1

6.1 Problems with weldments

before after

2

Outline

nn Applied stresses not well knownApplied stresses not well known

nn Complex geometriesComplex geometries

nn Sources of inherent scatterSources of inherent scatter

3

The stresses which a structure may experience in service are not precisely known.

Applied stresses

4

Loading Conditions

Fatigue resistance depends upon loading condition

5

Service stresses generally unknown

6

Measurement of service history

7

T

Strain Gage Pair

= +

Total Axial BendingT

Applied stresses

8

Region of Rapidly Rising Stress

Region of Linear Stress

Weld Toe

"Nominal" Component

"Geometric" Component

"Local" Component

Strain Gauge

Stress Distribution

Strain gage location?

9

σ/S

T S

x

rσ

x/T

1

2

3

0.1 0.2 0.3 0.4

r = 0.4 mm

r = 3 mm

Pure axial loading

Pure bending loading

Stress concentrating effects of weld toe persist for as much as x/t ˜ 0.3.

Less for bending and less for more generous weld toe radii.

Weld toe stress fields

10

104 105 106 107 1081

10

100

Fatigue Life, N (cycles)

∆

Ideal

Nominal

2x

­10x ­10x

Variability in applied stresses

11

Outline

nn Applied stresses not well known Applied stresses not well known

nn Complex geometriesComplex geometries

nn Sources of inherent scatterSources of inherent scatter

12

Geometries

There is an endless numberof weldmentgeometries.

13

Outline

nn Applied stresses not well known Applied stresses not well known nn Complex geometriesComplex geometriesnn Sources of inherent scatterSources of inherent scatternn Weld qualityWeld qualitynn Mean, fabrication and residual stressesMean, fabrication and residual stressesnn Stress concentrationsStress concentrationsnn Weldment Weldment sizesizenn Material propertiesMaterial properties

14

10 710 610 510 410 310 0

10 1

10 2

10 3

failuresrun outs

Fatigue life, N (cycles)

S = 159.20 * N^-0.16648, r^2 = 0.173, s = 0.1718

logN = 6.8974 - 1.03751*logS, r = -0.2387, s = 0.4293

99% survival with 50% confidence

99% survival with 95% confidence

Scatter in weldment fatigue data

15

Statistical variation

The geometry of a weldment may vary with location.

16

"Nominal" Weldment

Undercut, Slag Entrapment and/or other Discontinuities

Fatigue Cracks

Base Metal

"Ideal" Weldment

Weld Metal

HAZBase Metal

Weld Metal

HAZ

0.1 in

Weld quality

17

104 105 106 107 1081

10

100

Ideal: Sr = 36, S

fab = 36

Nominal: Sfab

= 36

Fatigue Life, N (cycles)

∆

Effect of weld quality

18

Mulitple failure scenarios

While the weldment may be simple, many different failure scenarios may exist.

19

²M²S

180 - θ

r

T

Sfab²Sa

²Sb

Remote Weld Done Last

+ +

Restraint

²M²S

σr

Mean, fabrication, and residual stresses

20

Thermal contractions after welding generate yield point residual stresses

Welding Residual Stresses

21

Subsequent welding may develop fabrication stresses which are superposed on the welding residual stresses.

Fabrication (residual) stresses

22

104 105 106 107 1081

10

100

Ideal: Sr = 0, S

fab = 0

Ideal: Sr = 36, S

fab = 0

Ideal: Sr = 36, S

fab = 36

Nominal: Sfab

= 0Nominal: S

fab = 36

Fatigue Life, N (cycles)

∆Mean, fabrication, and residual stresses

23

Mean, fabrication, and residual stresses

104 105 106 107 108

1

10

100

Ideal: T = 0.5 in., Sfab

= 0Ideal: T = 2.0 in., S

fab = 36

Nominal: T = 0.5 in., Sfab

= 0Nominal: T = 2.0 in., S

fab = 36

Fatigue Life, N (cycles)

Rem

ote

Stre

ss R

ange

, ∆S

(ksi

)

Light Industry, High Quality Welding

Heavy Industry, High Quality Welding

Light Industry, Low Quality Welding

Heavy Industry, Low Quality Welding

24

²S²S

180 - θ

r

t +

Applied stress

Induced Bending Stress

L

α

SB =32

α SALt

tanh ββ

β =Lt

3 SA4E

Mean, fabrication, and residual stresses

25

Weldment stress concentrations

Component with a stress-concentrating notch

Applied Stress

Notch

Unnotched Component

26

Weld toe is a stress concentration

27

Fatigue failure starting at weld toe

28

The mavericks

Tight fit-up: K- = 3.6

Normal fit-up: 3.6 < Kt < 6.4

Loose fit-up? Kt = 6.4

Example: T-joint with a variable fit-up

29

180 -θr

t

? M

? S

? M? S

180 - θr

t

? M

? S? M

? S

Longer Fatigue Life

Shorter Fatigue life

Region of high stress Same initial

size, crack grows to 2x initial length

Effect of weldment size

30

1

10

100

0.1 1 10

Ideal, Sfab=36

Ideal, Sfab=0

Nominal, Sfab=36

Nominal, Sfab=0

Thickness, T (in.)

-1/3-1/2

-1/7

Ideal weldments are predicted to be more affected by the size effect than Nominal weldments.

Predicted size effect

31

Base Metal (BM)Weld Metal (WM)

Heat Affected Zone (HAZ)

Fatigue cracks begin in WM or HAZ not in BM!

Material properties

32

Correlation between HAZ and BM hardness

33

Correlation between Syand hardness

34

Correlation between hardness and strain-controlled fatigue properties

35

Variation in fatigue crack growth behavior of steels

36

0

10

20

30

40

50

60

0 50 100 150 200

Fatig

ue S

tren

gth

at 1

E+0

7 cy

cles

, Sa

(ksi

)

Base Metal Ultimate Strength, SuBM

(ksi)

As Welded, Sfab

= 0

As Welded, Sfab

= SyBM

Over Stressed

Hot Rolled Steel Q & T Steel

Stress Relieved

Trends in “Ideal” 1.0-in plate thickness, non-load carrying cruciform weldments fatigue strength.

• R = 0

• Welding residual stresses = 50% of SYBM

• Sfab ˜ SYBM

Predicted effect of SuBM

37

Summary

n The variables influencing weldment fatigue life can be thought of as being only two: n the magnitude of the notch root stresses. n the properties of the notch root material.

n In this sense, the applied stresses, the degree of bending, the welding residual stresses, the fabrication stresses, the weldment geometry, the notch root weld defects all influence the magnitude of the notch root stresses.

38

Summary

n The fatigue behavior of a weldment is controlled by the local (notch root, hot-spot) stress strain history.

n For structural steel weldments: material properties are of minor importance except as they determine and limit the value of the residual stresses.