Top Banner
10001886:1999-12-20 1 2nd Workshop on Joints Modelling, Dartington Hall, Totnes, Devon, UK 26-29 April 2009 Quantifying Fretting Damage Using a Contact-Evolution Based Modelling Approach Jian (Kenny) Ding, Sean Leen, Phil Shipway, Tom Hyde School of Mechanical, Materials and Manufacturing Engineering, University Technology Centre in Gas Turbine Transmission Systems University of Nottingham, UK
15

Quantifying Fretting Damage Using a Contact-Evolution ......10001886:1999-12-20 3 2nd Workshop on Joints Modelling, Dartington Hall, Totnes, Devon, UK 26-29 April 2009 Contact-evolution

Jan 31, 2021

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • 10001886:1999-12-20

    12nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Quantifying Fretting Damage Using a Contact-Evolution Based Modelling Approach

    Jian (Kenny) Ding, Sean Leen, Phil Shipway, Tom Hyde

    School of Mechanical, Materials and Manufacturing Engineering, University Technology Centre in Gas Turbine Transmission Systems

    University of Nottingham, UK

  • 10001886:1999-12-20

    22nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Nottingham UTC in Gas Turbine Transmission Systems

    Study of Spline Couplings

    • Experimentally characterise the fretting behaviour of

    splines using scaled-down spline and/or representative

    specimens

    • Develop lifing methodologies for spline against fretting

    Spline Joints

    • Improved lifing methods

    • Fretting fatigue & wear modelling

    • Assessment of coatings & alternative materials

    Support structures

    • Super plastic forming

    • Robotic plasma arc welding

    • Weld distortion

    RFW Process Modelling

    • Dual-alloy shafts

    • Other components

    Shaft Plain Section

    • Buckling

    • Damage tolerant design

    • TiMMC technology

    Spline Joints

    • Improved lifing methods

    • Fretting fatigue & wear modelling

    • Assessment of coatings & alternative materials

    Support structures

    • Super plastic forming

    • Robotic plasma arc welding

    • Weld distortion

    RFW Process Modelling

    • Dual-alloy shafts

    • Other components

    Shaft Plain Section

    • Buckling

    • Damage tolerant design

    • TiMMC technology

    Spline fretting failure

  • 10001886:1999-12-20

    32nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Contact-evolution based lifing methodology

    • An approach that considers transient interaction between wear and fatigue under fretting, especially the effect of wear on fatigue life.

    • Contact-evolution based lifing approach comprises: o A finite element wear simulation tool to determine the evolution of

    contact geometry. o Damage Accumulation approach for crack nucleation.

    • Ongoing EPSEC project in collaboration with Oxford (total grant ~0.6 Millions)

    cracking

    (partial slip)

    Displacement amplitude, μm

    wear

    (gross sliding)

    cracking+ wear

    Norm

    al forc

    e, N

    (mixed regime)

    Fretting Map *

    cracking

    (partial slip)

    Displacement amplitude, μm

    wear

    (gross sliding)

    cracking+ wear

    Norm

    al forc

    e, N

    (mixed regime)

    Fretting Map *

    Displacement amplitude, μm

    wear

    (gross sliding)

    cracking+ wear

    Norm

    al forc

    e, N

    (mixed regime)

    Fretting Map *

    wear

    (gross sliding)

    cracking+ wear

    Norm

    al forc

    e, N

    (mixed regime)

    Fretting Map *

    Gross slidingPartial SlipStick Reciproc.sliding

    Slip amplitude ( m)�

    We

    ar

    rate

    (m/N

    m)

    3

    Fa

    tig

    ue

    life

    (cy

    cle

    s)

    107

    106

    105

    10-16

    10-15

    10-14

    1 3 10 30 100 300 1000

    Fatigue

    life

    (cyc

    les)

    Wea

    r ra

    te (

    m3/N

    m)

    Gross slidingPartial SlipStick Reciproc.sliding

    Slip amplitude ( m)�

    We

    ar

    rate

    (m/N

    m)

    3

    Fa

    tig

    ue

    life

    (cy

    cle

    s)

    107

    106

    105

    10-16

    10-15

    10-14

    1 3 10 30 100 300 1000

    Fatigue

    life

    (cyc

    les)

    Wea

    r ra

    te (

    m3/N

    m)

  • 10001886:1999-12-20

    42nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Modelling Framework

    The approach integrates a number of ‘tools’.

    o Fretting wear tool is central, which predicts the extent of wear damage and the concomitant change of contact geometry

    o For each wear step n, accumulated fatigue damage is calculated by fatigue parameter Smith-Woston-Topper; thus, total accumulated damage is given by

    All cycles completed ?

