VII Improving the fatigue life of...
Transcript of VII Improving the fatigue life of...
AM 11/03 1
VII Improving the fatigue life of weldments
AM 11/03 2
Outline
n Obvious things to don Problems the weld toen Fatigue life Improvement Strategiesn Light and heavy industry weldmentsn Improving the “bad” weldments
AM 11/03 3
Crude! (bad)
AM 11/03 4
Better
AM 11/03 5
Bad - planar weld discontinuities
AM 11/03 6
Outline
n Obvious things to do
n Problems the weld toen Fatigue life Improvement Strategiesn Light and heavy industry weldmentsn Improving the “bad” weldments
AM 11/03 7
Weld toe is a stress concentration
AM 11/03 8
Slag entrapments at toe?
Virtually eliminates fatigue crack initiation life NI
AM 11/03 9
Welding procedure A Welding procedure B
Cold-lap defects at weld toe
Cold laps virtually eliminate the fatigue crack initiation life (NI)
Such weldments may have an appreciable fatigue crack initiation life (NI)
“Nominal”“Perfect”
AM 11/03 10
Cold lap at a weld toe
Loading Direction
D
Weld Metal
Base MetalHeat Affected Zone
Weld Toe Location Without Cold-Lap Defect
Curved PathVertical Path
θ
rφ
AM 11/03 11
Effect of cold lap depth
1
10
0.001 0.01 0.1
D = 0.0 mm, MK CG
, curved path from weld toe locationD = 1.0 mm, M
K CG, curved path from
cold-lap defectD = 1.0 mm, M
K CG, vertical path from
cold-lap defect
K
Crack length / main plate thickness, (a/T)
2-D FEM
Wel
d ge
omet
ry c
orre
ctio
n fa
ctor
, MK
AM 11/03 12
Effect of flank angle
1
10
0.001 0.01 0.1
θ = 30o
θ = 45o
θ = 60o
K
Crack length / main plate thickness, (a/T)
2-D FEM
D = 1.0 mm
Wel
d ge
omet
ry c
orre
ctio
n fa
ctor
, MK
AM 11/03 13
Effect of cold root radius
1
10
0.001 0.01 0.1
D = 0.00 mm, weld toe locationD = 2.00 mm, r
cl = 0.146 mm
D = 2.00 mm, rcl
= 0.025 mm
K
Crack length / main plate thickness, (a/T)
Wel
d ge
omet
ry c
orre
ctio
n fa
ctor
, MK
AM 11/03 14
Recent study on rail welds
Toe Radius, R
Weld Reinforcement
Height, H
Base Metal Thickness, Tb
Weld Collar Width, W
Flank Angle, θ
AM 11/03 15
Geometric Parameters
Fin Length, Lf
Cold Lap Length, Lcl
Fin Thickness, Tf
Flank Angle, θ
AM 11/03 16
Weld with a Fin and a Cold Lap
AM 11/03 17
Deformed Shape
AM 11/03 18
Nominal Weld Geometry
1
1.5
2
2.5
3
3.5
4
4.5
0 10 20 30 40 50 60 70 80 90Flank Angle, θ (Deg.)
R=.5mm
R=1mm
R=3mm
R=6mm
AM 11/03 19
1
2
3
4
5
6
7
0 2 4 6 8 10 12 14
Fin Length, Lf (mm)
Ela
stic
Str
ess
Con
cent
ratio
n Fa
ctor
, Kt (
-)
Fin Thickness = .5mm
Fin Thickness=1mm
Fin Thickness=2mm
Fin length
AM 11/03 20
Fins and Cold Laps
1
2
3
4
5
6
7
-4 -2 0 2 4 6 8
Cold Lap Length, Lcl (mm)
Cold Lap; NO Fin
Cold Lap with Fin 0.5 mm Thick
Cold Lap with Fin 1 mm Thick
Cold Lap with Fin 2 mm Thick
Defect Free
For All Tests R=3mm and
θ=30°
AM 11/03 21
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
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
AM 11/03 22
Outline
n Obvious things to don Problems the weld toe
n Fatigue life Improvement Strategies
n Light and heavy industry weldmentsn Improving the “bad” weldments
AM 11/03 23
Good - grind off reinforcement
AM 11/03 24
Good - burr grind weld toe
AM 11/03 25
Very good - full face grinding
AM 11/03 26
Shot peened weld toe
AM 11/03 27
Remelted weld toe (laser)
AM 11/03 28
TWI suggestions as to weld improvement procedures
Improvement Strategies
AM 11/03 29
100
10
θ15°30°45°
15°30°45°
15°30°45°
1 10001
100Fa
tigue
Stre
ngth
(ksi
) at
106
cycl
es
Base metal UTS (ksi)
Plain plate
As Welded
Stress Relieved
Over Stressed
Improvement strategies
Shot Peen
Residual stress
Geometry
Base metal strength
AM 11/03 30
Example
IP model predictions and experimental data.
