Virtual Development of Drivetrains · CAx im Fahrzeug- und Motorenbau 1 07. Nov. 2016 Wintersemester 2016/17 Virtual Development of Drivetrains Dipl.-Ing. Johannes Quehenberger Magna

CAx im Fahrzeug- und Motorenbau 1

07. Nov. 2016

Wintersemester 2016/17

Virtual Development

of

Drivetrains

Dipl.-Ing. Johannes QuehenbergerMagna Powertrain

[email protected]


Contents

Introduction

Development Approach

Tools and Applications

Software Development

Virtual Production Development


Introduction

Product and Production


(Quelle: BMW Group, WPS Training)

Hand Craft

Combination of Man and

Machine

Mass Production FORD

Equipment Specialization and

Division of Work

Lean Manufacturing

The TOYOTA Production

System

• Tailor made

• Quality depending on Skill

of Craftsman

• High Cost and Price

• Long Lead Time

• Low Volume

• Low Customer Orientation

• Good Quality

• Dramatic Cost Reduction

• Reduction of Delivery Time

• High Volumes

• Optimization of all

Resources, Low Cost

• Focusing on Elimination of

Waste to achieve complete

Customer Satisfaction

• High Flexibility

• Low Volumes per Type

• High Total Volumes

Product Development


(Quelle: Lindemann, nach: Gausemeier, 2006)

Virtual Product

vs.

Virtual

Production

Simultaneous / Concurrent Engineering

Production

Pre-Production

Procurement

ProductionPlanning

Test

Design

Concept

Production

Pre-Prod’n

Procurement

ProductionPlanning

Test

Design

Concept Conventional Engineering

Simultaneous Engineering

Reduction

Fro

nt

Loadin

g

6CAx im Fahrzeug- und Motorenbau

Virtual Development


Geometry

Function

Hardware

Software

System


Development Approach


Requirements

from Vehicle

Requirements

Analysis

Modultest

KomponentenTest

Definition and Validation

Test Scripts

Detailed Functionality

Testing

Testing of

Functionality

Concept, Component

Specification, Software

Specification

Vehicle

(OEM)

In Vehicle

Testing

Vehicle

Transfer CaseTransfer Case

Tests

Component

DevelopmentComponent

Tests

Acceptance In Vehicle

Testing

(OEM)

Virtual Product Development

Development Process

Source: Magna Powertrain

System Integration


The V-Cycle will be run

through several times


Tools and Application

AVI2/Audi_TV_Deckel_rechts-ges1.avi

AVI2/Audi_TV_Deckel_rechts-ges1.avi

AVI2/Audi_Tv_Deckel_rechts_Erstarrung_2008-11-27.02.avi

AVI2/Audi_Tv_Deckel_rechts_Erstarrung_2008-11-27.02.avi

AVI2/EOL_1150Nm_bw_01.avi

AVI2/EOL_1150Nm_bw_01.avi

AVI2/pq75_diff_1800Nm_scale20_lin_view2.avi

AVI2/pq75_diff_1800Nm_scale20_lin_view2.avi

AVI2/VW_MQB_Radialspannung_v2_1050Nm.avi

AVI2/VW_MQB_Radialspannung_v2_1050Nm.avi

System Function: Simulation-Environment VeDyna


Vehicle: - Validated vehicle-models

Component: - 4x4 Algorithm

- Consumption-module

- Temperature-module

Driver Model: - VeDyna Advanced Driver 2

M4x4

Cycles

Handling

Maneuvers

Consumption

System-stress &

Comfort

Traction

&

Vehicle Dynamics

4x4

Algorithm


Component Function


Input

• Vehicle and drive unit data (masses, dimensions, ratios, stiffnesses damping…)

• Load case (impulse shift)

• Material

Multi Body Simulation of parking lock system with AMESIM

Output

• Torque

as function of time

in relevant positions of transmission

• Peak force on pawl

upstream torque at pawl (rotor)

downstream torque at pawl (road)

Model


Component Function


Multi Domain Simulation of Hydraulic Actuation System

Rotor gear profiles geometry

generation with

MATLAB/SIMULINK


Component Function


Evaluation with Reference

Pump AMESIM 1-D Hydraulic

simulation

Option: Pump stiffness and

gap variation, FEA with CATIA

Multi Domain Simulation of Hydraulic Actuation System


Analytical Strength Assessment


Model

• Beam model for shafts

• Notch geometry

choice according to FKM method

• Load spectra

adapted according to FKM (rainflow)

• Material data modification

(heat treatment)

• Bearing model: analytical

(integrated in tool)

KISSsoft and BEARINX: Dimensioning of shafts and bearings

example BEARINX

Hertzian pressure

on roller/outer ring

Input

• Shaft geometry

• Loads, load spectra

• Material

• Bearing type, dimension

• housing stiffness (option)

Output

• Safety figures (fatigue, static)

• Stresses, Hertzianpressure

example KISSsoft

Pinion model


System Structural Dynamics


Model Generation for FEA analysis: example eRAD




Determination of Modal System Behavior

First results extracted by the FEM

Solver

Natural Eigen-Frequencies and their

corresponding modeshapes

Structural weakness areas by use of

the modal strain energy distribution

Description and documentation of the

extracted natural frequencies and their

mode shapes in the so-called modal

map.

