University of Illinois at Urbana-Champaign Thermo

University of Illinois at Urbana-Champaign

Department of Mechanical and Industrial Engineering

ThermoThermo--Mechanical Fatigue of CastMechanical Fatigue of Cast319 Aluminum Alloys319 Aluminum Alloys

Huseyin Sehitoglu, 1 Carlos C. Engler-Pinto Jr. 2,Hans J. Maier 3 ,Tracy J. Foglesong4

1 University of Illinois at Urbana- Champaign, USA,2 Ford Motor Company, Dearborn, 3 Universität-GH Paderborn, Germany, 4 Exxon , Houston

Research Supported by Ford Motor Company



• Use of Aluminum Alloys in engine blocks and cylinder

heads

• Thermo-Mechanical Fatigue Results

• Summary

• Modeling Studies (Precipitation hardened aluminum

alloys)

OutlineOutline



Percentage of Vehicles with AluminumPercentage of Vehicles with AluminumEngine Blocks and Heads (*)Engine Blocks and Heads (*)

(*) Delphi VIII Study, 1996

20%10%5%Light trucks

50%30%13%Passenger cars

Blocks

60%40%20%Light trucks

95%85%78%Passenger cars

Heads

200520001994



Advantages of cast aluminumAdvantages of cast aluminum

• Lightweight

– V-8 Engine Block: 150 lbs Cast Iron vs. 68 lbsAluminum

• Cast into complex shapes

• Increased thermal conductivity



• Cylinder Heads

Inlet Valve Exhaust Valve

Spark Plug

Practical ApplicationPractical Application



Al319Al319--T7BT7B

• Nominal Composition in weight percentage

– (*) WAP319: max 0.4% Fe - EAP319: max 0.8% Fe

• Thermal treatment– solutionizing at 495°C for 8 hours followed by precipitating at

260°C for 4 hours)

*

Fe

0.05max

0.25max

0.25max

0.20-0.30

0.25-0.35

3.3-3.7

7.2-7.7

Bal.

CrTiZnMnMgCuSiAl



ThermoThermo--Mechanical Fatigue CyclesMechanical Fatigue Cycles

• Simultaneously changing strain andtemperature (T)

• In-Phase: max-strain at max-T

• Out-of-Phase: max-strain at min-T

300

200

100

T(¡

C)

-1

0

1

ε mec

h(%

)

6420time (min)

OP

IP

-1

0

1

ε mec

h(%

)

300200100T (¡C)

IP

OP



• Fatigue of materials subjected to simultaneously changingtemperature and strain.

• εtot = εth + εmech

• Terminology– in-phase

– out-of-phase

ThermoThermo--Mechanical FatigueMechanical Fatigue

Temp.

Mech. Strain

Tmax

Tmin

εmaxεmin

TMF IPTMF OP



Micro-Computer

HydraulicTesting

Machine

FT εεεεT vs. t

TMF TestingTMF Testing

pyrometer

extensometerInductionHeating

load cell

εεεε vs. t



• Isothermal LCF

– 20°C, 150°C, 250°C and 300°C

– 2×10-1 s-1 , 4×10-3 s-1 and 5×10-5 s-1

• Thermo-Mechanical Fatigue

– 100–300°C — 5×10-5 s-1

Experimental ProceduresExperimental Procedures



TMF LoopsTMF Loops

-0.4 -0.2 0.0 0.2 0.4

Mechanical Strain (%)

100¡C 300¡C

1

200

In-Phase∆εm = 0.54%

Nf = 39020

-200

-100

0

100

200

Stre

ss(M

Pa)

-0.4 -0.2 0.0 0.2 0.4

Mechanical Strain (%)

100¡C

300¡C

Cycle 1

200

1220

Out-of-Phase∆εm = 0.6%

Nf = 2460



TMF OP 100TMF OP 100--300°C 1.0%300°C 1.0%

-200

-100

0

100

200

Str

ess

(MP

a)

-6x10-3

-4 -2 0 2 4 6Strain

STRESS1-2fea Stress1-2STRESS300fea Stress300

100°C

300°C

200°C



Fatigue Life CriterionFatigue Life Criterion

LCF - 250°C - 0.3% - 0.5 hzWAP319-T7B

120

80

40

0

Max

imum

Stre

ss(M

Pa)

50000Number of Cycles

50% Load Drop

Nf



80

60

40

20

0

Yie

ldSt

reng

th(M

Pa)

120100806040200time (h)

PrecipitatePrecipitateCoarseningCoarsening

45000 cycles / ~25 hours(300°C, ∆εm = 0.2%).

