lecture 9 nlfm ctoa
TRANSCRIPT
-
7/28/2019 Lecture 9 Nlfm Ctoa
1/21
CEE 770 Meetings 9 and 10
Objectives of These Meetings
Introduce Issues and Approaches for
Non-Linear Fracture Mechanics (NLFM)
What makes NLFM more difficult than LEFM?
NLFM Approaches under Development
A NLFM Approach for Thin Metallics: CTOAc (Meeting 9)
A NLFM Approach for Many Other Situations:
Cohesive Fracture Mechanics (Meeting 10)
182
-
7/28/2019 Lecture 9 Nlfm Ctoa
2/21
What makes NLFM more difficult than LEFM?
1. The Process Zone is too large relative to the crack length:
Therefore, SIFs lose their meaning, and value. We no longer have a
single parameter that completely controls all action in the process zone.
We now have to solve a local, materially non-linear problem, and wemust expect that similar structural geometries and BCs will produce dis-
similar locally non-linear results. Therefore, toughness for a given
material is no longer a material state value.
rpa
183
-
7/28/2019 Lecture 9 Nlfm Ctoa
3/21
What makes NLFM more difficult than LEFM?
2. The Process Zone might be too large relative to other structural dimensions:The problem is now one of full (structural) scale inelasticity, and we can
expect conditions for crack growth to be very sensitive (more so than in case
1.) to changes in geometry and/or BCs.
Again, theoretically we would have to:
a. Compute local fields from non-linear
analysis, and
b. Measure resistance to crack extension
case-by-case!
What simplifications can we get away with?
184
-
7/28/2019 Lecture 9 Nlfm Ctoa
4/21
NLFM Approaches under Development
Since the late 1960s, many NLFM theories and approaches have been
proposed. All are of limited use. Few have any standardized testing or
computational methodologies associated with them.
For material states characterized by elasto-plastic process zone behavior,
the crack tip opening displacement (CTOD) and the EP J-Integral have seen
most development and use. (Read Chapter 3 of Anderson!).
For many other non-linear material states, the so-called cohesive fracturemechanics approach is seeing rapid development and use. (Not in
Anderson!).
185
-
7/28/2019 Lecture 9 Nlfm Ctoa
5/21
CEE 770 Meeting 9
Objectives of This Meeting
Learn the CTOAc concept for some types of NLFM problems:
Steady-state plastic zone evolution, and stable tearing
EP FEM methods for calibrating and then predicting
Example applications
Using FRAN2D/L and the CTOAc criterion.
186
-
7/28/2019 Lecture 9 Nlfm Ctoa
6/21
A Common NLFM Problem Involves EP
Crack Growth in Thin Sheet Structures
Like Aircraft FuselagesShip Hulls???187
-
7/28/2019 Lecture 9 Nlfm Ctoa
7/21
Features of Stable Crack Growth Under
NLFM Conditions
Active plastic zone
Residual plastic strain from
elastic unloading
Envelope of previous
plastic zones
Elastic-plastic
If this process can reach a steady-state before global (structural scale) crack
instability, maybe we can find a local feature to measure, and then compute
and use as a stable crack growth criterion. One possible feature is the CTOA.
188
-
7/28/2019 Lecture 9 Nlfm Ctoa
8/21
Crack Tip Opening Angle (CTOA) Fracture Criterion
d
CTOAc
CTOD
Optical microscopy
Digital image correlation
CTOAc
(degrees)
a (mm)
2.3 mm thick (B)
2024 aluminum alloy
When the CTOA at a specific distance,d, from the crack tip reaches a critical
value, the crack can grow:
Foil test
189
-
7/28/2019 Lecture 9 Nlfm Ctoa
9/21
3-D EP Finite Element Analyses Can Be Performed to Find a
Critical CTOA Value that Best Fits the Results of One Test
a
W
C(T)
Dawicke and Newman, Residual strength predictions for multiple site damage cracking using a three-dimensional finite element
analysis and a CTOA criterion, Fatigue and Fracture Mechanics: 29th Volume, ASTM STP 1332, Panontin and Sheppard Eds., 1999190
-
7/28/2019 Lecture 9 Nlfm Ctoa
10/21
This CTOAc Value Can Then Be Used to Predict the
Failure Load for Structures of Different Sizes and Shapes
191
a
W
C(T)
2a
W
M(T)
-
7/28/2019 Lecture 9 Nlfm Ctoa
11/21
The Critical CTOA
Must Be Calibrated
for:
Material Thickness
Distance behind the
crack-tip where it ismeasured
Convergence studies show that the
predicted failure load is relatively
insensitive to near-tip element sizes,provided that there is at least one
node between the crack tip and the
node where the CTOAc is measured.