Comparison of the determined natural

frequencies to acting excitation

frequencies and first assessment




Determination of System Response

Determination of system response

behavior due to acting excitation loads

Vibration motion behavior at excitation

frequencies up to 2500 Hz

Acceleration run over frequency at

structural points to identify resonance

areas

Surface velocity prediction to identify

structural areas with potential to

radiate sound




Example of Optimization Loop

Calculation of Response behavior for chain excitation load case

View vibration animation at resonance frequency and identify relevant mode shape or shapes out of the modal map

Together with the designer discuss and define countermeasures

Update the FEM model with the countermeasure and recalculate

compare the results to the initial design Source: Magna Powertrain

Ball-Ramp System


Simulation using Multi-Body Capabilities of Abaqus

Rotation Input to the Worm Gear

Motor Worm Gear

Fixed Ramp

Moving Ramp


Ball-Ramp System


Details: Rigid Body Parts with Contact Surfaces

Ball movement

In analysis, the ball is a perfect sphere. The

“mesh” is displayed only for visualizing motion.


Ball-Ramp System


Details: Contact Area and Contact Path Differences

Contact pressure evolution animation (shown for two cycles)

Ball No. 1 Ball No. 2

Ball No. 3


Ball-Ramp System


Contact Path Hysteresis with Aberration in End Position

Start (1st)

2nd

3rd

4th

End - 5th

Ball 1 motion (displayed for 5 cycles)


Rolling Gear Simulation - AUDI Sport Differential


Nonlinear Tooth Root Stress Analysis (Abaqus)


Rolling Gear Simulation - AUDI Sport Differential


Nonlinear Tooth Root Stress Analysis (Abaqus)

Only tooth root (tensile) stress

displayed.

Contact stress omitted.


Dynamic Deflections: Parking Pawl Engagement


Nonlinear FE Analysis (Abaqus) Source: Magna Powertrain


FEA Solver: NASTRAN

Model So

urc

e: M

agn

a P

ow

ert

rain


FEA Solver: NASTRAN

Nonlinear Loading

Taper roller

bearing cones

Shafts, wheel

body elements

Roller elements

(nonlinear)

Output torque Input torque Pair of contact

forces

So

urc

e: M

agn

a P

ow

ert

rain


FEA Solver: NASTRAN

Nonlinear Loading So

urc

e: M

agn

a P

ow

ert

rain


FEA Solver: NASTRAN

Results So

urc

e: M

agn

a P

ow

ert

rain


FEA Fatigue Analysis: FEMSITE

Stress plot Safety plot




Stress plot Safety plot




x

z

1307N

N1800

2075N

N1000

2118N

N1100

3493N

N12005455N

N1300

5830N

N1500

2232N

N1600

8864N

N1400

511N

N1700

675N

N1900

Shear load distribution

on screws

Flange separation under load



Hypoid Gears: GLEASON (T 900) and ANSYS


FEA-based strength analysis program by GLEASON

Heat dissipation - Thermal model of coupling: ANSYS



Software Application during Development

NASTRAN

ABAQUS

MEDINA

ANSYS

FEMSITE

FEMFAT

TOSCA

KISSSOFT / KISSSYS

CAGE for WINDOWS

AMESIM

VEDYNA

MATLAB / SIMULINK

ADAMS

Concept Design

PT 1 PT 2 PPAP


Control development

Mechanical development


Software Development


Software Development Process

System

Specification

Functional

Design

Hardware-in-the-

Loop

White-Box-Test

Software-in-the-Loop

OEM´s

Solution

Software-in-the-Loop

Rapid Controller

PT

Target Code

Generation

Control Design

System Integration

Calibration, Test

SW-Code- Implementation

Test Spec



Feasibility Study – „Proof of Concept“

All components that have influence on functionality

Estimation of functionality e.g. Maximum torque capacity

Control behaviour of torque build-up

System availability

...

Simulation model of Power supply

ECU PWM driver

Worm gear motor

Ball helicline

Clutch

Simulation based on Simulink simulation modules using

libraries

Approved on former programs

Actuatoric System of ATC



Transfer Case Functionality - Achieving Requirement Spec.