Exposure at 300°C

Initial



TMF LoopsTMF Loops

200

100

0

-100

OP

Stre

ss(M

Pa)

-0.6 -0.4 -0.2 0.0 0.2 0.4

OP Inelastic Strain (%)

-200

-100

0

100

IPStress

(MPa)

0.6 0.4 0.2 0.0 -0.2 -0.4IP Inelastic Strain (%)

OPIP

100¡C

300¡C



TMFTMF –– Peak StressesPeak Stresses300

250

200

150

100

50

Stre

ss(M

Pa)

2 3 4 5 6 7 8 90.1

2 3 4 5 6 7 8 91

2

Inelastic Strain Range (%)

OP IPσmax-σmin 100¡C

300¡C



TMFTMF –– Stress Range EvolutionStress Range Evolution350

300

250

200

150

100

50

Stre

ssR

ange

(MPa

)

25002000150010005000Number of Cycles

TMF OPTMF IP

EAP319-T7B

Initial Behavior

IPN f /2

OPN f /2



Cyclic StressCyclic Stress--Strain CurvesStrain Curves400

350

300

250

200

150

Stre

ssR

ange

(MPa

)

0.00012 3 4 5 6 7 8 9

0.0012 3 4 5 6 7 8 9

0.012

Inelastic Strain Range

WAP319TMF OPTMF IP

EAP319TMF OPTMF IP300¡C 0.5 hz

300¡C 5x10-5

s-1

WAP319

EAP319



TMF LifeTMF Life

0.001

2

3

4

56

0.01

2

3

4

5

Mec

hani

calS

trai

nR

ange

101

102

103

104

105

106

107

108

Cycles to Failure

150¡C 40 hz250¡C 40 hz250¡C 0.5 hz

250¡C 5x10-5

s-1

300¡C 5x10-5

s-1

300¡C 0.5 hzTMF OPTMF IP

RoomTemperature

EAP319-T7B



TMF LifeTMF Life400350300

250

200

150

100

50

Stre

ssR

ange

(MPa

)

101

102

103

104

105

106

107

108

Cycles to Failure

RoomTemperature

300¡C 5x10-5

s-1

TMF OP 5x10-5

s-1

TMF IP

5x10-5

s-1

300¡C 2x10-3

s-1

150¡C

2x10-1

s-1

250¡C

2x10-3

s-1

EAP319-T7B



0.0001

2

4

68

0.001

2

4

68

0.01

2

Inel

astic

Stra

inR

ange

101

102

103

104

105

Cycles to Failure

EAP319250¡C 0.5 hz

250¡C 5x10-5

s-1

300¡C 0.5 hz

300¡C 5x10-5

s-1

TMF OPTMF IP

OP + LCF

IP

WAP319TMF OPTMF IP

TMF LifeTMF Life



EAP319EAP319--T7BT7BTMFTMF--OPOP

100100––300°C300°C∆ε∆εmm=0.6%=0.6%NNff=2460 c.=2460 c.



EAP319EAP319--T7BT7BTMFTMF--IPIP

100100––300°C300°C∆ε∆εmm=0.54%=0.54%NNff=390 c.=390 c.



2

3

4

5

6

7

8

90.01

2

3

4

5

Meha

nica

l St

rain

Ran

ge

101

2 3 4 5 6 7 8 9

102

2 3 4 5 6 7 8 9

103

2 3 4 5 6 7 8 9

104

Nf

WAP EAP PredictionTMF-IPTMF-OP

a0 = 70 µm

300 µm



1. TMF stress-strain behavior is identical for both IP andOP loading conditions. TMF-IP lives are shorter thanTMF-OP (based on the mechanical or inelastic strainrange) lives.