    Calculate wear depth increment by modified Archard’s equation

    NO

    Yes

    Output results

    FE analysis of fretting contact

    Calculate cumulative fatigue damage of the load step

    Update contact geometry

    Initial parameters• Contact geometry• Applied loads• Wear coefficient & friction coefficient

    ∑Δ

    =

    =

    Δ=

    NN

    n

    n fnc

    T

    NND

    1∑Δ

    =

    =

    =NN

    n

    n ni

    T

    N1 ,

    Archard’s Wear equation:

    ),(),(),( txdstxpktxdh ii ××=

  • 10001886:1999-12-20

    52nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Normal load -120 N/mm Stoke - 20 μm

    Gross slip case

    Worn surface profile (after 5000 cycles)

    Original surface profile

    Partial slip case

    Normal load -120 N/mm Stoke - 5 μm

    Contact width increase markedly from Hertz

    prediction

    Fretting Wear Modelling

    (Ding et al, Int J of Fatigue, 2004)

    Little change of contact size, wear occurring at slip zone

  • 10001886:1999-12-20

    62nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    0200400600800

    100012001400

    -0.15 -0.1 -0.05 0 0.05 0.1 0.15

    Hoizontal position (mm)C

    onta

    ct p

    ress

    ure

    (MPa

    )

    0 cycle

    1000 cycles

    5000 cycles

    0100200300400500600700800900

    -0.4 -0.2 0 0.2 0.4

    Hoizontal position (mm)

    Con

    tact

    pre

    ssur

    e (M

    Pa)

    0 cycle

    1000 cycle

    5000 cycle

    Normal load -120 N/mm Stoke - 20 μm

    Partial slip case

    Normal load -120 N/mmStoke - 5 μm

    Fretting Wear Modelling

    Evolution of contact pressure

    Gross slip case

  • 10001886:1999-12-20

    72nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Contact-evolution based prediction of crack nucleation (I) gross sliding

    -25

    -20

    -15

    -10

    -5

    0

    5-800 -600 -400 -200 0 200 400 600 800

    surface height (μm)

    Horizontal position (μm)

    Predicted profileMeasured profile

    0

    0.5

    1

    1.5

    2

    2.5

    -6 -4 -2 0 2 4 6x/a

    SWT (MPa)0 cycle

    10000 cycle20000 cycle80000 cycle

    0

    0.5

    1

    1.5

    2

    2.5

    0 20000 40000 60000 80000Number of fretting cycles

    SWT (MPa)

    x/a=0.90

    x/a=1.24

    x/a=1.75

    SWT snapshots

    Sample SWT evolutions

    Ti-6Al-4V

  • 10001886:1999-12-20

    82nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Contact-evolution based prediction of crack nucleation (II) partial slip

    -0.005-0.004-0.003

    -0.002-0.001

    0-2 -1 0 1 2

    x/a

    Wear depth (mm)

    25, 000 cycle150,000 cycle300,000 cycle

    0

    2

    4

    6

    8

    10

    0 100000 200000 300000Number of cycles

    SWT (MPa)x/a = 1x/a = 0.7 x/a = 0.8 x/a = 0.4

    Sample SWT evolutionsWear scar evolution

    50μm

    Slip Stick Slip

    Slip Stick Slip

    Ti-6Al-4V

  • 10001886:1999-12-20

    92nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Contact-evolution based prediction of crack nucleation (II) partial slip

    0

    200

    400

    600

    800

    1000

    1200

    1400

    1600

    -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2

    Horizontal position (mm)

    Contact pressure (MPa)

    50 μm50 μm50 μm50 μm

    300000th cycle150000th cycle75000th cycle30000th cycle0th cycle

    Complex pressure evolution predicted due to plasticity effects

    Multiple cracking locationso x ≈ ± 0.11 mm (light blue)o x ≈ ± 0.04 mm (red)o x ≈ ± 0.13 mm (yellow)

    x ≈ ± 0.13 mmo Initial Hertzian contact edgeo early cycles, low COF: gross sliding (N < 3k)

    x ≈ ± 0.04 mmo ~ initial stick-slip boundarieso late cycles (N ≈ 150k-300k)

    x ≈ ± 0.11 mmo intermediate cycles (3k < N < 150k)o due to flat indenter type pressure peaks

  • 10001886:1999-12-20

    102nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Prediction vs. tests (Madge et al, 2007)

    Effect of slip amplitude on fretting fatigue is captured by taking into account how wear affects fatigue damage parameter.