Once calibrated, the CTOAc can be
used to predict the failure load instructures of a different size or with
different structural details.
192
-
7/28/2019 Lecture 9 Nlfm Ctoa
12/21
0 10 20 30 40 50
50
100
150
200
250
0 5 10 15
50
100
150
200
250
0 5 10 15 20 25
50
100
150
200
250
Experimental Measurements
Numerical Predictions
applied stress (MPa)
half crack extension, a (mm)
applied stress (MPa)
half crack extension, a (mm)
half crack extension, a (mm)
applied stress (MPa)
0 20 40 60 80 100
50
100
150
200
250
half crack extension, a (mm)
applied stress (MPa)
w = 76 mm (3 in.) w = 305 mm (12 in.)
w = 610 mm (24 in.) w = 1524 mm (60 in.)
Similar Analyses Can Be Performed Using Thin-Shell,
EP Finite Elements
w2
d = 0.04"
CTOA
Chen, Wawrzynek, and Ingraffea, Elastic-plastic crack growth simulation and residual strength predictions of thin plates
with single and multiple cracks, Fatigue and Fracture Mechanics: 29th Volume, ASTM STP 1332, Panontin and
Sheppard Eds., 1999 193
-
7/28/2019 Lecture 9 Nlfm Ctoa
13/21
Calibration with Tests 2024_FAA_TL3 & TL4
to determine critical CTOA on Thin Plate24
40
0 0.5 1 1.5 2 2.5 3
10
20
30
Half Crack Extension (inches)
4.04.55.05.5
Applied
Stress
(ksi)
one half plane-
strain core
heights0.04
194
-
7/28/2019 Lecture 9 Nlfm Ctoa
14/21
KC-135 Crack Configuration
Analyses performed under USAF Contract No. F9603-97-C-0349, Subcontract NCI-USAF-9128-001, Corrosion Damage Assessment Framework 195
-
7/28/2019 Lecture 9 Nlfm Ctoa
15/21
A Recent KC-135 Failure
196
-
7/28/2019 Lecture 9 Nlfm Ctoa
16/21
Plastic Zone Evolution and Stable Tearing in KC-135 Fuselage Panel
Under CTOAc Criterion
508 mm
254 mm
Effective Stress (ksi)
60
39
53
46
Yield
197
-
7/28/2019 Lecture 9 Nlfm Ctoa
17/21
Predicted KC-135 Residual Strength
(initial lead crack length = 10.0 inches)
Total Crack Extension (inches)0 1 2 3 4
2
4
6
8
10
12
14
16No MSD (RS = 15.3 psi)
MSD = 0.025 in. (RS = 11.3 psi)
MSD = 0.025 in., Corrosion (RS = 10.8 psi)
required residual strength (12.5 psi)
Pressure
(psi)
198
-
7/28/2019 Lecture 9 Nlfm Ctoa
18/21
7.14 or 10 in.
20 in.
Corrosion thinningMSD
Lead crack
12 DamageConfigurations
Lead crack--7.14 or 10 in.
MSD--0.025 in. or 0.046 in.
symmetric about and at three rivets in front
of lead crack
emanating from both sides of a rivet hole Corrosion--10% material thinning
uniform reduction in thickness of the upperskin199
-
7/28/2019 Lecture 9 Nlfm Ctoa
19/21
Predicted residual strength: summary
12.5
200
-
7/28/2019 Lecture 9 Nlfm Ctoa
20/21
Plane stress or thin shell analyses can accurately predict residual
strengths with the use of a plane strain core
plane stress
elements
3-D analyses show that, in the crack-tip region, the through-the-
thickness behavior is more similar to plane strain than to plane
stress.
Plane strain
core height
plate thickness
plane strain elements
Plane strain (or special thin shell elements) are placed in a small
region or strip along the crack path.
201
-
7/28/2019 Lecture 9 Nlfm Ctoa
21/21
Failure Load Predictions for Analyses Using
a Plane-Strain Core (PSC)
Dawicke, Residual strength predictions using a crack tip opening angle criterion, Proc. Of the FAA-NASA symposiumon the continued airworthiness of aircraft structures, Atlanta, GA, 1997
202