2. Creep damage dominates for TMF-IP loading and inthe high strain range regime.

3. The secondary alloy (EAP319) is softer than theprimary alloy (WAP319), but TMF lives are verysimilar.

SummarySummary



AluminumAluminum--Copper AlloysCopper Alloys• Precipitate-dislocation interactions

– Anisotropy on plastic flow behavior (Hosford & Zeisloft ‘72, Bate et al. ‘81, Barlat & Liu ‘98,Choi & Barlat ‘99)

– Bauschinger effect (Abel & Ham ‘66, Moan & Embury ‘79, Wilson ‘65)

• Coherent particles - GP zones and θ'' (Price and Kelly ‘64)

– Higher yield stress than Al shearing of particles

– Comparable work hardening rates and

deformation to Al

• Semi-coherent - θ' ( P & K ‘64, Russell & Ashby ‘70)

– High yield stress and high work hardening rates

• Incoherent particles - θ ( P & K ‘64, R & A ‘70)

– Low initial yield stress

– Highest rates of work hardening



Precipitate DevelopmentPrecipitate Development

GP zones * Peak aged, θ'

Over aged, θ' & fine θVery over aged, coarse θ*Sato & Takahashi, 1983



Limitations of current modelsLimitations of current models

• No implicit consideration of aging treatment.– Models were developed for one specific aging treatment

• Peak aged, θ'

• No inclusion of length scale– Volume fraction, precipitate size, mean free path etc. with aging treatment

• Empirical hardening models with a microstructural basis.



Proposed Hardening LawProposed Hardening Law• Single crystal formulation - one precipitate type

• Polycrystal formulation– More than one type of precipitate– Incorporate grain size length scale

Ýτ = Koα 2µ 2b

2 τ −τ o( )d+θo

τ s −ττ s −τ o

Ýγ kk∑

Ýτ = µ2b

2 τ −τ o( )α1

2 ′ K1

d1

+α 2

2 ′ K2

d2

+α 3

2 ′ K3

d3

+ θo

τ s −ττ s −τ o3

Ýγ kk∑

d3

d1

d2

grain

θ'

θ



Constitutive EquationsConstitutive Equations• Relate stress and strain rate at single crystal and

polycrystal level.

• Can be written in pseudo-linear form.

• Assume overall polycrystal response described by lawsimilar to that of single crystal.

Ýε iin = Ýγ o

mism j

s

τ cs

mks ′ σ kτ c

s

s =1

S

∑n−1

′ σ j wherei =1,5

Ýε iin = Mij

c (sec) ′ σ ( ) ′ σ j

ÝEiin = Mij

(sec) ( ′ Σ ) ′ Σ j



Hardening with PrecipitatesHardening with Precipitates• Start with dislocation evolution equation

• Combine with the Bailey-Hirsch relationship for flow stress.

Ýρ = Ko

db+ k1 ρ − k2ρ

k∑ Ýγ k

Geometric storage term dueto boundaries / obstacles

Statistical storage of dislocations

Dynamic recoveryof dislocations

τ = τ o +αµb ρ



500

400

300

200

100

0

Stre

ss, (

MP

a)

20151050

Inelastic Strain, (%)

Solid line - Experiment

Dashed line - Simulation

Experiment and Simulation for Al - 4% Cu Single Crystals[123] Orientation - Different Aging Conditions

190C, 24 hrs

190C, 3 hrs

No Aging



SummarySummary

• The hardening law including the effects ofprecipitates on the deformation behavior of binaryAl - Cu alloys is physically based and accounts forprecipitate size, orientation, and mean free path.

• The model incorporates hardening law andpredicts single crystal behavior of pure Al and Al-Cu alloys, it also predicts polycrystallineexperiments from knowledge of single crystalbehavior.

Documents

University of Illinois at Urbana-Champaign Thermo