    Effect of slip amplitude on fretting fatigue

  • 10001886:1999-12-20

    112nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Cyclic Plasticity in Fretting

    Prager linear kinematic hardeningMaterial: Ti-6Al-4V (α + β)Coefficient of friction – 0.9

    EE

    σ

    ε

    EE

    σ

    ε

    EE

    σ

    EE

    σ

    εε

    EE

    σratchetting strain,

    rεΔ

    σ2

    σ1

    EE

    σ2

    ε2|x|

    Initial yield locus Subsequent

    translated yield locus

    σ2

    σ1

    EE

    σ2

    ε2|x|

    Initial yield locus Subsequent

    translated yield locus

    0100200300400500600700800900

    1000

    0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018

    Strain

    Stre

    ss (M

    Pa)

    E = 115 GPa

    c = 7317 MPa

    0100200300400500600700800900

    1000

    0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018

    Strain

    Stre

    ss (M

    Pa)

    0100200300400500600700800900

    1000

    0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018

    Strain

    Stre

    ss (M

    Pa)

    E = 115 GPa

    c = 7317 MPa

    Plastic shakedownsteady reversed cyclic

    plastic strains

    Ratchettingplastic strain magnitude

    increases continually with load cycling

  • 10001886:1999-12-20

    122nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    -0.02

    -0.01

    0

    0.01

    0.02

    0.03

    0.04

    0.05

    0.06

    0.07

    0.08

    0 50000 100000 150000 200000 250000 300000

    Fretting cycles

    Plas

    tic s

    trai

    n

    p22ε

    p11ε

    p12ε

    p

    c

    N

    ir

    r

    εε∑ =Δ

    Cyclic Plasticity in Partial Slip

    Nominal Hertzian geometry elastic Wear simulation with plasticity ratchetting phenomenon Possibility of damage/cracking due to ductility exhaustion

    rij

    rijr εεε ΔΔ=Δ 3

    2

    Predicted wear-induced evolution of plastic strains at final stick-slip interface

  • 10001886:1999-12-20

    132nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    0.000

    0.001

    0.002

    0.003

    0.004

    -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1

    x (mm)

    p

    3000th cycle75000th cycle150000th cycle300000th cycle

    • Gross sliding: shear-dominant plasticity

    • g.s. plasticity take a W-shape

    • Wearing away of plasticity reduction in equivalent plastic strain

    FE-predicted plastic strain in surface layer

    Cyclic Plasticity in Gross Sliding

  • 10001886:1999-12-20

    142nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Conclusions and Future Challenges

    • Contact-evolution based fretting lifing methodology provides

    o an integrated solution for fretting wear and fatigue prediction.

    o a convincing explanation about the effects of slip amplitude on fretting fatigue

    • Future challenges:o Incorporate near-surface effects into fretting fatigue

    prediction, such as asperity, oxidation, plasticity and debris accumulation. How important are they for fretting crack nucleation?

    o Fretting contact mechanics under micro or nano scales.

  • 10001886:1999-12-20

    152nd Workshop on Joints Modelling, Dartington Hall,

    Totnes, Devon, UK 26-29 April 2009

    Contact-evolution based prediction of crack nucleation (I) gross sliding

    ∑==

    =

    NNn

    n ni

    T

    NNΔ Δϖ

    1 ,

    cbniff

    bni

    fnia )N'('εσ)N(E

    ')(σεσ ++= ,

    2,

    2

    ,max 22)( Δ

    Crack nucleation defined to occur at material point i when accumulated damage ω reaches value of 1, where ω is defined as :

    Each Ni,n is calculated based on a critical-plane fatigue damage parameter Smith-Watson-Topper (SWT). -25

    -20

    -15

    -10

    -5

    0

    5-800 -600 -400 -200 0 200 400 600 800

    surface height (μm)

    Horizontal position (μm)

    Predicted profileMeasured profile

    0

    0.5

    1

    1.5

    2

    2.5

    -6 -4 -2 0 2 4 6x/a

    SWT (MPa)0 cycle

    10000 cycle20000 cycle80000 cycle

    0

    0.5

    1

    1.5

    2

    2.5

    0 20000 40000 60000 80000Number of fretting cycles

    SWT (MPa)

    x/a=0.90

    x/a=1.24

    x/a=1.75

    SWT snapshots

    Sample SWT evolutions

    Ti-6Al-4V

    Quantifying Fretting Damage Using a Contact-Evolution Based Modelling ApproachNottingham UTC in Gas Turbine Transmission SystemsContact-evolution based lifing methodologyModelling FrameworkContact-evolution based prediction of crack nucleation �(I) gross slidingContact-evolution based prediction of crack nucleation �(II) partial slipContact-evolution based prediction of crack nucleation �(II) partial slipConclusions and Future ChallengesContact-evolution based prediction of crack nucleation �(I) gross sliding