AM 11/03 31
ASTM A 36 butt weldment
AM 11/03 32
ASTM A 514 butt weldment
AM 11/03 33
Outline
n Obvious things to don Problems the weld toen Fatigue life Improvement Strategies
n Light and heavy industry weldments
n Improving the “bad” weldments
AM 11/03 34
• Light industry weldments are presumed to be fabricated from 1/2” or smaller plate and not to have large fabrication stresses.
• Heavy industry weldments are presumed to be fabricated from larger than 1” thick plates and to possess large fabrication stresses.
Light, heavy industry weldments
AM 11/03 35
104 105 106 107 1081
10
100
Toe Ground (radius = 0.1 in.)Over StressedStress RelievedWeld Profile (flank angle 20 o)As Welded
Fatigue Life, NT (cycles)
Rem
ote
Stre
ss R
ange
, ∆S
(ksi
)
t = 0.5-in. (12mm); R = 0Without Fabrication Stresses
Nominal
Ideal
Light industry weldments
AM 11/03 36
104 105 106 107 108
1
10
100
Toe Ground (radius = 0.1 in.)Over StressedStress Relieved
Weld Profile (flank angle 20 o)As Welded
Fatigue Life, NT (cycles)
Rem
ote
Stre
ss R
ange
, ∆
S (k
si) t = 2.0-in. (50 mm); R = 0
With Fabrication Stresses
Nominal
Ideal
Heavy industry weldments
AM 11/03 37
Outline
n Obvious things to don Problems the weld toen Fatigue life Improvement Strategiesn Light and heavy industry weldments
n Improving the “bad” weldments
AM 11/03 38
Weldment with a transverse attachment
Transverse attachment
Good weldment
AM 11/03 39
Weldments with longitudinal attachments have a low fatigue resistance because of the presence of weld terminations. Starts and stops introduce weld discontinuities. Residual stresses very high. 3-D stress concentrations effects
C
Longitudinal attachment
Bad weldment
AM 11/03 40
Examples of terminations
C
C
a
b
C
L-PL-P
LP
C
CC
AM 11/03 41
Longitudinal attachment
Longitudinal attachment with stress diffusers
Stress diffuser
Placement of stress diffuser
AM 11/03 42
3-D FEM modeling
AM 11/03 43
Effectiveness of a stress diffuser
Longitudinal attachmentLongitudinal attachmentwith stress diffuser
AM 11/03 44
Effect on MK and NP
1
2
3
4
0.001 0.01 0.1
L.A.
L.A. with stress diffusers
Transverse attachment
Wel
d ge
omet
ry c
orre
ctio
n fa
ctor
, M K
Crack length / main plate thickness, (a/T)
105
106
107
0.002 0.02 0.2
L.A.L.A. with stress diffusersTransverse attachment
Fatig
ue C
rack
Pro
paga
tion
Life
, N P
(cyc
les)
Initial crack length / main plate thickness, (ao /T)
AM 11/03 45
Fatigue test results
40
60
80
100
300
500
105 106 107
L.A., Series 1 L.A., Series 2L.A., Series 3 L.A. with stress diffusers, Series 1 L.A. with stress diffusers, Series 3
Rem
ote
Stre
ss R
ange
, ∆S
(MPa
)
Fatigue Life, N (cycles)
AM 11/03 46
40
60
80
100
300
500
105 106 107
Transverse attachments, databaseLongitudinal attachments, databaseLA specimens, Procedures 1 and 2LA specimens, Procedure 3SD specimens, Procedure 1SD specimens, Procedure 3
∆
Fatigue Life, N (cycles)
Fatigue test results
AM 11/03 47
Summary
• The fatigue strength of “Ideal” weldments can be much improved; whereas, the fatigue strength of “Nominal”weldments cannot.
• Weld toe grinding or weld profile control works best for “Ideal” weldments at short lives. Beware of corrosion pitting.
• Smaller “Ideal” weldments are more susceptible to improvement than larger weldments.
• Fabrication stresses are critically important.
AM 11/03 48
• The behaviors of light and heavy industry weldments are dissimilar.
• Stress relief annealing and over-stressing works best for “Ideal”weldment at long lives. Beware of compressive overloads.
• Fatigue behavior of weldments and effective life improvement methods depends upon weldment size and weld quality
• Stress-diffuser can substantially improve the fatigue life of terminations without post-weld processing.
Summary