Detailed simulation

shows all components‘ influences

e.g. DC Motor

Temperature dependencies

Torque ripple of motor

Eddy current losses

Brush voltage drop

Friction of worm gear unit

...

Delivers accurate information about

Energy consumption

Torque set behavior and speed

disturbances due to

Temperature

Power supply

Aging of wiring harness

...

Time

Time

Current

Torque

Typical Results



Functional Approval Model

Approved simulation models tested using prototype parts

Control software included

New control strategies developed simultaneously

Allows to generate software specification

AudiS3

Reisinger/EEE-F, 02-Jun-1999 16:21vtlastw: MueD=1.0,eben, r= 50m, ab Lastw. fix, v0=36kmh->DKW:80° ->0° , T1=0.3s, V1.0: Haldex geregelt

15 16 17 18 19 20-500

0

500

1000

1500

Mo

me

nte

[N

m]

4WD: Momente,Schupf auf Kardanniveau

mspezgmgv

mgh

15 16 17 18 19 20-100

-50

0

50

Dn [U

/min

]

ZGv-ZG

h

ZGv-VA

eZG

h-HA

e

15 16 17 18 19 20

0

0.5

1

10

0%

HA

<--

->1

00

%V

A

Zeit [s]

AntriebsmomenteRadlast

Project ATCReisinger/ESA, 02-Feb-2004 16:21Control Program V1.0

AudiS3

Reisinger/EEE-F, 02-Jun-1999 16:21vtlastw: MueD=1.0,eben, r= 50m, ab Lastw. fix, v0=36kmh->DKW:80° ->0° , T1=0.3s, V1.0: Haldex geregelt

15 16 17 18 19 20-500

0

500

1000

1500

Mo

me

nte

[N

m]

4WD: Momente,Schupf auf Kardanniveau

mspezgmgv

mgh

15 16 17 18 19 20-100

-50

0

50

Dn [U

/min

]

ZGv-ZG

h

ZGv-VA

eZG

h-HA

e

15 16 17 18 19 20

0

0.5

1

10

0%

HA

<--

->1

00

%V

A

Zeit [s]

AntriebsmomenteRadlast

Project ATCReisinger/ESA, 02-Feb-2004 16:21Control Program V1.0

Control AlgorithmActuator-Model

Voltage

Test Spec

Current, Position, ...

Simulation of Mechatronic System



Functional Approval Model - Quality

Simulation of

Torque capacity

System dynamics

Current consumption

Detailed interface data

…

Motor Characteristics @ Temp.

0 5 10 15 20 25 30 350

10

20

30

40

50

60

70

Kennlinienvergleich Simulation - Motor - Erstmuster 03 / RT

M [Nm]

i[A

], n

[1/m

in]

0 5 10 15 20 250

10

20

30

40

50

60

70

80

Kennlinienvergleich Simulation - Motor - Erstmuster 03 / 125°C

M [Nm]

i[A

], n

[1/m

in]

0 5 10 15 20 25 30 35 400

10

20

30

40

50

60

70

Kennlinienvergleich Simulation - Motor - Erstmuster 03 / -40°C

M [Nm]

i[A

], n

[1/m

in]

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

1000

2000

t[s]M

Sperr

Ist[

Nm

]

MessungSimulation

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

50

t[s]

nS

M[1

/min

]

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

10

20

30

t[s]

iSM

[A

]

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.550

100

t[s]

phiS

M [

°]

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

5

10

t[s]

uS

M [

°]

Momentensprung 0 - 1400Nm @ 10V

Comparison Simulation vs. Measurement


Development Tools

MatlabSimulinkStateflow Functional software modeling

Hardware modeling

Simulation of ECU and plant

Test bench

Micro Autobox

Targetlink C-Code generation an optimization

Fixed-point simulation and validation

Supports interface to CANape (ASAP2)

CANape CANapeGraph

CANalyzer

Measurement Tools Temperature

Voltage, Current, ...


Software Development Process

specificationfunctional software

Model-in-the-Loop (MIL)

Host PC, highest precision

fixed point effectsSoftware-in-the-Loop (SIL)

Host PC

on-target verificationhardware-in-the-Loop (HIL)

execution time/code profiling

On Target

modeling

scaling

compiling

ECU /

component

integration

automated test

fixed point effects

automated test

functional software

automated test

hardware effects


Example for automated testing

53


Virtual Production








The End

Virtual Development of Drivetrains · CAx im Fahrzeug- und Motorenbau 1 07. Nov. 2016 Wintersemester 2016/17 Virtual Development of Drivetrains Dipl.-Ing. Johannes Quehenberger Magna

Documents