400 600 800 1000 12000

50

100

150

Zeit [s]

TH

_Z

W_

A0

2_

03

5 [°C

]

m.TH_ZW_A02_035

eSCV zu Mean

eSCV auf Mean

Tem

per

atur

e [°

C]

Time [s]

Cylinder Bridge Temperature, NEDC

0D/1D Simulation Friction

Total engine

GT-Suite

Simulation

data

Final result:

Impact on

CO2 - emissions

Cor

rela

tion

&

valid

atio

n

Predictive evaluation of concepts and trends in engine friction reduction.

„As simple as possible and as complex as necessary.“

other

lossesmech.

power

Piston

assembly

Crank-

shaft

Oil + vacuum

Belt driveCylinder head

+ chain drive

Measurement data (motored and fired)

Fri

ctio

n

Speed

pump

Friction

Low

Total

engine

fired

1-cylinder fired(floating liner)

Total engine /

strip-down

Single components

motored

Transient

cycles

Low High

Real boundary conditions

Data resolution &reproducibility

High

OBJECTIVE

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 5

Low

Total engine

fired

1-cylinder fired(floating liner)

Total engine /

strip-down

Single components

motored

Transient cycles

Low High

Closeness of basic

conditions to reality

Possible measuring

resolution & reproducibility

High

Fast prediction

Agile transferability

Relative comparison

Physical evaluation

Req

uire

men

ts

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 6

02MODELLING OF FRICTION RELATED COMPONENTS

IDENTIFICATION OF MAJOR FRICTION SOURCES

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 7

Chain drive & hydraulic components

Guides and tensioner

Oil & vacuum pump

VANOS

Chain

Cylinder head unit

Camshaft bearings

Valvetrain and VALVETRONIC

High pressure pump

Crank train

Piston – Liner system

Conrod and crankshaft bearings

Balancer shaft

Front end accessory drive

Multiple v-belt

Auxiliaries, e.g. alternator

CAPABILITIES IN GTSOME EXAMPLES

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 8

Surface propertiesInlet camshaft and VALVETRONIC

of a 3-cylinder engine

Crank train of a 4-cylinder engine

Dynamic beam modelling and

hydrodynamic bearing solution

Complex kinematics (var. valve lift)

& independent definition of each

friction contact

Detailed piston and liner

geometry

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 9

03VALIDATION

CRANKSHAFTVALIDATION AND CALIBRATION

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 10

Friction torque deviation between measurement and

simulation is shown (strip-down crankshaft).

At high speeds >3500 RPM the model lacks in accuracy more effort in calibration is necessary.

Good correlation in usual speed ranges (<2500 RPM) of driving cycles fuel consumption area of interest.

Model calibration on temperature, oil flow and friction

torque.

Model overestimates

friction

Model underestimates friction

Area of good correlation

0 1000 2000 3000 4000 5000 6000

Fri

ctio

n To

rque

[N

m]

Speed [RPM]

1

Messung

Simulation

CRANKSHAFTMODEL TRANSFERABILITY (CASE STUDY ON BEARING CLEARANCE)

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 11

0 1000 2000 3000 4000 5000 6000

Fri

ctio

n To

rque

[N

m]

Speed [RPM]

Friction Crankshaft only, 90°C

Messung

Simulation

Investigation on transferability at

changing boundary conditions

(bearing clearance).

Experiment on strip-down test rig and

variation of clearance:

• Minimum = 20 µm

• Maximum = 60 µm

GT setup:

• Bearing mobility method

• Crankshaft 3D beam dynamic

• Tuning of oil temperature model

Model shows good agreement of

frictional losses with changing

clearances.

Clearance 20µm

measured

Clearance 20µm

simulation

Clearance 60µm

simulation

Clearance 60µm

measured

PISTON ASSEMBLYTOTAL FRICTION OF PISTON SKIRT + RINGS (FLOATING LINER)

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 12

Study shows the importance of proper geometry definition in the piston-liner system (convexity, ovality, distortion).

Unsteady peaks occur due to high contact pressure on partial areas at the skirt.

Current investigation on thermal FE skirt calculation and development of a work around by modifying skirt geometry. also V2019 will include an elastic deformation model for the skirt to improve friction prediction.

Measurement

A – w/o bore distortion & skirt geom.

B – with bore distortion & skirt geom.

CYLINDER HEAD UNITEXAMPLE: EXHAUST CAMSHAFT AND VALVETRAIN

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 13

0 1000 2000 3000 4000 5000 6000

Fri

ctio

n To

rque

[N

m]

Speed [RPM]

Camshaft and Valvetrain Friction, exhaust side, 90°C

Strip-Messung Bezug AZ3.2 Simulation

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

750 1000 1500 2000 3000 4000 5000

Fri

ctio

n To

rque

dis

trib

utio

n [%

]

Speed [RPM]

Friction distribution, exhaust side, 90°C

Friction Torque - CamShaftBearings

Reibmoment HVA

Reibmoment Ventilfuß

Reibmoment Ventilführung

Friction Torque - Cam-FollowerContacts absolut

MWälz; Part Mittelwert

Measurement Simulation Camshaft Bearings

Hyd. Lash Adjuster

Valve Pad

Valve Guide

Cam Follower

Roller Bearing

State of measurement: camshafts are driven by chain, valvetrain rocker arms are partly removed dynamic chain movement might interfere result fidelity (4500/5000 RPM)

Percentage distribution of friction contacts is a current state estimation.

TOTAL ENGINESIMULATION AND ADDITIONAL MEASUREMENTS

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 14

Total Engine

Test Rig Transmission

Oil- & Vacuum – Pump

Water pump

Acc. Belt Drive

Inlet Camshaft + Valvetrain

Exhaust Camshaft + Valvetrain

Balancer Shaft

Piston Skirt

Piston Rings

Small End Bearing

Big End Bearing

Main Bearings

Seals

Sim

ulat

ion

Exp

eri-

men

t

Gap due to:

Missing simulation of chain friction

Belt drive friction needs a bit more tuning

Deviation in piston assembly

Strip-Down Test, 90°C

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 15

04CONCLUSION AND FURTHER DEVELOPMENTS

CONCLUSION AND FURTHER DEVELOPMENTS

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 16

Conclusion

GT offers a variety of capabilities to model entire engine friction losses.

Main models of friction related engine components are set up, validated and calibrated but some effort is necessary for physical reasonable interpretation of simulation results!

Good agreement on simulation and measurement data correlation.

Sophisticated fusion of subsystem

models.

Investigation on load and temperature and its impact on overall

engine friction.

Further measurements for

detailed subsystem friction analysis.

Slick workflowmodel set up,

parameter studies and post processing.

Optimizing studies on friction reduction.

Fur

ther

dev

elop

men

ts

Simulation and Evaluation of Engine Friction | Oleg Krecker | October 9th 2017 Slide 17

THANK YOU!