Chuong 8 - Tinh Toan Ket Cau
Chuong 8 - Tinh Toan Ket Cau
Chuong 8 - Tinh Toan Ket Cau
Chapter 8. Failure
Analysis
Introduction
• Structures such as bridges, aircraft, and machine components can fail in many
different ways.
• An overloaded structure may experience permanent deformation, which can
lead to compromised function or failure of the entire structure.
• When subjected to millions of small repeated loads, a structure may have a slow
growth of surface cracks that can cause material strength degradation and a
sudden failure.
• When a slender structure is loaded in compression, it may undergo an
unexpected large deformation and lose its ability to carry loads.
• Failure analysis plays an important role in improving the safety and reliability of
an engineering structure.
Static Failure
• Under static loading conditions, material failure can occur when a structure is
stressed beyond the elastic limit. There are two types of material failures from
static loading, namely, ductile failure and brittle failure.
• The main difference between the two failure types is the amount of plastic
deformation a material experiences before fracture.
• Ductile materials tend to have extensive plastic deformation before fracture,
while brittle materials are likely to experience no apparent plastic deformation
before fracture.
• Two common theories on ductile failure are the maximum shear stress theory
(or Tresca Criterion) and the distortion energy theory (or von Mises Criterion).
Brittle Failure
For brittle materials, the two common failure theories are the maximum normal
stress theory and the Mohr–Coulomb theory.
Maximum Normal Stress Theory
The maximum normal stress theory states that brittle failure occurs when the
maximum principal stress exceeds the ultimate tensile (or compressive) strength of
the material. Suppose that a factor of safety n is considered in the design. The safe
design conditions require that
Mohr–Coulomb Theory
The Mohr–Coulomb theory predicts brittle failure by comparing the maximum principal
stress with the ultimate tensile strength Sut and the minimum principal stress with the
ultimate compressive strength Suc. Suppose that a factor of safety n is considered in
the design. The safe design conditions require that the principal stresses lie within the
hexagonal failure envelope
Fatigue Failure
• Under dynamic or cyclic loading conditions, fatigue can occur at stress levels that
are considerably lower than the yield or ultimate strengths of the material.
• Fatigue failures are often sudden with no advanced warning and can lead to
catastrophic results.
• Design of components subjected to cyclic load involves the concept of mean and
alternating stresses.
• The mean stress, σm, is the average of the maximum and minimum stresses in
one cycle, that is σm = (σmax + σmin )/2
• The alternating stress, σa, is one-half of the stress range in one cycle, that is
σa = (σmax - σmin )/2
For most materials, there exist a fatigue limit, and parts having stress levels below this
limit are considered to have infinite fatigue life.
The fatigue limit is also referred to as the endurance limit, Se. In the following, three
commonly used fatigue failure theories, namely, Soderberg, Goodman, and Gerber
failure criteria
Soderberg Failure Criterion
This theory states that the structure is safe if
where σa is the alternating stress, σm the mean stress, Se the endurance limit, Sy the
yield strength, and n the factor of safety. When the alternating stress σa is plotted
versus the mean stress σm, the Soderberg line can be drawn between the points of
σa = Se/n and σm = Sy/n. If the stress is below the line, then the design is safe. This
is a conservative criteria based on the material yield strength Sy
Where Sut is the ultimate tensile strength and n is the factor of safety. When the
alternating stress σa is plotted versus the mean stress σm, the Goodman line can
be drawn between the points of σa = Se/n and σm = Sut/n . If the stress is below
the line, then the design is safe. This is a less conservative criteria based on the
material ultimate strength Sut
where Sut is the ultimate tensile strength and n is the factor of safety. When the
alternating stress σ a is plotted versus the mean stress σm, the Gerber parabola
can be drawn between the points of σa= Se/n and σm= Sut/n . If the stress is below
the parabola, then the design is safe. This is a less conservative criteria based on
the material ultimate strength Sut
• In buckling analysis, one may be interested in finding the critical load, that is, the
maximum compressive load a structure can support before buckling occurs, or the
buckling mode shapes, that is, shapes the structure takes while buckling.
• The critical load is a structure’s elastic stability limit, which is adequate to force the
structure to buckle into the first mode shape.
• A structure can potentially have infinitely many buckling modes, and higher critical
load values are required to trigger higher-order buckling modes.
• Because buckling failure can lead to catastrophic results, a high factor of safety (at
least >3) is often used for critical load calculations.
• When developing a model for buckling, one should pay attention to the boundary
conditions as they have a considerable effect on a structure’s buckling behavior.
• For example, a column with both ends pinned can buckle more freely than a
column with both ends fixed.
• Also note that bending moments acting on a structural member, for example, due
to eccentricity of an axial load, may play a role in affecting buckling and should not
be neglected in the analysis in general.
Next, right-click on Solution (A6) in the project Outline, and select Insert -> Stress
Tool -> Max Equivalent Stress. The initial yielding in the test sample may be
predicted by comparing the maximum von-Mises stress in the specimen with the
tensile yield strength of the specimen material. The Stress Tool is used here to
show the safety factor results.
Right-click on Solution (A6) and select Solve. The computed total deformation, von-
Mises stress and safety factor distributions are shown below. From the static
analysis results, it is apparent that the neck portion of the specimen will not yield if
loaded statically.
In the Details of “Fatigue Tool,” set the Mean Stress Theory to Goodman. Note that
the default loading type is Fully Reversed constant amplitude load, and that the
default analysis type is the Stress Life type using the von-Mises stress calculations.
Right-click on the Fatigue Tool in the Outline, and select Insert -> Life.
Next, right-click on the Fatigue Tool and select Insert -> Safety Factor. In the Details of
“Safety Factor,” change the Design life from the default value of 109 cycles to 106
cycles.
Finally, right-click on the Fatigue Tool and select Evaluate All Results. From the fatigue
analysis results, the shortest life is at the undercut fillets (19,079 cycles) followed by
the neck portion of the specimen. The neck portion of the specimen has a fairly small
safety factor with a minimum value of 0.3973. The results show that the specimen will
not survive the fatigue testing assuming a design life of 106 cycles.
An Eigenvalue buckling analysis system will be added, with the static structural
results being used as initial conditions. The engineering data, geometry, and model
will be shared by both analyses.
Click on the Analysis Settings under Eigenvalue Buckling (B5) in the Outline. In the
Details of “Analysis Settings,” set the Max Modes to Find to 3 and Include Negative
Load Multiplier to No
Right-click on Solution (B6) and select Solve to view the buckling modes. To use the
default window layout as shown below, select View -> Windows -> Reset Layout from
the top menu bar. Note that the load multiplier for the first buckling mode is found to
be 0.78173.
To find the load required to buckle the structure, multiply the applied load by the load
multiplier. For example, the first buckling load will be 39.0865 MPa (0.78173 × 50
MPa), thus the applied pressure of 50 MPa will cause the specimen to buckle. In the
Graph window, you can play the buckling animation.
The following figures show the first three buckling mode shapes. The corresponding
load multipliers for the first, second, and third mode shapes are 0.78173, 2.0094,
and 5.7321, respectively.
Note that the max value in the total deformation plots is scaled to 1 when displaying
the buckling mode shapes. Here, the deformation plot is used for mode shape
visualization, with the actual values of deformation carrying no physical meaning.
Problem 1
Problem 2
Problem 3
Problem 4
Homework 1
Homework 2
Homework 3